From owner-linux-arm-outgoing@vger.rutgers.edu Sat Apr 4 15:17:01 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id PAA09613 for ; Sat, 4 Apr 1998 15:17:00 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:29117 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390804-120>; Sat, 4 Apr 1998 17:16:35 +0300 Received: by vger.rutgers.edu id <971110-12780>; Sat, 4 Apr 1998 08:42:42 -0500 Received: from tazenda.demon.co.uk ([158.152.220.239]:4525 "HELO kings-cross.london.uk.eu.org" ident: "exim") by vger.rutgers.edu with SMTP id <971087-12780>; Sat, 4 Apr 1998 08:41:18 -0500 Received: from kings-cross.london.uk.eu.org [127.0.0.1] (phil) by kings-cross.london.uk.eu.org with esmtp (Exim 1.71 #1) id 0yLTeu-0004d3-00; Sat, 4 Apr 1998 15:11:00 +0100 X-Mailer: exmh version 2.0zeta 7/24/97 To: linux-arm@vger.rutgers.edu Subject: Processor type Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sat, 04 Apr 1998 15:10:59 +0100 From: Philip Blundell Message-Id: X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Hi. Attached is a patch to remove CONFIG_CPU_ARM6 (etc) from circulation. Instead, the configure script works out whether your hardware uses a 26-bit or 32-bit processor and defines CONFIG_CPU_{26,32} appropriately. You can still specify an exact CPU type but this is only used to work out what compiler options to use. Also, this patch implements an optimisation to reduce the impact of having to vector processor-specific options through the `processor' structure. Macros like xchg() now expand to a call to a function in proc-fn.c, which looks up the appropriate processor entry, calls the function and patches the calling procedure to transfer control directly to the right place in the future. At the moment this only works for ARM2/3 machines and a couple of functions but it should be easy to extend if anyone has the inclination. It might be possible to use a similar mechanism to handle architecture-dependent functions. This way, we could drop the A5k/Archimedes distinction and have a single kernel that runs on all 26-bit machines. Finally, I replaced the `CONFIG_CPU_ARM2' test in page_alloc.c with a general check for machines with large pages. I imagine this will be a bit more Torvalds-acceptable. Usual disclaimers apply. p. begin 644 linux-diff-980404.gz M'XL(".8^)C4"`VQI;G5X+61I9F8M.3@P-#`T`,4[:W?:2+*?X5?4.+-C"(B' MQ,/&Z]P0@A-O_+I`9F=V,E[NEMC8S(!:0&2Y#)]X7K&/)Z7)&D/,B;L3RC\6.A=7YV=OU-[-Q_5[N!2+I:7Q7PI!2BW.`B(1'\XO![DX!20 M13LO1:GG2J<@S>4\Q$B\.;_Z.#J_&*I7_7^*.7IQ),W1/4ENIL@8)--U9 MG"(?+_;`J7463W? MM:TI4D#2Z2M*Y];3"*&T,@F2$3Q51@$L%WL.>7@!-ZZM,\^S7>AB3#1\IOL+ ME_&HUY;+&/1:S7)+Q+SKRYL!RJ+_5GW?[[XE@Q3+Y:W$? M<05Q"L)@,>K_,NJ^?L@>\1B"$W;9'?3>JX.;GO"MRV[O_?E5/T?8 MKJ.+)%5OHZ9*]9H2+&U3"HU:+9!.1)'(Q*"+X!\+H^N;M^>#XCHSFL9MU315 M?::YZH.K>KI.FLXP`DS_GQ>>HQLXIA0+7W#]YA]O/U[><&9Q/8/KX5`E.N<7 M_:)]^^_Q8N[DQ\]55.BV-3&F%<."+,`3RXH0[TEU13I6H]-L=QJ-=6%Q5$8: M)?S;Y!K+PUPSK#FS%JKM^(9M`,?H`'0;:+]YF+` MY@?M.+0=%A!+3_9MA%#ZEM@NDI/L0A4@ZNR(1O>N>;0Z[ZOWPC/-Q+5N#NF][UZ<]_HA+71R^`T.?HRM]@`=Y6!Y`)(=AS4_ M9,)PJ2O8[R>D4RL/`&.L;BVFWMJV"1N4>M>#*U@&&85,46Z4ZQCNY799"2(C MSS6I)&B52'YB4'!,K&`ML@Q6`[''V952YPH3,G);(EL\,Z:8H^/`P8XCI MDJ5[*X-:>&BY&B&5_:M/5["3/.($KM'W-N%8?J@H>>A M]:&]^C,T:%VSP,?6`&YM?U;F!F_9#P)&Q1O@?/:M9YO,9T3)$G3QJ6G<,7-) M/L8,XJT"Q/4S2&F;PA.PT-`V)5S*E#`FJ^46,&9C*[`(U%*:U:XM,Z+.3'I< MADL:LJ()&\$!*Y\@,N3(-P^HTEZ[]JKV%GY+M!(P)82UJNT$M"6@0UZ3X>\- M>"#U$$ABV=]/(Z+?)OSM`U#\2W)%=#CZS]W@"NT1@RVW2D96[_+`/K-M"NLP MM]%''$RF!OY$&FC\(_(2,EP+#101QH3@,9^H$8V)J4T]RA>.YGD$>]?KE=&T M?4P$Z`)+S!IC5DF:LUA"RH)7&KW&M#$G/B;HD#B::[7$M=I*J(:#2''M.*@M M(BO73YK&"#M?$O))3THMK%X.U;* M=85'7]\U/%_SD;<;S44],Q,<+"7!6SCT>1B&XN[@YGHPBJ>2X/$!_(!KMF(I MP5'7U`%N>I+G+S&[8Z4W?M!<%J>-X1YB=$6HV*2S3M_9E"B5)TAAK3(!]EG3 M?;1+*AE6Z**^H=#.3.Q+R&!Q#6B>+A?%@X'%B[W2G/:@+7FI(@3$-4EU-U8N MW'AMBL:4`]ZR"<,PSG,&"BWT'OQYQYA5"?W]4R$A3Y5B;B!224N"15`0\$]% M"@Q91$YW$/DA)/+["9K;.H;PV(#BAN%ZHPG%>+;<)5.3>DT:NMR_!@[I])QT#[UKT:#7U5L7C!L MH)>Z2RG:'(JV]UH==0?O^B/HG(*8"9]+UX0U1!5UJ.$)Z13!L$EC%1L,]T_\ M&U3M],UW-9WA%P\UJ9GT9>FIR#]^H\J:/EPLV*F]^I[4.9(TL<3J,.QTPF7A M7-3Q!5-9AJ_ZFG?'!SZW%>'7.V\Y]RHZ[!JPGR6M\?>QI4QLI=&1(\U=O='@ M[5WPR6WJ%^Z1PU\OWUQ?J%?7/_<'P\*"Q*66( M(KRLYJ47&)5$@3!.[+C!7W^EP<1.%H;%=%1%+L9X*J@J+1=3D[-4)ZX]5XD9 M8BU[F&_O&(0Z<;Y6\!7O+\2>QG/;NX@NM-M4&<+N(?O9?)3"/E:?B:]TZG4DL;;[5AE_E5HK MFX>7]!\N[`?)Q([+A'O,Y9B<#E=# M93T0,;C7S`7S*A6L"=[P^L%E/V":?,;I.37!PB.GA\#3+=U@BV$V: MO0*,#KQ$3=G$0$1J#$)4@23TR&[Y/C)2H+HT3I[V@*H3.HS7_@-`#[.]J#P$K9]1<. M;Z.X%GDG6N6]#GZL&\X=1I"M9M?1,U0!?-!-+@F=7*Q4@6;F; MP_93)4?=1XE)1*Z^6C19MTA_^/P#9X[)J-\C)V( MQE5*NSU!587UMS4VL<7@JJ7X$%1.G-B5YB^HC5MB'$"JU%1K@&$37&9J%"_* M&%G\(.30G!YV-:;F&KZ!,>*)8>%M_\W'=VI_,+@>#'E82%C+F&$8H&A,"PF(9]FI$+%DE7I"@C^W7:VJTRS;"N'[V5>(GAO-%O`/S0*H0ZW> MJ:&I4'YKU[;8V!IY<\>_4>LTC]=V5BMC;JR7V_Q,K51]R5/$SN7@(#&PAP67 M:TQG/A1Z14X5;G"!A@-OS(4U9J9)XW@F"&CS71\N;OP4NSZTUV.XGL^+?[1# MG4X!$#K7QOQT@'IP6W3&X<&"Q(EY#M.-B:'#9&'Q-$<-]3]IQ\C79QR!:%'+ MS8.<9XLL3+L=8C)#Y,2IC1D10XR>,SCZFX687,<7&G)IM8<%!U8=G'7*/QI9)L\1V M@8/UB.W.-1Q2IGVP?R]01E,[G(*8%`F>CDB)FD:NC!AF)11[PM=T9U&56XFX MK7GSJL&]*/8PT'K@ER44'5I#,&_DH"?\=B+H3K#DW=PSQ*>B#(:;GGK9'7[( MY6J?:\J$_NF\4/98ZIC)9#U&G""3-<&980GCT<9C%Z;$OT]("XLOPT,AP<*B4(S8)M57JNFB?K").C#=@^)).'QS%`XZ MY4-_6DD@:^2M2>4=CGY9*&Q"7A8+2$""1K&(9$A^XJCU)(4*B0MI;%(HALJL MJ).38/9@OK]P<(%C24#3E/@T5?S+YZK5A+)66%N9.PV(GD3BP\1<>#,4NSYC M0>K"MAZ\!P,].C#J>]L8T^4'=DQBGVDIE)4Q9@;K\]V4)2TJ5!?*E8"*6)?+T/ZMC>7%QQ5P`J*X M6M]N#N5GX5!^)(?R'APVGH7#QB,Y;"0X_$Z=^6H/;5O!P>'[%QP/-7:YZM/00@]1IUC`>;]MOE;X/W.M]B^SV;B%]K6`=UW MS3+M*5O>K>9A[A]K^@F=/:DJ9F15A4(>#G)S5P>GWBQ#K0Q_P_]Z'?_7/ED' MS^[&'@84)].)`^A^+AP@[^/`::A*I]'JU(\B[EL3E[H"\Z&D!DF+B^X-?`>' M;-;X1G:3WYWC=T-$48K3>GZR-.5[3>K:],+*G>]P!\,IGF.W'!3T?.[=\MF[^Y+W0/KMYLU>0:VEM8Z.+W,_6R?]G+E8&.?*L_7PS/ M#\JY`[KV?)##IS_1M[05H4U_+6]0;CZ:K\IHB0M>?.8\)Z!:V4%5Q^Q`4H\IOS48J,6N$*6H^2?JO^ MU`GJH1(>,4.MUJY%EU"+SP`;,R`RS=$.%]'.GN+90YXXZLN,>0%XSZ`78.\5 M]=)PE4[MJ*/(Z[#7H*S96&_&)3<^,&_@WQ8,%I['3!,^8,._BH^/WBD+[P-1 MX2)4ST$K M\W3I%17HGD,'5ND#33!SQR5Q MT2`K+`70_:)2@+Q/4$I#Q9C4[C2.-TJQ-B_%PJAT&*-&=^8%J>$A78*V/;K2 MI`'?;I_RFTC&6.ST\\M+=($:"_HE)W9G!'>QL9@GZ>/H58NW^T3HN<(4O_B] MHR),ABH1TH-HU>"W:)JKBI4*/LTW=-X88,5G3)9T/J#?J8[-.YO87AAP(/;V M98[A&?]AU%=\MYH08875E/!W\;I(@][%()HAH,09@5<$5\3N7I$:&O$O%^Q8 M2/VS[L>+$;\I0)NKS^8\9%@8#JG>'FN^EO2?Q(`GNE`"/\6+E$POVH6M-#O- MR`4!A5Z^*2FK5W`VU3\:#JFLO>S>8"MP?7:FCH8?"CXZP9S-YQH/RTF$8??G M?FPXG;2G#SZ[Z:>,GSC\)F/%LR?^"OD[V5SJ@GNI*]:#)3]_/":=&;:DT7L? MR=/W3>@>QK1&?K(E9:(JK8X<-2-N16%CS(/H$\K$IQ:)%2B\ MKLY.1+#E5U+(W%[#@$#PMR^G_IEK*JPZOIF5P@DB4*<\Z?\X(W>6JR*5TDJ,WCLP,2(5I2,?;[16];JXE8@?K2!#9Q]`0G@`^>1SE,0)`&W6/YJ8 M_)S$&H\@%FYJJ:IAF?06AAH_4.3$'G/0$2M=("<.N+#0H$=4S*"%:QZ#>F=5 M.F0>%ZU/.E9(\DXD.8G4V(G4B)RI)-A+UW<"0=Z*("<1&EL1(BQA%,``HBU, MO\.'&A8&*V.,P`(?ZI'4N7AI[-<\?'WV^+`C*GQ;+-@2`>J[(D"&WRNU3J,5 M\7O>!N/?HVC"_(..6@]Y:ON#SEX/(V^+!CNU=""LBN^B=S'\X/J$9CYH2V\5 MV5=I-3AA6A\2\S-=?J0+:8>^$?CZM#\5._T,.HH>^/#.$^ORVO,Q_E6%+,Z# M*+C>R1XS!PLO>N76T5S?^V:3FF-TQ4RNTHM;>EC%QYX^SH!B2'&[.>HH1W&[ MV8ZB;.[`R<>R>"&4/L7KR71T!<;JFCJ]0?#4'+)=N@Q!U?I7#=_DKD'>, M.3!QF;CNN+YLX_F+R43@HUYF]+J:81GSQ9R(-(X$G4KXPIIO^_SJG.;>&MA> MNTMZ_^I;IJX(`MGT(<>KM)Q!5S70`E2ZC2_!3?==7\4*?=@?%>'5*RCP!\/W MYV>C4ON;^@4I%.BUN%Q,*%4%B]=6L+334VC6Y0_BOM78)O=[H`71>TQE@:H% M]W-X&4!1%WC@Z"<\E(( Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id PAA09643 for ; Sat, 4 Apr 1998 15:21:55 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:29117 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390801-120>; Sat, 4 Apr 1998 17:22:12 +0300 Received: by vger.rutgers.edu id <970996-12780>; Sat, 4 Apr 1998 08:47:53 -0500 Received: from snowcrash.cymru.net ([163.164.160.3]:1392 "EHLO snowcrash.cymru.net" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <971087-12780>; Sat, 4 Apr 1998 08:47:38 -0500 Received: from caramon.armlinux.org (dialup1.cymru.net [163.164.160.65]) by snowcrash.cymru.net (8.8.7/8.7.1) with ESMTP id PAA07777 for ; Sat, 4 Apr 1998 15:20:18 +0100 Received: from raistlin.armlinux.org (raistlin [192.168.0.3]) by caramon.armlinux.org (8.7.4/8.7.3) with ESMTP id PAA20316 for ; Sat, 4 Apr 1998 15:16:43 +0100 From: Russell King Received: (from rmk@localhost) by raistlin.armlinux.org (8.7.4/8.7.3) id PAA01278 for linux-arm@vger.rutgers.edu; Sat, 4 Apr 1998 15:16:40 +0100 Message-Id: <199804041416.PAA01278@raistlin.armlinux.org> Subject: ARM Linux console + keyboard advice To: linux-arm@vger.rutgers.edu Date: Sat, 4 Apr 1998 15:16:39 +0100 (BST) X-Location: london.england.earth.mulky-way.universe Reply-To: rmk@arm.uk.linux.org X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Hi all, I'm currently looking to integrate the console and keyboard stuff back into the Linux kernel, and I will have to change the keyboard RAW interface. You can maintain keyboard RAW compatability by using the MEDIUMRAW mode. Under ARM Linux, this shall always present a uniform view and behaviour of the keyboard, no matter what type keyboard is actually connected. I believe that X will be affected by this change (it uses RAW mode) and other console changes as well. If you have any questions, please mail me. _____ |_____| ------------------------------------------------- ---+---+- | | Russell King rmk@arm.uk.linux.org --- --- | | | | http://www.arm.uk.linux.org/~rmk/home.html / / | | +-+-+ --- -+- / | THE developer of ARM Linux |+| /|\ / | | | --- | +-+-+ ------------------------------------------------- /\\\ | unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Sat Apr 4 15:36:15 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id PAA09722 for ; Sat, 4 Apr 1998 15:36:14 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:27611 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390801-120>; Sat, 4 Apr 1998 17:37:03 +0300 Received: by vger.rutgers.edu id <971105-12780>; Sat, 4 Apr 1998 09:03:43 -0500 Received: from snowcrash.cymru.net ([163.164.160.3]:1746 "EHLO snowcrash.cymru.net" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <970996-12780>; Sat, 4 Apr 1998 09:03:31 -0500 Received: (from alan@localhost) by snowcrash.cymru.net (8.8.7/8.7.1) id PAA08063; Sat, 4 Apr 1998 15:34:54 +0100 From: Alan Cox Message-Id: <199804041434.PAA08063@snowcrash.cymru.net> Subject: Re: Processor type To: Philip.Blundell@pobox.com (Philip Blundell) Date: Sat, 4 Apr 1998 15:34:49 +0100 (BST) Cc: linux-arm@vger.rutgers.edu In-Reply-To: from "Philip Blundell" at Apr 4, 98 03:10:59 pm Content-Type: text X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO > Attached is a patch to remove CONFIG_CPU_ARM6 (etc) from circulation. > Instead, the configure script works out whether your hardware uses a 26-bit or > 32-bit processor and defines CONFIG_CPU_{26,32} appropriately. You can still What about cross compiling. > Finally, I replaced the `CONFIG_CPU_ARM2' test in page_alloc.c with a general > check for machines with large pages. I imagine this will be a bit more > Torvalds-acceptable. That is probably a good idea, if its written in terms of #if PAGE_SIZE > 8192 ... or similar unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Sat Apr 4 19:00:20 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id TAA10574 for ; Sat, 4 Apr 1998 19:00:19 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:7111 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390826-120>; Sat, 4 Apr 1998 20:59:51 +0300 Received: by vger.rutgers.edu id <971202-12780>; Sat, 4 Apr 1998 12:20:50 -0500 Received: from tazenda.demon.co.uk ([158.152.220.239]:1428 "HELO kings-cross.london.uk.eu.org" ident: "exim") by vger.rutgers.edu with SMTP id <971291-12780>; Sat, 4 Apr 1998 12:12:42 -0500 Received: from kings-cross.london.uk.eu.org [127.0.0.1] (phil) by kings-cross.london.uk.eu.org with esmtp (Exim 1.71 #1) id 0yLVbQ-0005up-00; Sat, 4 Apr 1998 17:15:32 +0100 X-Mailer: exmh version 2.0zeta 7/24/97 To: Alan Cox cc: linux-arm@vger.rutgers.edu Subject: Re: Processor type In-reply-to: Your message of "Sat, 04 Apr 1998 15:34:49 BST." <199804041434.PAA08063@snowcrash.cymru.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sat, 04 Apr 1998 17:15:32 +0100 From: Philip Blundell Message-Id: X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO >> Attached is a patch to remove CONFIG_CPU_ARM6 (etc) from circulation. >> Instead, the configure script works out whether your hardware uses a 26-bit > >What about cross compiling. Sorry - by `your hardware' I meant the machine type you're compiling for. It has no knowledge of what you're running on. >That is probably a good idea, if its written in terms of > >#if PAGE_SIZE > 8192 ... > >or similar I can change it to be like that if you want. At the moment it's like this /* * Except on machines with very big pages, enforce a minimum of 48. */ if (PAGE_SIZE < 32768 && i < 48) i = 48; It only affects the init code so it's not a performance issue. In any case gcc can optimise away a compare of one constant against another. p. unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Sat Apr 4 19:05:52 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id TAA10596 for ; Sat, 4 Apr 1998 19:05:51 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:7111 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390833-120>; Sat, 4 Apr 1998 21:05:01 +0300 Received: by vger.rutgers.edu id <971224-12780>; Sat, 4 Apr 1998 12:21:02 -0500 Received: from tazenda.demon.co.uk ([158.152.220.239]:1303 "HELO kings-cross.london.uk.eu.org" ident: "exim") by vger.rutgers.edu with SMTP id <971290-12780>; Sat, 4 Apr 1998 12:12:49 -0500 Received: from kings-cross.london.uk.eu.org [127.0.0.1] (phil) by kings-cross.london.uk.eu.org with esmtp (Exim 1.71 #1) id 0yLWz3-0007Re-00; Sat, 4 Apr 1998 18:44:01 +0100 X-Mailer: exmh version 2.0zeta 7/24/97 To: linux-arm@vger.rutgers.edu Subject: merging a5k and arc Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sat, 04 Apr 1998 18:44:01 +0100 From: Philip Blundell Message-Id: X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Hi. After what I wrote earlier I decided to investigate how difficult it would be to merge support for the a5k and arc architectures into one. It turns out that there were only a few differences in the files in the arch-arc and arch-a5k directories; this patch deletes the latter and includes its contents in the former. Support for the `arc' architecture was seriously broken before, so I doubt this will make things any worse. :-) p. begin 644 linux-diff-980404b.gz M'XL("(1B)C4"`VQI;G5X+61I9F8M.3@P-#`T8@#L//U7V[BR/YN_0@MWNW9Q MP`X!4FCWKDG2;;:$L$G8MGN[Q\?82N-+8J>V`V6W?7_[FY'\(3L.2G('>]; M@5T9N][LRS[_#>\C^#79[U@WE`$O*-^J5"IK$9+Z5D2,:4!(C>A')]7CD]H1 MT5^\J&_M[NZ2%9&J)P=5CO333Z2B:T?J$=G%/_H!^>FG+4(]QQVF?[9V=XAQ MJ&D:F5CVR/5H2,*1%5`2C2@QH"UW0ATHQ&:)XX*L(C^XW]O:=8?T,Y'_(7>, MQIOV14M1K<,;96O7Z#7>--L]22*O$&=KEXY#FB_.<*"6<;&[1;IGOS2O.I0\\/ M#^$/=GR+A/<3$.)->'*R1:1_R+V.0ES/'L\]U&J]_O]A16`MU?A60LNA5)0B\G`3R" MA%@O@PFI#-?HI;.Y6=S0P*/C?6=B`5E[SR9+ZE&5@&T9L<@PYEG1Z[O@7E-J>T.79M$?FQYR*,;@67-`C"5AQ'R M%IEB05/[:$8[\:"2E[S[H3VBSM[H1]2+K`Y&'`7!RO.EKE]2"/[`N0.?P.E4 M7"]BC9L!_3RC860"*=E!]V%&Q!Y9'LA9)5``K\_A#^AY2F^;-;L-K.ZN2T8E M-DB#50500&]-UP%]_PN4W!T2.<8@KUZA^,S?VKW!E7%NOC[O7EY^T,C7KP!' MX.=!.%UAHU^MJ2_(;O4X=AS%)KXK;^+9LT(3WRUHHB))`85A\XAV"B_,0V9E ME5;[XC?C_!2\@S0-0$HW1-YF@AH&E**43L!T;ZVQZZ0M?>]\]+;5Y%4YW2+? M<,1O?8!AJ-2SKL=T^4@]WOS+7$P%YH1]:\^?17LCLAK8,F>P#+_H$VHGNE;T M"4MI#$8S\HOE$:(333_1-/@'-(XU/J^H.DRHFJJA;E3VN8FOVKG8(\PYDJ.\ MYZB@.5>V*CONT'/HD)BFT>^8.!.8AMF]&IAOH`XJ8+XNK6.8'/%MJW?1.C?- M#*$_,!IOS4$7)EM9^Z+IEH8_2@9PWC[K&;T/)@#V!N:@]7[``#4[!>33=D5X M>@+%8:Z"K`+T:*5AV!NI#*>P1&&JU4QA2H84G43Y@,8U65W'>,\*C6:S!_.^ M),&X'+!AT1!L/YMP)CY.$A5R1\D(YZCHSE=)^X)8GD-`2V(5B^ZG%+FAWFP" M?JXB(?%.M]DR>RVCJ8H%[WKM06NK\@W]!1(WH],RSAIOC`M0N+YT(-9BC28! MN_DB'8KT?%'!JTK5(AE>(4DEP)E&DOWGI"!&[.\3:&DV.9*5(1^MKQF)$J75 M5U9:@K>;N]%'`PZN%$D)(;^@&9A93QZD`#%'1*RMYVM1+D4"18_7[HK*4>A"+)$87Y9O M_;$5H9;-O-#]Y%$GCBD5=*8'1X79L-UMG:V("KIX**+BJ,>^"5P0H[&$A%:U M\JU?BAV'>KW0\8XUG8(-091'K6*_`;-UT90XYI#]Y"HYV=P4(M3VV[]CKJ!O$PM'J,!HO]%KM2ZJG%L@5=6.ZSEV M$H!4$-4BPQQ"YR1$\'F0E(@^K)GGM%Y0-,ZQB7$56\^]&5+0761 MTS$>^]XGQ2+/"`92QVPL%+)++HV?6V;W]>M^:R#(%6.V/)T"(?F9I2@K$ONY M-4#7TH6`#H0C3Q5)%@`!39Y6?ISI>^'>U/I$S=#]DRK0T;34"TRL"!51*PV0 M0S^QBCPY[ M-I,L\'W*`-AU5H@J&-"C`PJ&O5$LP2DL"2-JM77#B+A;\Q&$S2*(X_+U4AB! M>[%AOG,]H$]-$YP.S,]`S(3QM&;CR'2#SS*^NWY`/\&Z]]H**6@?QKZXQL=7 MS`UH7_2A)B[5=1W7ZL*Z_=N"!C/:A89]$XG+F/=P85K^(K3*WEFSN>P`\K!B MJ\5NAA%E7G>NN>6T,&O`B:&7,T=W[M"<^D$4DIS@GD]54A2DBHR0I&W^]APD MGG15`$>*Y!7#.BU4V5$PCJN8AZAJ1PP$>\+6')(+U74LNQNA< MPL3]%2S9]H2M2L1L*W%\6&QX?D0P=>E" M@>7=0U0PMNYA@FYW55QD,]21Q55'%AP!^XH8,+.%PA@Q*2?H)[/-@6&=+"5R%6>,='518'GA&(<,\]O8'D@*;!8ZU^X"I0L_PLTHT&`+U"%' MD]$)1^X0D<=T&'U7",=,\[+;&[#%E_P%XI[O9/C#PJ:$%T41TR;WGC4!7PB+ MB2S[7@'*43R(DRG8>MN!5X1U76"2+YM"G MW5KC&7C.7!FZK=AIYB(_$M')%'V5:8))`5G33-83IBE#^7841M+W597L9)W\ MZ'V,MK%NXM]*WVLJ^;Z6%H%`Z6=>B/\/THHP"JZE[W65_(M5`,5Q."8[U3^P M]H1LOWH6;!,9V5%X`;ZRKB@J?V%]@.UM)9F<'A+C MW?\6,3Y.BE_9[Y[L,&'L[)#M\I%+P//#!\/P*AN\C.C)(X0P927>JUN7V\KN?)P M!)RH=^IVH1R57!WSXB1D$MB:*^)\"'-[(X[`D@@B-\D+"6@:)OLE/J9X)Y8= M^"&+;G"B_^&7'W@L%U@XM@!SYPW%$7$G$^JX$-$0/(4!<<[> M7%B"48#-55M-%"0>+7F)EL1J(EC$0V;.ZN.Y7$/;T]'V_A`,'$RBS$VF2H8% MOV`);SC3DG)5>PI>!$X6,/)-$=45+13$*PJV3++Y_"LH!:R;3_]FR8,'?*BW MW,7QII6D[T\F^L.!X=J*^IZ?UVYA\5VBO'7 MTG%YO$6LSDES4.J.7<0PA%O4**QR*W;A#-K#294XB.,#8R M(V838G_DYXMW3L%/MKDJ(VQ\%S,R7HV1C(HB[#$] M1=HY^+Q"WAF!'I]X1NS-,L^,PK*S>`?:7.IY68]6/X3'X4:6]XF._4]QQIE4 M:Q7M!6:!?X,FYA-4?<=7!6PQ"KC*;1T7:,0WR2$JSE< MWC!;K=$@F$TCXLTFUS0(86:980X4="`]MA$O51(N](JF(Z4ZHV3`2MT!B/#& MA/6(:=DWN`^6G9*)%?7LJGW>--N]7^50]=2)0E!3688(RG9VO)T=,^5%QG+E M-`,96F%DK@!W;3FK@$T#_YHN!LSEYV&UYDYH0#!/SX]+@W=),8JG(GJ_'F;D M)(FA8D.F@)$#AY7DH-7K75T.9$^1YEG*@%\;_4$>>H%4!)D;S3Q&N7R$#>M> M]ZR51UDD*^CYDI^,['M,CJNE3+7VJ)T^ZECQ5TZ=:^E07=M%_ MCXFKRW\MWI.RC,<'0&9X-,A@ MB2-^FD"!.+M.OA*9;1GJAXJBSB.<%1$XT.^R]HB'F/3K-B/]=U+&!]SK/!5C M$9:YH),IC#+^J?(MU069VVV)!5DL;W>(2;XT;_?1V\;J"%.`.J80XX)K**@. MXQ=AR0FJA'#C,"`[^E&"?NW:6091CPN%'$$Y%LM;ZFG>$DIA6MU_3AP?S[MZ M%*,5#TV7V&-J!?]D#H#1]6BRJX&4:S'!ZLE<_EAG&CE`23 MJ($H=D*VVQA;QOK2.._E](L'8#%.JII*EE[FJ:SWZ=/ORVSB`5N(DY8,EMI6 MX)B.&^)=`-S8SB<$3Q\"'[IEX/D0"GDA\58[VTA_1HZYIY9"F/3L$2_]\4=R MP'?K;=SAUTY04W$*DV$NRUDE[M7]!U)52:Z8[>Q?P[1[TPX2AE8Q82<797(4@J&]/M`(*5UH"A9NM@<+- M%T'A"@=PZFN?OXE[MNZ%)"$JONRQ(#1WJ0/+SXS!>?==[F('%O.CF.V+9NN] MA+]D*1:OGS0[K1ZFB0=EA3KDG14B.`[ MG5:S;0R`_G&^IO7^LF'TFJ!ODE3/5_6[5Q=-O,_R,Z"]*-2U>FWC'',WT-&" M`/`*0;/=?PLUI3*(ZZISC!@7_7;W`MF1]((8WK8^G'6A8@#"TVOE=3VL.Q2' M"7J5M&D,C-Q(856KT;UH#7*C@<5BWVH%C!R31X66NKT&$_`3V.MD,EMNK@SH MT=;*L#GH'F8;/#"N$/9'!,)*!$)@>)3A/F:#S MQPXLPNP176&2$F$?K?PBD8UL($=HB2G,6T)S-IGC'6J[>-/X;D3CK\;P M+VV0.Q=&)Z3L4R/(C$7P^P6P0)QY-WBZ]W;"S^]BZ\29!7B&>3*QIC^$>V5L M7%UTC,O+5I-?QY%3UL@^+")SH>E%>P`SE'&)N>2_R#.V-IE,5`(KB^Q*E\IO M2_4A5FLU5?);AUUXAC40/+&;SORQ];[54-EM`#6EM(=T='A\A:4:!D5=)",+DEJ@?'1YJ!@92MC7% M;`#:'8C%&;,KHR3RI[A=0ATLU_<.&94)_61=NU$(HK=]SSD%!B(^+\:W'7A8 MAAL;X#Q8FT6#11,%.[UJ$EG'SH,2[`,S.3/M#YKXQ298GA@_]XELX`K0/.MV M!Z9Q->C"PN0U6"`O[;]M7YJ#5G_`"+!]"_`^*#W2.'^;;&7@>H6PG"_^<++Q MC\A;KV\B9I_M-/V%3B'E%OW#P>NZ2G!C(\_?-Q73RU%TW]>`FE2.6UV&JR_$ M12Y50A[`K6Z`>_`(7)(@US9H^'`#W*,-<(\WP*UO@/MB`UQ=VP1Y$]72-]$M M_2")I9YN+AQ-IE9@35:8#5/(Q\^'*8G-9L2,S)(Y<>VKNZ-?II:':PD&AA&ZQ#H@ MY45*2K]H!_SZ"NXBR*S'\G.%]US1%+G`57O(+Q0ZKN/]$+$C_"K0@F`(7-@8 MUI[C^Y2'^(*QGNW./IF;PX,HRRV)0SW:CCCZ1E84DUB6VUG[(VE)SS;.<]8J M?3I=<#X+E@!=.WK<^2RMEK]J?#5UV*%&#.CQ*%;ZK4G6$SO-,CQP<2C9K4,, M?S@L&FQN)]?V9UZDJ_QO5<5[LM$LU).'ZNG<@=.8(K^_'A^,`0QV*I3MO?5: MOQJ(=FT%@0O1O)SNPY%D(VZ@G1N#QAME'HSS`\22;<.!UK@8G.-&HIPK>L/W M$4M(Q)P+-&*FE,VXJJ['57P6-T9]&4LZ_J(`\QI2LL";4(M]E0KL)_ZR&KJ0 M?^,7H486.W4BG+YB:-R#X'%8_A$<4+.XVWL<8I]]9P#:3Z3Q+#Y[=*BP3Q!( MZ3!&KLWVBS/1/"_KQ+K[A)##MPXHN$,X9YGB.T4J8CAG; M7<(9$W!:P0:.5&+F3G.S10S"CSGBHI2OF)%Z&*]KQQ8_U)7*F/.2L1)3>NZ8+GY\:2B>F*#U4""--;VEP3[(3D\G,S$!-!B:#BQ(3 M:^@:7&:"C%1OT-@C*'G'1WV!&0/O\8_OR5W@1C1M+;X"S]U:I]LO7U[/F-\Q M@\B6Q:/:?\IPDU\3P5W?*^2T7_V=JU/;2-) M_#/[5\SE0V(SPDCR`_-(;D-(LMS")1?VKNIJ-^5U;)EU82S*LHF]E?W?K[OG MH9$T>M@*1PC@Z7E/3T_W/'X=KO#S%#Y#56F*8_(OJ\FO_<]B\M+%`<%7SYG? M[6H9`2%7\=4"$>/5*]8W(_R4N7OPB_O^@Q(+9E/<]9#V$$?P$XH'6Y^28H41 MK@0JY'Y6."(-"D7@D6/7!4E#2R--9TUPW=PR?%E`EPJ`OF"R@,YG>E7B33`< MND>%MT5XF\*QQV2X3Q6"SE,1/0J`CF2)*I^_N1C\\F%P?OG/!I363`5!0>D@ M+",=!L6D@Z`@V9.2`^_OD'4:Q8/\9+<^L.QU-1UK75/)6G\'+6M=Y?[[+FK6 MVM"SLAC4)!K6&G=ZF'SID05K<004MO4T3@$7"MGS:1J-/KYQA!PR1#OTDW_] MTY]I0??FZ@,BU5["CT]T=R/&8WL"YEC-$>$##Y;+.<2,NS.;F)G4XI5$1F4, MXVY](I=L[%;'L0:*Y/7;:Q"+,=:A'$N0B7@+70.$MUHQ9OMX,<4CK$-,=#@! M$8G0X0<"@?QQ.@["`6@*(#MGJ_MYY!A!N'+I`$2T&RS"KZ^Y'H'DIEBP._')>ZA_^IW>V+=$?8LH\V8@?R0A<;0J'I9 M_5Q!ZV4ID.-X&MWEI+M'%6<1ALL%J!.SC24YT,;!8Y:0Z+6HB`X=EDM?YQ$L M5;D/[O7R-E@$M]G;HA&T9BQQR[+4X7@2$1]$69IE#(MC6'ID-5]%P1@'];!S MX-'R^->I<=W1-LC[]"G"IUY6LF!O@D$\)?8VCX@T>-3PX2%`$"2T'9%AY921 MY8I',*MEU#!A[R/+3LYL-B`MDIK84+I5$C!2X:>MG4VVDV*E:O]AN1#+]1I! MU*B5!Z^,0=]D@CO4Q/;QU];W-P(&(CIILRA`0-V@.92>F#3R`W^QG!L:\:E&86 MOF[NIUE,W-S=2_8YE`)*[;<&B9:SLWZSF:C].EE[66QMILN=GR9!$DWM9#U+)WOFR#2I%2)35.Q99IN2XE25"\*M-LQ%8&SIHE((=4@M+=W, M:NOR:7\QOP)$KEP#BKUU%=#]1J(&XM45A%8N&".7E*L=--E:7Q,YG"?";XXE(RH7G.C\ROT::+<'1G[E$)CB;T;)<9"EYO[>,G@>%M2W4CQ M;*TA0ID,E*_6F,X%7T#U.LU3IEYVZ5@>Q#(#I;ES8ND]ZR1,)/Y+[UVS-_B\ M54"380=]#&G?E[V;31_@1_CPHB5`9&F7V275[>KF_/#ZYMR!-0P_PI^'$"1Q M8%OJ6JBX\1723;*0$=S]<*;P3M^]$T`>=_/PZYP.!@*1\/4$2\?3$8?=!<$# M0G^,Z+;G=-EB)P=-BH7+Q>6X MFHTQF4Z.9B]>ED/;39Y6Z!UO@37.U$A>:4\O"3.8LE4HU;=E34[#EN.C#[+B=TL4ZO(*G3@\(0&6" M9Q'BX4,KMB;AF_J`\'=7-(:X]JL4="]WCD=5.%8*BW@9S"ESR@;#H)]HG$TSZ1>,(G$O^!Y_I$ZIC4 MV"<2S_I$XL4^D0K(WEX;E0;-7VGF`1V6.(V&FMC&PMQR.LO;K+BX3T"D3$"B MT&U2G#^I`Z/$Z-#[>-OH9-?J4UM52?7-GWI%1=.:4UZV6*95X>^AXOWXZ1R! M3FYI7.C"O,QL-0?C>CP=D7TCD;SDM+^"` MU*4)J5P8B`P^X&'OURFA]2\E3=SIC<]=!\*,(51M.I24M8]O72C>(%^5P!L' M1Y)MQ,; M2;[C'3'N.V#2"G^G6S0QUSMJ^L(.M[M0_0@=`^K*^0S1+F8S[39U%V]@[]SCH$XUMFB,+G]2@$WGU]D,F,5BAFM?H,]V&LV+L8,]RBGD&%)M-;=?(Q M0D^K>=3N'5%_`,:G;B_V32:JFO9-QJ5)B=?L+]]+YY'=G^/IIWV6<:F>LXSO M,F;Q7<8ROLN(N8[QI5^[YP#C"NXRJF3W+\;*_8OQ4O]BUE9^,FI.%P)>H]6M MTG6@`BGZ>9+N]BIX-LOTFM7#&2OW<,:>0I"5^Q52D7867S7]"N7D`-]N8F?' MZ^/.3MLGMI+SH5*[6*YC(2=[ELTK>""RQK'*MD,Q\[7D(+\?%R@1N"$PC-!R M26#W!\3*W1ZA>ILG"FSND'C"'9+FDNB:?=)>76 MBOQZ]YTVXUW?Z1++9%W^,,/E#Z3B.KNGF(=AA6D8UIF%8=U)F,W`=4_:/7,. MTN9J5\W`@PHS,!0;K)5-H?"ERM)H*JT`@U#\,(YFK:7Q1<@/^("X;3TW05_UPKG-$70<;1'#U<=43 M7U2IN(JHF MIYB^Y1!".M;N4[@FA%!>%FV8R4>)I=2G0Y+^-DNI:!JK=B/(81>XV_-^.ON" M_->X%7^,?QQ%+;#^6L-1:W77K'KJ8HMCYK^%L6%=D#_]Z\:V(LO@LB6Y6+$O M5[]3\$K<#J_$"^"5U+)5J@,;19W_^^:_6!S+`AU!A/<91">9[/6;G_?,+2E% M,BBR+LF!AV(%\%"L`![*.'?,PD.55CE.`ED+)DR9.DE@J=+^RP). MV7-,`4\Q*_`4J]B`"^_HR-_;2S3&Z/@GU/G*<:Q4I)V7C)HX5CDYU-?[1*MX M5;PKB^9G?P>F5#]/U`EAA;!2IFZ7CT[%JJ!3<3R_^#[H5'P'="J^`SJ5WA]] M"B:NC$Z5BKLS2W\?=*KBC#K')VTOP^`^<7C\4->T$N#_X,/5!>T2)1[L%D12 M!V';]-H6EU&JJ#SRSDI:Y5&5?&M3?)+$IU)_-"4+WL5W!._*"G6>OT:D=P9A MH;AY>T4@"`WO[,QMVB-X*H+7%/N=+O).QW6.I"`"X0'S[7$X>]F`)L%O]IPP M89O?('@\7`Y)"25[1S5Y,!2OO!I99%TG&22S:)X:;7S2V5\)FLB(M_.L+X`F M\BO/^)Q,O*,3KY9FJOD,H\HM?-;,"^"(N'R;#-V7' M/O[C?.\Z6*#?B-^AHB_H6/UWF!0O$E9SE+,Y8F(;L2+@(U93""3M^>UAD]2J M#Y.M!R/7=?Q>:MG?#6&(Y8NGW9"'N`5YB-=`'N(UD(=X#>0A7@-YB-=!'N(U MD(=X#>0A7@-YB-=`'N(UD(=X'>0A7@=YB-=!'N*[(P]56-9WF[?,,F_9+O64 AO4O'#SW4#'J>W)ZNT]^L3G^S^OT=JQC_`_:H/5]*F0`` ` end unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Sun Apr 5 05:31:27 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id FAA13188 for ; Sun, 5 Apr 1998 05:31:26 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:7226 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391112-18718>; Sun, 5 Apr 1998 07:32:16 +0300 Received: by vger.rutgers.edu id <971224-12780>; Sat, 4 Apr 1998 22:58:10 -0500 Received: from neon-best.transmeta.com ([206.184.214.10]:8329 "EHLO neon.transmeta.com" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <971221-12780>; Sat, 4 Apr 1998 22:57:54 -0500 Received: from gold-1.transmeta.com (mailhost.transmeta.com [10.1.1.79]) by neon.transmeta.com (8.8.8/8.8.4) with ESMTP id UAA23872 for ; Sat, 4 Apr 1998 20:31:13 -0800 Received: from palladium.transmeta.com (bin@palladium.transmeta.com [10.1.1.46]) by gold-1.transmeta.com (8.8.8/8.8.5) with ESMTP id UAA28848 for ; Sat, 4 Apr 1998 20:31:31 -0800 (PST) Received: (from bin@localhost) by palladium.transmeta.com (8.8.5/8.8.5) id UAA26738 for linux-arm@vger.rutgers.edu; Sat, 4 Apr 1998 20:31:31 -0800 X-Authentication-Warning: palladium.transmeta.com: bin set sender to news@transmeta.com using -f To: linux-arm@vger.rutgers.edu From: hpa@transmeta.com (H. Peter Anvin) Subject: Re: Processor type Date: 5 Apr 1998 04:31:30 GMT Organization: Transmeta Corporation, Santa Clara CA Message-ID: <6g71f2$q3d$1@palladium.transmeta.com> References: <199804041434.PAA08063@snowcrash.cymru.net> Reply-To: hpa@transmeta.com (H. Peter Anvin) Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Disclaimer: Not speaking for Transmeta in any way, shape, or form. Copyright: Copyright 1998 H. Peter Anvin - All Rights Reserved X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Followup to: By author: Philip Blundell In newsgroup: linux.dev.arm > > >That is probably a good idea, if its written in terms of > > > >#if PAGE_SIZE > 8192 ... > > > >or similar > > I can change it to be like that if you want. At the moment it's like this > > /* > * Except on machines with very big pages, enforce a minimum of 48. > */ > if (PAGE_SIZE < 32768 && i < 48) > i = 48; > > It only affects the init code so it's not a performance issue. In any case > gcc can optimise away a compare of one constant against another. > Yes; in those cases this is better to do it this way than with the preprocessor. -hpa -- PGP: 2047/2A960705 BA 03 D3 2C 14 A8 A8 BD 1E DF FE 69 EE 35 BD 74 See http://www.zytor.com/~hpa/ for web page and full PGP public key I am Bahá'í -- ask me about it or see http://www.bahai.org/ "To love another person is to see the face of God." -- Les Misérables unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Sun Apr 5 23:11:35 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id XAA17291 for ; Sun, 5 Apr 1998 23:11:33 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:3938 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391563-4882>; Mon, 6 Apr 1998 01:10:22 +0300 Received: by vger.rutgers.edu id <971240-12780>; Sun, 5 Apr 1998 17:22:49 -0400 Received: from snowcrash.cymru.net ([163.164.160.3]:1726 "EHLO snowcrash.cymru.net" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <971246-12780>; Sun, 5 Apr 1998 16:58:59 -0400 Received: from caramon.armlinux.org (dialup1.cymru.net [163.164.160.65]) by snowcrash.cymru.net (8.8.7/8.7.1) with ESMTP id WAA31857 for ; Sun, 5 Apr 1998 22:35:17 +0100 Received: from raistlin.armlinux.org (raistlin [192.168.0.3]) by caramon.armlinux.org (8.7.4/8.7.3) with ESMTP id UAA23960 for ; Sun, 5 Apr 1998 20:39:38 +0100 From: Russell King Received: (from rmk@localhost) by raistlin.armlinux.org (8.7.4/8.7.3) id UAA00477 for linux-arm@vger.rutgers.edu; Sun, 5 Apr 1998 20:39:35 +0100 Message-Id: <199804051939.UAA00477@raistlin.armlinux.org> Subject: Posting of patches to the linux-arm mailing list To: linux-arm@vger.rutgers.edu Date: Sun, 5 Apr 1998 20:39:34 +0100 (BST) X-Location: london.england.earth.mulky-way.universe Reply-To: rmk@arm.uk.linux.org X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Hi. Could we stop posting patches to the mailing list please? When things to wrong because of mis-administration of machines (there's a lot of supposed sysadmins who don't know what they're doing out there) on the Internet I tend to get flooded with email, and it doesn't help when I receive 10 copies of the same thing... A better solution would be to put them on a FTP site or web site somewhere, and mention where to get them in the email. Thanks a lot. _____ |_____| ------------------------------------------------- ---+---+- | | Russell King rmk@arm.uk.linux.org --- --- | | | | http://www.arm.uk.linux.org/~rmk/home.html / / | | +-+-+ --- -+- / | THE developer of ARM Linux |+| /|\ / | | | --- | +-+-+ ------------------------------------------------- /\\\ | unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Mon Apr 6 01:14:30 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id BAA18086 for ; Mon, 6 Apr 1998 01:14:29 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:7355 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391694-4882>; Mon, 6 Apr 1998 03:13:46 +0300 Received: by vger.rutgers.edu id <971102-12780>; Sun, 5 Apr 1998 19:32:09 -0400 Received: from snowcrash.cymru.net ([163.164.160.3]:1199 "EHLO snowcrash.cymru.net" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <971179-12780>; Sun, 5 Apr 1998 19:14:26 -0400 Received: (from alan@localhost) by snowcrash.cymru.net (8.8.7/8.7.1) id AAA00856; Mon, 6 Apr 1998 00:51:05 +0100 From: Alan Cox Message-Id: <199804052351.AAA00856@snowcrash.cymru.net> Subject: Re: Posting of patches to the linux-arm mailing list To: rmk@arm.uk.linux.org Date: Mon, 6 Apr 1998 00:51:02 +0100 (BST) Cc: linux-arm@vger.rutgers.edu In-Reply-To: <199804051939.UAA00477@raistlin.armlinux.org> from "Russell King" at Apr 5, 98 08:39:34 pm Content-Type: text X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO > A better solution would be to put them on a FTP site or web site somewhere, > and mention where to get them in the email. ftp.uk.linux.org keeps an opening incoming directory and a fast warez removal system so might be good place if folks have nowhere else to post diffs unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Mon Apr 6 03:55:50 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id DAA18792 for ; Mon, 6 Apr 1998 03:55:49 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:10731 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391829-4882>; Mon, 6 Apr 1998 05:55:32 +0300 Received: by vger.rutgers.edu id <971306-12780>; Sun, 5 Apr 1998 22:17:40 -0400 Received: from skywalker.comnet.co.nz ([131.203.1.204]:4249 "EHLO skywalker.comnet.co.nz" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <971305-12780>; Sun, 5 Apr 1998 22:17:33 -0400 Received: from gemini.comnet.co.nz (gemini.comnet.co.nz [131.203.1.215]) by skywalker.comnet.co.nz (8.8.8/8.8.8) with SMTP id OAA02208 for ; Mon, 6 Apr 1998 14:54:28 +1200 (NZST) Received: from [131.203.8.9] by gemini.comnet.co.nz (NTMail 3.02.13) with ESMTP id ea582040 for ; Mon, 6 Apr 1998 14:58:30 +1200 Received: from IRL_GRACE1/SpoolDir by irlgrace1.grace.cri.nz (Mercury 1.21); 6 Apr 98 14:55:35 GMT+12 Received: from SpoolDir by IRL_GRACE1 (Mercury 1.21); 6 Apr 98 14:55:10 GMT+12 From: "Glenn Ramsey" Organization: Industrial Research Limited To: linux-arm@vger.rutgers.edu Date: Mon, 6 Apr 1998 14:55:04 GMT+1200 MIME-Version: 1.0 Content-type: text/plain; charset=US-ASCII Content-transfer-encoding: 7BIT Subject: How to set up PPP? X-mailer: Pegasus Mail for Windows (v2.42a) Message-ID: <135BB23E2B@irlgrace1.grace.cri.nz> X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Hi, I have ARMLinux 2.0.31pre9 installed and running OK on an RPC700, but I can't figure out how to get PPP working using the files on the FTP site. There is a precompiled PPP module on the FTP site but it is for a different version of the kernel and won't load. I assume that I need to either compile PPP into the kernel or install it as a module. Do I need to download the source (using RiscOS) and compile my own kernel and modules or is there an way that I can avoid having to download the source? g -- Glenn Ramsey G . Ramsey @ irl . cri . nz unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Mon Apr 6 12:53:26 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id MAA21338 for ; Mon, 6 Apr 1998 12:53:25 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:22590 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390584-4882>; Mon, 6 Apr 1998 14:50:55 +0300 Received: by vger.rutgers.edu id <971351-10942>; Mon, 6 Apr 1998 07:06:01 -0400 Received: from odie.barnet.ac.uk ([194.82.202.98]:3112 "EHLO odie.barnet.ac.uk" ident: "willy") by vger.rutgers.edu with ESMTP id <971396-10942>; Mon, 6 Apr 1998 06:54:12 -0400 Received: (from willy@localhost) by odie.barnet.ac.uk (8.8.6/8.8.6) id MAA21196 for linux-arm@vger.rutgers.edu; Mon, 6 Apr 1998 12:32:50 +0100 From: Matthew Wilcox Message-Id: <199804061132.MAA21196@odie.barnet.ac.uk> Subject: iscafs: now reads discnames! To: linux-arm@vger.rutgers.edu Date: Mon, 6 Apr 1998 12:32:49 +0100 (BST) X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO A patch against 0.04 is now available from ftp.barnet.ac.uk/pub/Acorn/armlinux/iscafs/patch-0.04-discnames This patch enables iscafs to read the discnames, and also the disc cycle ID, though this isn't too important since it doesn't write floppies yet. I've added some more definitions for the superblock, though this doesn't include the very latest patches that Ted T'so sent to linux-kernel. I'll add those when I get a Round Tuit. -- Set Alias$Case Set Alias$[ |||| |MSet Alias$Otherwise Set Alias$[ \ Matthew "" |MSet Alias$When If %0=%%0 Then Set Alias$[ "" ||MIf %0=%%0 \ Wilcox Then Set Alias$Otherwise Set Alias$[ |||||||||||||||| ||MIf \ %0=%%0 Then Set Alias$When Set Alias$[ |||||||||||||||| unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Tue Apr 7 00:50:57 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id AAA25096 for ; Tue, 7 Apr 1998 00:50:56 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:12390 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390628-20058>; Tue, 7 Apr 1998 02:50:43 +0300 Received: by vger.rutgers.edu id <971005-15157>; Mon, 6 Apr 1998 19:10:10 -0400 Received: from aquila.EECS.Berkeley.EDU ([128.32.134.143]:2390 "EHLO aquila.EECS.Berkeley.EDU" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <970993-15157>; Mon, 6 Apr 1998 19:09:51 -0400 Received: from localhost (pmlai@localhost) by aquila.EECS.Berkeley.EDU (8.8.8/8.8.8) with SMTP id QAA06251 for ; Mon, 6 Apr 1998 16:49:33 -0700 (PDT) X-Authentication-Warning: aquila.EECS.Berkeley.EDU: pmlai owned process doing -bs Date: Mon, 6 Apr 1998 16:49:33 -0700 (PDT) From: Patrick Lai X-Sender: pmlai@aquila.EECS.Berkeley.EDU Reply-To: Patrick Lai To: linux-arm@vger.rutgers.edu Subject: bootp problem In-Reply-To: <01bd5ff0$ddfd59c0$d50581c0@avh.ins-coin.de> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Hi, I am trying to install arm linux on my EBSA110 evaluation board. However, the bootp on the board doesn't seems to work. It seems it is not boardcasting any packet (I run snoop on the bootp server yet cannot detect any packet from the board, and the TX light on the board is never on) Are there anyone who encountered this situation before ? Patrick unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Tue Apr 7 10:09:31 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id KAA27008 for ; Tue, 7 Apr 1998 10:09:30 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:28575 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390826-20058>; Tue, 7 Apr 1998 12:09:10 +0300 Received: by vger.rutgers.edu id <971156-15157>; Tue, 7 Apr 1998 04:27:48 -0400 Received: from monarch.ja.net ([128.86.16.37]:62803 "EHLO monarch.ja.net" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <970959-15157>; Tue, 7 Apr 1998 04:23:52 -0400 Received: from sun3.nsfnet-relay.ac.uk by monarch.ja.net with SPAM-CHECKED SMTP (PP); Tue, 7 Apr 1998 10:03:57 +0100 Received: from bright.ecs.soton.ac.uk by sun3.nsfnet-relay.ac.uk with NRS SMTP (PP); Tue, 7 Apr 1998 10:03:54 +0100 Received: from penelope.ecs.soton.ac.uk by bright.ecs.soton.ac.uk; Tue, 7 Apr 98 10:01:05 BST From: Russ King Message-Id: <4509.199804070907@penelope.ecs.soton.ac.uk> Subject: Re: bootp problem To: pmlai@po.EECS.Berkeley.EDU Date: Tue, 7 Apr 1998 10:07:17 +0100 (BST) Cc: linux-arm@vger.rutgers.edu In-Reply-To: from "Patrick Lai" at Apr 6, 98 04:49:33 pm X-Mailer: ELM [version 2.4 PL24] Content-Type: text X-Orcpt: rfc822;linux-arm@vger.rutgers.edu X-Envid: [/PRMD=UK.AC/ADMD= /C=GB/;<4509.199804070907@penelope.ecs.] Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Patrick Lai writes: > I am trying to install arm linux on my EBSA110 evaluation board. > However, the bootp on the board doesn't seems to work. It seems it is not > boardcasting any packet (I run snoop on the bootp server yet cannot detect > any packet from the board, and the TX light on the board is never on) Are > there anyone who encountered this situation before ? Which bootp client on the evaluation board are you using? unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Tue Apr 7 14:31:28 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id OAA28205 for ; Tue, 7 Apr 1998 14:31:27 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:18518 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390467-20058>; Tue, 7 Apr 1998 16:31:24 +0300 Received: by vger.rutgers.edu id <971226-15157>; Tue, 7 Apr 1998 08:50:08 -0400 Received: from foobar.starlab.net ([195.171.37.162]:7713 "EHLO downtown.zav.starlab.net" ident: "soggie") by vger.rutgers.edu with ESMTP id <971212-15157>; Tue, 7 Apr 1998 08:49:45 -0400 Received: from localhost (soggie@localhost) by downtown.zav.starlab.net (8.8.5/8.8.5) with SMTP id RAA03508 for ; Tue, 7 Apr 1998 17:30:21 +0200 Date: Tue, 7 Apr 1998 17:30:21 +0200 (MET DST) From: Ivo Clarysse To: linux-arm@vger.rutgers.edu Subject: SA-1100 eval. board Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO I've just gotten a StrongARM SA-1100 evaluation board ('Brutus'). Anyone tried running Linux-ARM on that ? Ivo. ------------------------------------------------------------------------- Ivo Clarysse TEL +32-2-7215454 FAX +32-2-7215380 Research Assistant Starlab - http://www.starlab.org/ PGP ID: 0x50154325 // A0A0 EDBC 37B2 C574 CC4D F1F0 FF94 04B9 5015 4325 unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Tue Apr 7 18:18:06 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id SAA29147 for ; Tue, 7 Apr 1998 18:18:04 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:18520 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390554-20058>; Tue, 7 Apr 1998 20:17:01 +0300 Received: by vger.rutgers.edu id <971106-23205>; Tue, 7 Apr 1998 12:13:25 -0400 Received: from mail1.digital.com ([204.123.2.50]:3244 "EHLO mail1.digital.com" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <971018-23205>; Tue, 7 Apr 1998 12:04:43 -0400 Received: from pobox1.pa.dec.com (pobox1.pa.dec.com [16.1.240.19]) by mail1.digital.com (8.8.8/8.8.8/WV1.0d) with SMTP id JAA25977; Tue, 7 Apr 1998 09:45:27 -0700 (PDT) Received: from stowe.pa.dec.com by pobox1.pa.dec.com (5.65v3.2/1.1.10.5/07Nov97-1157AM) id AA03057; Tue, 7 Apr 1998 09:45:27 -0700 Received: by stowe.pa.dec.com; (5.65v4.0/1.1.8.2/06Jun96-0357PM) id AA12199; Tue, 7 Apr 1998 09:45:48 -0700 Date: Tue, 7 Apr 1998 09:45:48 -0700 Message-Id: <199804071645.AA12199@stowe.pa.dec.com> From: Deborah Wallach To: soggie@starlab.net Cc: linux-arm@vger.rutgers.edu In-Reply-To: (message from Ivo Clarysse on Tue, 7 Apr 1998 17:30:21 +0200 (MET DST)) Subject: Re: SA-1100 eval. board Address: Digital Equipment Corporation; Western Research Laboratory; 250 University Avenue; Palo Alto, CA 94301; USA Phone: 650-617-3315 X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO I've just gotten a StrongARM SA-1100 evaluation board ('Brutus'). Anyone tried running Linux-ARM on that ? Yes. We've been working on a Linux port to the SA1100 (first to Brutus, the SA1100 evaluation board, and then later to our own SA1100-based hardware). Before we moved from Brutus to our own hardware, we got a preliminary version up and running on Brutus (console, keyboard, serial ports only), based on Russell's August 2.0.30 distribution. The source code still needs cleaning, and it's nowhere near ready for distribution, but if you're interested in developing code for Brutus and want to start with working code, you are welcome to start with our Brutus cutoff source. This source can be found at: http://www.research.digital.com/wrl/people/kerr/linux-brutus.html Please let me know if you have any problems with it, although I can't guarantee that anything will be fixed. -Deborah Wallach unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Tue Apr 7 22:14:38 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id WAA30969 for ; Tue, 7 Apr 1998 22:14:36 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:17321 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390723-4357>; Wed, 8 Apr 1998 00:13:04 +0300 Received: by vger.rutgers.edu id <970990-23205>; Tue, 7 Apr 1998 16:31:20 -0400 Received: from aquila.EECS.Berkeley.EDU ([128.32.134.143]:1261 "EHLO aquila.EECS.Berkeley.EDU" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <971055-23205>; Tue, 7 Apr 1998 16:21:31 -0400 Received: from localhost (pmlai@localhost) by aquila.EECS.Berkeley.EDU (8.8.8/8.8.8) with SMTP id OAA01848; Tue, 7 Apr 1998 14:02:26 -0700 (PDT) X-Authentication-Warning: aquila.EECS.Berkeley.EDU: pmlai owned process doing -bs Date: Tue, 7 Apr 1998 14:02:25 -0700 (PDT) From: Patrick Lai X-Sender: pmlai@aquila.EECS.Berkeley.EDU To: Russ King cc: linux-arm@vger.rutgers.edu Subject: Re: bootp problem In-Reply-To: <4509.199804070907@penelope.ecs.soton.ac.uk> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO On Tue, 7 Apr 1998, Russ King wrote: > Which bootp client on the evaluation board are you using? I am using the bootp client that comes with the disk 'EBSA-110 firmware Database - Rev 1.0'. The bootp.c is dated as 05/10/96. Is there newer version of bootp out there ? Any known bug of this version ? Thanks, Patrick unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Wed Apr 8 10:35:47 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id KAA01002 for ; Wed, 8 Apr 1998 10:35:46 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:22713 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390833-4357>; Wed, 8 Apr 1998 12:33:31 +0300 Received: by vger.rutgers.edu id <971032-23205>; Wed, 8 Apr 1998 04:42:33 -0400 Received: from yalta.NL.net ([193.78.240.21]:3986 "EHLO yalta.NL.net" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <970947-23205>; Wed, 8 Apr 1998 04:40:01 -0400 Received: from alushta.NL.net ([193.78.240.22]:29956 "EHLO alushta.NL.net" ident: "IDENT-NOT-QUERIED") by yalta.NL.net with ESMTP id <21214-25053>; Wed, 8 Apr 1998 11:21:34 +0200 Received: from ichfsnl by alushta.NL.net with UUCP id <742-9173>; Wed, 8 Apr 1998 11:20:56 +0200 Received: from ichsun5.noname (ichsun5 [192.3.2.14]) by ichfsnl.ich.nl (8.6.12/8.6.12) with SMTP id LAA13413 for ; Wed, 8 Apr 1998 11:27:24 +0200 Received: by ichsun5.noname (4.1/SMI-4.1) id AA25460; Wed, 8 Apr 98 11:18:10 +0200 Message-Id: <19980408111809.A25453@ich.nl> Date: Wed, 8 Apr 1998 11:18:09 +0200 From: Hans Bogaards To: ARM Linux Mailing List Subject: Is there an archive of this list? Reply-To: Hans Bogaards Mail-Followup-To: ARM Linux Mailing List Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.90.11i X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Hi, I'm new on this list. Is there an archive of this list somewhere, so I could look there first before asking some faq here? -- Tot mails! "You can't help respecting someone Hans who can spell Tuesday, even if he spells it wrong" Winnie the Pooh unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Wed Apr 8 13:02:41 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id NAA01602 for ; Wed, 8 Apr 1998 13:02:40 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:23352 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390939-4357>; Wed, 8 Apr 1998 15:01:58 +0300 Received: by vger.rutgers.edu id <971025-23205>; Wed, 8 Apr 1998 07:18:53 -0400 Received: from odie.barnet.ac.uk ([194.82.202.98]:6764 "EHLO odie.barnet.ac.uk" ident: "TIMEDOUT") by vger.rutgers.edu with ESMTP id <971015-23205>; Wed, 8 Apr 1998 07:18:38 -0400 Received: (from willy@localhost) by odie.barnet.ac.uk (8.8.6/8.8.6) id MAA01573; Wed, 8 Apr 1998 12:59:30 +0100 From: Matthew Wilcox Message-Id: <199804081159.MAA01573@odie.barnet.ac.uk> Subject: Re: Is there an archive of this list? To: hansbo@stack.nl Date: Wed, 8 Apr 1998 12:59:29 +0100 (BST) Cc: linux-arm@vger.rutgers.edu In-Reply-To: <19980408111809.A25453@ich.nl> from "Hans Bogaards" at Apr 8, 98 11:18:09 am X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO > I'm new on this list. Is there an archive of this list somewhere, so I could > look there first before asking some faq here? Yes there is. It's a trifle ad-hoc, and I haven't updated it in a while, but try ftp://ftp.barnet.ac.uk/pub/Acorn/armlinux/. There's some other stuff there too, such as iscafs and a few replacement RPMS that I cooked up. I'll do a couple more months of mailing list archives RSN. Someone volunteered to do a web variant of this, but that's not materialised yet to the best of my knowledge. -- Set Alias$Case Set Alias$[ |||| |MSet Alias$Otherwise Set Alias$[ \ Matthew "" |MSet Alias$When If %0=%%0 Then Set Alias$[ "" ||MIf %0=%%0 \ Wilcox Then Set Alias$Otherwise Set Alias$[ |||||||||||||||| ||MIf \ %0=%%0 Then Set Alias$When Set Alias$[ |||||||||||||||| unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Fri Apr 10 00:46:05 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id AAA09942 for ; Fri, 10 Apr 1998 00:46:04 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:24672 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391994-1806>; Fri, 10 Apr 1998 02:45:07 +0300 Received: by vger.rutgers.edu id <970791-8479>; Thu, 9 Apr 1998 18:59:46 -0400 Received: from odin.waw.com ([194.51.88.249]:3619 "EHLO odin.waw.com" ident: "vincent") by vger.rutgers.edu with ESMTP id <971166-8479>; Thu, 9 Apr 1998 18:58:08 -0400 Received: from localhost (vincent@localhost) by odin.waw.com (8.8.7/8.8.7/waw) with SMTP id BAA19985 for ; Fri, 10 Apr 1998 01:42:37 +0100 Date: Fri, 10 Apr 1998 01:42:36 +0100 (GMT+0100) From: Vincent Renardias Reply-To: Vincent Renardias To: linux-arm@vger.rutgers.edu Subject: glibc status on arm? Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Hello, I'm trying to compile glibc-2.0.92 for arm (cross-compiling from i386 with binutils-2.8.1.0.30, gcc-2.8.1+pblundell-patches & kernel includes "vger-CVS-current") and found the following problems: In the file sysdeps/unix/sysv/linux/arm/profil-counter.h (line 25), gcc complains that sc.reg.APM_pc is not a valid structure member; changing it into "sc.arm_pc" (as defined in sigcontext.h) makes gcc happy. This allows the build process to continue for a while and then fail on dl-machine.h (which has not been ported to arm yet). There also seems to be a binutils related problem when compiling csu/crti.S; the corresponding lines are: (line 23) ldr sl,L40 (line 27) ldr r3, L40+4 and the error message is: "Error: Can not represent OFFSET_IMM relocation in this object file format (1)" Also, is the source used to build the current arm-aout-libs-4.6.27-1a6.arm.rpm from ftp.arm.uk.linux.org available somewhere? I couldn't find it in the SRPMS/ dir. Cordialement, -- - Vincent RENARDIAS vincent@{waw.com,pipo.com,debian.org} - - Debian/GNU Linux: Pipo: WAW: - - http://www.fr.debian.org http://www.pipo.com http://www.waw.com - --------------------------------------------------------------------------- - "La fonctionnalite Son Visuel vous delivre des avertissements visuels." - - [Message durant l'installation de Windows95]:wq unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Fri Apr 10 10:59:47 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id KAA11760 for ; Fri, 10 Apr 1998 10:59:45 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:30834 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390835-1806>; Fri, 10 Apr 1998 13:00:26 +0300 Received: by vger.rutgers.edu id <970944-8479>; Fri, 10 Apr 1998 05:14:36 -0400 Received: from tazenda.demon.co.uk ([158.152.220.239]:2224 "HELO kings-cross.london.uk.eu.org" ident: "exim") by vger.rutgers.edu with SMTP id <970909-8479>; Fri, 10 Apr 1998 05:14:18 -0400 Received: from kings-cross.london.uk.eu.org [127.0.0.1] (phil) by kings-cross.london.uk.eu.org with esmtp (Exim 1.71 #1) id 0yNaYP-0001Eq-00; Fri, 10 Apr 1998 10:57:01 +0100 X-Mailer: exmh version 2.0zeta 7/24/97 To: Vincent Renardias cc: linux-arm@vger.rutgers.edu Subject: Re: glibc status on arm? In-reply-to: Your message of "Fri, 10 Apr 1998 01:42:36 BST." Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 10 Apr 1998 10:57:00 +0100 From: Philip Blundell Message-Id: X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO >In the file sysdeps/unix/sysv/linux/arm/profil-counter.h (line 25), gcc >complains that sc.reg.APM_pc is not a valid structure member; changing it >into "sc.arm_pc" (as defined in sigcontext.h) makes gcc happy. This is fixed in the very latest glibc sources. >There also seems to be a binutils related problem when compiling >csu/crti.S; >the corresponding lines are: >(line 23) ldr sl,L40 >(line 27) ldr r3, L40+4 >and the error message is: >"Error: Can not represent OFFSET_IMM relocation in this object file format >(1)" I haven't seen that. I'll check it out. p. unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Fri Apr 10 15:25:06 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id PAA12426 for ; Fri, 10 Apr 1998 15:25:05 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:12479 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391208-1806>; Fri, 10 Apr 1998 17:24:27 +0300 Received: by vger.rutgers.edu id <971231-8479>; Fri, 10 Apr 1998 09:29:35 -0400 Received: from foobar.starlab.net ([195.171.37.162]:16930 "EHLO downtown.zav.starlab.net" ident: "soggie") by vger.rutgers.edu with ESMTP id <971269-8479>; Fri, 10 Apr 1998 09:28:50 -0400 Received: from localhost (soggie@localhost) by downtown.zav.starlab.net (8.8.5/8.8.5) with SMTP id QAA02170 for ; Fri, 10 Apr 1998 16:14:33 +0200 Date: Fri, 10 Apr 1998 16:14:32 +0200 (MET DST) From: Ivo Clarysse To: linux-arm@vger.rutgers.edu Subject: Memory mapping SA-1100 PCMCIA Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO I've succesfully built Linux for the Brutus SA-1100 evaluation board, using the source tree pointed to by Deborah Wallach on this list. (I built it on a vanilla gcc-2.8.1 and a vanilla binutils-2.8.1.0.23, on a vanilla (+ upgrade packages) RedHat5.0/Intel system - 2 small Makefile changes were necessary, probably due to GCC2.8.1) Anyway. Now that this somewhat works, I'd like to work on support for the built-in SA-1100 PCMCIA controller. According to the SA-1100 Technical Reference Manual, the PCMCIA cards are physically mapped like this: 0x20000000 - 0x23FFFFFF Socket 1 I/O Space 0x28000000 - 0x2BFFFFFF Socket 1 Attribute Space 0x2C000000 - 0x2FFFFFFF Socket 1 Memory Space 0x30000000 - 0x33FFFFFF Socket 1 I/O Space 0x38000000 - 0x3BFFFFFF Socket 1 Attribute Space 0x3C000000 - 0x3FFFFFFF Socket 1 Memory Space I assume I have to map this to virtual memory space somewhere. The current memory map for Brutus, according to include/asm-arm/proc-armv/mm-init.h, is: PHYS VIRT c0000000 00000000 Zero page c00003ff 000003ff Zero page end c0000000 c0000000 Kernel and all physical memory [physical non-contiguous] d83fffff c0ffffff End of physical (16MB) 80000000 e0000000 IO start (n.b. not all IO space is mapped) bfffffff ffffffff IO end Is it OK for me to map the entire 0x20000000-0x3FFFFFFF block as: PHYS VIRT 20000000 20000000 PCMCIA 3FFFFFFF 3FFFFFFF PCMCIA end ? Ivo. ------------------------------------------------------------------------- Ivo Clarysse TEL +32-2-7215454 FAX +32-2-7215380 Research Assistant Starlab - http://www.starlab.org/ PGP ID: 0x50154325 // A0A0 EDBC 37B2 C574 CC4D F1F0 FF94 04B9 5015 4325 unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Sat Apr 11 02:28:30 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id CAA14790 for ; Sat, 11 Apr 1998 02:28:29 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:2725 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <392295-16469>; Sat, 11 Apr 1998 04:29:02 +0300 Received: by vger.rutgers.edu id <971462-24968>; Fri, 10 Apr 1998 20:13:55 -0400 Received: from post-31.mail.demon.net ([194.217.242.41]:36160 "HELO post.mail.demon.net" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with SMTP id <971598-24968>; Fri, 10 Apr 1998 19:53:31 -0400 Received: from (tardis.home.dave) [194.222.210.232] by post.mail.demon.net with smtp (Exim 1.82 #2) id 0yNoM7-0005k6-00; Sat, 11 Apr 1998 01:41:15 +0100 Date: Sat, 11 Apr 1998 01:36:41 +0100 (BST) From: Dave Gilbert X-Sender: gilbertd@tardis.home.dave To: linux-arm@vger.rutgers.edu cc: egcs-bugs@cygnus.com, rearnsha@arm.com, nickc@cygnus.com Subject: egcs-980406 (and earlier)/ARM ldrsb - incorrect offset range Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Hi, In egcs-980406 (and earlier versions) the address offset range of the ldrsb instruction (introduced in ARM architecture v4) is set to be the same as normal ARM ldr (load register) instructions when it is actually a lot smaller. The following example excersises this: ------------------------------------------------------------------------ struct foo { int padding[200]; signed char toget; }; signed char bar(struct foo* p) { return p->toget; }; ------------------------------------------------------------------------- gcc -mcpu=strongarm110 egbug.c -O2 /tmp/cca06251.s: Assembler messages: /tmp/cca06251.s:19: Error: address offset too large _bar: @ args = 0, pretend = 0, frame = 0 @ frame_needed = 1, current_function_anonymous_args = 0 mov ip, sp stmfd sp!, {fp, ip, lr, pc} sub fp, ip, #4 ldrsb r0, [r0, #800] ******** Range should be +/- 255 ldmea fp, {fp, sp, pc}^ (This cropped up while trying to compile the tulip (DEC 21x ether) driver in Linux for StrongARM, while cross building from Alpha) Dave -------------------------------------------------------------------- / Dr. David Alan Gilbert | Running Linux on Alpha & ARM \ \ gro.gilbert @ treblig.org | ------- Happy in hex ------- / \____________________________|___ http://www.treblig.demon.co.uk __/ unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Sat Apr 11 11:04:49 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id LAA16267 for ; Sat, 11 Apr 1998 11:04:48 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:25810 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390428-16469>; Sat, 11 Apr 1998 13:05:33 +0300 Received: by vger.rutgers.edu id <970803-24968>; Sat, 11 Apr 1998 05:15:08 -0400 Received: from hikmail2.fzk.de ([141.52.2.5]:50615 "HELO hikmail2.fzk.de" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with SMTP id <970975-24968>; Sat, 11 Apr 1998 05:14:43 -0400 Received: from windose (actually dialin17.fzk.de) by hikmail2 with Local-SMTP (PP); Sat, 11 Apr 1998 12:04:44 +0200 From: Roland Fritz Organization: Rudolph&Rohnacher To: linux-arm@vger.rutgers.edu Date: Sat, 11 Apr 1998 12:04:38 +0100 MIME-Version: 1.0 Content-type: text/plain; charset=US-ASCII Content-transfer-encoding: 7BIT Subject: Linux on A310? Reply-to: roland@hik.fzk.de In-reply-to: Message-ID: <"hikmail2.f.556:11.04.98.10.04.45"@hikmail2.fzk.de> X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Hi all! For me as newbe to ARM-Linux, I want to start on my old A310/4MB. Following questions: Is the OAK SCSI interface supported? I only have SCSI disks on my A310. Is 4 MB enough to start with? The A310 can't be fitted with more (unless providing a second MEMC1a)? Is 200MB HD enough? On my PC I use 500MB, but with all you can eat (ISDN, X, FTP, WWW, SAMBA). TIA Roland. ========================================================= Roland Fritz roland@hik.fzk.de Magdeburger Ring 29 76344 Eggenstein-Leopoldshafen Germany ========================================================= unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Sat Apr 11 14:44:20 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id OAA17321 for ; Sat, 11 Apr 1998 14:44:19 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:10130 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390903-16469>; Sat, 11 Apr 1998 16:44:59 +0300 Received: by vger.rutgers.edu id <971581-24968>; Sat, 11 Apr 1998 08:53:39 -0400 Received: from odie.barnet.ac.uk ([194.82.202.98]:8759 "EHLO odie.barnet.ac.uk" ident: "willy") by vger.rutgers.edu with ESMTP id <971591-24968>; Sat, 11 Apr 1998 08:52:10 -0400 Received: (from willy@localhost) by odie.barnet.ac.uk (8.8.6/8.8.6) id OAA17300 for linux-arm@vger.rutgers.edu; Sat, 11 Apr 1998 14:41:39 +0100 From: Matthew Wilcox Message-Id: <199804111341.OAA17300@odie.barnet.ac.uk> Subject: strange goings-on with a PC-card To: linux-arm@vger.rutgers.edu Date: Sat, 11 Apr 1998 14:41:38 +0100 (BST) X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Has anyone had problems with a RiscPC and copro card? I added an SX-33 about a week ago and I've had several really nasty problems since. Might it be overheating? Does ARMLinux do something nasty to the pc card? I ask since it seems to undo the effect of PCSleep. [oh, in case anyone's interested, I'm currently working on drivers for Linux on the PC card. I've asked Mark Brinicombe to send me the source to his drivers, but he's not responded yet. Prolly because he's in the states. Stock 2.1.92 boots from an initrd without trouble but my driver for accessing floppies/hard discs/cdroms is not even nearly working yet.] -- Set Alias$Case Set Alias$[ |||| |MSet Alias$Otherwise Set Alias$[ \ Matthew "" |MSet Alias$When If %0=%%0 Then Set Alias$[ "" ||MIf %0=%%0 \ Wilcox Then Set Alias$Otherwise Set Alias$[ |||||||||||||||| ||MIf \ %0=%%0 Then Set Alias$When Set Alias$[ |||||||||||||||| unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Sun Apr 12 11:40:51 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id LAA21390 for ; Sun, 12 Apr 1998 11:40:50 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:21142 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390678-29081>; Sun, 12 Apr 1998 13:41:49 +0300 Received: by vger.rutgers.edu id <971780-24968>; Sun, 12 Apr 1998 05:46:33 -0400 Received: from tazenda.demon.co.uk ([158.152.220.239]:1085 "HELO kings-cross.london.uk.eu.org" ident: "exim") by vger.rutgers.edu with SMTP id <971771-24968>; Sun, 12 Apr 1998 05:46:19 -0400 Received: from kings-cross.london.uk.eu.org [127.0.0.1] (phil) by kings-cross.london.uk.eu.org with esmtp (Exim 1.71 #1) id 0yOK7Q-0003tb-00; Sun, 12 Apr 1998 11:36:12 +0100 X-Mailer: exmh version 2.0zeta 7/24/97 To: roland@hik.fzk.de cc: linux-arm@vger.rutgers.edu Subject: Re: Linux on A310? In-reply-to: Your message of "Sat, 11 Apr 1998 12:04:38 BST." <"hikmail2.f.556:11.04.98.10.04.45"@hikmail2.fzk.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sun, 12 Apr 1998 11:36:11 +0100 From: Philip Blundell Message-Id: X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO >Is the OAK SCSI interface supported? I only have SCSI disks on my >A310. A driver exists. I've no idea how well it works. >Is 4 MB enough to start with? The A310 can't be fitted with more >(unless providing a second MEMC1a)? 4MB is just about sufficient, yes. You will probably find your machine swaps a lot, but it will work. I'm still hoping to get 2MB machines running one day. >Is 200MB HD enough? On my PC I use 500MB, but with all you can eat >(ISDN, X, FTP, WWW, SAMBA). 200MB entirely dedicated to Linux would be fine (people use far smaller disks than that with no problems). If you want to share it with RISC OS it might be getting a little tight. It really depends on what you want to do with the system. p. unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Sun Apr 12 13:09:37 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id NAA21574 for ; Sun, 12 Apr 1998 13:09:36 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:20641 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390695-29081>; Sun, 12 Apr 1998 15:10:37 +0300 Received: by vger.rutgers.edu id <971797-214>; Sun, 12 Apr 1998 08:11:06 -0400 Received: from snowcrash.cymru.net ([163.164.160.3]:1797 "EHLO snowcrash.cymru.net" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <971791-214>; Sun, 12 Apr 1998 08:10:47 -0400 Received: from caramon.armlinux.org (dialup1.cymru.net [163.164.160.65]) by snowcrash.cymru.net (8.8.7/8.7.1) with ESMTP id NAA15078 for ; Sun, 12 Apr 1998 13:09:33 +0100 Received: from raistlin.armlinux.org (raistlin [192.168.0.3]) by caramon.armlinux.org (8.7.4/8.7.3) with ESMTP id JAA02005 for ; Sun, 12 Apr 1998 09:58:37 +0100 Received: (from linux@localhost) by raistlin.armlinux.org (8.7.4/8.7.3) id JAA00344 for linux-arm@vger.rutgers.edu; Sun, 12 Apr 1998 09:53:48 +0100 Received: from caramon (caramon [192.168.0.1]) by caramon.armlinux.org (8.7.4/8.7.3) with ESMTP id JAA01907 for ; Sun, 12 Apr 1998 09:29:39 +0100 X-POP3-Rcpt: linux@snowcrash Received: from snowcrash.cymru.net by caramon (fetchmail-4.3.2 POP3 run by linux:arm.uk.linux.org) for (single-drop); Sun Apr 12 09:29:39 1998 Received: from teapot06.onaustralia.com.au (teapot06.bigpond.com [139.134.5.237]) by snowcrash.cymru.net (8.8.7/8.7.1) with SMTP id FAA09638 for ; Sun, 12 Apr 1998 05:14:36 +0100 Received: from BN-A01-pool-31.tmns.net.au (BN-A01-pool-31.tmns.net.au [139.134.41.31]) by teapot06.onaustralia.com.au (NTMail 3.02.13) with ESMTP id ya607306 for ; Sun, 12 Apr 1998 14:14:23 +1100 From: rmk@arm.uk.linux.org (Dan Todd) To: linux@arm.uk.linux.org Subject: Installing ARMLinux from Acorn Aug 97 CDROM Date: Sun, 12 Apr 1998 14:04:19 GMT Message-ID: <19980412.140419.98@dantodd.bigpond.com> Organization: The House of Dan X-Mailer: TTFN version 0.41 (Acorn RISC OS) X-Posting-Agent: RISC OS Newsbase 0.59i X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Hello all I've been trying to install ARMLinux from the Acorn Aug 97 CDROM. I have downloaded Phil Normans archive in order to get all the files I need. I have followed all the instructions (i think) from both the installation guide available from www.arm.uk.linux.org and from Phil's archive. I have no problems except RedHat csn not find the install path on my harddrive. I am using 1 hardrive that I believe is partitioned correctly using the latest !partman. Any ideas what I can try next? Will this install take place from 1 hardrive? -- Dan Todd unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From linux@arm.uk.linux.org Mon Apr 13 09:53:49 1998 Return-Path: Received: from snowcrash.cymru.net (snowcrash.cymru.net [163.164.160.3]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id JAA25475 for ; Mon, 13 Apr 1998 09:53:48 +0100 Received: from caramon.armlinux.org (dialup2.cymru.net [163.164.160.66]) by snowcrash.cymru.net (8.8.7/8.7.1) with ESMTP id JAA29202; Mon, 13 Apr 1998 09:53:51 +0100 Received: from raistlin.armlinux.org (raistlin [192.168.0.3]) by caramon.armlinux.org (8.7.4/8.7.3) with ESMTP id JAA07907; Mon, 13 Apr 1998 09:46:46 +0100 From: Russell King - ARM Linux Admin Received: (from linux@localhost) by raistlin.armlinux.org (8.7.4/8.7.3) id JAA00480; Mon, 13 Apr 1998 09:39:49 +0100 Message-Id: <199804130839.JAA00480@raistlin.armlinux.org> Subject: Re: strange goings-on with a PC-card To: willy@odie.barnet.ac.uk (Matthew Wilcox) Date: Mon, 13 Apr 1998 09:39:48 +0100 (BST) Cc: linux-arm@vger.rutgers.edu In-Reply-To: <199804111341.OAA17300@odie.barnet.ac.uk> from "Matthew Wilcox" at Apr 11, 98 02:41:38 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Status: RO Matthew Wilcox writes: > Has anyone had problems with a RiscPC and copro card? I added an SX-33 > about a week ago and I've had several really nasty problems since. Might > it be overheating? Does ARMLinux do something nasty to the pc card? > I ask since it seems to undo the effect of PCSleep. ARM Linux does not touch anything near the second processor interface. The only thing is that ARM Linux will trample over all RAM, including the RAM that the PC card is using... > [oh, in case anyone's interested, I'm currently working on drivers for > Linux on the PC card. I've asked Mark Brinicombe to send me the source > to his drivers, but he's not responded yet. Prolly because he's in the > states. Stock 2.1.92 boots from an initrd without trouble but my driver > for accessing floppies/hard discs/cdroms is not even nearly working yet.] >From what I remember a couple of years ago, this information is only available under NDA. You're not the first person to try to port Linux to the PC card - others include Mark Brinicombe and Martin Ebourne... In each case, they haven't been able to release the driver sources for their port. _____ |_____| ------------------------------------------------- ---+---+- | | Russell King rmk@ecs.soton.ac.uk --- --- | | | | http://www.arm.uk.linux.org/~rmk/home.html / / | | +-+-+ --- -+- / | THE developer of ARM Linux |+| /|\ / | | | --- | +-+-+ ------------------------------------------------- /\\\ | From owner-linux-arm-outgoing@vger.rutgers.edu Mon Apr 13 10:56:45 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id KAA25794 for ; Mon, 13 Apr 1998 10:56:43 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:30782 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390805-13337>; Mon, 13 Apr 1998 12:56:22 +0300 Received: by vger.rutgers.edu id <970996-214>; Mon, 13 Apr 1998 05:46:49 -0400 Received: from odie.barnet.ac.uk ([194.82.202.98]:28103 "EHLO odie.barnet.ac.uk" ident: "willy") by vger.rutgers.edu with ESMTP id <971037-214>; Mon, 13 Apr 1998 05:19:50 -0400 Received: (from willy@localhost) by odie.barnet.ac.uk (8.8.6/8.8.6) id KAA25679; Mon, 13 Apr 1998 10:20:17 +0100 From: Matthew Wilcox Message-Id: <199804130920.KAA25679@odie.barnet.ac.uk> Subject: Re: strange goings-on with a PC-card To: linux@arm.uk.linux.org (Russell King - ARM Linux Admin) Date: Mon, 13 Apr 1998 10:20:16 +0100 (BST) Cc: linux-arm@vger.rutgers.edu In-Reply-To: <199804130839.JAA00480@raistlin.armlinux.org> from "Russell King - ARM Linux Admin" at Apr 13, 98 09:39:48 am X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Russell King - ARM Linux Admin > > Matthew Wilcox writes: > > Has anyone had problems with a RiscPC and copro card? I added an SX-33 > > about a week ago and I've had several really nasty problems since. Might > > it be overheating? Does ARMLinux do something nasty to the pc card? > > I ask since it seems to undo the effect of PCSleep. > > ARM Linux does not touch anything near the second processor interface. > The only thing is that ARM Linux will trample over all RAM, including the > RAM that the PC card is using... That's not the problem. I have quit !PC before booting Linux. For some reason the CPU heats up again. I get random crashes, and I just had a nasty bout with fsck ;-( Fortunately nothing important got damaged. If I can get the source released in a usable form, how about integrating PCSleep into the kernel? > > [oh, in case anyone's interested, I'm currently working on drivers for > > Linux on the PC card. I've asked Mark Brinicombe to send me the source > > to his drivers, but he's not responded yet. Prolly because he's in the > > states. Stock 2.1.92 boots from an initrd without trouble but my driver > > for accessing floppies/hard discs/cdroms is not even nearly working yet.] > > From what I remember a couple of years ago, this information is only > available under NDA. You're not the first person to try to port Linux > to the PC card - others include Mark Brinicombe and Martin Ebourne... > In each case, they haven't been able to release the driver sources > for their port. Ah, I don't remember MJE having a go at it. Yes, I've had to sign an NDA, but they accept that the source will be released under the GPL and are happy with this. I wouldn't've agreed to do it otherwise. -- Set Alias$Case Set Alias$[ |||| |MSet Alias$Otherwise Set Alias$[ \ Matthew "" |MSet Alias$When If %0=%%0 Then Set Alias$[ "" ||MIf %0=%%0 \ Wilcox Then Set Alias$Otherwise Set Alias$[ |||||||||||||||| ||MIf \ %0=%%0 Then Set Alias$When Set Alias$[ |||||||||||||||| unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From alan@snowcrash.cymru.net Mon Apr 13 15:22:54 1998 Return-Path: Received: from snowcrash.cymru.net (snowcrash.cymru.net [163.164.160.3]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id PAA26732 for ; Mon, 13 Apr 1998 15:22:53 +0100 Received: (from alan@localhost) by snowcrash.cymru.net (8.8.7/8.7.1) id PAA01182; Mon, 13 Apr 1998 15:22:59 +0100 From: Alan Cox Message-Id: <199804131422.PAA01182@snowcrash.cymru.net> Subject: Re: strange goings-on with a PC-card To: linux@arm.uk.linux.org (Russell King - ARM Linux Admin) Date: Mon, 13 Apr 1998 15:22:59 +0100 (BST) Cc: willy@odie.barnet.ac.uk, linux-arm@vger.rutgers.edu In-Reply-To: <199804130839.JAA00480@raistlin.armlinux.org> from "Russell King - ARM Linux Admin" at Apr 13, 98 09:39:48 am Content-Type: text Status: RO > available under NDA. You're not the first person to try to port Linux > to the PC card - others include Mark Brinicombe and Martin Ebourne... > In each case, they haven't been able to release the driver sources > for their port. In which case you ae probably better reverse engineering the system but doing it legally. Or better yet tell Acorn where to stick their policy and buy ARM hardware and PC hardware from more helpful sources. Alan From owner-linux-arm-outgoing@vger.rutgers.edu Mon Apr 13 15:29:51 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id PAA26775 for ; Mon, 13 Apr 1998 15:29:50 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:14795 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390972-13337>; Mon, 13 Apr 1998 17:28:02 +0300 Received: by vger.rutgers.edu id <970794-214>; Mon, 13 Apr 1998 10:26:49 -0400 Received: from irwell.zetnet.co.uk ([194.247.47.48]:12038 "EHLO irwell.zetnet.co.uk" ident: "root") by vger.rutgers.edu with ESMTP id <970806-214>; Mon, 13 Apr 1998 10:26:23 -0400 Received: from [194.247.40.152] (man-119.dialup.zetnet.co.uk [194.247.40.152]) by irwell.zetnet.co.uk (8.8.7/8.8.5) with SMTP id PAA22149 for ; Mon, 13 Apr 1998 15:27:13 +0100 To: linux-arm@vger.rutgers.edu From: jjf@zetnet.co.uk (JJF) Subject: Re: [Fwd: virus warning] In-Reply-To: Date: Mon, 13 Apr 1998 13:37:55 +0100 Message-ID: <19980413.133755.68@zetnet.co.uk> Organization: Internet Access supplied by Zetnet X-Mailer: TTFN version 0.41 (Acorn RISC OS) X-Posting-Agent: RISC OS Newsbase 0.60g X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Just in case anyone doesn't know, this are all hoax viruses. It is *not* possible for simply reading an email message to install a virus in your computer (*unless* you have a dumb mail reader that either tries to execute attachments automatically, or supports MS Word macros). And, of course, since we all use Acorns, we'd be okay even if such an impossible virus did exist in the DOS world. Public service announcement ends, I now return you to your regularly scheduled email... JJF. -- ***************************************************************************** * Doctor Who: * * The Missing Internet Adventures * * http://www.users.zetnet.co.uk/jjf/drwho/mia/ * unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Mon Apr 13 15:37:24 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id PAA26839 for ; Mon, 13 Apr 1998 15:37:23 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:32001 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390983-13337>; Mon, 13 Apr 1998 17:37:10 +0300 Received: by vger.rutgers.edu id <970794-214>; Mon, 13 Apr 1998 10:35:43 -0400 Received: from odie.barnet.ac.uk ([194.82.202.98]:30780 "EHLO odie.barnet.ac.uk" ident: "willy") by vger.rutgers.edu with ESMTP id <970835-214>; Mon, 13 Apr 1998 10:35:33 -0400 Received: (from willy@localhost) by odie.barnet.ac.uk (8.8.6/8.8.6) id PAA26822; Mon, 13 Apr 1998 15:36:14 +0100 From: Matthew Wilcox Message-Id: <199804131436.PAA26822@odie.barnet.ac.uk> Subject: Re: strange goings-on with a PC-card To: alan@cymru.net (Alan Cox) Date: Mon, 13 Apr 1998 15:36:13 +0100 (BST) Cc: linux@arm.uk.linux.org, linux-arm@vger.rutgers.edu In-Reply-To: <199804131422.PAA01182@snowcrash.cymru.net> from "Alan Cox" at Apr 13, 98 03:22:59 pm X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Alan Cox > > > available under NDA. You're not the first person to try to port Linux > > to the PC card - others include Mark Brinicombe and Martin Ebourne... > > In each case, they haven't been able to release the driver sources > > for their port. > > In which case you ae probably better reverse engineering the system but > doing it legally. Or better yet tell Acorn where to stick their policy and > buy ARM hardware and PC hardware from more helpful sources. Calm down! I've been speaking to MattBee from Aleph-1, he phoned up Laurie van Someren and we now have permission to use the source to PCSleep in ARMLinux. I have said source and will look at it tonight. There's a certain amount more openness than there used to be - once I've got this out of the way I have another proposition to put to them. The NDA that I have signed will allow me to release the work I have done on the driver under the GPL. I have not yet done so because it doesn't work. Yet ;-) -- Set Alias$Case Set Alias$[ |||| |MSet Alias$Otherwise Set Alias$[ \ Matthew "" |MSet Alias$When If %0=%%0 Then Set Alias$[ "" ||MIf %0=%%0 \ Wilcox Then Set Alias$Otherwise Set Alias$[ |||||||||||||||| ||MIf \ %0=%%0 Then Set Alias$When Set Alias$[ |||||||||||||||| unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From fmpatino@fps.biblos.unal.edu.co Mon Apr 13 17:55:35 1998 Return-Path: Received: from fps.biblos.unal.edu.co ([168.176.37.11] (may be forged)) by odie.barnet.ac.uk (8.8.6/8.8.6) with SMTP id RAA27321 for ; Mon, 13 Apr 1998 17:52:26 +0100 Received: from localhost by fps.biblos.unal.edu.co (AIX 4.1/UCB 5.64/4.03) id AA04516; Mon, 13 Apr 1998 11:39:03 -0500 Date: Mon, 13 Apr 1998 11:39:03 -0500 (EST) From: "FABIAN PATINO (BIANCCIO)" To: Matthew Wilcox Cc: Alan Cox , linux@arm.uk.linux.org, linux-arm@vger.rutgers.edu Subject: Re: strange goings-on with a PC-card In-Reply-To: <199804131436.PAA26822@odie.barnet.ac.uk> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Status: RO Some body can help me? i got problems with my linux webserver, when i try to connect linuux give me a erro message : server_log acces denied whats is the problem. what can i do ? On Mon, 13 Apr 1998, Matthew Wilcox wrote: > Alan Cox > > > > > available under NDA. You're not the first person to try to port Linux > > > to the PC card - others include Mark Brinicombe and Martin Ebourne... > > > In each case, they haven't been able to release the driver sources > > > for their port. > > > > In which case you ae probably better reverse engineering the system but > > doing it legally. Or better yet tell Acorn where to stick their policy and > > buy ARM hardware and PC hardware from more helpful sources. > > Calm down! I've been speaking to MattBee from Aleph-1, he phoned up > Laurie van Someren and we now have permission to use the source to > PCSleep in ARMLinux. I have said source and will look at it tonight. > There's a certain amount more openness than there used to be - once I've > got this out of the way I have another proposition to put to them. > > The NDA that I have signed will allow me to release the work I have done > on the driver under the GPL. I have not yet done so because it doesn't > work. Yet ;-) > > -- > Set Alias$Case Set Alias$[ |||| |MSet Alias$Otherwise Set Alias$[ \ Matthew > "" |MSet Alias$When If %0=%%0 Then Set Alias$[ "" ||MIf %0=%%0 \ Wilcox > Then Set Alias$Otherwise Set Alias$[ |||||||||||||||| ||MIf \ > %0=%%0 Then Set Alias$When Set Alias$[ |||||||||||||||| > unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu > From owner-linux-arm-outgoing@vger.rutgers.edu Mon Apr 13 18:30:33 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id SAA27523 for ; Mon, 13 Apr 1998 18:30:31 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:31878 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <392755-13337>; Mon, 13 Apr 1998 20:28:13 +0300 Received: by vger.rutgers.edu id <970932-214>; Mon, 13 Apr 1998 13:24:39 -0400 Received: from odie.barnet.ac.uk ([194.82.202.98]:31802 "EHLO odie.barnet.ac.uk" ident: "willy") by vger.rutgers.edu with ESMTP id <970791-214>; Mon, 13 Apr 1998 13:24:13 -0400 Received: (from willy@localhost) by odie.barnet.ac.uk (8.8.6/8.8.6) id SAA27447; Mon, 13 Apr 1998 18:20:21 +0100 From: Matthew Wilcox Message-Id: <199804131720.SAA27447@odie.barnet.ac.uk> Subject: Re: strange goings-on with a PC-card To: fmpatino@fps.biblos.unal.edu.co (FABIAN PATINO) Date: Mon, 13 Apr 1998 18:20:21 +0100 (BST) Cc: linux-arm@vger.rutgers.edu In-Reply-To: from "FABIAN PATINO" at Apr 13, 98 11:39:03 am X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO FABIAN PATINO > Some body can help me? i got problems with my linux webserver, when i try > to connect linuux give me a erro message : server_log acces denied > > whats is the problem. what can i do ? You could try asking in the appropriate place. This list is specifically for discussing the Linux port to ARM processors. You could also attempt to provide more useful information when you do ask for help in a more appropriate place. Give a clear description of a sequence of steps which lead to the error occurring and report the error message exactly. Some details of your setup would also be a good idea. Try the newsgroup comp.infosystems.www.servers.unix. -- Set Alias$Case Set Alias$[ |||| |MSet Alias$Otherwise Set Alias$[ \ Matthew "" |MSet Alias$When If %0=%%0 Then Set Alias$[ "" ||MIf %0=%%0 \ Wilcox Then Set Alias$Otherwise Set Alias$[ |||||||||||||||| ||MIf \ %0=%%0 Then Set Alias$When Set Alias$[ |||||||||||||||| unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Mon Apr 13 20:13:51 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id UAA29788 for ; Mon, 13 Apr 1998 20:13:49 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:14724 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <393068-13337>; Mon, 13 Apr 1998 22:11:36 +0300 Received: by vger.rutgers.edu id <970909-214>; Mon, 13 Apr 1998 14:33:15 -0400 Received: from runyon.cygnus.com ([205.180.230.5]:32947 "EHLO cygnus.com" ident: "SOCKWRITE-65") by vger.rutgers.edu with ESMTP id <971103-214>; Mon, 13 Apr 1998 14:29:49 -0400 Received: from elmo.cygnus.com (elmo.cygnus.com [205.180.230.137]) by runyon.cygnus.com (8.8.7-cygnus/8.8.7) with ESMTP id LAA03750; Mon, 13 Apr 1998 11:31:00 -0700 (PDT) Received: (nickc@localhost) by elmo.cygnus.com (8.8.5/8.6.4) id LAA30433; Mon, 13 Apr 1998 11:30:29 -0700 Date: Mon, 13 Apr 1998 11:30:29 -0700 Message-Id: <199804131830.LAA30433@elmo.cygnus.com> From: Nick Clifton To: gilbertd@treblig.org CC: linux-arm@vger.rutgers.edu, egcs-bugs@cygnus.com, rearnsha@arm.com In-reply-to: (message from Dave Gilbert on Sat, 11 Apr 1998 01:36:41 +0100 (BST)) Subject: Re: egcs-980406 (and earlier)/ARM ldrsb - incorrect offset range X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO : Date: Sat, 11 Apr 1998 01:36:41 +0100 (BST) : From: Dave Gilbert : : Hi, : In egcs-980406 (and earlier versions) the address offset range of the : ldrsb instruction (introduced in ARM architecture v4) is set to be the : same as normal ARM ldr (load register) instructions when it is actually : a lot smaller. The following example excersises this: : : ------------------------------------------------------------------------ : struct foo { : int padding[200]; : signed char toget; : }; : : signed char bar(struct foo* p) { : return p->toget; : }; : ------------------------------------------------------------------------- : gcc -mcpu=strongarm110 egbug.c -O2 : /tmp/cca06251.s: Assembler messages: : /tmp/cca06251.s:19: Error: address offset too large : : _bar: : @ args = 0, pretend = 0, frame = 0 : @ frame_needed = 1, current_function_anonymous_args = 0 : mov ip, sp : stmfd sp!, {fp, ip, lr, pc} : sub fp, ip, #4 : ldrsb r0, [r0, #800] ******** Range should be +/- 255 : ldmea fp, {fp, sp, pc}^ : Applying this patch fixes the problem: Index: arm.h =================================================================== RCS file: /egcs/carton/cvsfiles/egcs/gcc/config/arm/arm.h,v retrieving revision 1.5 diff -p -r1.5 arm.h *** arm.h 1998/04/08 06:19:00 1.5 --- arm.h 1998/04/13 18:30:02 *************** do \ *** 1408,1414 **** && INTVAL (op) <= 31) \ goto LABEL; \ } \ ! range = (MODE) == HImode ? (arm_arch4 ? 256 : 4095) : 4096; \ if (code == CONST_INT && INTVAL (INDEX) < range \ && INTVAL (INDEX) > -range) \ goto LABEL; \ --- 1408,1416 ---- && INTVAL (op) <= 31) \ goto LABEL; \ } \ ! /* NASTY: Since this limits the addressing of unsigned byte loads */ \ ! range = ((MODE) == HImode || (MODE) == QImode) \ ! ? (arm_arch4 ? 256 : 4095) : 4096; \ if (code == CONST_INT && INTVAL (INDEX) < range \ && INTVAL (INDEX) > -range) \ goto LABEL; \ The code now produced looks like this: _bar: mov ip, sp stmfd sp!, {fp, ip, lr, pc} mov r3, #800 sub fp, ip, #4 ldrsb r0, [r0, r3] ldmea fp, {fp, sp, pc} Nick unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Mon Apr 13 21:26:23 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id VAA30345 for ; Mon, 13 Apr 1998 21:26:22 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:14724 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390445-13337>; Mon, 13 Apr 1998 23:26:14 +0300 Received: by vger.rutgers.edu id <970998-214>; Mon, 13 Apr 1998 16:18:01 -0400 Received: from tazenda.demon.co.uk ([158.152.220.239]:1066 "HELO kings-cross.london.uk.eu.org" ident: "exim") by vger.rutgers.edu with SMTP id <971050-214>; Mon, 13 Apr 1998 15:50:31 -0400 Received: from kings-cross.london.uk.eu.org [127.0.0.1] (phil) by kings-cross.london.uk.eu.org with esmtp (Exim 1.71 #1) id 0yOp6g-0000P1-00; Mon, 13 Apr 1998 20:41:30 +0100 X-Mailer: exmh version 2.0zeta 7/24/97 To: rmk@arm.uk.linux.org cc: linux-arm@vger.rutgers.edu Subject: loader Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 13 Apr 1998 20:41:30 +0100 From: Philip Blundell Message-Id: X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Hi Russell, Do you have any plans to make the source for the bootloader available? I couldn't find it on the ftp site. p. unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Tue Apr 14 10:38:22 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id KAA00685 for ; Tue, 14 Apr 1998 10:38:21 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:30256 "EHLO vger.rutgers.edu" ident: "TIMEDOUT") by orava.funet.fi with ESMTP id <390758-28518>; Tue, 14 Apr 1998 12:36:52 +0300 Received: by vger.rutgers.edu id <971003-214>; Tue, 14 Apr 1998 05:33:54 -0400 Received: from violet.csi.cam.ac.uk ([131.111.8.58]:36856 "HELO violet.csi.cam.ac.uk" ident: "exim") by vger.rutgers.edu with SMTP id <970806-214>; Tue, 14 Apr 1998 05:27:49 -0400 Received: from mup.pet.cam.ac.uk [131.111.209.211] (kjm25) by violet.csi.cam.ac.uk with smtp (Exim 1.73 #1) id 0yP22K-0003Vu-00; Tue, 14 Apr 1998 10:29:52 +0100 Date: Tue, 14 Apr 1998 10:30:02 +0100 (BST) From: Kieran Mansley Subject: gdb To: linux-arm@vger.rutgers.edu Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII X-Organization: Peterhouse, Cambridge. X-Mailer: ANT RISCOS Marcel [ver 1.41] X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO I'm trying to get a working copy of gdb. Looking through past postings to this list I see in November that Russell mentioned that he had a working binary (and source) for it, but couldn't compile it up for the rpm distribution. Has any progress been made? Is this binary more widely available? Does anyone have a working version? I have tried compiling gdb straight but failed. Thanks Kieran -- Kieran Mansley, Peterhouse, Cambridge. CB2 1RD kjm25@cam.ac.uk http://mup.pet.cam.ac.uk/ I tell all my friends but they never believe me. They think that I've finally lost it completely. unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Tue Apr 14 12:34:49 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id MAA01004 for ; Tue, 14 Apr 1998 12:34:48 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:5365 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390705-28518>; Tue, 14 Apr 1998 14:33:41 +0300 Received: by vger.rutgers.edu id <970959-214>; Tue, 14 Apr 1998 07:30:36 -0400 Received: from aa175.du.pipex.com ([193.130.240.175]:24731 "EHLO seneca.collegium.co.uk" ident: "chris") by vger.rutgers.edu with ESMTP id <970917-214>; Tue, 14 Apr 1998 07:27:15 -0400 Received: from localhost (chris@localhost) by seneca.collegium.co.uk (8.8.7/8.8.7) with SMTP id MAA25947 for ; Tue, 14 Apr 1998 12:29:21 +0100 Date: Tue, 14 Apr 1998 12:29:21 +0100 (BST) From: Chris Rutter To: linux-arm@vger.rutgers.edu Subject: Building a 2.0.33 kernel with the latest patches In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO I've configured a 2.0.33 kernel (RiscPC/StrongARM), but on trying to build it this occurs: make zImage ... make[3]: Entering directory `/usr/src/linux-2.0.33/arch/arm/lib' make[3]: `constants.h' is up to date. make[3]: Leaving directory `/usr/src/linux-2.0.33/arch/arm/lib' gcc -D__KERNEL__ -I/usr/src/linux-2.0.33/include -m6 -Wall -Wstrict-prototypes -O2 -fomit-frame-pointer -fno-strength-reduce -pipe -D__ASSEMBLY__ -E proc-arm6,7.S | tr ';$' '\n#' > ..tmp.s proc-arm6,7.S:11: constants.h: No such file or directory gcc -D__KERNEL__ -I/usr/src/linux-2.0.33/include -m6 -Wall -Wstrict-prototypes -O2 -fomit-frame-pointer -fno-strength-reduce -pipe -c -o proc-arm6,7.o ..tmp.s proc-arm6,7.S: Assembler messages: proc-arm6,7.S:73: Error: Internal_relocation (type 159) not fixed up (OFFSET_IMM) proc-arm6,7.S:76: Error: Internal_relocation (type 159) not fixed up (OFFSET_IMM) proc-arm6,7.S:77: Error: Internal_relocation (type 159) not fixed up (OFFSET_IMM) make[2]: *** [proc-arm6,7.o] Error 1 make[2]: Leaving directory `/usr/src/linux-2.0.33/arch/arm/mm' make[1]: *** [first_rule] Error 2 make[1]: Leaving directory `/usr/src/linux-2.0.33/arch/arm/mm' make: *** [linuxsubdirs] Error 2 This may be relevant: find . -name constants.h ./drivers/scsi/constants.h ./arch/arm/drivers/scsi/constants.h ./arch/arm/lib/constants.h Any ideas? Chris. unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Tue Apr 14 12:50:28 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id MAA01077 for ; Tue, 14 Apr 1998 12:50:27 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:17746 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390825-28518>; Tue, 14 Apr 1998 14:49:34 +0300 Received: by vger.rutgers.edu id <970955-214>; Tue, 14 Apr 1998 07:43:10 -0400 Received: from odie.barnet.ac.uk ([194.82.202.98]:8933 "EHLO odie.barnet.ac.uk" ident: "willy") by vger.rutgers.edu with ESMTP id <971039-214>; Tue, 14 Apr 1998 07:42:51 -0400 Received: (from willy@localhost) by odie.barnet.ac.uk (8.8.6/8.8.6) id MAA01047; Tue, 14 Apr 1998 12:44:42 +0100 From: Matthew Wilcox Message-Id: <199804141144.MAA01047@odie.barnet.ac.uk> Subject: Re: Building a 2.0.33 kernel with the latest patches To: chris@collegium.co.uk (Chris Rutter) Date: Tue, 14 Apr 1998 12:44:40 +0100 (BST) Cc: linux-arm@vger.rutgers.edu In-Reply-To: from "Chris Rutter" at Apr 14, 98 12:29:21 pm X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Chris Rutter > proc-arm6,7.S:11: constants.h: No such file or directory This is the important error. > find . -name constants.h > ./drivers/scsi/constants.h > ./arch/arm/drivers/scsi/constants.h > ./arch/arm/lib/constants.h > > Any ideas? Yep, this is the remanifestation of a bug which occurred with the previous 2.0.33 patch. I fixed this by editing all the proc-*.S files in #include .../constants.h instead of merely constants.h. -- Set Alias$Case Set Alias$[ |||| |MSet Alias$Otherwise Set Alias$[ \ Matthew "" |MSet Alias$When If %0=%%0 Then Set Alias$[ "" ||MIf %0=%%0 \ Wilcox Then Set Alias$Otherwise Set Alias$[ |||||||||||||||| ||MIf \ %0=%%0 Then Set Alias$When Set Alias$[ |||||||||||||||| unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Tue Apr 14 13:55:52 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id NAA01276 for ; Tue, 14 Apr 1998 13:55:51 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:4758 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390475-28518>; Tue, 14 Apr 1998 15:55:37 +0300 Received: by vger.rutgers.edu id <970814-214>; Tue, 14 Apr 1998 08:52:41 -0400 Received: from snowcrash.cymru.net ([163.164.160.3]:1289 "EHLO snowcrash.cymru.net" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <970806-214>; Tue, 14 Apr 1998 08:52:16 -0400 Received: from caramon.armlinux.org (dialup1.cymru.net [163.164.160.65]) by snowcrash.cymru.net (8.8.7/8.7.1) with ESMTP id NAA24313; Tue, 14 Apr 1998 13:54:26 +0100 Received: from raistlin.armlinux.org (raistlin [192.168.0.3]) by caramon.armlinux.org (8.7.4/8.7.3) with ESMTP id OAA12426; Tue, 14 Apr 1998 14:03:59 +0100 From: Russell King - ARM Linux Admin Received: (from linux@localhost) by raistlin.armlinux.org (8.7.4/8.7.3) id NAA06194; Tue, 14 Apr 1998 13:54:31 +0100 Message-Id: <199804141254.NAA06194@raistlin.armlinux.org> Subject: Re: Building a 2.0.33 kernel with the latest patches To: chris@collegium.co.uk (Chris Rutter) Date: Tue, 14 Apr 1998 13:54:30 +0100 (BST) Cc: linux-arm@vger.rutgers.edu In-Reply-To: from "Chris Rutter" at Apr 14, 98 12:29:21 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Chris Rutter writes: > I've configured a 2.0.33 kernel (RiscPC/StrongARM), but on trying to > build it this occurs: Which source is this? Is it from the one I released yesterday? _____ |_____| ------------------------------------------------- ---+---+- | | Russell King rmk@ecs.soton.ac.uk --- --- | | | | http://www.arm.uk.linux.org/~rmk/home.html / / | | +-+-+ --- -+- / | THE developer of ARM Linux |+| /|\ / | | | --- | +-+-+ ------------------------------------------------- /\\\ | unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Tue Apr 14 15:25:58 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id PAA01593 for ; Tue, 14 Apr 1998 15:25:56 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:26863 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390754-28518>; Tue, 14 Apr 1998 17:23:23 +0300 Received: by vger.rutgers.edu id <971043-214>; Tue, 14 Apr 1998 09:46:47 -0400 Received: from odie.barnet.ac.uk ([194.82.202.98]:9812 "EHLO odie.barnet.ac.uk" ident: "willy") by vger.rutgers.edu with ESMTP id <971144-214>; Tue, 14 Apr 1998 09:40:52 -0400 Received: (from willy@localhost) by odie.barnet.ac.uk (8.8.6/8.8.6) id OAA01432; Tue, 14 Apr 1998 14:42:49 +0100 From: Matthew Wilcox Message-Id: <199804141342.OAA01432@odie.barnet.ac.uk> Subject: Re: Building a 2.0.33 kernel with the latest patches To: linux@arm.uk.linux.org (Russell King - ARM Linux Admin) Date: Tue, 14 Apr 1998 14:42:48 +0100 (BST) Cc: chris@collegium.co.uk, linux-arm@vger.rutgers.edu In-Reply-To: <199804141254.NAA06194@raistlin.armlinux.org> from "Russell King - ARM Linux Admin" at Apr 14, 98 01:54:30 pm X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Russell King - ARM Linux Admin > > Chris Rutter writes: > > I've configured a 2.0.33 kernel (RiscPC/StrongARM), but on trying to > > build it this occurs: > > Which source is this? Is it from the one I released yesterday? Yep. -- Set Alias$Case Set Alias$[ |||| |MSet Alias$Otherwise Set Alias$[ \ Matthew "" |MSet Alias$When If %0=%%0 Then Set Alias$[ "" ||MIf %0=%%0 \ Wilcox Then Set Alias$Otherwise Set Alias$[ |||||||||||||||| ||MIf \ %0=%%0 Then Set Alias$When Set Alias$[ |||||||||||||||| unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Tue Apr 14 19:00:54 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id TAA03146 for ; Tue, 14 Apr 1998 19:00:52 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:29104 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390569-28518>; Tue, 14 Apr 1998 20:49:20 +0300 Received: by vger.rutgers.edu id <971172-214>; Tue, 14 Apr 1998 13:20:40 -0400 Received: from snowcrash.cymru.net ([163.164.160.3]:1850 "EHLO snowcrash.cymru.net" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <971240-214>; Tue, 14 Apr 1998 12:58:20 -0400 Received: from caramon.armlinux.org (dialup2.cymru.net [163.164.160.66]) by snowcrash.cymru.net (8.8.7/8.7.1) with ESMTP id SAA29130; Tue, 14 Apr 1998 18:00:24 +0100 Received: from raistlin.armlinux.org (raistlin [192.168.0.3]) by caramon.armlinux.org (8.7.4/8.7.3) with ESMTP id QAA13307; Tue, 14 Apr 1998 16:23:51 +0100 From: Russell King - ARM Linux Admin Received: (from linux@localhost) by raistlin.armlinux.org (8.7.4/8.7.3) id QAA07167; Tue, 14 Apr 1998 16:14:10 +0100 Message-Id: <199804141514.QAA07167@raistlin.armlinux.org> Subject: Re: Building a 2.0.33 kernel with the latest patches To: chris@collegium.co.uk (Chris Rutter) Date: Tue, 14 Apr 1998 16:14:09 +0100 (BST) Cc: linux-arm@vger.rutgers.edu In-Reply-To: from "Chris Rutter" at Apr 14, 98 12:29:21 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Chris Rutter writes: > I've configured a 2.0.33 kernel (RiscPC/StrongARM), but on trying to > build it this occurs: Ok fixed. In arch/arm/mm/proc-armv.S, change: -#include "constants.h" +#include "../lib/constants.h" This will be in the next patch (probably for 2.0.34). _____ |_____| ------------------------------------------------- ---+---+- | | Russell King rmk@ecs.soton.ac.uk --- --- | | | | http://www.arm.uk.linux.org/~rmk/home.html / / | | +-+-+ --- -+- / | THE developer of ARM Linux |+| /|\ / | | | --- | +-+-+ ------------------------------------------------- /\\\ | unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Wed Apr 15 14:02:55 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id OAA07508 for ; Wed, 15 Apr 1998 14:02:52 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:12520 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390941-16497>; Wed, 15 Apr 1998 16:02:47 +0300 Received: by vger.rutgers.edu id <970794-214>; Wed, 15 Apr 1998 08:58:12 -0400 Received: from [143.89.86.254] ([143.89.86.254]:1210 "EHLO sqz519.ust.hk" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <970862-214>; Wed, 15 Apr 1998 08:56:40 -0400 Received: (from otfried@localhost) by sqz519.ust.hk (8.8.7/8.8.7) id VAA01520; Wed, 15 Apr 1998 21:00:15 +0800 Date: Wed, 15 Apr 1998 21:00:15 +0800 Message-Id: <199804151300.VAA01520@sqz519.ust.hk> From: To: linux-arm@vger.rutgers.edu Subject: Setting up cross-compiler MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: GNU Emacs 20.2.1 X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Excuse me for what is probably a very basic question: Is there any more information about setting up gcc as a cross-compiler for ARM? The instructions at "http://www.arm.uk.linux.org/armlinux/tools.html" simply suggest to apply the patches to gcc-2.7.2.2 (by the way, why is the directory in gcc-2.7.2.2.patch called gcc-2.7.2, and why is one of the patches already in the gcc-2.7.2.2 distribution?). What I don't understand is how to get around the problem of making libgcc1.a. I'm building with --target=arm-unknown-linuxaout --host=i586-redhat-linux, and the build stops with "You must find a way to make libgcc1.a". I've read the gcc info files about building cross compilers, but that leaves me even more confused. Otfried unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Wed Apr 15 14:28:01 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id OAA07680 for ; Wed, 15 Apr 1998 14:28:00 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:12520 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390785-16497>; Wed, 15 Apr 1998 16:28:12 +0300 Received: by vger.rutgers.edu id <970801-214>; Wed, 15 Apr 1998 09:24:10 -0400 Received: from odie.barnet.ac.uk ([194.82.202.98]:19714 "EHLO odie.barnet.ac.uk" ident: "TIMEDOUT") by vger.rutgers.edu with ESMTP id <970790-214>; Wed, 15 Apr 1998 09:23:56 -0400 Received: (from willy@localhost) by odie.barnet.ac.uk (8.8.6/8.8.6) id OAA07648; Wed, 15 Apr 1998 14:25:54 +0100 From: Matthew Wilcox Message-Id: <199804151325.OAA07648@odie.barnet.ac.uk> Subject: Re: Setting up cross-compiler To: otfried@cs.ust.hk Date: Wed, 15 Apr 1998 14:25:53 +0100 (BST) Cc: linux-arm@vger.rutgers.edu In-Reply-To: <199804151300.VAA01520@sqz519.ust.hk> from "otfried@cs.ust.hk" at Apr 15, 98 09:00:15 pm X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO otfried@cs.ust.hk > Excuse me for what is probably a very basic question: Is there any > more information about setting up gcc as a cross-compiler for ARM? It's not a basic question at all, crosscompiling to the ARM with 2.7.x was Hard. 2.8.x is significantly easier. > What I don't understand is how to get around the problem of making > libgcc1.a. I'm building with --target=arm-unknown-linuxaout > --host=i586-redhat-linux, and the build stops with "You must find a > way to make libgcc1.a". I've read the gcc info files about building > cross compilers, but that leaves me even more confused. Grab the libgcc.a that comes with the distribution. Should be in the gcc RPM. Alternatively, build gcc 2.8.1 for crosscompilation, without applying any patches to it. This _should_ work. 2.8.0 miscompiles ARMLinux kernels for me (and apparently it also miscompiles Linux for other architectures); I haven't tried 2.8.1 yet. -- Set Alias$Case Set Alias$[ |||| |MSet Alias$Otherwise Set Alias$[ \ Matthew "" |MSet Alias$When If %0=%%0 Then Set Alias$[ "" ||MIf %0=%%0 \ Wilcox Then Set Alias$Otherwise Set Alias$[ |||||||||||||||| ||MIf \ %0=%%0 Then Set Alias$When Set Alias$[ |||||||||||||||| unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Wed Apr 15 16:57:56 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id QAA09322 for ; Wed, 15 Apr 1998 16:57:54 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:5366 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390476-16497>; Wed, 15 Apr 1998 18:55:45 +0300 Received: by vger.rutgers.edu id <970936-214>; Wed, 15 Apr 1998 11:37:25 -0400 Received: from nexusel.demon.co.uk ([158.152.30.195]:17963 "HELO globe" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with SMTP id <970977-214>; Wed, 15 Apr 1998 11:35:14 -0400 Received: from (spring.nexus.co.uk) [192.0.0.3] (root) by globe with smtp (Exim 1.82 #1) id 0yPUC5-0001Vn-00; Wed, 15 Apr 1998 16:33:49 +0100 Received: from localhost (spring.nexus.co.uk) [127.0.0.1] (pb) by spring.nexus.co.uk with esmtp (Exim 1.82 #1) id 0yPUC4-0008QE-00; Wed, 15 Apr 1998 16:33:48 +0100 X-Mailer: exmh version 2.0zeta 7/24/97 To: otfried@cs.ust.hk cc: linux-arm@vger.rutgers.edu Subject: Re: Setting up cross-compiler In-reply-to: Your message of "Wed, 15 Apr 1998 21:00:15 +0800." <199804151300.VAA01520@sqz519.ust.hk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 15 Apr 1998 16:33:47 +0100 From: Philip Blundell Message-Id: X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO >libgcc1.a. I'm building with --target=arm-unknown-linuxaout >--host=i586-redhat-linux, and the build stops with "You must find a >way to make libgcc1.a". I've read the gcc info files about building >cross compilers, but that leaves me even more confused. Get gcc 2.8.1. It knows how to build libgcc1.a from the assembler source. I've posted the patch to make 2.7 work to this list a few times and you can probably find it in Matthew's archive. p. unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Wed Apr 15 20:14:54 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id UAA10192 for ; Wed, 15 Apr 1998 20:14:53 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:12860 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390943-16497>; Wed, 15 Apr 1998 22:15:14 +0300 Received: by vger.rutgers.edu id <970958-214>; Wed, 15 Apr 1998 15:01:44 -0400 Received: from snowcrash.cymru.net ([163.164.160.3]:1995 "EHLO snowcrash.cymru.net" ident: "TIMEDOUT") by vger.rutgers.edu with ESMTP id <971039-214>; Wed, 15 Apr 1998 14:41:24 -0400 Received: from caramon.armlinux.org (dialup1.cymru.net [163.164.160.65]) by snowcrash.cymru.net (8.8.7/8.7.1) with ESMTP id TAA21987; Wed, 15 Apr 1998 19:41:31 +0100 Received: from raistlin.armlinux.org (raistlin [192.168.0.3]) by caramon.armlinux.org (8.7.4/8.7.3) with ESMTP id TAA17241; Wed, 15 Apr 1998 19:53:34 +0100 From: Russell King - ARM Linux Admin Received: (from linux@localhost) by raistlin.armlinux.org (8.7.4/8.7.3) id TAA00388; Wed, 15 Apr 1998 19:40:39 +0100 Message-Id: <199804151840.TAA00388@raistlin.armlinux.org> Subject: Re: Setting up cross-compiler + New web pages To: pb@nexus.co.uk (Philip Blundell) Date: Wed, 15 Apr 1998 19:40:38 +0100 (BST) Cc: otfried@cs.ust.hk, linux-arm@vger.rutgers.edu In-Reply-To: from "Philip Blundell" at Apr 15, 98 04:33:47 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Philip Blundell writes: > probably find it in Matthew's archive. There are also other archive sites - have a look on my (new) web pages ;) Also, of interest to developers (and may I recommend others do the same) is a web page of what I'm working on. (Idea from Alan Cox). I know that Phil's going to appreciate it ;) _____ |_____| ------------------------------------------------- ---+---+- | | Russell King rmk@ecs.soton.ac.uk --- --- | | | | http://www.arm.uk.linux.org/~rmk/home.html / / | | +-+-+ --- -+- / | THE developer of ARM Linux |+| /|\ / | | | --- | +-+-+ ------------------------------------------------- /\\\ | unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Wed Apr 15 21:39:06 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id VAA10972 for ; Wed, 15 Apr 1998 21:39:05 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:20645 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391139-16497>; Wed, 15 Apr 1998 23:38:11 +0300 Received: by vger.rutgers.edu id <971024-214>; Wed, 15 Apr 1998 16:27:34 -0400 Received: from snowcrash.cymru.net ([163.164.160.3]:1258 "EHLO snowcrash.cymru.net" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <971043-214>; Wed, 15 Apr 1998 16:25:07 -0400 Received: (from alan@localhost) by snowcrash.cymru.net (8.8.7/8.7.1) id VAA23744; Wed, 15 Apr 1998 21:07:43 +0100 From: Alan Cox Message-Id: <199804152007.VAA23744@snowcrash.cymru.net> Subject: Re: Setting up cross-compiler + New web pages To: linux@arm.uk.linux.org (Russell King - ARM Linux Admin) Date: Wed, 15 Apr 1998 21:07:41 +0100 (BST) Cc: pb@nexus.co.uk, otfried@cs.ust.hk, linux-arm@vger.rutgers.edu In-Reply-To: <199804151840.TAA00388@raistlin.armlinux.org> from "Russell King - ARM Linux Admin" at Apr 15, 98 07:40:38 pm Content-Type: text X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO > > Also, of interest to developers (and may I recommend others do the same) > is a web page of what I'm working on. (Idea from Alan Cox). I know that > Phil's going to appreciate it ;) Is 'Annoying Phil by doing the same job' on the top of it 8) unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Thu Apr 16 08:52:30 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id IAA14513 for ; Thu, 16 Apr 1998 08:52:29 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:7688 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390608-1442>; Thu, 16 Apr 1998 10:52:19 +0300 Received: by vger.rutgers.edu id <970794-214>; Thu, 16 Apr 1998 03:17:11 -0400 Received: from aa137.du.pipex.com ([193.130.240.137]:28819 "EHLO seneca.collegium.co.uk" ident: "TIMEDOUT2") by vger.rutgers.edu with ESMTP id <971328-214>; Thu, 16 Apr 1998 02:26:40 -0400 Received: from carmel.collegium.co.uk (carmel.collegium.co.uk [192.168.0.3]) by seneca.collegium.co.uk (8.8.7/8.8.7) with SMTP id AAA30086 for ; Thu, 16 Apr 1998 00:25:54 +0100 Date: Thu, 16 Apr 1998 00:25:49 +0100 (BST) From: Chris Rutter Subject: EtherH driver broken in latest 2.0.33 patches? To: linux-arm@vger.rutgers.edu In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII X-Mailer: ANT RISCOS Marcel [ver 1.45] X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO On Tue 14 Apr, Chris Rutter wrote: > I've configured a 2.0.33 kernel (RiscPC/StrongARM), but on trying to > build it this occurs: And now it builds. `eth0' detected fine, but no traffic through it at all. Upon reboot (`/sbin/reboot') I get errors about `heartbeat failure' from the EtherH driver under RISC OS. A power-off/on fixes this. The driver that came with the latest pre-built 2.0.31 worked fine. Any ideas? -- Chris Rutter ( http://www.fluff.org/widget/ ) unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Thu Apr 16 09:38:13 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id JAA14677 for ; Thu, 16 Apr 1998 09:38:12 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:28776 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390607-1442>; Thu, 16 Apr 1998 11:37:09 +0300 Received: by vger.rutgers.edu id <970932-214>; Thu, 16 Apr 1998 04:22:46 -0400 Received: from odie.barnet.ac.uk ([194.82.202.98]:31296 "EHLO odie.barnet.ac.uk" ident: "willy") by vger.rutgers.edu with ESMTP id <971114-214>; Thu, 16 Apr 1998 04:06:23 -0400 Received: (from willy@localhost) by odie.barnet.ac.uk (8.8.6/8.8.6) id JAA14560; Thu, 16 Apr 1998 09:10:22 +0100 From: Matthew Wilcox Message-Id: <199804160810.JAA14560@odie.barnet.ac.uk> Subject: Re: EtherH driver broken in latest 2.0.33 patches? To: chris@fluff.org (Chris Rutter) Date: Thu, 16 Apr 1998 09:10:21 +0100 (BST) Cc: linux-arm@vger.rutgers.edu In-Reply-To: from "Chris Rutter" at Apr 16, 98 00:25:49 am X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Chris Rutter > > On Tue 14 Apr, Chris Rutter wrote: > > > I've configured a 2.0.33 kernel (RiscPC/StrongARM), but on trying to > > build it this occurs: > > And now it builds. `eth0' detected fine, but no traffic through it at > all. Upon reboot (`/sbin/reboot') I get errors about `heartbeat failure' > from the EtherH driver under RISC OS. A power-off/on fixes this. The > driver that came with the latest pre-built 2.0.31 worked fine. Any ideas? There are several types of EtherH. Do you have the 500, 600 or 600A? -- Set Alias$Case Set Alias$[ |||| |MSet Alias$Otherwise Set Alias$[ \ Matthew "" |MSet Alias$When If %0=%%0 Then Set Alias$[ "" ||MIf %0=%%0 \ Wilcox Then Set Alias$Otherwise Set Alias$[ |||||||||||||||| ||MIf \ %0=%%0 Then Set Alias$When Set Alias$[ |||||||||||||||| unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From chris@fluff.org Thu Apr 16 10:50:39 1998 Return-Path: Received: from seneca.collegium.co.uk (root@aa156.du.pipex.com [193.130.240.156]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id KAA15041 for ; Thu, 16 Apr 1998 10:50:35 +0100 Received: from carmel.collegium.co.uk (carmel.collegium.co.uk [192.168.0.3]) by seneca.collegium.co.uk (8.8.7/8.8.7) with SMTP id KAA31201; Thu, 16 Apr 1998 10:50:59 +0100 Date: Thu, 16 Apr 1998 10:50:56 +0100 (BST) From: Chris Rutter Subject: Re: EtherH driver broken in latest 2.0.33 patches? To: Matthew Wilcox cc: linux-arm@vger.rutgers.edu In-Reply-To: <199804160810.JAA14560@odie.barnet.ac.uk> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII X-Mailer: ANT RISCOS Marcel [ver 1.45] Status: RO > There are several types of EtherH. Do you have the 500, 600 or 600A? Yes, oops, I forgot to say. An `EtherLan 600A', I believe; an Acorn re- badged NIC slot card, at any rate, with 10b2 and 10bT. -- Chris Rutter ( http://www.fluff.org/widget/ ) From owner-linux-arm-outgoing@vger.rutgers.edu Thu Apr 16 14:08:06 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id OAA17327 for ; Thu, 16 Apr 1998 14:08:05 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:9686 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390703-1442>; Thu, 16 Apr 1998 16:07:53 +0300 Received: by vger.rutgers.edu id <970800-214>; Thu, 16 Apr 1998 08:47:03 -0400 Received: from snowcrash.cymru.net ([163.164.160.3]:1044 "EHLO snowcrash.cymru.net" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <971052-214>; Thu, 16 Apr 1998 08:38:13 -0400 Received: from caramon.armlinux.org (dialup2.cymru.net [163.164.160.66]) by snowcrash.cymru.net (8.8.7/8.7.1) with ESMTP id NAA07804 for ; Thu, 16 Apr 1998 13:43:30 +0100 Received: from raistlin.armlinux.org (raistlin [192.168.0.3]) by caramon.armlinux.org (8.7.4/8.7.3) with ESMTP id NAA19096 for ; Thu, 16 Apr 1998 13:55:41 +0100 From: Russell King - ARM Linux Admin Received: (from linux@localhost) by raistlin.armlinux.org (8.7.4/8.7.3) id NAA01436 for linux-arm@vger.rutgers.edu; Thu, 16 Apr 1998 13:41:37 +0100 Message-Id: <199804161241.NAA01436@raistlin.armlinux.org> Subject: Request for patches... To: linux-arm@vger.rutgers.edu Date: Thu, 16 Apr 1998 13:41:36 +0100 (BST) X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Hi. If you have any useful ARM Linux kernel patches for either 2.0.33 or 2.1.xx, could you please get them to me by Saturday (18th) morning if you wish them to be in the Sunday release please? _____ |_____| ------------------------------------------------- ---+---+- | | Russell King linux@arm.uk.linux.org --- --- | | | | http://www.arm.uk.linux.org/~rmk/armlinux.html / / | | +-+-+ --- -+- / | THE developer of ARM Linux |+| /|\ / | | | --- | +-+-+ ------------------------------------------------- /\\\ | unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Thu Apr 16 14:32:11 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id OAA17490 for ; Thu, 16 Apr 1998 14:32:09 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:9686 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390723-1442>; Thu, 16 Apr 1998 16:31:53 +0300 Received: by vger.rutgers.edu id <970928-214>; Thu, 16 Apr 1998 09:09:57 -0400 Received: from alushta.NL.net ([193.78.240.22]:3660 "EHLO alushta.NL.net" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <971069-214>; Thu, 16 Apr 1998 09:03:24 -0400 Received: from ichfsnl by alushta.NL.net with UUCP id <14729-24507>; Thu, 16 Apr 1998 15:08:26 +0200 Received: from ichsun5.noname (ichsun5 [192.3.2.14]) by ichfsnl.ich.nl (8.6.12/8.6.12) with SMTP id PAA02236 for ; Thu, 16 Apr 1998 15:12:10 +0200 Received: by ichsun5.noname (4.1/SMI-4.1) id AA09109; Thu, 16 Apr 98 15:05:02 +0200 Message-Id: <19980416150502.B9068@ich.nl> Date: Thu, 16 Apr 1998 15:05:02 +0200 From: Hans Bogaards To: ARM Linux Mailing List Subject: Installing on an 4 MB A5000 Reply-To: Hans Bogaards Mail-Followup-To: ARM Linux Mailing List Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.91i X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Hello, I'm trying to install ARM-Linux on my A5000 with 4 MB RAM. I'm using the a5k-kernel version 2.0.31 and the newest RPC-root disk and supplemental disk. I tried to install with giving the kernel the options 'root=/dev/fd0 ramdisk=0'. By doing this the root floppy gets mounted as root-disk and now the RedHat installer starts. But because I'm installing from harddisk the installer asks for the supplemental-disk and then it hangs. As far as I can see, this method could be used when installing from CD-ROM or NFS (the ones that don't need the supplemental-disk), but my question is, is there a way to use a method like this (or maybe something totally different) to install ARM-Linux on a 4 MB machine? -- Tot mails! "You can't help respecting someone Hans who can spell Tuesday, even if he spells it wrong" Winnie the Pooh unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Thu Apr 16 15:14:10 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id PAA17754 for ; Thu, 16 Apr 1998 15:14:09 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:32709 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390707-1442>; Thu, 16 Apr 1998 17:12:57 +0300 Received: by vger.rutgers.edu id <970878-214>; Thu, 16 Apr 1998 10:00:12 -0400 Received: from aa129.du.pipex.com ([193.130.240.129]:30769 "EHLO seneca.collegium.co.uk" ident: "TIMEDOUT2") by vger.rutgers.edu with ESMTP id <970940-214>; Thu, 16 Apr 1998 09:52:00 -0400 Received: from localhost (chris@localhost) by seneca.collegium.co.uk (8.8.7/8.8.7) with SMTP id OAA32704; Thu, 16 Apr 1998 14:55:12 +0100 Date: Thu, 16 Apr 1998 14:55:12 +0100 (BST) From: Chris Rutter To: Russell King - ARM Linux Admin cc: Chris Rutter , linux-arm@vger.rutgers.edu Subject: Re: EtherH driver broken in latest 2.0.33 patches? In-Reply-To: <199804161154.MAA01151@raistlin.armlinux.org> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO On Thu, 16 Apr 1998, Russell King - ARM Linux Admin wrote: > Yep - the driver attempts to do some autodetection (which works if you specify > xcvr=1 on the module command line). However, if compiled into the kernel or > inserted without this option, it doesn't autodetect properly. Ta; rebuilt 8390.o and etherh.o as modules and used `xcvr=1' and it now works fine. -- Chris Rutter ( http://www.fluff.org/widget ) unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Thu Apr 16 15:38:59 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id PAA17975 for ; Thu, 16 Apr 1998 15:38:58 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:18780 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390746-1442>; Thu, 16 Apr 1998 17:38:37 +0300 Received: by vger.rutgers.edu id <970891-214>; Thu, 16 Apr 1998 10:23:15 -0400 Received: from odie.barnet.ac.uk ([194.82.202.98]:2918 "EHLO odie.barnet.ac.uk" ident: "willy") by vger.rutgers.edu with ESMTP id <970868-214>; Thu, 16 Apr 1998 10:13:00 -0400 Received: (from willy@localhost) by odie.barnet.ac.uk (8.8.6/8.8.6) id PAA17782; Thu, 16 Apr 1998 15:17:39 +0100 From: Matthew Wilcox Message-Id: <199804161417.PAA17782@odie.barnet.ac.uk> Subject: Re: Installing on an 4 MB A5000 To: hansbo@stack.nl Date: Thu, 16 Apr 1998 15:17:38 +0100 (BST) Cc: linux-arm@vger.rutgers.edu In-Reply-To: <19980416150502.B9068@ich.nl> from "Hans Bogaards" at Apr 16, 98 03:05:02 pm X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Hans Bogaards > I'm trying to install ARM-Linux on my A5000 with 4 MB RAM. I'm using the > a5k-kernel version 2.0.31 and the newest RPC-root disk and supplemental disk. > > I tried to install with giving the kernel the options > 'root=/dev/fd0 ramdisk=0'. By doing this the root floppy gets mounted as > root-disk and now the RedHat installer starts. But because I'm installing > from harddisk the installer asks for the supplemental-disk and then it hangs. Yes. You've just removed the root filesystem from underneath the kernel. It cannot continue. > As far as I can see, this method could be used when installing from CD-ROM > or NFS (the ones that don't need the supplemental-disk), but my question is, > is there a way to use a method like this (or maybe something totally > different) to install ARM-Linux on a 4 MB machine? Not right now. It could probably be done by building you a special kernel and bootdiscs. If IscaFS ever gains the ability to write to ext2 partitions, then this would allow people to install without a special bootdisc. Let's see what Phil Norman's managed to achieve over Easter ;-) If you're interested in helping, please feel free to download the source to the read-only IscaFS from ftp.barnet.ac.uk/pub/Acorn/armlinux/iscafs/ -- Set Alias$Case Set Alias$[ |||| |MSet Alias$Otherwise Set Alias$[ \ Matthew "" |MSet Alias$When If %0=%%0 Then Set Alias$[ "" ||MIf %0=%%0 \ Wilcox Then Set Alias$Otherwise Set Alias$[ |||||||||||||||| ||MIf \ %0=%%0 Then Set Alias$When Set Alias$[ |||||||||||||||| unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Thu Apr 16 18:11:08 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id SAA18774 for ; Thu, 16 Apr 1998 18:11:07 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:4001 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390827-1442>; Thu, 16 Apr 1998 20:09:52 +0300 Received: by vger.rutgers.edu id <970795-214>; Thu, 16 Apr 1998 13:02:41 -0400 Received: from snowcrash.cymru.net ([163.164.160.3]:1587 "EHLO snowcrash.cymru.net" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <970867-214>; Thu, 16 Apr 1998 13:00:42 -0400 Received: from caramon.armlinux.org (dialup1.cymru.net [163.164.160.65]) by snowcrash.cymru.net (8.8.7/8.7.1) with ESMTP id SAA12981; Thu, 16 Apr 1998 18:06:20 +0100 Received: from raistlin.armlinux.org (raistlin [192.168.0.3]) by caramon.armlinux.org (8.7.4/8.7.3) with ESMTP id RAA19864; Thu, 16 Apr 1998 17:41:36 +0100 From: Russell King - ARM Linux Admin Received: (from linux@localhost) by raistlin.armlinux.org (8.7.4/8.7.3) id RAA01805; Thu, 16 Apr 1998 17:29:02 +0100 Message-Id: <199804161629.RAA01805@raistlin.armlinux.org> Subject: Re: Sound To: chris@collegium.co.uk (Chris Rutter) Date: Thu, 16 Apr 1998 17:29:00 +0100 (BST) Cc: linux@arm.uk.linux.org, linux-arm@vger.rutgers.edu In-Reply-To: from "Chris Rutter" at Apr 16, 98 03:20:41 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Chris Rutter writes: > Okay, next problem. I've enabled all the sound stuff (`y' in the kernel > config), but `echo woo >/dev/audio' just produces `bash: /dev/audio: No such > device'. This is a 2.0.33 kernel with the latest patches applied. What does /proc/devices say? You should see a line: 14 sound If you see that, check the following /dev entries, and their major/minor numbers are correct: crw--w--w- 1 root root 14, 4 Jun 20 1997 /dev/audio crw--w--w- 1 root root 14, 3 Jun 20 1997 /dev/dsp crw--w--w- 1 root root 14, 5 Jun 26 1997 /dev/dsp16 crw--w--w- 1 root root 14, 2 Jun 20 1997 /dev/midi crw--w--w- 1 root root 14, 1 Jun 20 1997 /dev/sequencer If these are incorrect, remake them. Next, check /dev/sndstat. It should look something like: Sound Driver:3.5.4-960630 (Tue Apr 14 22:37:28 BST 1998 src, Linux raistlin 2.0.33 #63 Sat Mar 14 20:58:46 GMT 1998 sa110) Kernel: Linux raistlin 2.0.33 #5 Wed Apr 15 23:30:40 BST 1998 sa110 Config options: 0 Installed drivers: Type 28: ARM VIDC D/A Card config: ARM VIDC D/A irq 20 drq 7,0 Audio devices: 0: VIDCsound Synth devices: 0: VIDCsound Midi devices: NOT ENABLED IN CONFIG Timers: 0: System clock Mixers: 0: VIDCsound *note* mixer and synth support are no where complete! _____ |_____| ------------------------------------------------- ---+---+- | | Russell King linux@arm.uk.linux.org --- --- | | | | http://www.arm.uk.linux.org/~rmk/armlinux.html / / | | +-+-+ --- -+- / | THE developer of ARM Linux |+| /|\ / | | | --- | +-+-+ ------------------------------------------------- /\\\ | unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Thu Apr 16 18:59:48 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id SAA19157 for ; Thu, 16 Apr 1998 18:59:47 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:15804 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390837-1442>; Thu, 16 Apr 1998 20:58:05 +0300 Received: by vger.rutgers.edu id <970904-214>; Thu, 16 Apr 1998 13:45:24 -0400 Received: from violet.csi.cam.ac.uk ([131.111.8.58]:50120 "HELO violet.csi.cam.ac.uk" ident: "exim") by vger.rutgers.edu with SMTP id <970865-214>; Thu, 16 Apr 1998 13:35:30 -0400 Received: from ajb85.acad.cai.cam.ac.uk [131.111.136.66] (root) by violet.csi.cam.ac.uk with smtp (Exim 1.73 #1) id 0yPseq-0002oT-00; Thu, 16 Apr 1998 18:41:08 +0100 Received: from ajb85 by ajb85.acad.cai.cam.ac.uk with smtp (Exim 1.82 #1) id 0yPsem-0000nX-00; Thu, 16 Apr 1998 18:41:04 +0100 Date: Thu, 16 Apr 1998 18:41:03 +0100 (BST) From: Andrew Bower X-Sender: ajb85@ajb85.acad.cai.cam.ac.uk Reply-To: Andrew Bower To: linux-arm@vger.rutgers.edu Subject: fd0* device for ADFS shape discs under i386 Linux Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO I cannot find a device suitable for reading ADFS discs among the standard shapes on my i386 distribution. How can this be done please? I have compiled in ADFS support, but I can't even read the raw data off the floppies. Out of interest, is anyone working on write support for ADFS? Andrew Bower unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Thu Apr 16 20:09:01 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id UAA19457 for ; Thu, 16 Apr 1998 20:08:59 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:18002 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390685-1442>; Thu, 16 Apr 1998 22:09:24 +0300 Received: by vger.rutgers.edu id <970821-214>; Thu, 16 Apr 1998 15:03:10 -0400 Received: from dns1.enterprise.net ([194.72.192.3]:3888 "EHLO dns1.enterprise.net" ident: "root") by vger.rutgers.edu with ESMTP id <970868-214>; Thu, 16 Apr 1998 15:02:35 -0400 Received: from stop.reading.the.headers.numptie (ppp321.enterprise.net [194.72.196.67]) by dns1.enterprise.net (8.8.5/8.8.5) with SMTP id UAA26877; Thu, 16 Apr 1998 20:08:16 +0100 (BST) Date: Thu, 16 Apr 1998 14:04:16 +0100 From: micm To: ramseyg@irlgrace1.grace.cri.nz Cc: linux-arm@vger.rutgers.edu Subject: Re: X server configuration Message-ID: <931fb23748%micm@micm.enterprise.net> In-Reply-To: Reply-To: Mike.M@ed.ac.uk Organization: The Organisation Trained To Yap (TOTTY) X-Mailer: Messenger v1.34c for RISC OS MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Posting-Agent: RISC OS Newsbase 0.60g X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO In message you wrote: > Hi, > > I'm having a little trouble figuring out how to configure the > armlinux X server. > > When starting up the server reports that the screen resolution is > 640x480 and that it can use 2, 4, or 8 bit colour (or something like > that). > > Is it possible to change this screen resolution? If so, where do I > find the configuration options or file? I dunno if the screen res can be changed whilst you are running linux or not, but the way I got round this was to edit the file called !Linux.SetModeNew. Simply edit the X and Y sizes in the mode line of that BASIC file. SetModeOld is (AFAIK) used on pre RiscPC machines. Setting the screen size in this file will affect both the shell and the X screen size. HTH, Mike. -- From Michael Marshall, using an Acorn StrongARM RiscPC @ 202.4MHz Running RiscOS / ARMLinux / DOS & Win95 Mail me at: mailto:micm@enterprise.net Visit the NUMPTIE XONE : http://come.to/numptie "Danger could be my middle name.....but it's James" unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Thu Apr 16 20:19:50 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id UAA19527 for ; Thu, 16 Apr 1998 20:19:49 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:13303 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390820-1442>; Thu, 16 Apr 1998 22:19:38 +0300 Received: by vger.rutgers.edu id <970801-214>; Thu, 16 Apr 1998 15:13:11 -0400 Received: from hillend.cent.gla.ac.uk ([130.209.16.102]:47599 "EHLO hillend.cent.gla.ac.uk" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <970900-214>; Thu, 16 Apr 1998 15:12:07 -0400 Received: from pop-student.gla.ac.uk (actually host mailsrv1.cent.gla.ac.uk) by hillend.cent.gla.ac.uk with SMTP-STU (MMTA) with ESMTP; Thu, 16 Apr 1998 20:16:59 +0100 Received: from MAILSRV1/SpoolDir by pop-student.gla.ac.uk (Mercury 1.40); 16 Apr 98 20:16:59 +0000 Received: from SpoolDir by MAILSRV1 (Mercury 1.40); 16 Apr 98 20:16:41 +0000 Received: from darkside.hacknet.uk (194.93.134.133) by pop-student.gla.ac.uk (Mercury 1.40); 16 Apr 98 20:16:37 +0000 Date: Thu, 16 Apr 1998 20:16:13 +0000 (GMT) From: James Craig X-Sender: root@darkside.hacknet.uk To: Andrew Bower cc: linux-arm@vger.rutgers.edu Subject: Re: fd0* device for ADFS shape discs under i386 Linux In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Orcpt: rfc822;linux-arm@vger.rutgers.edu X-Envid: [/PRMD=UK.AC/ADMD= /C=GB/;I cannot find a device suitable for reading ADFS discs among the >standard shapes on my i386 distribution. How can this be done please? >I have compiled in ADFS support, but I can't even read the raw data off >the floppies. The bad news? You can't right now. The ADFS support right now is basically for hard-drives, although if you can make up a disk image file, it should manage with that just fine too. I've patched up the floppy drivers on past versions of linux (1.2.xx) to manage the trick, but looking over the current floppy drivers they've changed a bit. I haven't got time to fix the current 2.1.x ones to handle ADFS disks myself - exams and the like - but it's certainly doable if someone is feeling brave enough to tackle it. The one problem I had when I was working on 1.2.xx kernels is that the kernel doesn't manage to autodetect the sector size, so I was forced to make either a 512-byte-sector-kernel or a 1024-byte-sector-kernel - if someone does the patches on the current kernel, a more elegant solution would be nice too. hint for anyone brave enough to tackle it (from what I vaguely remember from a few years ago): the FD_SIZECODE stuff in /usr/include/linux/fd.h is the really important bit, but you'll need to watch buffer sizes and stuff too. Have fun all, -- James Craig unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Fri Apr 17 00:11:22 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id AAA21624 for ; Fri, 17 Apr 1998 00:11:20 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:1902 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391118-13857>; Fri, 17 Apr 1998 02:10:48 +0300 Received: by vger.rutgers.edu id <971014-214>; Thu, 16 Apr 1998 18:55:17 -0400 Received: from snowcrash.cymru.net ([163.164.160.3]:1179 "EHLO snowcrash.cymru.net" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <971054-214>; Thu, 16 Apr 1998 18:53:03 -0400 Received: from caramon.armlinux.org (dialup2.cymru.net [163.164.160.66]) by snowcrash.cymru.net (8.8.7/8.7.1) with ESMTP id XAA18850; Thu, 16 Apr 1998 23:58:57 +0100 Received: from raistlin.armlinux.org (raistlin [192.168.0.3]) by caramon.armlinux.org (8.7.4/8.7.3) with ESMTP id XAA20932; Thu, 16 Apr 1998 23:55:20 +0100 From: Russell King Received: (from rmk@localhost) by raistlin.armlinux.org (8.7.4/8.7.3) id XAA05357; Thu, 16 Apr 1998 23:42:12 +0100 Message-Id: <199804162242.XAA05357@raistlin.armlinux.org> Subject: Re: Sound To: chris@collegium.co.uk (Chris Rutter) Date: Thu, 16 Apr 1998 23:42:12 +0100 (BST) Cc: linux@arm.uk.linux.org, linux-arm@vger.rutgers.edu In-Reply-To: from "Chris Rutter" at Apr 16, 98 03:20:41 pm X-Location: london.england.earth.mulky-way.universe Reply-To: rmk@arm.uk.linux.org X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Chris Rutter writes: > Okay, next problem. I've enabled all the sound stuff (`y' in the kernel > config), but `echo woo >/dev/audio' just produces `bash: /dev/audio: No such > device'. This is a 2.0.33 kernel with the latest patches applied. Ok, found & fixed... Missing init call in arch/arm/drivers/char/mem.c... _____ |_____| ------------------------------------------------- ---+---+- | | Russell King rmk@arm.uk.linux.org --- --- | | | | http://www.arm.uk.linux.org/~rmk/home.html / / | | +-+-+ --- -+- / | THE developer of ARM Linux |+| /|\ / | | | --- | +-+-+ ------------------------------------------------- /\\\ | unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Fri Apr 17 07:54:29 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id HAA23708 for ; Fri, 17 Apr 1998 07:54:28 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:28833 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391239-13857>; Fri, 17 Apr 1998 09:54:58 +0300 Received: by vger.rutgers.edu id <971031-29197>; Fri, 17 Apr 1998 02:47:48 -0400 Received: from odie.barnet.ac.uk ([194.82.202.98]:11001 "EHLO odie.barnet.ac.uk" ident: "willy") by vger.rutgers.edu with ESMTP id <971026-29197>; Fri, 17 Apr 1998 02:47:41 -0400 Received: (from willy@localhost) by odie.barnet.ac.uk (8.8.6/8.8.6) id HAA23693 for linux-arm@vger.rutgers.edu; Fri, 17 Apr 1998 07:53:17 +0100 From: Matthew Wilcox Message-Id: <199804170653.HAA23693@odie.barnet.ac.uk> Subject: memfastset.S To: linux-arm@vger.rutgers.edu Date: Fri, 17 Apr 1998 07:53:16 +0100 (BST) X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO I've optimised Russell's code a bit. I've changed rather a lot in fact, but I've stuck to his structure rather than try to work out whether a 64-byte inner loop would be better, or whether it would make sense to sacrifice more registers to make the inner loop smaller. Here's the first version, which I have tested in BBC BASIC's assembler and then rewritten in GAS style, so typos may exist though I checked it three times. It is definitely faster for large blocks of data since I removed an instruction from the inner loop. I suspect it to be faster for small blocks of data as well, but have not tested this hypothesis. /* * linux/arch/arm/lib/memfastset.S * * Copyright (C) 1995, 1996 Russell King * Copyright 1998 Matthew Wilcox */ #include .text @ Prototype: void memsetl (unsigned long *d, unsigned long c, size_t n); .global _memsetl @ If this is quad-word aligned then Lloop_32 is in one cache line _memsetl: stmfd sp!, {lr} mov lr, r1 subs r2, r2, #16 blt Lmod_4 mov r3, r1 mov ip, r1 subs r2, r2, #16 bmi Lmod_16 Lloop_32: stmia r0!, {r1, r3, ip, lr} stmia r0!, {r1, r3, ip, lr} subs r2, r2, #32 bge Lloop_32 @ cmn r2, #32 @ LOADREGS(eqfd, sp!, {pc}) @ at this point, we are up to 28 bytes away from the end of the buffer, @ r2 is n-32 bytes, ie {-4, -8, ..., -28}; *OR* we've finished and r2 @ contains -32 Lmod_16: cmn r2, #16 stmgeia r0!, {r1, r3, ip, lr} LOADREGS(eqfd, sp!, {pc}) addlt r2, r2, #16 @ now we are up to 12 bytes from end of buffer; r2 can be {-4, -8, -12, -16} Lmod_4: adds r2, r2, #8 stmgeia r0!, {r1, lr} @ -8 & -16 are already done. If it's non-zero, then it wasn't -8, if @ it's -16, it won't be taken above 0. addnes r2, r2, #6 strgt r1, [r0] LOADREGS(fd, sp, {pc}) And here's a completely untested optimisation which I came up with on the tram on the way into work. It should improve performance for 16-byte aligned code. /* * linux/arch/arm/lib/memfastset.S * * Copyright (C) 1995, 1996 Russell King * Copyright 1998 Matthew Wilcox */ #include .text @ Prototype: void memsetl (unsigned long *d, unsigned long c, size_t n); .global _memsetl @ If this is quad-word aligned then Lloop_32 is in one cache line _memsetl: stmfd sp!, {lr} mov lr, r1 subs r2, r2, #16 blt Lmod_4 mov r3, r1 mov ip, r1 subs r2, r2, #16 bmi Lmod_16 Lloop_32: stmia r0!, {r1, r3, ip, lr} stmia r0!, {r1, r3, ip, lr} subs r2, r2, #32 bge Lloop_32 @ cmn r2, #32 @ LOADREGS(eqfd, sp!, {pc}) @ at this point, we are up to 28 bytes away from the end of the buffer, @ r2 is n-32 bytes, ie {-4, -8, ..., -28}; *OR* we've finished and r2 @ contains -32 Lmod_16: cmn r2, #16 stmgeia r0!, {r1, r3, ip, lr} addlt r2, r2, #16 @ now we are up to 12 bytes from end of buffer; r2 can be {-4, -8, -12, -16} Lmod_4: cmn r2, #16 LOADREGS(eqfd, sp!, {pc}) adds r2, r2, #8 stmgeia r0!, {r1, lr} strne r1, [r0] LOADREGS(fd, sp, {pc}) -- Set Alias$Case Set Alias$[ |||| |MSet Alias$Otherwise Set Alias$[ \ Matthew "" |MSet Alias$When If %0=%%0 Then Set Alias$[ "" ||MIf %0=%%0 \ Wilcox Then Set Alias$Otherwise Set Alias$[ |||||||||||||||| ||MIf \ %0=%%0 Then Set Alias$When Set Alias$[ |||||||||||||||| unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Fri Apr 17 08:36:50 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id IAA23851 for ; Fri, 17 Apr 1998 08:36:48 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:9709 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391246-13857>; Fri, 17 Apr 1998 10:37:19 +0300 Received: by vger.rutgers.edu id <970807-29197>; Fri, 17 Apr 1998 03:30:18 -0400 Received: from odie.barnet.ac.uk ([194.82.202.98]:11372 "EHLO odie.barnet.ac.uk" ident: "root") by vger.rutgers.edu with ESMTP id <971038-29197>; Fri, 17 Apr 1998 03:29:34 -0400 Received: from netgate.wincoll.ac.uk (chris@wiccamica.demon.co.uk [193.237.12.173]) by odie.barnet.ac.uk (8.8.6/8.8.6) with SMTP id IAA23835; Fri, 17 Apr 1998 08:35:03 +0100 Date: Fri, 17 Apr 1998 08:33:04 +0100 (BST) From: Chris Rutter X-Sender: chris@netgate.wincoll.ac.uk To: Russell King cc: Chris Rutter , linux@arm.uk.linux.org, linux-arm@vger.rutgers.edu Subject: Re: Sound In-Reply-To: <199804162242.XAA05357@raistlin.armlinux.org> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO > Ok, found & fixed... Missing init call in arch/arm/drivers/char/mem.c... Great; ta. :) Will `sound' build as a module and work correctly in the version I've got, though? Could you p'raps supply a diff? :) -- Chris Rutter ( http://www.fluff.org/widget ) unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From linux@arm.uk.linux.org Fri Apr 17 12:23:46 1998 Return-Path: Received: from snowcrash.cymru.net (snowcrash.cymru.net [163.164.160.3]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id MAA24890 for ; Fri, 17 Apr 1998 12:23:44 +0100 Received: from caramon.armlinux.org (dialup2.cymru.net [163.164.160.66]) by snowcrash.cymru.net (8.8.7/8.7.1) with ESMTP id MAA30214 for ; Fri, 17 Apr 1998 12:24:39 +0100 Received: from raistlin.armlinux.org (raistlin [192.168.0.3]) by caramon.armlinux.org (8.7.4/8.7.3) with ESMTP id KAA21450 for ; Fri, 17 Apr 1998 10:07:37 +0100 From: Russell King - ARM Linux Admin Received: (from linux@localhost) by raistlin.armlinux.org (8.7.4/8.7.3) id JAA00401 for willy@odie.barnet.ac.uk; Fri, 17 Apr 1998 09:52:06 +0100 Message-Id: <199804170852.JAA00401@raistlin.armlinux.org> Subject: Re: memfastset.S To: willy@odie.barnet.ac.uk (Matthew Wilcox) Date: Fri, 17 Apr 1998 09:52:05 +0100 (BST) In-Reply-To: <199804170653.HAA23693@odie.barnet.ac.uk> from "Matthew Wilcox" at Apr 17, 98 07:53:16 am X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Status: RO Matthew Wilcox writes: > I've optimised Russell's code a bit. I've changed rather a lot in fact, > but I've stuck to his structure rather than try to work out whether a > 64-byte inner loop would be better, or whether it would make sense to > sacrifice more registers to make the inner loop smaller. Here's the first > version, which I have tested in BBC BASIC's assembler and then rewritten in > GAS style, so typos may exist though I checked it three times. > It is definitely faster for large blocks of data since I removed an > instruction from the inner loop. I suspect it to be faster for small > blocks of data as well, but have not tested this hypothesis. I presume that you did check out where it was used ;) You have to be careful how many registers you use in LDMs/STMs on the A5000. I have found that generally using more than 4 causes floppy FIQ underruns (since a LDM/STM effectively locks out FIQs for (worst case) (n+1)S+1N cycles, plus on top of that you've got the VIDC DMA, sound DMA and DRAM refresh. This is enough to push the FIQ response time over the limit. To prove this, I went to the trouble of using a logic analyser on my A5000! Hence, I think that sticking to 4 registers is a must. It'll give us a bit more speed on the console though ;) The critical one is in string.S - memset, which is used for general purpose memory setting, and memzero for pages and other things needing a quick zero'ing of an aligned piece of memory with 32-byte granularity. _____ |_____| ------------------------------------------------- ---+---+- | | Russell King linux@arm.uk.linux.org --- --- | | | | http://www.arm.uk.linux.org/~rmk/armlinux.html / / | | +-+-+ --- -+- / | THE developer of ARM Linux |+| /|\ / | | | --- | +-+-+ ------------------------------------------------- /\\\ | From linux@arm.uk.linux.org Fri Apr 17 12:23:50 1998 Return-Path: Received: from snowcrash.cymru.net (snowcrash.cymru.net [163.164.160.3]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id MAA24905 for ; Fri, 17 Apr 1998 12:23:47 +0100 Received: from caramon.armlinux.org (dialup2.cymru.net [163.164.160.66]) by snowcrash.cymru.net (8.8.7/8.7.1) with ESMTP id MAA30218; Fri, 17 Apr 1998 12:24:42 +0100 Received: from raistlin.armlinux.org (raistlin [192.168.0.3]) by caramon.armlinux.org (8.7.4/8.7.3) with ESMTP id LAA21694; Fri, 17 Apr 1998 11:45:05 +0100 From: Russell King - ARM Linux Admin Received: (from linux@localhost) by raistlin.armlinux.org (8.7.4/8.7.3) id LAA00648; Fri, 17 Apr 1998 11:29:25 +0100 Message-Id: <199804171029.LAA00648@raistlin.armlinux.org> Subject: Re: memfastset.S To: willy@odie.barnet.ac.uk (Matthew Wilcox) Date: Fri, 17 Apr 1998 11:29:25 +0100 (BST) Cc: linux-arm@vger.rutgers.edu In-Reply-To: <199804170653.HAA23693@odie.barnet.ac.uk> from "Matthew Wilcox" at Apr 17, 98 07:53:16 am X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Status: RO Matthew Wilcox writes: > Lmod_16: > cmn r2, #16 > stmgeia r0!, {r1, r3, ip, lr} > addlt r2, r2, #16 > > @ now we are up to 12 bytes from end of buffer; r2 can be {-4, -8, -12, -16} > Lmod_4: > cmn r2, #16 > > LOADREGS(eqfd, sp!, {pc}) > adds r2, r2, #8 > stmgeia r0!, {r1, lr} > strne r1, [r0] > > LOADREGS(fd, sp, {pc}) Yet another thought: @ now we are up to 12 bytes from end of buffer; r2 can be {-4, -8, -12, -16} Lmod_4: tst r2, #8 stmneia r0!, {r1, lr} tst r2, #4 strne r1, [r0] LOADREGS(fd, sp!, {pc}) That's one more instruction removed ;) _____ |_____| ------------------------------------------------- ---+---+- | | Russell King linux@arm.uk.linux.org --- --- | | | | http://www.arm.uk.linux.org/~rmk/armlinux.html / / | | +-+-+ --- -+- / | THE developer of ARM Linux |+| /|\ / | | | --- | +-+-+ ------------------------------------------------- /\\\ | From linux@arm.uk.linux.org Fri Apr 17 12:23:52 1998 Return-Path: Received: from snowcrash.cymru.net (snowcrash.cymru.net [163.164.160.3]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id MAA24910 for ; Fri, 17 Apr 1998 12:23:50 +0100 Received: from caramon.armlinux.org (dialup2.cymru.net [163.164.160.66]) by snowcrash.cymru.net (8.8.7/8.7.1) with ESMTP id MAA30224; Fri, 17 Apr 1998 12:24:44 +0100 Received: from raistlin.armlinux.org (raistlin [192.168.0.3]) by caramon.armlinux.org (8.7.4/8.7.3) with ESMTP id LAA21606; Fri, 17 Apr 1998 11:10:17 +0100 From: Russell King - ARM Linux Admin Received: (from linux@localhost) by raistlin.armlinux.org (8.7.4/8.7.3) id KAA00561; Fri, 17 Apr 1998 10:54:26 +0100 Message-Id: <199804170954.KAA00561@raistlin.armlinux.org> Subject: Re: memfastset.S To: willy@odie.barnet.ac.uk (Matthew Wilcox) Date: Fri, 17 Apr 1998 10:54:25 +0100 (BST) Cc: linux-arm@vger.rutgers.edu In-Reply-To: <199804170653.HAA23693@odie.barnet.ac.uk> from "Matthew Wilcox" at Apr 17, 98 07:53:16 am X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Status: RO Matthew Wilcox writes: > @ -8 & -16 are already done. If it's non-zero, then it wasn't -8, if > @ it's -16, it won't be taken above 0. > addnes r2, r2, #6 ^^^ is this correct? > strgt r1, [r0] > > LOADREGS(fd, sp, {pc}) was ^^^ sp!, {pc}) is what you meant here? Surely you can make this optimisation as well: > @ at this point, we are up to 28 bytes away from the end of the buffer, > @ r2 is n-32 bytes, ie {-4, -8, ..., -28}; *OR* we've finished and r2 > @ contains -32 > Lmod_16: > cmn r2, #16 > stmgeia r0!, {r1, r3, ip, lr} > addlt r2, r2, #16 > > @ now we are up to 12 bytes from end of buffer; r2 can be {-4, -8, -12, -16} > cmn r2, #16 > LOADREGS(eqfd, sp!, {pc}) > @ now we are up to 12 bytes from end of buffer; r2 can be {-4, -8, -12} > Lmod_4: > adds r2, r2, #8 > stmgeia r0!, {r1, lr} > strne r1, [r0] > > LOADREGS(fd, sp!, {pc}} was ^^^ sp!, {pc}) is what you meant here? _____ |_____| ------------------------------------------------- ---+---+- | | Russell King linux@arm.uk.linux.org --- --- | | | | http://www.arm.uk.linux.org/~rmk/armlinux.html / / | | +-+-+ --- -+- / | THE developer of ARM Linux |+| /|\ / | | | --- | +-+-+ ------------------------------------------------- /\\\ | From owner-linux-arm-outgoing@vger.rutgers.edu Fri Apr 17 13:20:58 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id NAA25131 for ; Fri, 17 Apr 1998 13:20:57 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:7623 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391307-13857>; Fri, 17 Apr 1998 15:21:35 +0300 Received: by vger.rutgers.edu id <971145-29197>; Fri, 17 Apr 1998 08:14:11 -0400 Received: from odie.barnet.ac.uk ([194.82.202.98]:12900 "EHLO odie.barnet.ac.uk" ident: "willy") by vger.rutgers.edu with ESMTP id <971163-29197>; Fri, 17 Apr 1998 08:13:50 -0400 Received: (from willy@localhost) by odie.barnet.ac.uk (8.8.6/8.8.6) id NAA25120; Fri, 17 Apr 1998 13:19:38 +0100 From: Matthew Wilcox Message-Id: <199804171219.NAA25120@odie.barnet.ac.uk> Subject: Re: memfastset.S To: linux@arm.uk.linux.org (Russell King - ARM Linux Admin) Date: Fri, 17 Apr 1998 13:19:38 +0100 (BST) Cc: linux-arm@vger.rutgers.edu In-Reply-To: <199804170954.KAA00561@raistlin.armlinux.org> from "Russell King - ARM Linux Admin" at Apr 17, 98 10:54:25 am X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Russell King - ARM Linux Admin > > Matthew Wilcox writes: > > @ -8 & -16 are already done. If it's non-zero, then it wasn't -8, if > > @ it's -16, it won't be taken above 0. > > addnes r2, r2, #6 > ^^^ is this correct? It is, yes. It makes both -4 and -12 positive, while leaving -16 below zero and -8 on zero. So greater than stores 4 bytes more. Bingo. Remember, that's the tested one ;-) For maximum confuse-a-catness, you could make it 5 or 7 instead. > > strgt r1, [r0] > > > > LOADREGS(fd, sp, {pc}) > was ^^^ sp!, {pc}) is what you meant here? Almost certainly. As i explained, I converted it from a BBC Basic assembler format, so there are typos. I should've supplied the BBC Basic version as well, I guess. Sorry. > Surely you can make this optimisation as well: Depends if you ever call it with zero as a parameter. Your previous code coped with that, so I didn't want to rock the boat. > > @ at this point, we are up to 28 bytes away from the end of the buffer, > > @ r2 is n-32 bytes, ie {-4, -8, ..., -28}; *OR* we've finished and r2 > > @ contains -32 > > Lmod_16: > > cmn r2, #16 > > stmgeia r0!, {r1, r3, ip, lr} > > addlt r2, r2, #16 > > > > @ now we are up to 12 bytes from end of buffer; r2 can be {-4, -8, -12, -16} > > cmn r2, #16 > > LOADREGS(eqfd, sp!, {pc}) > > @ now we are up to 12 bytes from end of buffer; r2 can be {-4, -8, -12} > > Lmod_4: > > adds r2, r2, #8 > > stmgeia r0!, {r1, lr} > > strne r1, [r0] > > > > LOADREGS(fd, sp!, {pc}} > was ^^^ sp!, {pc}) is what you meant here? yep. Sorry. -- Set Alias$Case Set Alias$[ |||| |MSet Alias$Otherwise Set Alias$[ \ Matthew "" |MSet Alias$When If %0=%%0 Then Set Alias$[ "" ||MIf %0=%%0 \ Wilcox Then Set Alias$Otherwise Set Alias$[ |||||||||||||||| ||MIf \ %0=%%0 Then Set Alias$When Set Alias$[ |||||||||||||||| unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Fri Apr 17 13:35:33 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id NAA25214 for ; Fri, 17 Apr 1998 13:35:32 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:32318 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390735-13857>; Fri, 17 Apr 1998 15:36:02 +0300 Received: by vger.rutgers.edu id <971163-29197>; Fri, 17 Apr 1998 08:28:21 -0400 Received: from odie.barnet.ac.uk ([194.82.202.98]:12910 "EHLO odie.barnet.ac.uk" ident: "willy") by vger.rutgers.edu with ESMTP id <971168-29197>; Fri, 17 Apr 1998 08:28:02 -0400 Received: (from willy@localhost) by odie.barnet.ac.uk (8.8.6/8.8.6) id NAA25201; Fri, 17 Apr 1998 13:33:50 +0100 From: Matthew Wilcox Message-Id: <199804171233.NAA25201@odie.barnet.ac.uk> Subject: Re: memfastset.S To: linux@arm.uk.linux.org (Russell King - ARM Linux Admin) Date: Fri, 17 Apr 1998 13:33:49 +0100 (BST) Cc: linux-arm@vger.rutgers.edu In-Reply-To: <199804171029.LAA00648@raistlin.armlinux.org> from "Russell King - ARM Linux Admin" at Apr 17, 98 11:29:25 am X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Russell King - ARM Linux Admin > > Yet another thought: > > @ now we are up to 12 bytes from end of buffer; r2 can be {-4, -8, -12, -16} > Lmod_4: tst r2, #8 > stmneia r0!, {r1, lr} > tst r2, #4 > strne r1, [r0] > LOADREGS(fd, sp!, {pc}) > > That's one more instruction removed ;) Yes, though it's slower for the case where it's quad-word aligned. Mmm, forgot to optimise for the ARM2's 1N+3S properties. Anyone think this is important? ;-) -- Set Alias$Case Set Alias$[ |||| |MSet Alias$Otherwise Set Alias$[ \ Matthew "" |MSet Alias$When If %0=%%0 Then Set Alias$[ "" ||MIf %0=%%0 \ Wilcox Then Set Alias$Otherwise Set Alias$[ |||||||||||||||| ||MIf \ %0=%%0 Then Set Alias$When Set Alias$[ |||||||||||||||| unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Fri Apr 17 13:49:03 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id NAA25285 for ; Fri, 17 Apr 1998 13:49:02 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:7687 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391541-13857>; Fri, 17 Apr 1998 15:48:30 +0300 Received: by vger.rutgers.edu id <971168-29197>; Fri, 17 Apr 1998 08:40:50 -0400 Received: from alushta.NL.net ([193.78.240.22]:2420 "EHLO alushta.NL.net" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <971217-29197>; Fri, 17 Apr 1998 08:40:10 -0400 Received: from ichfsnl by alushta.NL.net with UUCP id <14379-12127>; Fri, 17 Apr 1998 14:46:47 +0200 Received: from ichsun5.noname (ichsun5 [192.3.2.14]) by ichfsnl.ich.nl (8.6.12/8.6.12) with SMTP id OAA08164 for ; Fri, 17 Apr 1998 14:52:08 +0200 Received: by ichsun5.noname (4.1/SMI-4.1) id AA12381; Fri, 17 Apr 98 14:44:46 +0200 Message-Id: <19980417144445.A12345@ich.nl> Date: Fri, 17 Apr 1998 14:44:45 +0200 From: Hans Bogaards To: ARM Linux Mailing List Subject: Re: Installing on an 4 MB A5000 Reply-To: Hans Bogaards Mail-Followup-To: ARM Linux Mailing List References: <19980416150502.B9068@ich.nl> <199804161417.PAA17782@odie.barnet.ac.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.91i In-Reply-To: <199804161417.PAA17782@odie.barnet.ac.uk>; from Matthew Wilcox on Thu, Apr 16, 1998 at 03:17:38PM +0100 X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Hello, On Thu, Apr 16, 1998 at 03:17:38PM +0100, Matthew Wilcox wrote: >> I'm trying to install ARM-Linux on my A5000 with 4 MB RAM. I'm using the >> a5k-kernel version 2.0.31 and the newest RPC-root disk and supplemental >> disk. >> >> I tried to install with giving the kernel the options 'root=/dev/fd0 >> ramdisk=0'. By doing this the root floppy gets mounted as root-disk and now >> the RedHat installer starts. But because I'm installing from harddisk the >> installer asks for the supplemental-disk and then it hangs. > > Yes. You've just removed the root filesystem from underneath the kernel. > It cannot continue. I realised that, but I was wondering would it work with two diskdrives? One for the root-disk and one for the supplemental disk Further, is it possible to (temporarily) install a HD diskdrive from a PC in my A5000 with a standard PC-diskdrive cable or do I need special Acorn stuff for this? > Not right now. It could probably be done by building you a special > kernel and bootdiscs. If IscaFS ever gains the ability to write to ext2 > partitions, then this would allow people to install without a special > bootdisc. Let's see what Phil Norman's managed to achieve over Easter ;-) > If you're interested in helping, please feel free to download the source > to the read-only IscaFS from ftp.barnet.ac.uk/pub/Acorn/armlinux/iscafs/ I would like to help, but my knowledge of Linux and Risc OS isn't sufficient, I fear. But good luck with developing it further! I'll follow the progress closely! -- Tot mails! "You can't help respecting someone Hans who can spell Tuesday, even if he spells it wrong" Winnie the Pooh unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Fri Apr 17 14:26:04 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id OAA25462 for ; Fri, 17 Apr 1998 14:26:03 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:4196 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391203-13857>; Fri, 17 Apr 1998 16:24:34 +0300 Received: by vger.rutgers.edu id <971195-29197>; Fri, 17 Apr 1998 09:13:32 -0400 Received: from odie.barnet.ac.uk ([194.82.202.98]:13353 "EHLO odie.barnet.ac.uk" ident: "willy") by vger.rutgers.edu with ESMTP id <971194-29197>; Fri, 17 Apr 1998 09:11:45 -0400 Received: (from willy@localhost) by odie.barnet.ac.uk (8.8.6/8.8.6) id OAA25406; Fri, 17 Apr 1998 14:17:40 +0100 From: Matthew Wilcox Message-Id: <199804171317.OAA25406@odie.barnet.ac.uk> Subject: Re: Installing on an 4 MB A5000 To: hansbo@stack.nl Date: Fri, 17 Apr 1998 14:17:39 +0100 (BST) Cc: linux-arm@vger.rutgers.edu In-Reply-To: <19980417144445.A12345@ich.nl> from "Hans Bogaards" at Apr 17, 98 02:44:45 pm X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Hans Bogaards > I realised that, but I was wondering would it work with two diskdrives? One > for the root-disk and one for the supplemental disk It /should/ work. It will probably require a slight change to the rootdisc, which I'm willing to do for you. Actually, it might give you the opportunity to put the disc in a different drive, it's a while since I ran the installer. > Further, is it possible to (temporarily) install a HD diskdrive from a PC > in my A5000 with a standard PC-diskdrive cable or do I need special Acorn > stuff for this? Should work straight off. I never did - I had a second hard disc in place of the 2nd floppy disc drive. > I would like to help, but my knowledge of Linux and Risc OS isn't > sufficient, I fear. But good luck with developing it further! I'll follow > the progress closely! Ah well, always worth blagging someone into helping ;-) Linux knowledge isn't really required to help with IscaFS, but understanding of Image filesystems is. -- Set Alias$Case Set Alias$[ |||| |MSet Alias$Otherwise Set Alias$[ \ Matthew "" |MSet Alias$When If %0=%%0 Then Set Alias$[ "" ||MIf %0=%%0 \ Wilcox Then Set Alias$Otherwise Set Alias$[ |||||||||||||||| ||MIf \ %0=%%0 Then Set Alias$When Set Alias$[ |||||||||||||||| unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Fri Apr 17 14:53:55 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id OAA25623 for ; Fri, 17 Apr 1998 14:53:53 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:8299 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391529-13857>; Fri, 17 Apr 1998 16:54:46 +0300 Received: by vger.rutgers.edu id <971211-29197>; Fri, 17 Apr 1998 09:47:05 -0400 Received: from nexusel.demon.co.uk ([158.152.30.195]:26243 "HELO globe" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with SMTP id <971194-29197>; Fri, 17 Apr 1998 09:46:45 -0400 Received: from (spring.nexus.co.uk) [192.0.0.3] (root) by globe with smtp (Exim 1.82 #1) id 0yQBFo-0004a5-00; Fri, 17 Apr 1998 14:32:32 +0100 Received: from localhost (spring.nexus.co.uk) [127.0.0.1] (pb) by spring.nexus.co.uk with esmtp (Exim 1.82 #1) id 0yQBFn-0001YJ-00; Fri, 17 Apr 1998 14:32:31 +0100 X-Mailer: exmh version 2.0zeta 7/24/97 To: Hans Bogaards cc: ARM Linux Mailing List Subject: Re: Installing on an 4 MB A5000 In-reply-to: Your message of "Fri, 17 Apr 1998 14:44:45 +0200." <19980417144445.A12345@ich.nl> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 17 Apr 1998 14:32:31 +0100 From: Philip Blundell Message-Id: X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO >Further, is it possible to (temporarily) install a HD diskdrive from a PC >in my A5000 with a standard PC-diskdrive cable or do I need special Acorn >stuff for this? It should work. The A5000 is pretty good in that respect. p. unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Fri Apr 17 21:47:39 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id VAA31703 for ; Fri, 17 Apr 1998 21:47:38 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:19554 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390666-13857>; Fri, 17 Apr 1998 23:48:01 +0300 Received: by vger.rutgers.edu id <971318-29197>; Fri, 17 Apr 1998 16:10:41 -0400 Received: from odie.barnet.ac.uk ([194.82.202.98]:20522 "EHLO odie.barnet.ac.uk" ident: "willy") by vger.rutgers.edu with ESMTP id <971397-29197>; Fri, 17 Apr 1998 16:04:40 -0400 Received: (from willy@localhost) by odie.barnet.ac.uk (8.8.6/8.8.6) id VAA31519; Fri, 17 Apr 1998 21:11:03 +0100 From: Matthew Wilcox Message-Id: <199804172011.VAA31519@odie.barnet.ac.uk> Subject: Re: fd0* device for ADFS shape discs under i386 Linux To: ajb85@cam.ac.uk Date: Fri, 17 Apr 1998 21:11:02 +0100 (BST) Cc: linux-arm@vger.rutgers.edu In-Reply-To: from "Andrew Bower" at Apr 16, 98 06:41:03 pm X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Andrew Bower > Out of interest, is anyone working on write support for ADFS? I've looked at the current code. It would be hard to add extendable files to it, though in principle, files which are newly created and stay the same length should be able to work quite easily. The trouble is that ADFS (or more truthfully filecore) does not have the concept of the inode. Inodes are what Unix filesystems are built of. They have the following properties: Each file has an inode The inode remains the same during the life of the file Inodes do not map to more than one file. The inode contains all of the metadata for the file except its name. I cannot think of a good way of faking inodes for ADFS filesystems *except* the following rather horrible idea: Create a structure in memory for each ADFS filesystem mounted. This can be done lazily. Assign each file an inode based on some scheme which will allow for easy mapping between it and the name of the ADFS file. Perhaps hash buckets. Note that it's okay to reuse inodes once a file has been deleted. Someone should look at other filing systems to see how they cope with filesystems that don't provide nice inodes. When Dickon Hood & I were experimenting with NFS servers under RISC OS, we had a structure that worked well; it was a tree, and the inode we returned was the address within that list. You've got me intrigued now, I might just do it. The way Russell has done it is faster than this method for read-only but will not extend to writable files. -- Set Alias$Case Set Alias$[ |||| |MSet Alias$Otherwise Set Alias$[ \ Matthew "" |MSet Alias$When If %0=%%0 Then Set Alias$[ "" ||MIf %0=%%0 \ Wilcox Then Set Alias$Otherwise Set Alias$[ |||||||||||||||| ||MIf \ %0=%%0 Then Set Alias$When Set Alias$[ |||||||||||||||| unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From alan@snowcrash.cymru.net Fri Apr 17 21:57:10 1998 Return-Path: Received: from snowcrash.cymru.net (snowcrash.cymru.net [163.164.160.3]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id VAA31754 for ; Fri, 17 Apr 1998 21:57:09 +0100 Received: (from alan@localhost) by snowcrash.cymru.net (8.8.7/8.7.1) id VAA08472; Fri, 17 Apr 1998 21:58:08 +0100 From: Alan Cox Message-Id: <199804172058.VAA08472@snowcrash.cymru.net> Subject: Re: fd0* device for ADFS shape discs under i386 Linux To: willy@odie.barnet.ac.uk (Matthew Wilcox) Date: Fri, 17 Apr 1998 21:58:05 +0100 (BST) Cc: ajb85@cam.ac.uk, linux-arm@vger.rutgers.edu In-Reply-To: <199804172011.VAA31519@odie.barnet.ac.uk> from "Matthew Wilcox" at Apr 17, 98 09:11:02 pm Content-Type: text Status: RO > Someone should look at other filing systems to see how they cope with > filesystems that don't provide nice inodes. When Dickon Hood & I were > experimenting with NFS servers under RISC OS, we had a structure that > worked well; it was a tree, and the inode we returned was the address > within that list. You've got me intrigued now, I might just do it. Several schemes are used. One of the cleanest if you stick to 2.1.8x+ is to issue inode numbers as you put things in the dcache and revoke them as your inode is flushed. All open file inodes are in the dcache. Alan From vincent@waw.com Fri Apr 17 22:00:17 1998 Return-Path: Received: from odin.waw.com (vincent@odin.waw.com [194.51.88.249]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id WAA31772 for ; Fri, 17 Apr 1998 22:00:05 +0100 Received: from localhost (vincent@localhost) by odin.waw.com (8.8.7/8.8.7/waw) with SMTP id XAA22470; Fri, 17 Apr 1998 23:01:37 +0100 Date: Fri, 17 Apr 1998 23:01:36 +0100 (GMT+0100) From: Vincent Renardias To: Matthew Wilcox cc: linux-arm@vger.rutgers.edu Subject: Re: fd0* device for ADFS shape discs under i386 Linux In-Reply-To: <199804172011.VAA31519@odie.barnet.ac.uk> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Status: RO On Fri, 17 Apr 1998, Matthew Wilcox wrote: > > Out of interest, is anyone working on write support for ADFS? [snip] > The trouble is that ADFS (or more truthfully filecore) does not have > the concept of the inode. Inodes are what Unix filesystems are built of. > They have the following properties: > > Each file has an inode > The inode remains the same during the life of the file > Inodes do not map to more than one file. > The inode contains all of the metadata for the file except its name. > > I cannot think of a good way of faking inodes for ADFS filesystems *except* > the following rather horrible idea: [snip] MSDOS' FAT system doesn't have i-nodes either, but they can be mounted r/w anyway, so you may want to look at the kludge used there. (or in the other non I-node based FSes recognized by Linux) Cordialement, -- - Vincent RENARDIAS vincent@{waw.com,pipo.com,debian.org} - - Debian/GNU Linux: Pipo: WAW: - - http://www.fr.debian.org http://www.pipo.com http://www.waw.com - --------------------------------------------------------------------------- - "La fonctionnalite Son Visuel vous delivre des avertissements visuels." - - [Message durant l'installation de Windows95]:wq From alan@snowcrash.cymru.net Sat Apr 18 01:11:40 1998 Return-Path: Received: from snowcrash.cymru.net (snowcrash.cymru.net [163.164.160.3]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id BAA00537 for ; Sat, 18 Apr 1998 01:11:40 +0100 Received: (from alan@localhost) by snowcrash.cymru.net (8.8.7/8.7.1) id BAA11015; Sat, 18 Apr 1998 01:12:24 +0100 From: Alan Cox Message-Id: <199804180012.BAA11015@snowcrash.cymru.net> Subject: Re: fd0* device for ADFS shape discs under i386 Linux To: vincent@waw.com (Vincent Renardias) Date: Sat, 18 Apr 1998 01:12:22 +0100 (BST) Cc: willy@odie.barnet.ac.uk, linux-arm@vger.rutgers.edu In-Reply-To: from "Vincent Renardias" at Apr 17, 98 11:01:36 pm Content-Type: text Status: RO > other non I-node based FSes recognized by Linux) A lot of kludging is done using block numbers, which works fine until you hit an FS where blocks get moved around. smbfs implements a inodes on a fs that has only path names stable From root@mad.scientist.com Sat Apr 18 08:24:12 1998 Return-Path: Received: from hillend.cent.gla.ac.uk ([130.209.16.102] (may be forged)) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id IAA01916 for ; Sat, 18 Apr 1998 08:24:11 +0100 Received: from pop-student.gla.ac.uk (actually host mailsrv1.cent.gla.ac.uk) by hillend.cent.gla.ac.uk with SMTP-STU (MMTA) with ESMTP; Sat, 18 Apr 1998 08:25:17 +0100 Received: from MAILSRV1/SpoolDir by pop-student.gla.ac.uk (Mercury 1.40); 18 Apr 98 08:25:16 +0000 Received: from SpoolDir by MAILSRV1 (Mercury 1.40); 18 Apr 98 08:25:13 +0000 Received: from darkside.hacknet.uk (194.93.134.137) by pop-student.gla.ac.uk (Mercury 1.40); 18 Apr 98 08:25:06 +0000 Date: Sat, 18 Apr 1998 08:24:41 +0000 (GMT) From: James Craig X-Sender: root@darkside.hacknet.uk To: Matthew Wilcox cc: ajb85@cam.ac.uk, linux-arm@vger.rutgers.edu Subject: Re: fd0* device for ADFS shape discs under i386 Linux In-Reply-To: <199804172011.VAA31519@odie.barnet.ac.uk> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Status: RO On Fri, 17 Apr 1998, Matthew Wilcox wrote: >Andrew Bower >> Out of interest, is anyone working on write support for ADFS? > >I've looked at the current code. It would be hard to add extendable >files to it, though in principle, files which are newly created and stay >the same length should be able to work quite easily. > >The trouble is that ADFS (or more truthfully filecore) does not have >the concept of the inode. Inodes are what Unix filesystems are built of. >They have the following properties: > >Each file has an inode >The inode remains the same during the life of the file >Inodes do not map to more than one file. >The inode contains all of the metadata for the file except its name. umm. ADFS doesn't have inodes in the same way UNIX does, but couldn't you get away with the file's ID that appears in the free space map? (He said, trying to talk knowledgably, despite the fact that it's 5 years since he last wrote an ADFS-E-format-filing-system-reader). The IDs still only map to one file, aren't likely to change over it's lifetime, and the only problem you have is that the length info is kept in the directory info. >I cannot think of a good way of faking inodes for ADFS filesystems *except* >the following rather horrible idea: > >Create a structure in memory for each ADFS filesystem mounted. This can >be done lazily. Assign each file an inode based on some scheme which will >allow for easy mapping between it and the name of the ADFS file. Perhaps >hash buckets. Note that it's okay to reuse inodes once a file has been >deleted. Ur, YUK!! That would be horrible. I mean, mapping names to inodes is gonna break as soon as you move a mile about the FS somewhere. >Someone should look at other filing systems to see how they cope with >filesystems that don't provide nice inodes. When Dickon Hood & I were >experimenting with NFS servers under RISC OS, we had a structure that >worked well; it was a tree, and the inode we returned was the address >within that list. You've got me intrigued now, I might just do it. Good man, writable ADFS would be nice. :) >The way Russell has done it is faster than this method for read-only >but will not extend to writable files. I haven't looked at it yet, but what's he done that stops you writing files the same way? >-- >Set Alias$Case Set Alias$[ |||| |MSet Alias$Otherwise Set Alias$[ \ Matthew >"" |MSet Alias$When If %0=%%0 Then Set Alias$[ "" ||MIf %0=%%0 \ Wilcox >Then Set Alias$Otherwise Set Alias$[ |||||||||||||||| ||MIf \ >%0=%%0 Then Set Alias$When Set Alias$[ |||||||||||||||| >unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu > From owner-linux-arm-outgoing@vger.rutgers.edu Sat Apr 18 08:26:19 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id IAA01929 for ; Sat, 18 Apr 1998 08:26:18 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:31173 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390659-29480>; Sat, 18 Apr 1998 10:25:44 +0300 Received: by vger.rutgers.edu id <970959-319>; Sat, 18 Apr 1998 03:18:51 -0400 Received: from hillend.cent.gla.ac.uk ([130.209.16.102]:37770 "EHLO hillend.cent.gla.ac.uk" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <970954-319>; Sat, 18 Apr 1998 03:16:23 -0400 Received: from pop-student.gla.ac.uk (actually host mailsrv1.cent.gla.ac.uk) by hillend.cent.gla.ac.uk with SMTP-STU (MMTA) with ESMTP; Sat, 18 Apr 1998 08:15:22 +0100 Received: from MAILSRV1/SpoolDir by pop-student.gla.ac.uk (Mercury 1.40); 18 Apr 98 08:15:21 +0000 Received: from SpoolDir by MAILSRV1 (Mercury 1.40); 18 Apr 98 08:14:51 +0000 Received: from darkside.hacknet.uk (194.93.134.137) by pop-student.gla.ac.uk (Mercury 1.40); 18 Apr 98 08:14:49 +0000 Date: Sat, 18 Apr 1998 08:14:25 +0000 (GMT) From: James Craig X-Sender: root@darkside.hacknet.uk To: Philip Blundell cc: Hans Bogaards , ARM Linux Mailing List Subject: Re: Installing on an 4 MB A5000 In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Orcpt: rfc822;linux-arm@vger.rutgers.edu X-Envid: [/PRMD=UK.AC/ADMD= /C=GB/;>Further, is it possible to (temporarily) install a HD diskdrive from a PC >>in my A5000 with a standard PC-diskdrive cable or do I need special Acorn >>stuff for this? > >It should work. The A5000 is pretty good in that respect. Yep, should be fine on an A5000. Don't try it on an A300/400/3000 tho, it usually doesn't work well at all, and technically you need a buffer-board when doing it on those machines. Apparently the Risc PCs, A3010s, A3020s, and A4000s don't have the drive select lines wired up either, so you're stuck with one drive on them. BTW - There's some comments in the 1772 FDC driver for ARM linux that suggests it supports two drives - but the hardware in old archies can actually support four drives, which I've actually got on one of my machines. Dave, how's the multiple drive stuff working this weather? ;) >p. ^^^^^ Shortest sig block since the dawn of time. :) -- James Craig unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Sat Apr 18 08:40:22 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id IAA01979 for ; Sat, 18 Apr 1998 08:40:21 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:3270 "EHLO vger.rutgers.edu" ident: "TIMEDOUT") by orava.funet.fi with ESMTP id <390690-29480>; Sat, 18 Apr 1998 10:39:29 +0300 Received: by vger.rutgers.edu id <970963-319>; Sat, 18 Apr 1998 03:27:14 -0400 Received: from hillend.cent.gla.ac.uk ([130.209.16.102]:37805 "EHLO hillend.cent.gla.ac.uk" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <970972-319>; Sat, 18 Apr 1998 03:20:46 -0400 Received: from pop-student.gla.ac.uk (actually host mailsrv1.cent.gla.ac.uk) by hillend.cent.gla.ac.uk with SMTP-STU (MMTA) with ESMTP; Sat, 18 Apr 1998 08:19:49 +0100 Received: from MAILSRV1/SpoolDir by pop-student.gla.ac.uk (Mercury 1.40); 18 Apr 98 08:19:49 +0000 Received: from SpoolDir by MAILSRV1 (Mercury 1.40); 18 Apr 98 08:19:27 +0000 Received: from darkside.hacknet.uk (194.93.134.137) by pop-student.gla.ac.uk (Mercury 1.40); 18 Apr 98 08:19:25 +0000 Date: Sat, 18 Apr 1998 08:18:58 +0000 (GMT) From: James Craig X-Sender: root@darkside.hacknet.uk To: linux@treblig.org cc: linux-arm@vger.rutgers.edu Subject: Re: Installing on an 4 MB A5000 In-Reply-To: <35375D6F.852E7F5B@cogency.co.uk> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Orcpt: rfc822;linux-arm@vger.rutgers.edu X-Envid: [/PRMD=UK.AC/ADMD= /C=GB/;Matthew Wilcox wrote: >> >> Hans Bogaards >> > I realised that, but I was wondering would it work with two diskdrives? One >> > for the root-disk and one for the supplemental disk >> >> It /should/ work. It will probably require a slight change to the >> rootdisc, which I'm willing to do for you. Actually, it might give you >> the opportunity to put the disc in a different drive, it's a while since >> I ran the installer. > >It might work; but it won't let you install - 4MB of RAM just isn't enough >for the standard RedHat installer to do its stuff. It can't even turn swapping >on before it runs out. > >At the moment there is no easy method of installing on a 4MB machine. >Non-easy methods include installing over the network using a kernel which >has NFS-root and putting a mkswap and swapon at the start of the /etc/rc's >and installing using a PC onto a PC hard drive and transferring the drive; >you'll have to do the same swap trick. >Neither of these are for the faint hearted; but it is possible - just. Here's a serious thought. The Slackware installer works in 4M on the ix86 architecture - anyone feel like recompiling Slack for the ARM? It should just be a case of setting up the cross-compiler as the default compiler, and then doing a SlackBuild command in the root of the slack source tree. The other thing is that you can install slackware packages without using the installer, if you really need, since they are just .tar.gz's with a few extra files included. Much nicer than that horrible RPM thingy. :) >> > Further, is it possible to (temporarily) install a HD diskdrive from a PC >> > in my A5000 with a standard PC-diskdrive cable or do I need special Acorn >> > stuff for this? >> >> Should work straight off. I never did - I had a second hard disc in place >> of the 2nd floppy disc drive. > >Be careful - the Acorn ADFS is pretty hopeless compatibility wise; I've >seen it get pretty confused with some harddrives - to the point where RiscOS >won't come up. I think its the luck of the draw as to whether a particular >manufacturers drive will work or not. Um, IDE, ST506, or SCSI? All the ST506s I've tried have been fine, and I don't remember seeing an IDE drive break except as a 2nd drive (some IDE HDDs just /do not/ work together. Not nice, but it's hit me 3 times so far) - so is it an ADFS thingy (in which case it shouldn't affect linux once you've got it booted up) or is it a dodgy-HDDs-not-working-together-job? >Dave > >-- >---------------------------------------------------------------------- >- Dr. David Alan Gilbert - WARNING! This is a beta release .signature- >- Work: dg @ cogency.co.uk - +44-(0)161-428-9444 - >- Home: gro.gilbert @ treblig.org - - >---------------------------------------------------------------------- >unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu > -- James Craig unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From linux@arm.uk.linux.org Sat Apr 18 09:03:09 1998 Return-Path: Received: from snowcrash.cymru.net (snowcrash.cymru.net [163.164.160.3]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id JAA02042 for ; Sat, 18 Apr 1998 09:03:08 +0100 Received: from caramon.armlinux.org (dialup1.cymru.net [163.164.160.65]) by snowcrash.cymru.net (8.8.7/8.7.1) with ESMTP id JAA16998; Sat, 18 Apr 1998 09:03:22 +0100 Received: from raistlin.armlinux.org (raistlin [192.168.0.3]) by caramon.armlinux.org (8.7.4/8.7.3) with ESMTP id JAA23389; Sat, 18 Apr 1998 09:21:01 +0100 From: Russell King - ARM Linux Admin Received: (from linux@localhost) by raistlin.armlinux.org (8.7.4/8.7.3) id JAA00406; Sat, 18 Apr 1998 09:03:25 +0100 Message-Id: <199804180803.JAA00406@raistlin.armlinux.org> Subject: ADFS format (was Re: fd0* device for ADFS shape discs under i386 Linux) To: alan@cymru.net (Alan Cox) Date: Sat, 18 Apr 1998 09:03:23 +0100 (BST) Cc: vincent@waw.com, willy@odie.barnet.ac.uk, linux-arm@vger.rutgers.edu In-Reply-To: <199804180012.BAA11015@snowcrash.cymru.net> from "Alan Cox" at Apr 18, 98 01:12:22 am X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Status: RO Alan Cox writes: > A lot of kludging is done using block numbers, which works fine until you > hit an FS where blocks get moved around. smbfs implements a inodes on a > fs that has only path names stable Ok, I think what I'll do at this point is supply a brief overview of the structure of ADFS/filecore filesystem, such that everyone knows the structure of the filesystem we're talking about. ---------------------------------------------------------------------------- Filecore, like DOS has two distinct structures, the directory and effectively a map indicating allocation on the disk. Under both, the directory contains all information to do with a file, and the map tells you where to find it. This is, however, where the two diverge. Filecore directories -------------------- Under filecore, the directory contains: * Name * Load address \ These days, they're combined into a * Execution address / filetype/datestamp. * Length * Attributes * File ID number Directory entries are alphanumerically sorted and can have no gaps (ie, you delete a file, and all the other entries get shifted up one position. The ID number is the `key' to finding where the file is stored. The ID number on it's own does not indicate where the file is. You need the map to locate the file. ID numbers refer to disk objects (fragment id) and offsets within a disk object (sector offset). Disk objects ------------ A disk object is similar to a file, but has a minimum size. Several files (normally with a directory) can be stored in a disk object, but they don't have to be. The map ------- The map is basically a bitstream. The map is partitioned into sector sized zones. Each map bit, relative to the first map bit indicates a set of sectors. (this is defined in the boot sector, in the log2bpmb/log2_bits_per_map_bit). A map entry consists of: +-------------+-------------+---+ | fragment id | zero bits | 1 | +-------------+-------------+---+ The fragment ID not only defines which entries to find, but which zone to start searching the map. The zero bits define the size of this fragment. Free blocks are indicated by a linked list of these blocks. In this case, the `fragment id' is the offset to the next free block in this zone, or zero if this is the last. ---------------------------------------------------------------------------- This allows filecore to automatically defrag the disk on the fly, since it doesn't need to find the directory associated with the file - it just changes the map. The reason that the Linux adfs isn't able to be writable is indeed the way that it allocates inode numbers. Inode numbers are currently defined to be the directories fragment ID << 8 and the entry offset within that directory. ---------------------------------------------------------------------------- There you have it... Hope it's useful. ;) _____ |_____| ------------------------------------------------- ---+---+- | | Russell King linux@arm.uk.linux.org --- --- | | | | http://www.arm.uk.linux.org/~rmk/armlinux.html / / | | +-+-+ --- -+- / | THE developer of ARM Linux |+| /|\ / | | | --- | +-+-+ ------------------------------------------------- /\\\ | From owner-linux-arm-outgoing@vger.rutgers.edu Sat Apr 18 12:19:16 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id MAA02563 for ; Sat, 18 Apr 1998 12:19:15 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:15041 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391092-29480>; Sat, 18 Apr 1998 14:17:42 +0300 Received: by vger.rutgers.edu id <971090-319>; Sat, 18 Apr 1998 07:18:04 -0400 Received: from chaos.oregan.net ([195.173.86.1]:20631 "EHLO chaos.oregan.net" ident: "root") by vger.rutgers.edu with ESMTP id <971046-319>; Sat, 18 Apr 1998 07:13:26 -0400 Received: from splodge.oregan.net (splodge.oregan.net [195.173.86.42]) by chaos.oregan.net (8.8.8/8.8.8) with SMTP id MAA04860 for ; Sat, 18 Apr 1998 12:12:37 +0100 Date: Sat, 18 Apr 1998 12:05:57 +0100 From: Dickon Hood To: linux-arm@vger.rutgers.edu Subject: Re: Installing on an 4 MB A5000 Message-ID: <81f6ae3848%dickon-ml@splurge.fluff.org> In-Reply-To: Organization: The Fluff Disorganisation X-Mailer: Messenger v1.34c for RISC OS X-Editor: Zap, using ZapEmail =?ISO-8859-1?Q?0.21=DF?= MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Posting-Agent: RISC OS Newsbase 0.60h X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO In message you wrote: : On Fri, 17 Apr 1998, David Alan Gilbert wrote: : Here's a serious thought. The Slackware installer works in 4M on the ix86 : architecture - anyone feel like recompiling Slack for the ARM? Would be nice... IMHO, RH is too SysVile. : >Be careful - the Acorn ADFS is pretty hopeless compatibility wise; I've : >seen it get pretty confused with some harddrives - to the point where : >RiscOS won't come up. I think its the luck of the draw as to whether a : >particular manufacturers drive will work or not. : Um, IDE, ST506, or SCSI? : All the ST506s I've tried have been fine, and I don't remember seeing an : IDE drive break except as a 2nd drive (some IDE HDDs just /do not/ work : together. Not nice, but it's hit me 3 times so far) - so is it an ADFS : thingy (in which case it shouldn't affect linux once you've got it booted : up) or is it a dodgy-HDDs-not-working-together-job? IIRC, from talking to Simes, it is an ADFS problem (not specifically Filecore) which was fixed with the 'Large filecore' released in Risc OS 3.6 (not to be confused with the 'Large filecore' to be released with RPCII). I bought a drive which wouldn't work under the older one; AFAICR we bodged around it by using a pull-[up|down] resistor across two pins. Very dodgy; don't try this at home ;-) ObLinux: Linux should cope with such drives, assuming I've got that right. -- Dickon Hood I've now managed to find my .signature file, normal service will be resumed when Connex South Central get their act together. We apologise for the inconvenience in the mean time. unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Sat Apr 18 21:42:28 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id VAA04430 for ; Sat, 18 Apr 1998 21:42:27 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:28679 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390935-29480>; Sat, 18 Apr 1998 23:43:18 +0300 Received: by vger.rutgers.edu id <971191-319>; Sat, 18 Apr 1998 16:43:16 -0400 Received: from snowcrash.cymru.net ([163.164.160.3]:1254 "EHLO snowcrash.cymru.net" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <971048-319>; Sat, 18 Apr 1998 16:43:07 -0400 Received: from caramon.armlinux.org (dialup1.cymru.net [163.164.160.65]) by snowcrash.cymru.net (8.8.7/8.7.1) with ESMTP id VAA26665; Sat, 18 Apr 1998 21:41:55 +0100 Received: from raistlin.armlinux.org (raistlin [192.168.0.3]) by caramon.armlinux.org (8.7.4/8.7.3) with ESMTP id JAA23450; Sat, 18 Apr 1998 09:32:05 +0100 From: Russell King - ARM Linux Admin Received: (from linux@localhost) by raistlin.armlinux.org (8.7.4/8.7.3) id JAA00438; Sat, 18 Apr 1998 09:14:28 +0100 Message-Id: <199804180814.JAA00438@raistlin.armlinux.org> Subject: Re: Installing on an 4 MB A5000 To: root@mad.scientist.com (James Craig) Date: Sat, 18 Apr 1998 09:14:28 +0100 (BST) Cc: linux@treblig.org, linux-arm@vger.rutgers.edu In-Reply-To: from "James Craig" at Apr 18, 98 08:18:58 am X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO James Craig writes: > Here's a serious thought. > The Slackware installer works in 4M on the ix86 architecture - anyone feel > like recompiling Slack for the ARM? It should just be a case of setting up > the cross-compiler as the default compiler, and then doing a SlackBuild > command in the root of the slack source tree. > The other thing is that you can install slackware packages without using > the installer, if you really need, since they are just .tar.gz's with a > few extra files included. Much nicer than that horrible RPM thingy. :) Here's a few facts: * The RedHat installer works in 4MB on the x86 architecture. * The RedHat installer works in 4MB on the RiscPC architecture. * The RedHat installer does not work in 4MB on the A5000 architecture. The reason? The old machines have a limited number of pages (128 if you've got 4MB) and with a 32k page size. This all adds up to not a lot of pages per process after the kernel has built, and hence the problems. In 4MB without swap, you can just about get away with a shell and a daemon before you hit the out of memory. There just aren't enough pages at a fine enough granularity. No matter what installer you use, you're going to get this problem. Besides, RPMs are much nicer... The binary ones contain all info to do with the package, including install and uninstall scripts. They also contain a description of the package, and a signature (mine don't) to verify that the packages are still sound. The source RPMs contain all the info to rebuild the binary RPM. In fact, you cannot have a source RPM without having a binary RPM. The RPMs have version and release numbers in, and they have interdependencies, such that the installer can say to itself: 'so you want to install this perl script do you? have you installed perl already ?' If you haven't, it'll warn you. It'll also warn you if you attempt to remove a package (eg perl) that has other packages using it. I've had this discussion with someone at work, and they saw the light... ;) _____ |_____| ------------------------------------------------- ---+---+- | | Russell King linux@arm.uk.linux.org --- --- | | | | http://www.arm.uk.linux.org/~rmk/armlinux.html / / | | +-+-+ --- -+- / | THE developer of ARM Linux |+| /|\ / | | | --- | +-+-+ ------------------------------------------------- /\\\ | unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From linux@arm.uk.linux.org Sun Apr 19 11:50:25 1998 Return-Path: Received: from snowcrash.cymru.net (snowcrash.cymru.net [163.164.160.3]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id LAA07730 for ; Sun, 19 Apr 1998 11:50:24 +0100 Received: from caramon.armlinux.org (dialup1.cymru.net [163.164.160.65]) by snowcrash.cymru.net (8.8.7/8.7.1) with ESMTP id LAA04717; Sun, 19 Apr 1998 11:51:34 +0100 Received: from raistlin.armlinux.org (raistlin [192.168.0.3]) by caramon.armlinux.org (8.7.4/8.7.3) with ESMTP id MAA24752; Sun, 19 Apr 1998 12:06:39 +0100 From: Russell King - ARM Linux Admin Received: (from linux@localhost) by raistlin.armlinux.org (8.7.4/8.7.3) id LAA01114; Sun, 19 Apr 1998 11:46:42 +0100 Message-Id: <199804191046.LAA01114@raistlin.armlinux.org> Subject: Re: fd0* device for ADFS shape discs under i386 Linux To: root@mad.scientist.com (James Craig) Date: Sun, 19 Apr 1998 11:46:42 +0100 (BST) Cc: willy@odie.barnet.ac.uk, ajb85@cam.ac.uk, linux-arm@vger.rutgers.edu In-Reply-To: from "James Craig" at Apr 18, 98 08:24:41 am X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Status: RO James Craig writes: > On Fri, 17 Apr 1998, Matthew Wilcox wrote: > >Andrew Bower > >> Out of interest, is anyone working on write support for ADFS? > > > >I've looked at the current code. It would be hard to add extendable > >files to it, though in principle, files which are newly created and stay > >the same length should be able to work quite easily. > > > >The trouble is that ADFS (or more truthfully filecore) does not have > >the concept of the inode. Inodes are what Unix filesystems are built of. > >They have the following properties: > > > >Each file has an inode > >The inode remains the same during the life of the file > >Inodes do not map to more than one file. > >The inode contains all of the metadata for the file except its name. > > umm. ADFS doesn't have inodes in the same way UNIX does, but couldn't you > get away with the file's ID that appears in the free space map? (He said, > trying to talk knowledgably, despite the fact that it's 5 years since he > last wrote an ADFS-E-format-filing-system-reader). I did toy with that idea, after all, it is quite unique. However, you stumble on a small problem: Under Unix, a filename translates to an inode. An inode then gives you the file data, and attributes. Under Filecore, a filename translates to a fragment ID + attributes. A fragment ID translates to file data. Hence, if you do use the fragment ID for the inode number, then yes, you can access the file's data, but you can't get the inodes attributes, since you don't know where these are stored on the disk. I did at one point use a combination of the fragment ID for the parent directory and the fragment ID of the file, but 32-bits isn't enough to store this and the sector offset required. This method still isn't acceptable for a writable ADFS - if the file is moved from the directory to another, then it's parent ID changes... > The IDs still only map to one file, aren't likely to change over it's > lifetime, and the only problem you have is that the length info is kept in > the directory info. Exactly. > >I cannot think of a good way of faking inodes for ADFS filesystems *except* > >the following rather horrible idea: > > > >Create a structure in memory for each ADFS filesystem mounted. This can > >be done lazily. Assign each file an inode based on some scheme which will > >allow for easy mapping between it and the name of the ADFS file. Perhaps > >hash buckets. Note that it's okay to reuse inodes once a file has been > >deleted. > > Ur, YUK!! That would be horrible. I mean, mapping names to inodes is gonna > break as soon as you move a mile about the FS somewhere. I think that this is horrible, but may be the only way. > >Someone should look at other filing systems to see how they cope with > >filesystems that don't provide nice inodes. When Dickon Hood & I were > >experimenting with NFS servers under RISC OS, we had a structure that > >worked well; it was a tree, and the inode we returned was the address > >within that list. You've got me intrigued now, I might just do it. > > Good man, writable ADFS would be nice. :) > > >The way Russell has done it is faster than this method for read-only > >but will not extend to writable files. > > I haven't looked at it yet, but what's he done that stops you writing > files the same way? The problem is that the inode number depends on the position of the directory on the disk and the offset into the directory. If you add a file, then as far as the Linux VFS is concerned, all the inodes on that disk have changed, and all inodes must be refreshed. NFS will go mad (since it depends on inodes staying the same), as will things like `pwd'. Linux needs to be able to map an inode number directly to a file on ADFS - this is the first problem that someone's going to have to crack when allowing ADFS to be writable. _____ |_____| ------------------------------------------------- ---+---+- | | Russell King linux@arm.uk.linux.org --- --- | | | | http://www.arm.uk.linux.org/~rmk/armlinux.html / / | | +-+-+ --- -+- / | THE developer of ARM Linux |+| /|\ / | | | --- | +-+-+ ------------------------------------------------- /\\\ | From owner-linux-arm-outgoing@vger.rutgers.edu Sun Apr 19 11:52:38 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id LAA07743 for ; Sun, 19 Apr 1998 11:52:37 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:26154 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390606-14253>; Sun, 19 Apr 1998 13:53:27 +0300 Received: by vger.rutgers.edu id <970827-319>; Sun, 19 Apr 1998 06:51:37 -0400 Received: from snowcrash.cymru.net ([163.164.160.3]:1961 "EHLO snowcrash.cymru.net" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <970806-319>; Sun, 19 Apr 1998 06:50:28 -0400 Received: from caramon.armlinux.org (dialup1.cymru.net [163.164.160.65]) by snowcrash.cymru.net (8.8.7/8.7.1) with ESMTP id LAA04713; Sun, 19 Apr 1998 11:51:29 +0100 Received: from raistlin.armlinux.org (raistlin [192.168.0.3]) by caramon.armlinux.org (8.7.4/8.7.3) with ESMTP id LAA24724; Sun, 19 Apr 1998 11:54:20 +0100 From: Russell King - ARM Linux Admin Received: (from linux@localhost) by raistlin.armlinux.org (8.7.4/8.7.3) id LAA01083; Sun, 19 Apr 1998 11:34:23 +0100 Message-Id: <199804191034.LAA01083@raistlin.armlinux.org> Subject: Re: Installing on an 4 MB A5000 To: dickon-ml@fluff.org (Dickon Hood) Date: Sun, 19 Apr 1998 11:34:23 +0100 (BST) Cc: linux-arm@vger.rutgers.edu In-Reply-To: <81f6ae3848%dickon-ml@splurge.fluff.org> from "Dickon Hood" at Apr 18, 98 12:05:57 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Dickon Hood writes: > : On Fri, 17 Apr 1998, David Alan Gilbert wrote: > > : Here's a serious thought. The Slackware installer works in 4M on the ix86 > : architecture - anyone feel like recompiling Slack for the ARM? > > Would be nice... IMHO, RH is too SysVile. >From what I remember of Slackware, it was worse than RH. Last time I looked at Slackware, it had all sorts of problems *because* it wasn't intelligent enough with the init daemon. For instance, when you boot, it would mount the local filesystems, turn on swap etc before entering a run level. Then, if you shutdown to single user mode, it would turn swap off and unmount all filesystems. Then when returning to a run level, it totally and utterly forgot that it had unmounted the filesystems. The RH init style may be SysVile-like, but at least *it works* and it is *extendable*. However, maybe Slackware has improved since I last looked at it. > IIRC, from talking to Simes, it is an ADFS problem (not specifically > Filecore) which was fixed with the 'Large filecore' released in Risc OS 3.6 > (not to be confused with the 'Large filecore' to be released with RPCII). I > bought a drive which wouldn't work under the older one; AFAICR we bodged > around it by using a pull-[up|down] resistor across two pins. Very dodgy; > don't try this at home ;-) Indeed, some drives (notably Western Digital drives) to something very silly - if you tell the drive to disable interrupts, it will !!assert!! it's interrupt line. This would then cause RiscOS to hang. I'm not certain, but I think that Linus may use this to it's advantage when resetting the drive... ;) _____ |_____| ------------------------------------------------- ---+---+- | | Russell King linux@arm.uk.linux.org --- --- | | | | http://www.arm.uk.linux.org/~rmk/armlinux.html / / | | +-+-+ --- -+- / | THE developer of ARM Linux |+| /|\ / | | | --- | +-+-+ ------------------------------------------------- /\\\ | unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Sun Apr 19 21:09:38 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id VAA10559 for ; Sun, 19 Apr 1998 21:09:37 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:23405 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <392181-14253>; Sun, 19 Apr 1998 23:08:37 +0300 Received: by vger.rutgers.edu id <971070-319>; Sun, 19 Apr 1998 16:05:26 -0400 Received: from methan.in-berlin.de ([160.45.10.13]:4518 "HELO methan.in-berlin.de" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with SMTP id <971075-319>; Sun, 19 Apr 1998 16:01:28 -0400 Received: by methan.in-berlin.de (Smail3.2.0.98) from bolzen.in-berlin.de (194.94.235.1) with esmtp id ; Sun, 19 Apr 1998 22:03:40 +0200 (MET DST) Received: by bolzen.in-berlin.de (Smail3.2) from yoda with uucp id ; Sun, 19 Apr 1998 22:03:23 +0200 (MET DST) Received: by yoda.in-berlin.de (Smail3.1.28.1 #0) id m0yR2CN-000000C; Sun, 19 Apr 98 20:04 GMT Date: Sun, 19 Apr 1998 21:46:16 +0200 From: Stefan Voss To: linux-arm@vger.rutgers.edu Subject: PowerTec initialization problem Message-ID: <826f623948%voss@yoda.in-berlin.de> Organization: Individual Network (IN), Berlin, Germany X-Mailer: Messenger v1.34c for RISC OS MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Posting-Agent: RISC OS Newsbase 0.60d X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Hi folks, I compiled a new kernel with PowerTec SCSI support from the latest release of the kernel patches (April 13th). When booting this kernel my PowerTec SCSI II card is recognized ok but shortly after that the following error message is displayed: scsi: aborting command due to timeout: pid 0, scsi0, channel 0, id 0, Test Unit Ready 00 00 00 00 scsi0: fas126_abort: command executing SCSI host 0 abort (pid 0) timed out resetting SCSI bus is being reset for host 0 channel 0 scsi0: fas126_reset The above messages are repeated in an endless loop thus hanging the boot process. My SCSI configuration: id 7: PowerTec SCSI II card id 2: Sony CDROM CDU-76S id 1: IBM harddisk DPES-31080 How can I avoid the loop during boot up? Regards, Stefan -- Stefan Voss (voss@yoda.in-berlin.de) unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Mon Apr 20 10:45:42 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id KAA14808 for ; Mon, 20 Apr 1998 10:45:41 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:28827 "EHLO vger.rutgers.edu" ident: "TIMEDOUT") by orava.funet.fi with ESMTP id <391175-29392>; Mon, 20 Apr 1998 12:45:27 +0300 Received: by vger.rutgers.edu id <971155-319>; Mon, 20 Apr 1998 05:41:07 -0400 Received: from lenzie.cent.gla.ac.uk ([130.209.16.18]:41825 "EHLO lenzie.cent.gla.ac.uk" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <971072-319>; Mon, 20 Apr 1998 05:40:51 -0400 Received: from localhost (9606585c@localhost) by lenzie.cent.gla.ac.uk (8.8.4/8.8.8) with SMTP id KAA23698; Mon, 20 Apr 1998 10:44:03 +0100 (BST) Date: Mon, 20 Apr 1998 10:44:00 +0100 (BST) From: James Craig <9606585c@udcf.gla.ac.uk> X-Sender: 9606585c@lenzie.cent.gla.ac.uk To: Russell King - ARM Linux Admin cc: James Craig , linux@treblig.org, linux-arm@vger.rutgers.edu Subject: Re: Installing on an 4 MB A5000 In-Reply-To: <199804180814.JAA00438@raistlin.armlinux.org> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO -- James Craig <9606585c@udcf.gla.ac.uk> On Sat, 18 Apr 1998, Russell King - ARM Linux Admin wrote: > James Craig writes: > > Here's a serious thought. > > The Slackware installer works in 4M on the ix86 architecture - anyone feel > > like recompiling Slack for the ARM? It should just be a case of setting up > > the cross-compiler as the default compiler, and then doing a SlackBuild > > command in the root of the slack source tree. > > The other thing is that you can install slackware packages without using > > the installer, if you really need, since they are just .tar.gz's with a > > few extra files included. Much nicer than that horrible RPM thingy. :) > > Here's a few facts: > * The RedHat installer works in 4MB on the x86 architecture. > * The RedHat installer works in 4MB on the RiscPC architecture. > * The RedHat installer does not work in 4MB on the A5000 architecture. It's well dodgy on 4M on an x86. I've tried it. :) > The reason? The old machines have a limited number of pages (128 if you've > got 4MB) and with a 32k page size. This all adds up to not a lot of pages > per process after the kernel has built, and hence the problems. In 4MB without > swap, you can just about get away with a shell and a daemon before you > hit the out of memory. There just aren't enough pages at a fine enough > granularity. Yep, I know about this problem. It's a bit of a problem (understatement of the century) but we're more likely to be able to manage when the installer is just a tar+gunzip, instead of some weird format that stores huge heaps of completely useless info. > No matter what installer you use, you're going to get this problem. > > Besides, RPMs are much nicer... The binary ones contain all info to do > with the package, including install and uninstall scripts. They also > contain a description of the package, and a signature (mine don't) to > verify that the packages are still sound. The slackware packages include the scripts too - If you look closely at a slackware package, it contains a install/script file usually which handles installation. Most slackware packages uninstall fairly cleanly too. The other *BIG*, in fact *HUGE* advantage is that you can get away with no installer whatsoever. The good old fashioned sash shell (stand-alone shell) includes tar in it - and compiles perfectly for the ARM leaving bucketloads of RAM free even on a 4M machine. You can just use it's built-in tar command to unpack the slackware packages - it works great. > The source RPMs contain all the info to rebuild the binary RPM. In fact, > you cannot have a source RPM without having a binary RPM. Woopee. Hmm... methinks slackware source dir and the magic SlackBuild files. :) > The RPMs have version and release numbers in, and they have > interdependencies, such that the installer can say to itself: 'so you want > to install this perl script do you? have you installed perl already ?' > If you haven't, it'll warn you. It'll also warn you if you attempt to > remove a package (eg perl) that has other packages using it. Interdependencies are much overrated. Installing RH and debian, I spend half my time overriding the bloody things because stupid people saw fit to make, e.g. tcl-7.4 and tcl-7.5, completely seperate from a dependencies point of view. Right now, ARM linux is a real pain in the butt to get working on older machines. I don't really think people who are smart enough to get it to even *boot* are going to have much trouble figuring out that the need to install perl before they can use a perl script. Not to forget, of course, RPM is truly hideous to use anyway. Trying to do anything other than one of the predefined action is seriously fiddly, and you don't stand much change of getting an RPM port to other OSs to let you modify packages before installing them - e.g. seperating individual files out to fit them onto 720K floppies.. Also - removing a package which other programs are using isn't smart usually - but you need to do it to make a decent job of upgrading versions on something. I'd much rather it just shut up and got on with it. :) > I've had this discussion with someone at work, and they saw the light... ;) My commiserations to your friend, losing his soul to the dark side of Linux. He's fallen in with evil companions. ;) -- James Craig unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Mon Apr 20 10:50:04 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id KAA14862 for ; Mon, 20 Apr 1998 10:50:02 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:28827 "EHLO vger.rutgers.edu" ident: "TIMEDOUT") by orava.funet.fi with ESMTP id <391180-29392>; Mon, 20 Apr 1998 12:49:25 +0300 Received: by vger.rutgers.edu id <971155-319>; Mon, 20 Apr 1998 05:45:02 -0400 Received: from lenzie.cent.gla.ac.uk ([130.209.16.18]:42609 "EHLO lenzie.cent.gla.ac.uk" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <971072-319>; Mon, 20 Apr 1998 05:44:47 -0400 Received: from localhost (9606585c@localhost) by lenzie.cent.gla.ac.uk (8.8.4/8.8.8) with SMTP id KAA26235; Mon, 20 Apr 1998 10:48:19 +0100 (BST) Date: Mon, 20 Apr 1998 10:48:16 +0100 (BST) From: James Craig <9606585c@udcf.gla.ac.uk> X-Sender: 9606585c@lenzie.cent.gla.ac.uk To: Russell King - ARM Linux Admin cc: Dickon Hood , linux-arm@vger.rutgers.edu Subject: Re: Installing on an 4 MB A5000 In-Reply-To: <199804191034.LAA01083@raistlin.armlinux.org> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO On Sun, 19 Apr 1998, Russell King - ARM Linux Admin wrote: > Dickon Hood writes: > > : On Fri, 17 Apr 1998, David Alan Gilbert wrote: > > > > : Here's a serious thought. The Slackware installer works in 4M on the ix86 > > : architecture - anyone feel like recompiling Slack for the ARM? > > > > Would be nice... IMHO, RH is too SysVile. > > >From what I remember of Slackware, it was worse than RH. Last time I looked at > Slackware, it had all sorts of problems *because* it wasn't intelligent enough > with the init daemon. For instance, when you boot, it would mount the local > filesystems, turn on swap etc before entering a run level. Then, if you shutdown > to single user mode, it would turn swap off and unmount all filesystems. Then > when returning to a run level, it totally and utterly forgot that it had unmounted > the filesystems. It's also so nice and uncomplicated that it causes no trouble to 99% of people. BTW, that bug is one I haven't encountered, and I've been using Slackware since 1993. :) > The RH init style may be SysVile-like, but at least *it works* and it is > *extendable*. Um, slack isn't non-extendable just because you have to add stuff to the end of /etc/rc.d/rc.local occasionally. And, of course, it *DOES* work. :) > However, maybe Slackware has improved since I last looked at it. Probably. Slackware is so simple in it's setup though that it would be hard to improve on. > > IIRC, from talking to Simes, it is an ADFS problem (not specifically > > Filecore) which was fixed with the 'Large filecore' released in Risc OS 3.6 > > (not to be confused with the 'Large filecore' to be released with RPCII). I > > bought a drive which wouldn't work under the older one; AFAICR we bodged > > around it by using a pull-[up|down] resistor across two pins. Very dodgy; > > don't try this at home ;-) > > Indeed, some drives (notably Western Digital drives) to something very silly - > if you tell the drive to disable interrupts, it will !!assert!! it's interrupt > line. This would then cause RiscOS to hang. > > I'm not certain, but I think that Linus may use this to it's advantage when > resetting the drive... ;) Pass. I wouldn't know. The last time I tried writing IDE device drivers, I managed to send the drive a format command instead of a verify one and totaled the first 2 tracks. I'll get around to doing them again, but not until I have a spare hard-drive I don't mind losing everything on. :) -- James Craig unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From 9606585c@udcf.gla.ac.uk Mon Apr 20 10:57:49 1998 Return-Path: <9606585c@udcf.gla.ac.uk> Received: from lenzie.cent.gla.ac.uk (lenzie.cent.gla.ac.uk [130.209.16.18]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id KAA14984 for ; Mon, 20 Apr 1998 10:57:48 +0100 Received: from localhost (9606585c@localhost) by lenzie.cent.gla.ac.uk (8.8.4/8.8.8) with SMTP id KAA03216; Mon, 20 Apr 1998 10:57:39 +0100 (BST) Date: Mon, 20 Apr 1998 10:57:37 +0100 (BST) From: James Craig <9606585c@udcf.gla.ac.uk> X-Sender: 9606585c@lenzie.cent.gla.ac.uk Reply-To: James Craig <9606585c@udcf.gla.ac.uk> To: Russell King - ARM Linux Admin cc: James Craig , willy@odie.barnet.ac.uk, ajb85@cam.ac.uk, linux-arm@vger.rutgers.edu Subject: Re: fd0* device for ADFS shape discs under i386 Linux In-Reply-To: <199804191046.LAA01114@raistlin.armlinux.org> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Status: RO On Sun, 19 Apr 1998, Russell King - ARM Linux Admin wrote: > James Craig writes: > > On Fri, 17 Apr 1998, Matthew Wilcox wrote: > > >Andrew Bower > > >> Out of interest, is anyone working on write support for ADFS? > > > > > >I've looked at the current code. It would be hard to add extendable > > >files to it, though in principle, files which are newly created and stay > > >the same length should be able to work quite easily. > > > > > >The trouble is that ADFS (or more truthfully filecore) does not have > > >the concept of the inode. Inodes are what Unix filesystems are built of. > > >They have the following properties: > > > > > >Each file has an inode > > >The inode remains the same during the life of the file > > >Inodes do not map to more than one file. > > >The inode contains all of the metadata for the file except its name. > > > > umm. ADFS doesn't have inodes in the same way UNIX does, but couldn't you > > get away with the file's ID that appears in the free space map? (He said, > > trying to talk knowledgably, despite the fact that it's 5 years since he > > last wrote an ADFS-E-format-filing-system-reader). > > I did toy with that idea, after all, it is quite unique. However, you stumble > on a small problem: > > Under Unix, a filename translates to an inode. An inode then gives you the > file data, and attributes. > > Under Filecore, a filename translates to a fragment ID + attributes. A > fragment ID translates to file data. > > Hence, if you do use the fragment ID for the inode number, then yes, you can > access the file's data, but you can't get the inodes attributes, since you > don't know where these are stored on the disk. > > I did at one point use a combination of the fragment ID for the parent directory > and the fragment ID of the file, but 32-bits isn't enough to store this and > the sector offset required. > > This method still isn't acceptable for a writable ADFS - if the file is moved from > the directory to another, then it's parent ID changes... Yeah, I can quite definately see the problem. However, could we store a hash table of recently-accessed-inodes and just cache their directory info? It's not like you can't search and ADFS filesystem for a particular ID, it just takes a while. For small enough numbers of files, (say a floppy) you could do it at mount/umount time. > > The IDs still only map to one file, aren't likely to change over it's > > lifetime, and the only problem you have is that the length info is kept in > > the directory info. > > Exactly. > > >Create a structure in memory for each ADFS filesystem mounted. This can > > >be done lazily. Assign each file an inode based on some scheme which will > > >allow for easy mapping between it and the name of the ADFS file. Perhaps > > >hash buckets. Note that it's okay to reuse inodes once a file has been > > >deleted. > > > > Ur, YUK!! That would be horrible. I mean, mapping names to inodes is gonna > > break as soon as you move a mile about the FS somewhere. > > I think that this is horrible, but may be the only way. How would it handle mv /bin/testfile /tmp/testfile, though, if the file was still being read by another process? > > >Someone should look at other filing systems to see how they cope with > > >filesystems that don't provide nice inodes. When Dickon Hood & I were > > >experimenting with NFS servers under RISC OS, we had a structure that > > >worked well; it was a tree, and the inode we returned was the address > > >within that list. You've got me intrigued now, I might just do it. > > > > Good man, writable ADFS would be nice. :) > > > > >The way Russell has done it is faster than this method for read-only > > >but will not extend to writable files. > > > > I haven't looked at it yet, but what's he done that stops you writing > > files the same way? > > The problem is that the inode number depends on the position of the directory > on the disk and the offset into the directory. If you add a file, then > as far as the Linux VFS is concerned, all the inodes on that disk have changed, > and all inodes must be refreshed. NFS will go mad (since it depends on inodes > staying the same), as will things like `pwd'. > > Linux needs to be able to map an inode number directly to a file on ADFS - > this is the first problem that someone's going to have to crack when allowing > ADFS to be writable. Truly. I've written ADFS file readers and writers before, but I've never tried integrating one into a totally different OS's idea of a filing system. In the meantime, we could probably fairly quickly code up an ADFS equivalent of the mtools for DOS - The old IscaFS that Phil Norman wrote would do the job just fine if we could manage to fiddle it to run under Linux. :) -- James Craig From owner-linux-arm-outgoing@vger.rutgers.edu Mon Apr 20 11:06:37 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id LAA15097 for ; Mon, 20 Apr 1998 11:06:36 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:28827 "EHLO vger.rutgers.edu" ident: "TIMEDOUT") by orava.funet.fi with ESMTP id <391196-29392>; Mon, 20 Apr 1998 13:06:17 +0300 Received: by vger.rutgers.edu id <971155-319>; Mon, 20 Apr 1998 06:01:55 -0400 Received: from violet.csi.cam.ac.uk ([131.111.8.58]:42154 "HELO violet.csi.cam.ac.uk" ident: "exim") by vger.rutgers.edu with SMTP id <971072-319>; Mon, 20 Apr 1998 06:01:45 -0400 Received: from ajb85.acad.cai.cam.ac.uk [131.111.136.66] (root) by violet.csi.cam.ac.uk with smtp (Exim 1.73 #1) id 0yRDRz-0007Fx-00; Mon, 20 Apr 1998 11:05:23 +0100 Received: from ajb85 by ajb85.acad.cai.cam.ac.uk with smtp (Exim 1.82 #1) id 0yRDRy-0002tv-00; Mon, 20 Apr 1998 11:05:22 +0100 Date: Mon, 20 Apr 1998 11:05:22 +0100 (BST) From: Andrew Bower X-Sender: ajb85@ajb85.acad.cai.cam.ac.uk Reply-To: Andrew Bower To: James Craig <9606585c@udcf.gla.ac.uk> cc: linux-arm@vger.rutgers.edu Subject: Slackware rocks (was Re: Installing on an 4 MB A5000) In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO On Mon, 20 Apr 1998, James Craig wrote: > It's well dodgy on 4M on an x86. I've tried it. :) And me:) I totally agree. Slackware whips Redhat: It works. A slackware distribution for ARM Linux would be a very good thing. Especially since most arcs are underspecified compared with PCs. I'd be happy to help too, although no guarantees. (i.e. I don't have my arc with me and it doesn't have enough hd :( ) Andrew unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Mon Apr 20 11:13:14 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id LAA15155 for ; Mon, 20 Apr 1998 11:13:13 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:28827 "EHLO vger.rutgers.edu" ident: "TIMEDOUT") by orava.funet.fi with ESMTP id <391195-29392>; Mon, 20 Apr 1998 13:12:53 +0300 Received: by vger.rutgers.edu id <971155-319>; Mon, 20 Apr 1998 06:08:40 -0400 Received: from snowcrash.cymru.net ([163.164.160.3]:1763 "EHLO snowcrash.cymru.net" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <971163-319>; Mon, 20 Apr 1998 06:08:21 -0400 Received: (from alan@localhost) by snowcrash.cymru.net (8.8.7/8.7.1) id LAA24388; Mon, 20 Apr 1998 11:11:48 +0100 From: Alan Cox Message-Id: <199804201011.LAA24388@snowcrash.cymru.net> Subject: Re: Installing on an 4 MB A5000 To: 9606585c@udcf.gla.ac.uk (James Craig) Date: Mon, 20 Apr 1998 11:11:46 +0100 (BST) Cc: linux@arm.uk.linux.org, root@mad.scientist.com, linux@treblig.org, linux-arm@vger.rutgers.edu In-Reply-To: from "James Craig" at Apr 20, 98 10:44:00 am Content-Type: text X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO > > * The RedHat installer works in 4MB on the x86 architecture. > > * The RedHat installer works in 4MB on the RiscPC architecture. > > * The RedHat installer does not work in 4MB on the A5000 architecture. > It's well dodgy on 4M on an x86. I've tried it. :) Its fine in 4Mbytes on x86 unless you use ramdisks. Its a concious decision that RH chooses to ship 8Mbyte minimum setups. > Yep, I know about this problem. It's a bit of a problem (understatement of > the century) but we're more likely to be able to manage when the installer > is just a tar+gunzip, instead of some weird format that stores huge heaps > of completely useless info. Actually for a minimal installer rpm is just cpio+gzip (tar is unsuitable for several reasons including path length limitations and taking more space) > other *BIG*, in fact *HUGE* advantage is that you can get away with no > installer whatsoever. The good old fashioned sash shell (stand-alone Your installer is /bin/sh and gnu tar. Thats _bigger_ than the minimal rpm installer, even before I point out you need awk,sed, and other bits > Interdependencies are much overrated. Installing RH and debian, I spend > half my time overriding the bloody things because stupid people saw fit to > make, e.g. tcl-7.4 and tcl-7.5, completely seperate from a dependencies > point of view. Right now, ARM linux is a real pain in the butt to get Which is correct. They aren't compatible. > to forget, of course, RPM is truly hideous to use anyway. Trying to do Run an ISP with 30+ machines to maintain or deal with code porting and cross platform maintenance and you'll understand why RPM is an absolute gift. At home I just type autorpm and all my boxes are up to date. For a code point of view the SRPM packages contain ChangeLogs and patch sets as well as per architecture rules. Updating an original package change normally involves flipping the tar file to the latest from the author, removing any patches now accepted by them and telling it to rebuild. > anything other than one of the predefined action is seriously fiddly, and > you don't stand much change of getting an RPM port to other OSs to let > you modify packages before installing them - e.g. seperating individual > files out to fit them onto 720K floppies.. Also - removing a package which RPM uses zlib to read data streams. Since RPM streams data you can install RPM files direct of tape - that means you can add a deblocker and install an rpm split over floppies quite easily. Alan unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Mon Apr 20 11:19:09 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id LAA15182 for ; Mon, 20 Apr 1998 11:19:08 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:28827 "EHLO vger.rutgers.edu" ident: "TIMEDOUT") by orava.funet.fi with ESMTP id <391199-29392>; Mon, 20 Apr 1998 13:17:11 +0300 Received: by vger.rutgers.edu id <971163-319>; Mon, 20 Apr 1998 06:10:03 -0400 Received: from snowcrash.cymru.net ([163.164.160.3]:1837 "EHLO snowcrash.cymru.net" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <971169-319>; Mon, 20 Apr 1998 06:09:51 -0400 Received: (from alan@localhost) by snowcrash.cymru.net (8.8.7/8.7.1) id LAA24415; Mon, 20 Apr 1998 11:13:27 +0100 From: Alan Cox Message-Id: <199804201013.LAA24415@snowcrash.cymru.net> Subject: Re: Installing on an 4 MB A5000 To: 9606585c@udcf.gla.ac.uk (James Craig) Date: Mon, 20 Apr 1998 11:13:25 +0100 (BST) Cc: linux@arm.uk.linux.org, dickon-ml@fluff.org, linux-arm@vger.rutgers.edu In-Reply-To: from "James Craig" at Apr 20, 98 10:48:16 am Content-Type: text X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO > > The RH init style may be SysVile-like, but at least *it works* and it is > > *extendable*. > Um, slack isn't non-extendable just because you have to add stuff to the > end of /etc/rc.d/rc.local occasionally. And, of course, it *DOES* work. :) Except for people who don't know or wish to know how to edit configuration files, people who want to manage boot order and what boots when. There is a reason everything in the world but Slackware uses Sys5 style init. Alan unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Mon Apr 20 11:19:10 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id LAA15184 for ; Mon, 20 Apr 1998 11:19:08 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:28827 "EHLO vger.rutgers.edu" ident: "TIMEDOUT") by orava.funet.fi with ESMTP id <391213-29392>; Mon, 20 Apr 1998 13:17:51 +0300 Received: by vger.rutgers.edu id <971072-319>; Mon, 20 Apr 1998 06:13:24 -0400 Received: from snowcrash.cymru.net ([163.164.160.3]:1994 "EHLO snowcrash.cymru.net" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <971184-319>; Mon, 20 Apr 1998 06:11:47 -0400 Received: (from alan@localhost) by snowcrash.cymru.net (8.8.7/8.7.1) id LAA24562; Mon, 20 Apr 1998 11:15:33 +0100 From: Alan Cox Message-Id: <199804201015.LAA24562@snowcrash.cymru.net> Subject: Re: Slackware rocks (was Re: Installing on an 4 MB A5000) To: ajb85@cam.ac.uk Date: Mon, 20 Apr 1998 11:15:31 +0100 (BST) Cc: 9606585c@udcf.gla.ac.uk, linux-arm@vger.rutgers.edu In-Reply-To: from "Andrew Bower" at Apr 20, 98 11:05:22 am Content-Type: text X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO > A slackware distribution for ARM Linux would be a very good thing. > Especially since most arcs are underspecified compared with PCs. This I have to see.. > I'd be happy to help too, although no guarantees. > (i.e. I don't have my arc with me and it doesn't have enough hd :( ) You need about 4Gig of free disk to work effeciently on a distribution and preferably a second disk and machine to keep feeding your latest attempts too. You also need to budget people for support and maintaining of each package you ship. unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Mon Apr 20 11:20:47 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id LAA15221 for ; Mon, 20 Apr 1998 11:20:46 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:28827 "EHLO vger.rutgers.edu" ident: "TIMEDOUT") by orava.funet.fi with ESMTP id <391207-29392>; Mon, 20 Apr 1998 13:20:26 +0300 Received: by vger.rutgers.edu id <971172-319>; Mon, 20 Apr 1998 06:16:02 -0400 Received: from odie.barnet.ac.uk ([194.82.202.98]:12439 "EHLO odie.barnet.ac.uk" ident: "willy") by vger.rutgers.edu with ESMTP id <971174-319>; Mon, 20 Apr 1998 06:15:38 -0400 Received: (from willy@localhost) by odie.barnet.ac.uk (8.8.6/8.8.6) id LAA15190 for linux-arm@vger.rutgers.edu; Mon, 20 Apr 1998 11:19:13 +0100 From: Matthew Wilcox Message-Id: <199804201019.LAA15190@odie.barnet.ac.uk> Subject: memsetl (improved) To: linux-arm@vger.rutgers.edu Date: Mon, 20 Apr 1998 11:19:10 +0100 (BST) X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO I've left this one embedded in its BASIC testbed, to ensure no transcription errors this time :-) 17 instructions and the removal of a fair bit of cruft. REM memsetl : DIM code &200 start=&20000 FOR i% = 0 TO 2 STEP 2 P%=code [ OPT i% MOV r0, #start LDR r1, data LDR r2, len B memsetl ..data EQUD &12345678 ..len EQUD 0 ] P% = (P% + 16) AND NOT 15 [ OPT i% ..memsetl stmfd r13!, {r14} mov r14, r1 subs r2, r2, #16 blt mod_4 mov r3, r1 mov r12, r1 tst r2, #16 addeq r2, r2, #16 ..loop_32 stmneia r0!, {r1, r3, r12, r14} stmia r0!, {r1, r3, r12, r14} subs r2, r2, #32 bgt loop_32 ..mod_4 tst r2, #8 stmneia r0!, {r1, r14} tst r2, #4 strne r1, [r0] ldmfd r13!, {pc} ] NEXT i% : FOR i% = 0 TO 128 STEP 4 !len = i% start!i% = &c0ffee00 CALL code IF start!i% <> &c0ffee00 PRINT "bug in "+STR$i%+" at end" IF i% = 0 NEXT FOR j% = 0 TO i%-4 STEP 4 IF start!j% <> &12345678 PRINT "bug in "+STR$i%+" at "+STR$j% start!j% = &00decaff NEXT j% NEXT i% END -- Set Alias$Case Set Alias$[ |||| |MSet Alias$Otherwise Set Alias$[ \ Matthew "" |MSet Alias$When If %0=%%0 Then Set Alias$[ "" ||MIf %0=%%0 \ Wilcox Then Set Alias$Otherwise Set Alias$[ |||||||||||||||| ||MIf \ %0=%%0 Then Set Alias$When Set Alias$[ |||||||||||||||| unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Mon Apr 20 11:54:30 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id LAA15427 for ; Mon, 20 Apr 1998 11:54:29 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:12035 "EHLO vger.rutgers.edu" ident: "TIMEDOUT") by orava.funet.fi with ESMTP id <391222-29392>; Mon, 20 Apr 1998 13:54:09 +0300 Received: by vger.rutgers.edu id <971169-319>; Mon, 20 Apr 1998 06:49:39 -0400 Received: from odie.barnet.ac.uk ([194.82.202.98]:12572 "EHLO odie.barnet.ac.uk" ident: "TIMEDOUT") by vger.rutgers.edu with ESMTP id <971178-319>; Mon, 20 Apr 1998 06:49:22 -0400 Received: (from willy@localhost) by odie.barnet.ac.uk (8.8.6/8.8.6) id LAA15415; Mon, 20 Apr 1998 11:53:02 +0100 From: Matthew Wilcox Message-Id: <199804201053.LAA15415@odie.barnet.ac.uk> Subject: strchr.S To: linux-arm@vger.rutgers.edu Date: Mon, 20 Apr 1998 11:52:59 +0100 (BST) Cc: mdw@ebi.ac.uk X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Here's a 39-instruction of strchr which works word-at-a-time. I have neither tested nor benchmarked it. Phil, does glibc have a testsuite for strchr? If not, I ought to write one - there are so many cases to take care of. -- Set Alias$Case Set Alias$[ |||| |MSet Alias$Otherwise Set Alias$[ \ Matthew "" |MSet Alias$When If %0=%%0 Then Set Alias$[ "" ||MIf %0=%%0 \ Wilcox Then Set Alias$Otherwise Set Alias$[ |||||||||||||||| ||MIf \ %0=%%0 Then Set Alias$When Set Alias$[ |||||||||||||||| strchr: tst r0, #3 beq 2f 1: ldrb r2, [r0], #1 cmp r2, r1 tstne r0, #3 bne 1b subcs r0, r0, #1 movcs pc, r14 2: orr r1, r1, r1, LSL #8 orr r1, r1, r1, LSL #16 3: ldr r2, [r0], #4 eor r3, r2, r1 tst r3, #&000000ff tstne r3, #&0000ff00 tstne r3, #&00ff0000 tstne r3, #&ff000000 beq 5f tst r2, #&000000ff tstne r2, #&0000ff00 tstne r2, #&00ff0000 tstne r2, #&ff000000 bne 3b 4: mov r0, #0 mov pc, r14 5: tst r2, #&000000ff beq 4b tst r3, #&000000ff subeq r0, r0, #1 tstne r3, #&0000ff00 subeq r0, r0, #2 moveq pc, r14 tst r2, #&0000ff00 tstne r2, #&00ff0000 beq 4b tst r3, #&00ff0000 subeq r0, r0, #1 tstne r3, #&ff000000 moveq pc, r14 b 4b unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From pb@nexus.co.uk Mon Apr 20 12:26:58 1998 Return-Path: Received: from globe (nexusel.demon.co.uk [158.152.30.195]) by odie.barnet.ac.uk (8.8.6/8.8.6) with SMTP id MAA15613 for ; Mon, 20 Apr 1998 12:26:56 +0100 Received: from (spring.nexus.co.uk) [192.0.0.3] (root) by globe with smtp (Exim 1.82 #1) id 0yREcS-00004h-00; Mon, 20 Apr 1998 12:20:16 +0100 Received: from localhost (spring.nexus.co.uk) [127.0.0.1] (pb) by spring.nexus.co.uk with esmtp (Exim 1.82 #1) id 0yREXv-00009W-00; Mon, 20 Apr 1998 12:15:35 +0100 X-Mailer: exmh version 2.0zeta 7/24/97 To: Matthew Wilcox cc: linux-arm@vger.rutgers.edu, mdw@ebi.ac.uk Subject: Re: strchr.S In-reply-to: Your message of "Mon, 20 Apr 1998 11:52:59 BST." <199804201053.LAA15415@odie.barnet.ac.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 20 Apr 1998 12:15:34 +0100 From: Philip Blundell Message-Id: Status: RO >Here's a 39-instruction of strchr which works word-at-a-time. I have >neither tested nor benchmarked it. Phil, does glibc have a testsuite >for strchr? If not, I ought to write one - there are so many cases to >take care of. Yes, we do have one. Here's the code. If you can think of additional tests to add then please do. void test_strchr (void) { it = "strchr"; check (strchr ("abcd", 'z') == NULL, 1); /* Not found. */ (void) strcpy (one, "abcd"); check (strchr (one, 'c') == one+2, 2); /* Basic test. */ check (strchr (one, 'd') == one+3, 3); /* End of string. */ check (strchr (one, 'a') == one, 4); /* Beginning. */ check (strchr (one, '\0') == one+4, 5); /* Finding NUL. */ (void) strcpy (one, "ababa"); check (strchr (one, 'b') == one+1, 6); /* Finding first. */ (void) strcpy (one, ""); check (strchr (one, 'b') == NULL, 7); /* Empty string. */ check (strchr (one, '\0') == one, 8); /* NUL in empty string. */ { char buf[4096]; int i; char *p; for (i=0; i < 0x100; i++) { p = (char *) ((unsigned long int) (buf + 0xff) & ~0xff) + i; strcpy (p, "OK"); strcpy (p+3, "BAD/WRONG"); check (strchr (p, '/') == NULL, 9+i); } } } p. From owner-linux-arm-outgoing@vger.rutgers.edu Mon Apr 20 12:42:57 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id MAA15711 for ; Mon, 20 Apr 1998 12:42:55 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:7676 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390458-29392>; Mon, 20 Apr 1998 14:42:09 +0300 Received: by vger.rutgers.edu id <971300-319>; Mon, 20 Apr 1998 07:27:53 -0400 Received: from red.csi.cam.ac.uk ([131.111.8.70]:62922 "HELO red.csi.cam.ac.uk" ident: "exim") by vger.rutgers.edu with SMTP id <971350-319>; Mon, 20 Apr 1998 07:23:13 -0400 Received: from dmf20 by red.csi.cam.ac.uk with smtp (Exim 1.73 #1) id 0yREiI-0007jy-00; Mon, 20 Apr 1998 12:26:18 +0100 Date: Mon, 20 Apr 1998 12:26:18 +0100 (BST) From: David X-Sender: dmf20@red.csi.cam.ac.uk To: Alan Cox cc: James Craig <9606585c@udcf.gla.ac.uk>, linux@arm.uk.linux.org, root@mad.scientist.com, linux@treblig.org, linux-arm@vger.rutgers.edu Subject: Re: Installing on an 4 MB A5000 In-Reply-To: <199804201011.LAA24388@snowcrash.cymru.net> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO > Actually for a minimal installer rpm is just cpio+gzip (tar is unsuitable > for several reasons including path length limitations and taking more > space) Can you extract RPMs with standard UNIX commands? If so, could you be a bit more explicit as to how you do this...I've been trying for ages. > Your installer is /bin/sh and gnu tar. Thats _bigger_ than the minimal rpm > installer, even before I point out you need awk,sed, and other bits It's perhaps not the wisest thing to say but: I've installed NetBSD/arm32 on my system and all the files there were .tar.gz. Which has two key advantages for me: (a) I know instantly how to unpack the things (b) I can unpack them under RISCOS if necessary. Also, for many things I haven't noticed the need for awk, sed, perl etc. (But I can get them running under RISCOS too, if I need to.) > RPM uses zlib to read data streams. Since RPM streams data you can install > RPM files direct of tape - that means you can add a deblocker and install > an rpm split over floppies quite easily. Can you not do this with tar? Just a thought, David. unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Mon Apr 20 12:49:09 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id MAA15746 for ; Mon, 20 Apr 1998 12:49:07 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:9826 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390606-29392>; Mon, 20 Apr 1998 14:47:24 +0300 Received: by vger.rutgers.edu id <971262-319>; Mon, 20 Apr 1998 07:38:42 -0400 Received: from red.csi.cam.ac.uk ([131.111.8.70]:33365 "HELO red.csi.cam.ac.uk" ident: "exim") by vger.rutgers.edu with SMTP id <971299-319>; Mon, 20 Apr 1998 07:37:55 -0400 Received: from dmf20 by red.csi.cam.ac.uk with smtp (Exim 1.73 #1) id 0yREx7-0001mM-00; Mon, 20 Apr 1998 12:41:37 +0100 Date: Mon, 20 Apr 1998 12:41:37 +0100 (BST) From: David X-Sender: dmf20@red.csi.cam.ac.uk To: Alan Cox cc: ajb85@cam.ac.uk, 9606585c@udcf.gla.ac.uk, linux-arm@vger.rutgers.edu Subject: Re: Slackware rocks (was Re: Installing on an 4 MB A5000) In-Reply-To: <199804201015.LAA24562@snowcrash.cymru.net> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO > > A slackware distribution for ARM Linux would be a very good thing. > > Especially since most arcs are underspecified compared with PCs. > > This I have to see.. Perhaps you might. I happen to think it'd be a good idea too. I think I'll have to look at the details, but if anyone's going to do this, count me in. (Well, after exams, anyway.) > You need about 4Gig of free disk to work effeciently on a distribution > and preferably a second disk and machine to keep feeding your latest > attempts too. Not a problem as far as disk space goes - 6Gb, and *two* spare machines. I'm currently in Cambridge and would be looking at doing this over the summer/beginning of next term. Is anyone *seriously* interested in this project? Anyone willing to help? David. unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Mon Apr 20 13:28:34 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id NAA15884 for ; Mon, 20 Apr 1998 13:28:33 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:21774 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390772-29392>; Mon, 20 Apr 1998 15:26:20 +0300 Received: by vger.rutgers.edu id <971301-319>; Mon, 20 Apr 1998 08:18:51 -0400 Received: from snowcrash.cymru.net ([163.164.160.3]:1084 "EHLO snowcrash.cymru.net" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <971296-319>; Mon, 20 Apr 1998 08:17:03 -0400 Received: (from alan@localhost) by snowcrash.cymru.net (8.8.7/8.7.1) id NAA26837; Mon, 20 Apr 1998 13:21:00 +0100 From: Alan Cox Message-Id: <199804201221.NAA26837@snowcrash.cymru.net> Subject: Re: Installing on an 4 MB A5000 To: dmf20@hermes.cam.ac.uk (David) Date: Mon, 20 Apr 1998 13:20:58 +0100 (BST) Cc: alan@cymru.net, 9606585c@udcf.gla.ac.uk, linux@arm.uk.linux.org, root@mad.scientist.com, linux@treblig.org, linux-arm@vger.rutgers.edu In-Reply-To: from "David" at Apr 20, 98 12:26:18 pm Content-Type: text X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO > Can you extract RPMs with standard UNIX commands? If so, could you be a > bit more explicit as to how you do this...I've been trying for ages. Take a look at rpm2cpio. Basically you pull off a header block, gunzip it and you get a cpio > advantages for me: (a) I know instantly how to unpack the things (b) I can > unpack them under RISCOS if necessary. Also, for many things I haven't tar cannot handle some things - try a 4K file path in tar for example. THats why cpio is used. 'pax' is a portable tar/cpio. > noticed the need for awk, sed, perl etc. (But I can get them running > under RISCOS too, if I need to.) For install scripts... > > an rpm split over floppies quite easily. > Can you not do this with tar? Yes, and cpio 8) unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Mon Apr 20 13:29:50 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id NAA15893 for ; Mon, 20 Apr 1998 13:29:49 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:21774 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391162-29392>; Mon, 20 Apr 1998 15:28:41 +0300 Received: by vger.rutgers.edu id <971286-319>; Mon, 20 Apr 1998 08:23:18 -0400 Received: from nexusel.demon.co.uk ([158.152.30.195]:2210 "HELO globe" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with SMTP id <971307-319>; Mon, 20 Apr 1998 08:22:17 -0400 Received: from (spring.nexus.co.uk) [192.0.0.3] (root) by globe with smtp (Exim 1.82 #1) id 0yRFHe-0000BE-00; Mon, 20 Apr 1998 13:02:50 +0100 Received: from localhost (spring.nexus.co.uk) [127.0.0.1] (pb) by spring.nexus.co.uk with esmtp (Exim 1.82 #1) id 0yRFM8-0002Jz-00; Mon, 20 Apr 1998 13:07:28 +0100 X-Mailer: exmh version 2.0zeta 7/24/97 To: David cc: linux-arm@vger.rutgers.edu Subject: Re: Installing on an 4 MB A5000 In-reply-to: Your message of "Mon, 20 Apr 1998 12:26:18 BST." Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 20 Apr 1998 13:07:28 +0100 From: Philip Blundell Message-Id: X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO >Can you extract RPMs with standard UNIX commands? If so, could you be a >bit more explicit as to how you do this...I've been trying for ages. rpm2cpio will strip the header off and leave you with a standard cpio archive. I'm sure you could engineer your own program to do this if you wanted to. I haven't looked at the rpm format in detail but I doubt it's hard - you might even be able to do it with dd, od and a bit of shell script. p. unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Mon Apr 20 13:29:59 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id NAA15902 for ; Mon, 20 Apr 1998 13:29:58 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:21774 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391166-29392>; Mon, 20 Apr 1998 15:28:51 +0300 Received: by vger.rutgers.edu id <971289-319>; Mon, 20 Apr 1998 08:23:28 -0400 Received: from snowcrash.cymru.net ([163.164.160.3]:1341 "EHLO snowcrash.cymru.net" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <971310-319>; Mon, 20 Apr 1998 08:23:03 -0400 Received: (from alan@localhost) by snowcrash.cymru.net (8.8.7/8.7.1) id NAA26909; Mon, 20 Apr 1998 13:25:50 +0100 From: Alan Cox Message-Id: <199804201225.NAA26909@snowcrash.cymru.net> Subject: Re: Slackware rocks (was Re: Installing on an 4 MB A5000) To: dmf20@hermes.cam.ac.uk (David) Date: Mon, 20 Apr 1998 13:25:49 +0100 (BST) Cc: alan@cymru.net, ajb85@cam.ac.uk, 9606585c@udcf.gla.ac.uk, linux-arm@vger.rutgers.edu In-Reply-To: from "David" at Apr 20, 98 12:41:37 pm Content-Type: text X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO > > > Especially since most arcs are underspecified compared with PCs. > > This I have to see.. > Perhaps you might. I happen to think it'd be a good idea too. I think > I'll have to look at the details, but if anyone's going to do this, count > me in. (Well, after exams, anyway.) I think you are out of your tree personally, but I wish you best of luck anyway unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Mon Apr 20 14:36:59 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id OAA17311 for ; Mon, 20 Apr 1998 14:36:35 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:5051 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391334-29392>; Mon, 20 Apr 1998 16:30:45 +0300 Received: by vger.rutgers.edu id <971313-319>; Mon, 20 Apr 1998 09:22:37 -0400 Received: from [194.201.189.33] ([194.201.189.33]:32837 "HELO mailhost.cogency.co.uk" ident: "SOCKWRITE-65") by vger.rutgers.edu with SMTP id <971346-319>; Mon, 20 Apr 1998 09:18:41 -0400 Received: from cogency.co.uk by mailhost.cogency.co.uk (SMI-8.6/PIPEX simple 1.27) id OAA01534; Mon, 20 Apr 1998 14:23:59 +0100 Message-ID: <353B4C5B.94DF4D02@cogency.co.uk> Date: Mon, 20 Apr 1998 14:23:39 +0100 From: David Alan Gilbert Organization: Cogency Technology Inc X-Mailer: Mozilla 4.04 [en] (X11; I; SunOS 5.5 sun4m) MIME-Version: 1.0 To: linux-arm@vger.rutgers.edu Subject: Re: Installing on an 4 MB A5000 References: <199804201221.NAA26837@snowcrash.cymru.net> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Hi All, While steering clear of the holy RH/Slackware war I think one soloution to the RH install on a 4MB machine would be to have a pre-install disc which just has a hacked up swapon and mkswap and perhaps enough to load a module for the SCSI discs. So you boot that, it sets up the swap and then you flip the RH install disc in and boot that. What do you reckon? RH is a nice installer - when it works; but it really has a nasty habit of coming to the end of the installation having completely ignored an important error and leaving the system half installed. And also it tends to install in no useful order; it doesn't install critical stuff first, so when you do get an error it tends to be something useful like init which is missing (but at least emacs was installed - hmm...) Slakware just works; its still the installer I believe is most likely to work given a-random-machine. Dave -- ---------------------------------------------------------------------- - Dr. David Alan Gilbert - WARNING! This is a beta release .signature- - Work: dg @ cogency.co.uk - +44-(0)161-428-9444 - - Home: gro.gilbert @ treblig.org - - ---------------------------------------------------------------------- unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Mon Apr 20 14:53:01 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id OAA17412 for ; Mon, 20 Apr 1998 14:53:00 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:24633 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391345-29392>; Mon, 20 Apr 1998 16:48:27 +0300 Received: by vger.rutgers.edu id <971311-319>; Mon, 20 Apr 1998 09:43:11 -0400 Received: from dent.axion.bt.co.uk ([132.146.16.161]:37723 "HELO dent.axion.bt.co.uk" ident: "TIMEDOUT") by vger.rutgers.edu with SMTP id <971326-319>; Mon, 20 Apr 1998 09:41:41 -0400 Received: from lspace.org (actually etmsibm3.gfms.bt.co.uk) by dent.axion.bt.co.uk with SMTP (PP); Mon, 20 Apr 1998 14:43:37 +0100 Message-ID: <353B5065.252D26F@lspace.org> Date: Mon, 20 Apr 1998 14:40:53 +0100 From: Andy Fawcett Organization: Being Lazy At Home (BLAH!) X-Mailer: Mozilla 4.04 [en] (Win95; I) MIME-Version: 1.0 To: David Alan Gilbert CC: linux-arm@vger.rutgers.edu Subject: Re: Installing on an 4 MB A5000 References: <199804201221.NAA26837@snowcrash.cymru.net> <353B4C5B.94DF4D02@cogency.co.uk> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO David Alan Gilbert wrote: > > Hi All, > While steering clear of the holy RH/Slackware war I think one > soloution to the RH install on a 4MB machine would be to > have a pre-install disc which just has a hacked up swapon > and mkswap and perhaps enough to load a module for the SCSI > discs. So you boot that, it sets up the swap and then you > flip the RH install disc in and boot that. > > What do you reckon? Well, since I can't currently install onto my 4MB A5000 for just these reasons, I think it's a good idea. I'm going to be upgrading to 8MB one of these days, but the absence of all installation disks for the A5K still makes installation problematical. I, for one, would welcome any work done to help us poor memory-starved individuals... -- Andy Fawcett Writing from work, but writing for himself http://www.afawcett.demon.co.uk/ unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Mon Apr 20 15:16:09 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id PAA17524 for ; Mon, 20 Apr 1998 15:16:07 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:31627 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391348-29392>; Mon, 20 Apr 1998 17:06:18 +0300 Received: by vger.rutgers.edu id <971343-319>; Mon, 20 Apr 1998 09:57:47 -0400 Received: from post-31.mail.demon.net ([194.217.242.41]:60248 "HELO post.mail.demon.net" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with SMTP id <971334-319>; Mon, 20 Apr 1998 09:55:42 -0400 Received: from (globe) [158.152.30.195] by post.mail.demon.net with smtp (Exim 1.82 #2) id 0yRH72-0006Y5-00; Mon, 20 Apr 1998 15:00:00 +0100 Received: from (spring.nexus.co.uk) [192.0.0.3] (root) by globe with smtp (Exim 1.82 #1) id 0yRGoc-0000Ma-00; Mon, 20 Apr 1998 14:40:58 +0100 Received: from localhost (spring.nexus.co.uk) [127.0.0.1] (pb) by spring.nexus.co.uk with esmtp (Exim 1.82 #1) id 0yRGob-0003Pg-00; Mon, 20 Apr 1998 14:40:57 +0100 X-Mailer: exmh version 2.0zeta 7/24/97 To: David Alan Gilbert cc: linux-arm@vger.rutgers.edu Subject: Re: Installing on an 4 MB A5000 In-reply-to: Your message of "Mon, 20 Apr 1998 14:23:39 BST." <353B4C5B.94DF4D02@cogency.co.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 20 Apr 1998 14:40:57 +0100 From: Philip Blundell Message-Id: X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO > While steering clear of the holy RH/Slackware war I think one >soloution to the RH install on a 4MB machine would be to >have a pre-install disc which just has a hacked up swapon >and mkswap and perhaps enough to load a module for the SCSI >discs. So you boot that, it sets up the swap and then you >flip the RH install disc in and boot that. Trouble is you need to have already partitioned the disks. If you can get fdisk, mkswap and swapon to run in the amount of real RAM you have then you ought to be in business - you could even roll that functionality into one binary if you got really desperate, say on 2MB machines. :-) p. unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Mon Apr 20 15:16:10 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id PAA17526 for ; Mon, 20 Apr 1998 15:16:08 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:31627 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391176-29392>; Mon, 20 Apr 1998 17:06:40 +0300 Received: by vger.rutgers.edu id <971345-319>; Mon, 20 Apr 1998 09:57:58 -0400 Received: from snowcrash.cymru.net ([163.164.160.3]:1619 "EHLO snowcrash.cymru.net" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <971342-319>; Mon, 20 Apr 1998 09:57:25 -0400 Received: (from alan@localhost) by snowcrash.cymru.net (8.8.7/8.7.1) id PAA28683; Mon, 20 Apr 1998 15:01:29 +0100 From: Alan Cox Message-Id: <199804201401.PAA28683@snowcrash.cymru.net> Subject: Re: Installing on an 4 MB A5000 To: dg@cogency.co.uk (David Alan Gilbert) Date: Mon, 20 Apr 1998 15:01:27 +0100 (BST) Cc: linux-arm@vger.rutgers.edu In-Reply-To: <353B4C5B.94DF4D02@cogency.co.uk> from "David Alan Gilbert" at Apr 20, 98 02:23:39 pm Content-Type: text X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO > While steering clear of the holy RH/Slackware war I think one > soloution to the RH install on a 4MB machine would be to > have a pre-install disc which just has a hacked up swapon > and mkswap and perhaps enough to load a module for the SCSI > discs. So you boot that, it sets up the swap and then you > flip the RH install disc in and boot that. The RH install disk runs off a ramdisk because it needs to load some things from floppy then allow you to change floppy. If your root is only fd0 you can't of course do this. Putting it on an ADFS partition or using a minimal initial bootstrap install should work for a 4Mbyte machine. Alan unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Mon Apr 20 15:16:11 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id PAA17529 for ; Mon, 20 Apr 1998 15:16:09 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:31627 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391304-29392>; Mon, 20 Apr 1998 17:10:00 +0300 Received: by vger.rutgers.edu id <971327-319>; Mon, 20 Apr 1998 10:04:19 -0400 Received: from [194.201.189.33] ([194.201.189.33]:32921 "HELO mailhost.cogency.co.uk" ident: "SOCKWRITE-65") by vger.rutgers.edu with SMTP id <971341-319>; Mon, 20 Apr 1998 10:04:03 -0400 Received: from man05.cogency.co.uk by mailhost.cogency.co.uk (SMI-8.6/PIPEX simple 1.27) id PAA01769; Mon, 20 Apr 1998 15:09:09 +0100 Received: by man05.cogency.co.uk (SMI-8.6/SMI-SVR4) id PAA09883; Mon, 20 Apr 1998 15:08:49 +0100 Date: Mon, 20 Apr 1998 15:08:49 +0100 From: dg@cogency.co.uk (David Gilbert) Message-Id: <199804201408.PAA09883@man05.cogency.co.uk> To: pb@nexus.co.uk Subject: Re: Installing on an 4 MB A5000 Cc: linux-arm@vger.rutgers.edu X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Philip Blundell spoke thus: Trouble is you need to have already partitioned the disks. If you can get fdisk, mkswap and swapon to run in the amount of real RAM you have then you ought to be in business - you could even roll that functionality into one binary if you got really desperate, say on 2MB machines. :-) Nah - you partition from RiscOS using Russ's wonderful !PartMan - actually it would be nice to be able to make a swap partition from RISC OS as well. Dave unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Mon Apr 20 15:20:18 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id PAA17576 for ; Mon, 20 Apr 1998 15:20:17 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:31627 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391350-29392>; Mon, 20 Apr 1998 17:15:04 +0300 Received: by vger.rutgers.edu id <971332-319>; Mon, 20 Apr 1998 10:10:14 -0400 Received: from odie.barnet.ac.uk ([194.82.202.98]:14982 "EHLO odie.barnet.ac.uk" ident: "willy") by vger.rutgers.edu with ESMTP id <971327-319>; Mon, 20 Apr 1998 10:09:50 -0400 Received: (from willy@localhost) by odie.barnet.ac.uk (8.8.6/8.8.6) id PAA17518; Mon, 20 Apr 1998 15:13:41 +0100 From: Matthew Wilcox Message-Id: <199804201413.PAA17518@odie.barnet.ac.uk> Subject: Re: Installing on an 4 MB A5000 To: dg@cogency.co.uk (David Alan Gilbert) Date: Mon, 20 Apr 1998 15:13:40 +0100 (BST) Cc: linux-arm@vger.rutgers.edu In-Reply-To: <353B4C5B.94DF4D02@cogency.co.uk> from "David Alan Gilbert" at Apr 20, 98 02:23:39 pm X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO David Alan Gilbert > > Hi All, > While steering clear of the holy RH/Slackware war I think one > soloution to the RH install on a 4MB machine would be to > have a pre-install disc which just has a hacked up swapon > and mkswap and perhaps enough to load a module for the SCSI > discs. So you boot that, it sets up the swap and then you > flip the RH install disc in and boot that. > > What do you reckon? I reckon the idea you mentioned previously in private mail is better. It should be almost trivial to look through the getpartns utility supplied with IscaFS, figure out where the swappartition is and write the two words there which Linux requires. Hmm.. then does the RH installer swapon early enough? Or can it be made to? -- Set Alias$Case Set Alias$[ |||| |MSet Alias$Otherwise Set Alias$[ \ Matthew "" |MSet Alias$When If %0=%%0 Then Set Alias$[ "" ||MIf %0=%%0 \ Wilcox Then Set Alias$Otherwise Set Alias$[ |||||||||||||||| ||MIf \ %0=%%0 Then Set Alias$When Set Alias$[ |||||||||||||||| unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Mon Apr 20 15:47:15 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id PAA17715 for ; Mon, 20 Apr 1998 15:47:11 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:10515 "EHLO vger.rutgers.edu" ident: "TIMEDOUT") by orava.funet.fi with ESMTP id <391607-29392>; Mon, 20 Apr 1998 17:45:13 +0300 Received: by vger.rutgers.edu id <971345-319>; Mon, 20 Apr 1998 10:37:22 -0400 Received: from nexusel.demon.co.uk ([158.152.30.195]:2469 "HELO globe" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with SMTP id <971332-319>; Mon, 20 Apr 1998 10:25:34 -0400 Received: from (spring.nexus.co.uk) [192.0.0.3] (root) by globe with smtp (Exim 1.82 #1) id 0yRHZf-0000St-00; Mon, 20 Apr 1998 15:29:35 +0100 Received: from localhost (spring.nexus.co.uk) [127.0.0.1] (pb) by spring.nexus.co.uk with esmtp (Exim 1.82 #1) id 0yRHZd-0003Rf-00; Mon, 20 Apr 1998 15:29:33 +0100 X-Mailer: exmh version 2.0zeta 7/24/97 To: dg@cogency.co.uk (David Gilbert) cc: linux-arm@vger.rutgers.edu Subject: Re: Installing on an 4 MB A5000 In-reply-to: Your message of "Mon, 20 Apr 1998 15:08:49 BST." <199804201408.PAA09883@man05.cogency.co.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 20 Apr 1998 15:29:33 +0100 From: Philip Blundell Message-Id: X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO >Nah - you partition from RiscOS using Russ's wonderful !PartMan - actually >it would be nice to be able to make a swap partition from RISC OS as well. I've never managed to get PartMan to work, to be honest. By the way, I couldn't find a mkswap binary on the ftp site that worked with 26-bit machines - I ended up having to roll my own, which was a bit gruesome. p. unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Mon Apr 20 16:31:08 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id QAA17941 for ; Mon, 20 Apr 1998 16:31:05 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:10515 "EHLO vger.rutgers.edu" ident: "TIMEDOUT") by orava.funet.fi with ESMTP id <391605-29392>; Mon, 20 Apr 1998 18:27:51 +0300 Received: by vger.rutgers.edu id <971454-319>; Mon, 20 Apr 1998 11:07:50 -0400 Received: from [194.201.189.33] ([194.201.189.33]:32971 "HELO mailhost.cogency.co.uk" ident: "SOCKWRITE-65") by vger.rutgers.edu with SMTP id <971529-319>; Mon, 20 Apr 1998 10:52:36 -0400 Received: from cogency.co.uk by mailhost.cogency.co.uk (SMI-8.6/PIPEX simple 1.27) id PAA01804; Mon, 20 Apr 1998 15:57:52 +0100 Message-ID: <353B625B.752FEE2D@cogency.co.uk> Date: Mon, 20 Apr 1998 15:57:31 +0100 From: David Alan Gilbert Organization: Cogency Technology Inc X-Mailer: Mozilla 4.04 [en] (X11; I; SunOS 5.5 sun4m) MIME-Version: 1.0 To: Philip Blundell , linux-arm@vger.rutgers.edu Subject: Re: Installing on an 4 MB A5000 References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Philip Blundell wrote: > > >Nah - you partition from RiscOS using Russ's wonderful !PartMan - actually > >it would be nice to be able to make a swap partition from RISC OS as well. > > I've never managed to get PartMan to work, to be honest. By the way, I > couldn't find a mkswap binary on the ftp site that worked with 26-bit machines > - I ended up having to roll my own, which was a bit gruesome. Have a look on ftp.compsoc.man.ac.uk/pub/ARMLinux at mkswap.old and swapon.old (they are from way back in the 1.3.x days before they were broken :-) Dave -- ---------------------------------------------------------------------- - Dr. David Alan Gilbert - WARNING! This is a beta release .signature- - Work: dg @ cogency.co.uk - +44-(0)161-428-9444 - - Home: gro.gilbert @ treblig.org - - ---------------------------------------------------------------------- unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Mon Apr 20 18:17:37 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id SAA18422 for ; Mon, 20 Apr 1998 18:17:36 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:10515 "EHLO vger.rutgers.edu" ident: "TIMEDOUT") by orava.funet.fi with ESMTP id <391858-29392>; Mon, 20 Apr 1998 20:17:24 +0300 Received: by vger.rutgers.edu id <971337-319>; Mon, 20 Apr 1998 13:11:50 -0400 Received: from newmail.virgin.net ([194.168.54.44]:42138 "EHLO newmail.virgin.net" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <971373-319>; Mon, 20 Apr 1998 13:11:34 -0400 Received: from AV10114 ([150.204.96.114]) by newmail.virgin.net (Post.Office MTA v3.1.2 release (PO203-101c) ID# 1-55555U125000L125000S0) with SMTP id AAA25219 for ; Mon, 20 Apr 1998 17:16:14 +0000 Message-ID: <353BF399.236B@virgin.net> Date: Mon, 20 Apr 1998 18:17:13 -0700 From: JP Stanford X-Mailer: Mozilla 3.01 (Win16; I) MIME-Version: 1.0 To: linux-arm@vger.rutgers.edu Subject: ARM Linux and Castle SCSI cards Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Hi, Does anyone know if you can install arm linux on a scsi drive connected to a castle scsi card? If so how do you go about installing and partitioning the scsi drive? From John-Paul Stanford unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Mon Apr 20 18:21:37 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id SAA18434 for ; Mon, 20 Apr 1998 18:21:36 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:4345 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391844-29392>; Mon, 20 Apr 1998 20:20:38 +0300 Received: by vger.rutgers.edu id <971382-319>; Mon, 20 Apr 1998 13:15:31 -0400 Received: from beech.sucs.soton.ac.uk ([152.78.129.138]:34861 "EHLO beech.sucs.soton.ac.uk" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <971374-319>; Mon, 20 Apr 1998 13:14:06 -0400 Received: from bright.ecs.soton.ac.uk (bright.ecs.soton.ac.uk [152.78.64.201]) by beech.sucs.soton.ac.uk (8.8.8/relay-02) with SMTP id SAA06516; Mon, 20 Apr 1998 18:18:42 +0100 (BST) Received: from penelope.ecs.soton.ac.uk by bright.ecs.soton.ac.uk; Mon, 20 Apr 98 18:15:52 BST From: Russ King Message-Id: <5955.199804201723@penelope.ecs.soton.ac.uk> Subject: Re: Installing on an 4 MB A5000 To: pb@nexus.co.uk (Philip Blundell) Date: Mon, 20 Apr 1998 18:23:11 +0100 (BST) Cc: dmf20@hermes.cam.ac.uk, linux-arm@vger.rutgers.edu In-Reply-To: from "Philip Blundell" at Apr 20, 98 01:07:28 pm X-Mailer: ELM [version 2.4 PL24] Content-Type: text X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Philip Blundell writes: > rpm2cpio will strip the header off and leave you with a standard cpio archive. > I'm sure you could engineer your own program to do this if you wanted to. I > haven't looked at the rpm format in detail but I doubt it's hard - you might > even be able to do it with dd, od and a bit of shell script. Indeed, you can do it with standard unix commands - dd, gzip and cpio. I originally did want a slackware-like distribution, but after going through Slackware's install for about a total of 10 hours on various machines, I got very pissed off with it. When I used it, it required complete attention - no chance to do anything else for 2 hours. This is not what I call a viable distribution. RedHat on the other hand does things the *right* way - it asks you what you want to install and then installs it. Meanwhile, you can leave the machine to it's own devices.. The other point I'd like to make is that there was *absolutely* no way to create the boot disk for slackware on the A5000 - all the necessary binaries *just* *did* *not* *fit*. It was so bad, that I even contemplated writing a separate installer for it. Someone then suggested using RedHat, and initially I didn't like it. However, it's not as horrible as you might think it is. The package system is logical and far better than tar. It uses less memory space and disk space than slackware, all of which a 4MB machine needs. Dave's concerns on it not reporting errors I believe to be a bug - it now reports errors if it can't install everything. However the installer does have a few bugs in. Because of this bugs, I have disabled the automatic reboot in the init daemon, to prevent it looking although it's installed itself. As far as I am concerned, the differences between RedHat and slackware is the same as between RiscOS and Windoze. RedHat, like RiscOS performs it's function, and does it well (although I will admit to RedHat having bugs in). Slackware, like windowze, is less flexable once installed, and requires your complete attention at every stage. For all of these reasons, I would see time spent on the package front of Slackware as time wasted *until* some one fixes the root problem - how to get a machine up and running in 4MB. So could we stop this 'distribution war' and concentrate on the main problem please, rather than ranting on about irrelevent topics? Thankyou. (after I've found that I've got 400 mail messages in one day, with a 2MB mail folder, consisting of ***loads*** of failed mails because system admins just can't). unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Mon Apr 20 19:20:38 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id TAA18869 for ; Mon, 20 Apr 1998 19:20:36 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:4345 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <392163-29392>; Mon, 20 Apr 1998 21:18:37 +0300 Received: by vger.rutgers.edu id <971477-319>; Mon, 20 Apr 1998 14:00:39 -0400 Received: from tazenda.demon.co.uk ([158.152.220.239]:1965 "HELO kings-cross.london.uk.eu.org" ident: "exim") by vger.rutgers.edu with SMTP id <971524-319>; Mon, 20 Apr 1998 13:58:30 -0400 Received: from kings-cross.london.uk.eu.org [127.0.0.1] (phil) by kings-cross.london.uk.eu.org with esmtp (Exim 1.71 #1) id 0yRKq5-00028G-00; Mon, 20 Apr 1998 18:58:45 +0100 X-Mailer: exmh version 2.0zeta 7/24/97 To: Russ King cc: dmf20@hermes.cam.ac.uk, linux-arm@vger.rutgers.edu Subject: Re: Installing on an 4 MB A5000 In-reply-to: Your message of "Mon, 20 Apr 1998 18:23:11 BST." <5955.199804201723@penelope.ecs.soton.ac.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 20 Apr 1998 18:58:45 +0100 From: Philip Blundell Message-Id: X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO >Someone then suggested using RedHat, and initially I didn't like it. I remember it well. Interesting to see how you've seen the light now. :-) p. unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Mon Apr 20 23:37:56 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id XAA19937 for ; Mon, 20 Apr 1998 23:37:55 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:3053 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391184-15032>; Tue, 21 Apr 1998 01:35:41 +0300 Received: by vger.rutgers.edu id <971659-319>; Mon, 20 Apr 1998 17:38:27 -0400 Received: from snowcrash.cymru.net ([163.164.160.3]:1719 "EHLO snowcrash.cymru.net" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <971712-319>; Mon, 20 Apr 1998 17:07:06 -0400 Received: from caramon.armlinux.org (dialup2.cymru.net [163.164.160.66]) by snowcrash.cymru.net (8.8.7/8.7.1) with ESMTP id WAA03584 for ; Mon, 20 Apr 1998 22:12:28 +0100 Received: from raistlin.armlinux.org (raistlin [192.168.0.3]) by caramon.armlinux.org (8.7.4/8.7.3) with ESMTP id VAA26930 for ; Mon, 20 Apr 1998 21:46:30 +0100 From: Russell King - ARM Linux Admin Received: (from linux@localhost) by raistlin.armlinux.org (8.7.4/8.7.3) id VAA00571 for linux-arm@vger.rutgers.edu; Mon, 20 Apr 1998 21:23:29 +0100 Message-Id: <199804202023.VAA00571@raistlin.armlinux.org> Subject: *** IMPORTANT *** READ THIS NOW *** To: linux-arm@vger.rutgers.edu Date: Mon, 20 Apr 1998 21:23:28 +0100 (BST) X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Hi, For those of you who don't know, I am both subscribed to this list, and maintainer of the list. As such, I would like those of you who post to the list to take this into account. Since I am subscribed to the list, I do get it's emails. As such, there is no point in CC:'ing the mail to me at any of my addresses whatsoever. It does start to become a little annoying receiving 2 or maybe 3 copies of the same email... Thank you for your co-operation in this matter. _____ |_____| ------------------------------------------------- ---+---+- | | Russell King linux@arm.uk.linux.org --- --- | | | | http://www.arm.uk.linux.org/~rmk/armlinux.html / / | | +-+-+ --- -+- / | THE developer of ARM Linux |+| /|\ / | | | --- | +-+-+ ------------------------------------------------- /\\\ | unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Mon Apr 20 23:37:57 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id XAA19939 for ; Mon, 20 Apr 1998 23:37:56 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:3053 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391220-15032>; Tue, 21 Apr 1998 01:36:22 +0300 Received: by vger.rutgers.edu id <971670-319>; Mon, 20 Apr 1998 17:38:38 -0400 Received: from snowcrash.cymru.net ([163.164.160.3]:1788 "EHLO snowcrash.cymru.net" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <971716-319>; Mon, 20 Apr 1998 17:07:37 -0400 Received: from caramon.armlinux.org (dialup2.cymru.net [163.164.160.66]) by snowcrash.cymru.net (8.8.7/8.7.1) with ESMTP id WAA03595; Mon, 20 Apr 1998 22:12:38 +0100 Received: from raistlin.armlinux.org (raistlin [192.168.0.3]) by caramon.armlinux.org (8.7.4/8.7.3) with ESMTP id WAA27010; Mon, 20 Apr 1998 22:23:48 +0100 From: Russell King - ARM Linux Admin Received: (from linux@localhost) by raistlin.armlinux.org (8.7.4/8.7.3) id WAA01164; Mon, 20 Apr 1998 22:00:44 +0100 Message-Id: <199804202100.WAA01164@raistlin.armlinux.org> Subject: Re: Installing on an 4 MB A5000 To: 9606585c@udcf.gla.ac.uk (James Craig) Date: Mon, 20 Apr 1998 22:00:43 +0100 (BST) Cc: root@mad.scientist.com, linux@treblig.org, linux-arm@vger.rutgers.edu In-Reply-To: from "James Craig" at Apr 20, 98 10:44:00 am X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Hi. Maybe I ought to make my position quite clear on this subject: 1) I believe that RedHat offers the best compromise for ARM Linux. 2) The RedHat installer cannot be used on floppy to install on machines with 4MB of ram. With a future PartMan, it will be possible to copy a small partition image into the drive, thus allowing non-floppy based installation. 3) If someone wants to get Slackware going, then they are welcome to try, but IMHO it could be effort best spent in other areas of ARM Linux that are more lacking. 4) The 2.1.xx kernels currently are going to require a lot of changes to get them to run properly on the older architectures: a) The MEMC is quite unlike any other memory manager in the world, and doesn't fit any sensible memory management model that Linux could provide. Therefore, the MM is less than optimal. b) With the advent of ELF, and it's requirement to map the same data in two logical locations, this will require more modifications to the kernel. c) The large page size requires numerous modifications to the way memory is allocated in several parts of Linux. d) The kernel's malloc() routines need a hack to reduce the inherent overheads that it produces on large page machines. I'm not saying that it'll be impossible to do, but more that I'm beginning to wonder if the old machines are really worth the effort, and whether we're going to be able to continue integrating the architecture-specifics for these machines with obsolete processors into Linus' kernel source tree. 5) It is my intention to continue integrating the sources as is or as patches allow for these old machines. I shall be working on this area, but since my resources are required for other purposes, this will not happen with any great speed. If someone wishes to take over the kernel admin/hacking for the A5000, they are welcome. If you want to know more details about this, then please mail me. _____ |_____| ------------------------------------------------- ---+---+- | | Russell King linux@arm.uk.linux.org --- --- | | | | http://www.arm.uk.linux.org/~rmk/armlinux.html / / | | +-+-+ --- -+- / | THE developer of ARM Linux |+| /|\ / | | | --- | +-+-+ ------------------------------------------------- /\\\ | unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Mon Apr 20 23:37:59 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id XAA19942 for ; Mon, 20 Apr 1998 23:37:56 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:3053 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391256-15032>; Tue, 21 Apr 1998 01:36:41 +0300 Received: by vger.rutgers.edu id <971591-319>; Mon, 20 Apr 1998 17:38:54 -0400 Received: from snowcrash.cymru.net ([163.164.160.3]:1802 "EHLO snowcrash.cymru.net" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <971718-319>; Mon, 20 Apr 1998 17:07:52 -0400 Received: from caramon.armlinux.org (dialup2.cymru.net [163.164.160.66]) by snowcrash.cymru.net (8.8.7/8.7.1) with ESMTP id WAA03601; Mon, 20 Apr 1998 22:12:46 +0100 Received: from raistlin.armlinux.org (raistlin [192.168.0.3]) by caramon.armlinux.org (8.7.4/8.7.3) with ESMTP id VAA26945; Mon, 20 Apr 1998 21:52:04 +0100 From: Russell King - ARM Linux Admin Received: (from linux@localhost) by raistlin.armlinux.org (8.7.4/8.7.3) id VAA00591; Mon, 20 Apr 1998 21:29:04 +0100 Message-Id: <199804202029.VAA00591@raistlin.armlinux.org> Subject: Re: Installing on an 4 MB A5000 To: dmf20@hermes.cam.ac.uk (David) Date: Mon, 20 Apr 1998 21:29:03 +0100 (BST) Cc: alan@cymru.net, 9606585c@udcf.gla.ac.uk, linux@arm.uk.linux.org, root@mad.scientist.com, linux@treblig.org, linux-arm@vger.rutgers.edu In-Reply-To: from "David" at Apr 20, 98 12:26:18 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO David writes: > It's perhaps not the wisest thing to say but: I've installed NetBSD/arm32 > on my system and all the files there were .tar.gz. Which has two key > advantages for me: (a) I know instantly how to unpack the things (b) I can > unpack them under RISCOS if necessary. Also, for many things I haven't > noticed the need for awk, sed, perl etc. (But I can get them running > under RISCOS too, if I need to.) Hang on a moment - RiscOS may be able to unpack tars, but can it create them? Ok, it can create files that look like tar files, but can it handle the unix permissions? Can it handle the uid/gids? I don't think so. Therefore, to unpack a unix tar to use it on a unix system is a very silly thing to do... The slackware installation system uses awk, sed, strings and many others. (Slackware 2.3 was the version that I used btw, which indeed had a broken install). _____ |_____| ------------------------------------------------- ---+---+- | | Russell King linux@arm.uk.linux.org --- --- | | | | http://www.arm.uk.linux.org/~rmk/armlinux.html / / | | +-+-+ --- -+- / | THE developer of ARM Linux |+| /|\ / | | | --- | +-+-+ ------------------------------------------------- /\\\ | unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Mon Apr 20 23:48:50 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id XAA20054 for ; Mon, 20 Apr 1998 23:48:48 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:4878 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391230-15032>; Tue, 21 Apr 1998 01:46:34 +0300 Received: by vger.rutgers.edu id <971725-319>; Mon, 20 Apr 1998 17:39:06 -0400 Received: from snowcrash.cymru.net ([163.164.160.3]:1792 "EHLO snowcrash.cymru.net" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <971719-319>; Mon, 20 Apr 1998 17:08:05 -0400 Received: from caramon.armlinux.org (dialup2.cymru.net [163.164.160.66]) by snowcrash.cymru.net (8.8.7/8.7.1) with ESMTP id WAA03598; Mon, 20 Apr 1998 22:12:42 +0100 Received: from raistlin.armlinux.org (raistlin [192.168.0.3]) by caramon.armlinux.org (8.7.4/8.7.3) with ESMTP id VAA26901; Mon, 20 Apr 1998 21:36:24 +0100 From: Russell King - ARM Linux Admin Received: (from linux@localhost) by raistlin.armlinux.org (8.7.4/8.7.3) id VAA00529; Mon, 20 Apr 1998 21:13:24 +0100 Message-Id: <199804202013.VAA00529@raistlin.armlinux.org> Subject: Re: Installing on an 4 MB A5000 To: 9606585c@udcf.gla.ac.uk (James Craig) Date: Mon, 20 Apr 1998 21:13:22 +0100 (BST) Cc: linux@arm.uk.linux.org, root@mad.scientist.com, linux@treblig.org, linux-arm@vger.rutgers.edu In-Reply-To: from "James Craig" at Apr 20, 98 10:44:00 am X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO James Craig writes: > Yep, I know about this problem. It's a bit of a problem (understatement of > the century) but we're more likely to be able to manage when the installer > is just a tar+gunzip, instead of some weird format that stores huge heaps > of completely useless info. That's where you're wrong. You're talking about tar + gunzip + sh + installer. With RedHat, you've just got the installer and cpio. There's nothing more to it. > The slackware packages include the scripts too - If you look closely at a > slackware package, it contains a install/script file usually which handles > installation. Most slackware packages uninstall fairly cleanly too. The > other *BIG*, in fact *HUGE* advantage is that you can get away with no > installer whatsoever. The good old fashioned sash shell (stand-alone > shell) includes tar in it - and compiles perfectly for the ARM leaving > bucketloads of RAM free even on a 4M machine. You can just use it's > built-in tar command to unpack the slackware packages - it works great. And if you want to teach Joe Bloggs how to use it, then go ahead. But that's not the end of it - it's all well untaring, but what about these installation scripts you mentioned. You have mentioned that you can unpack the sources, but what about the next stage. > I don't really think people who are smart enough to get it to even *boot* > are going to have much trouble figuring out that the need to install perl > before they can use a perl script. Not to forget, of course, RPM is truly > hideous to use anyway. Trying to do anything other than one of the > predefined action is seriously fiddly, and you don't stand much change > of getting an RPM port to other OSs to let you modify packages before > installing them - e.g. seperating individual files out to fit them onto > 720K floppies.. Also - removing a package which other programs are using > isn't smart usually - but you need to do it to make a decent job of upgrading > versions on something. I'd much rather it just shut up and got on with it. :) Well then, you should just make sure you use the --nodeps option everytime. The dependencies are provided for people who may not realise the ramifications of the action that they are about to perform. I'd like to ask you a couple of questions at this point. Have you tried creating a distribution? Have you tried making a RedHat package? If the answer to these questions is no, then I would suggest that you try this *before* commenting on it, since you're trying to talk about a subject that you have only half if not less of the total facts about. You forget, I *have* tried the tar method. I *have* tried the RPM method. I know what a pain the two are. I know the problems that each one cause. I know the benefits of each. I know this because I have had *experience* of them both. As someone who has spent half a year creating the ARM Linux distribution, I really *KNOW* how difficult it is to create a distribution. It is not a trivial matter, and without the RPM system, it would not be here today. I *KNOW* where the problems lie. Now, stop arguing about something that is IRRELLEVENT and discuss SENSIBLY THE REAL PROBLEM AT HAND - how to get AN installer to work in 4MB, not I DON'T LIKE REDHAT AND THATS WHY IT DOESN'T WORK. > My commiserations to your friend, losing his soul to the dark side of > Linux. He's fallen in with evil companions. ;) Not at all - he was going to use a tar file, until I explained RPMs to him. He immediately saw the enormous advantages that a RPM gives you over a simple tar.gz file. _____ |_____| ------------------------------------------------- ---+---+- | | Russell King linux@arm.uk.linux.org --- --- | | | | http://www.arm.uk.linux.org/~rmk/armlinux.html / / | | +-+-+ --- -+- / | THE developer of ARM Linux |+| /|\ / | | | --- | +-+-+ ------------------------------------------------- /\\\ | unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Tue Apr 21 00:48:16 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id AAA20773 for ; Tue, 21 Apr 1998 00:48:14 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:10305 "EHLO vger.rutgers.edu" ident: "TIMEDOUT2") by orava.funet.fi with ESMTP id <392526-15032>; Tue, 21 Apr 1998 02:47:50 +0300 Received: by vger.rutgers.edu id <971519-319>; Mon, 20 Apr 1998 19:04:14 -0400 Received: from red.csi.cam.ac.uk ([131.111.8.70]:51647 "HELO red.csi.cam.ac.uk" ident: "exim") by vger.rutgers.edu with SMTP id <971621-319>; Mon, 20 Apr 1998 17:42:15 -0400 Received: from dmf20 by red.csi.cam.ac.uk with smtp (Exim 1.73 #1) id 0yROPc-000235-00; Mon, 20 Apr 1998 22:47:40 +0100 Date: Mon, 20 Apr 1998 22:47:40 +0100 (BST) From: David X-Sender: dmf20@red.csi.cam.ac.uk To: Russell King - ARM Linux Admin cc: alan@cymru.net, 9606585c@udcf.gla.ac.uk, root@mad.scientist.com, linux@treblig.org, linux-arm@vger.rutgers.edu Subject: Re: Installing on an 4 MB A5000 In-Reply-To: <199804202029.VAA00591@raistlin.armlinux.org> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO > Hang on a moment - RiscOS may be able to unpack tars, but can it create them? > Ok, it can create files that look like tar files, but can it handle the > unix permissions? Can it handle the uid/gids? I don't think so. > Therefore, to unpack a unix tar to use it on a unix system is a very > silly thing to do... Well, I don't want to follow this too much further, but RISCOS tar *can* create files, of sorts, but that's not the point. (It's got lots of really handy features too.) I can see what's in them, extract and examine individual files and so on. I don't loose any filenames or anything like that because I can put them in an X-files image. (In principle, I suppose I ought to be able to extract to my NetBSD partitions using UnixFS, but I haven't done it, and that might cause a problem with uid/gids - I'd have to reset them when I next booted NetBSD.) David. unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Tue Apr 21 02:02:37 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id CAA21164 for ; Tue, 21 Apr 1998 02:02:35 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:32303 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <392573-15032>; Tue, 21 Apr 1998 04:01:57 +0300 Received: by vger.rutgers.edu id <971805-319>; Mon, 20 Apr 1998 19:06:17 -0400 Received: from snowcrash.cymru.net ([163.164.160.3]:1701 "EHLO snowcrash.cymru.net" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <971914-319>; Mon, 20 Apr 1998 18:46:14 -0400 Received: (from alan@localhost) by snowcrash.cymru.net (8.8.7/8.7.1) id XAA05456; Mon, 20 Apr 1998 23:51:55 +0100 From: Alan Cox Message-Id: <199804202251.XAA05456@snowcrash.cymru.net> Subject: Re: Installing on an 4 MB A5000t To: linux@arm.uk.linux.org (Russell King - ARM Linux Admin) Date: Mon, 20 Apr 1998 23:51:53 +0100 (BST) Cc: 9606585c@udcf.gla.ac.uk, root@mad.scientist.com, linux@treblig.org, linux-arm@vger.rutgers.edu In-Reply-To: <199804202100.WAA01164@raistlin.armlinux.org> from "Russell King - ARM Linux Admin" at Apr 20, 98 10:00:43 pm Content-Type: text X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO > 2) The RedHat installer cannot be used on floppy to install on machines > with 4MB of ram. With a future PartMan, it will be possible to copy > a small partition image into the drive, thus allowing non-floppy > based installation. There is also the SGI and MacLinux installer work which uses libext2fs under the native OS to unpack a base cpio for the install system (making it do RPM is on the job list) > a) The MEMC is quite unlike any other memory manager in the world, > and doesn't fit any sensible memory management model that Linux > could provide. Therefore, the MM is less than optimal. DaveM is having a think on the diffs btw > I'm beginning to wonder if the old machines are really worth the > effort, and whether we're going to be able to continue integrating > the architecture-specifics for these machines with obsolete > processors into Linus' kernel source tree. Rephrase that to 'who with an old ARM cares enough to fix it' ;) Alan unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Tue Apr 21 05:54:53 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id FAA22086 for ; Tue, 21 Apr 1998 05:54:52 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:15942 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <392591-15032>; Tue, 21 Apr 1998 07:53:24 +0300 Received: by vger.rutgers.edu id <971584-319>; Tue, 21 Apr 1998 00:10:07 -0400 Received: from odie.barnet.ac.uk ([194.82.202.98]:23053 "EHLO odie.barnet.ac.uk" ident: "willy") by vger.rutgers.edu with ESMTP id <972210-319>; Mon, 20 Apr 1998 23:24:16 -0400 Received: (from willy@localhost) by odie.barnet.ac.uk (8.8.6/8.8.6) id EAA21711; Tue, 21 Apr 1998 04:30:56 +0100 From: Matthew Wilcox Message-Id: <199804210330.EAA21711@odie.barnet.ac.uk> Subject: Re: Installing on an 4 MB A5000 To: linux@arm.uk.linux.org (Russell King - ARM Linux Admin) Date: Tue, 21 Apr 1998 04:30:55 +0100 (BST) Cc: linux-arm@vger.rutgers.edu In-Reply-To: <199804202100.WAA01164@raistlin.armlinux.org> from "Russell King - ARM Linux Admin" at Apr 20, 98 10:00:43 pm X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Russell King - ARM Linux Admin > > Hi. > > Maybe I ought to make my position quite clear on this subject: > > 1) I believe that RedHat offers the best compromise for ARM Linux. That's your belief and you're entitled to hold it. Since you created the distribution, others should accept that. I still like the idea of creating a Debian distribution, but I think that time spent working on that would be wasted before we have libc 6. > 2) The RedHat installer cannot be used on floppy to install on machines > with 4MB of ram. With a future PartMan, it will be possible to copy > a small partition image into the drive, thus allowing non-floppy > based installation. That will be really sexy. > 4) The 2.1.xx kernels currently are going to require a lot of changes > to get them to run properly on the older architectures: > a) The MEMC is quite unlike any other memory manager in the world, > and doesn't fit any sensible memory management model that Linux > could provide. Therefore, the MM is less than optimal. You can say that again! > b) With the advent of ELF, and it's requirement to map the same > data in two logical locations, this will require more > modifications to the kernel. I *thought* it had been agreed that that wasn't actually necessary, it was merely the way the x86 decided to do it? > c) The large page size requires numerous modifications to the > way memory is allocated in several parts of Linux. Yes, I noticed that in the patches. Are there not also issues on Alpha's with their 8k page sizes? I guess they tend to have so much more memory that this is a moot point. > d) The kernel's malloc() routines need a hack to reduce the > inherent overheads that it produces on large page machines. > I'm not saying that it'll be impossible to do, but more that > I'm beginning to wonder if the old machines are really worth the > effort, and whether we're going to be able to continue integrating > the architecture-specifics for these machines with obsolete > processors into Linus' kernel source tree. > > 5) It is my intention to continue integrating the sources as is or as > patches allow for these old machines. I shall be working on this > area, but since my resources are required for other purposes, this > will not happen with any great speed. If someone wishes to take > over the kernel admin/hacking for the A5000, they are welcome. If > you want to know more details about this, then please mail me. IMO, it would make more sense to separate out the port into an arm26 and an arm32. Yeah, I know this is a really bad time to mention it :-) Of course, there's nothing to stop arm32 machines executing arm26 binaries. This might encourage someone to take responsibility for the arm26 tree off your hands. -- Set Alias$Case Set Alias$[ |||| |MSet Alias$Otherwise Set Alias$[ \ Matthew "" |MSet Alias$When If %0=%%0 Then Set Alias$[ "" ||MIf %0=%%0 \ Wilcox Then Set Alias$Otherwise Set Alias$[ |||||||||||||||| ||MIf \ %0=%%0 Then Set Alias$When Set Alias$[ |||||||||||||||| unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Tue Apr 21 06:28:49 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id GAA22176 for ; Tue, 21 Apr 1998 06:28:48 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:15942 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <392661-15032>; Tue, 21 Apr 1998 08:25:17 +0300 Received: by vger.rutgers.edu id <971096-319>; Tue, 21 Apr 1998 00:52:51 -0400 Received: from odie.barnet.ac.uk ([194.82.202.98]:23115 "EHLO odie.barnet.ac.uk" ident: "willy") by vger.rutgers.edu with ESMTP id <971093-319>; Tue, 21 Apr 1998 00:16:51 -0400 Received: (from willy@localhost) by odie.barnet.ac.uk (8.8.6/8.8.6) id FAA21878; Tue, 21 Apr 1998 05:23:36 +0100 From: Matthew Wilcox Message-Id: <199804210423.FAA21878@odie.barnet.ac.uk> Subject: Re: fd0* device for ADFS shape discs under i386 Linux To: alan@cymru.net (Alan Cox) Date: Tue, 21 Apr 1998 05:23:35 +0100 (BST) Cc: ajb85@cam.ac.uk, linux-arm@vger.rutgers.edu In-Reply-To: <199804172058.VAA08472@snowcrash.cymru.net> from "Alan Cox" at Apr 17, 98 09:58:05 pm X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Alan Cox > > > Someone should look at other filing systems to see how they cope with > > filesystems that don't provide nice inodes. When Dickon Hood & I were > > experimenting with NFS servers under RISC OS, we had a structure that > > worked well; it was a tree, and the inode we returned was the address > > within that list. You've got me intrigued now, I might just do it. > > Several schemes are used. One of the cleanest if you stick to 2.1.8x+ > is to issue inode numbers as you put things in the dcache and revoke > them as your inode is flushed. All open file inodes are in the dcache. I don't have a problem with something being only valid for 2.1.8x+ linux, after all 2.2 will be soon enough. However, would this scheme not screw NFS rather badly? I thought inodes had to persist for the lifetime of the file. NFS has the rather nasty requirement of having to preserve them over reboots. [I've implemented an NFS server for RISC OS. It is nasty. You may see it released rsn if certain other things happen.] -- Set Alias$Case Set Alias$[ |||| |MSet Alias$Otherwise Set Alias$[ \ Matthew "" |MSet Alias$When If %0=%%0 Then Set Alias$[ "" ||MIf %0=%%0 \ Wilcox Then Set Alias$Otherwise Set Alias$[ |||||||||||||||| ||MIf \ %0=%%0 Then Set Alias$When Set Alias$[ |||||||||||||||| unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Tue Apr 21 06:42:43 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id GAA22225 for ; Tue, 21 Apr 1998 06:42:41 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:15942 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <392767-15032>; Tue, 21 Apr 1998 08:39:27 +0300 Received: by vger.rutgers.edu id <971025-319>; Tue, 21 Apr 1998 00:58:52 -0400 Received: from odie.barnet.ac.uk ([194.82.202.98]:23127 "EHLO odie.barnet.ac.uk" ident: "willy") by vger.rutgers.edu with ESMTP id <971223-319>; Tue, 21 Apr 1998 00:29:44 -0400 Received: (from willy@localhost) by odie.barnet.ac.uk (8.8.6/8.8.6) id FAA21957; Tue, 21 Apr 1998 05:36:31 +0100 From: Matthew Wilcox Message-Id: <199804210436.FAA21957@odie.barnet.ac.uk> Subject: Re: fd0* device for ADFS shape discs under i386 Linux To: 9606585c@udcf.gla.ac.uk Date: Tue, 21 Apr 1998 05:36:30 +0100 (BST) Cc: linux-arm@vger.rutgers.edu In-Reply-To: from "James Craig" at Apr 20, 98 10:57:37 am X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO James Craig > Yeah, I can quite definately see the problem. However, could we store a > hash table of recently-accessed-inodes and just cache their directory > info? It's not like you can't search and ADFS filesystem for a particular > ID, it just takes a while. For small enough numbers of files, (say a > floppy) you could do it at mount/umount time. Here's the system I used for TermiteNFS. It was optimal for that application; I'm not sure which operations are applied to Linux's inodes: pointer to parent pointer to younger sibling pointer to first child leafname The inode is simply the offset within the block. You can find the full pathname of a file in O(log n), and you can enumerate a directory in a straight-forward manner. Renaming an object does not change its inode and requires changing 3 pointers. Directory information is already cached in the dcache, it's a matter of being able to find it efficiently when it's no longer there. Now, if the ADFS filesystem is not modified over a reboot, this data could be stored on disc and we've managed to fulfill the requirements of NFS. Yuck. > Truly. I've written ADFS file readers and writers before, but I've never > tried integrating one into a totally different OS's idea of a filing > system. In the meantime, we could probably fairly quickly code up an ADFS > equivalent of the mtools for DOS - The old IscaFS that Phil Norman wrote > would do the job just fine if we could manage to fiddle it to run under > Linux. :) Uhh.. IscaFS reads ext2 filesystems from RISC OS. So does ExeterFS (IscaFS is an image filesystem and is extremely sexy. ExeterFS is more like Spark and has a number of bugs.) -- Set Alias$Case Set Alias$[ |||| |MSet Alias$Otherwise Set Alias$[ \ Matthew "" |MSet Alias$When If %0=%%0 Then Set Alias$[ "" ||MIf %0=%%0 \ Wilcox Then Set Alias$Otherwise Set Alias$[ |||||||||||||||| ||MIf \ %0=%%0 Then Set Alias$When Set Alias$[ |||||||||||||||| unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Tue Apr 21 08:01:27 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id IAA22505 for ; Tue, 21 Apr 1998 08:01:26 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:20629 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <392683-15032>; Tue, 21 Apr 1998 09:59:04 +0300 Received: by vger.rutgers.edu id <970790-319>; Tue, 21 Apr 1998 02:44:26 -0400 Received: from odie.barnet.ac.uk ([194.82.202.98]:23624 "EHLO odie.barnet.ac.uk" ident: "willy") by vger.rutgers.edu with ESMTP id <970822-319>; Tue, 21 Apr 1998 02:34:19 -0400 Received: (from willy@localhost) by odie.barnet.ac.uk (8.8.6/8.8.6) id HAA22434 for linux-arm@vger.rutgers.edu; Tue, 21 Apr 1998 07:41:20 +0100 From: Matthew Wilcox Message-Id: <199804210641.HAA22434@odie.barnet.ac.uk> Subject: strrchr To: linux-arm@vger.rutgers.edu Date: Tue, 21 Apr 1998 07:41:17 +0100 (BST) X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Assuming (1) I've understood the code right; (2) K&R haven't misdocumented the behaviour of strrchr and (3) Linux strrchr is supposed to be the same as ANSI C strrchr, then the current implementation of strrchr has an out by one error, which leads me to ask if anything actually uses it? I have most of a strrchr drafted, but it currently suffers from a nasty end-effect. -- Set Alias$Case Set Alias$[ |||| |MSet Alias$Otherwise Set Alias$[ \ Matthew "" |MSet Alias$When If %0=%%0 Then Set Alias$[ "" ||MIf %0=%%0 \ Wilcox Then Set Alias$Otherwise Set Alias$[ |||||||||||||||| ||MIf \ %0=%%0 Then Set Alias$When Set Alias$[ |||||||||||||||| unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From pb@nexus.co.uk Tue Apr 21 09:38:49 1998 Return-Path: Received: from globe (nexusel.demon.co.uk [158.152.30.195]) by odie.barnet.ac.uk (8.8.6/8.8.6) with SMTP id JAA22745 for ; Tue, 21 Apr 1998 09:38:43 +0100 Received: from (spring.nexus.co.uk) [192.0.0.3] (root) by globe with smtp (Exim 1.82 #1) id 0yRYXy-0001r3-00; Tue, 21 Apr 1998 09:36:58 +0100 Received: from localhost (spring.nexus.co.uk) [127.0.0.1] (pb) by spring.nexus.co.uk with esmtp (Exim 1.82 #1) id 0yRYXx-0005mv-00; Tue, 21 Apr 1998 09:36:57 +0100 X-Mailer: exmh version 2.0zeta 7/24/97 To: Matthew Wilcox cc: linux-arm@vger.rutgers.edu Subject: Re: strrchr In-reply-to: Your message of "Tue, 21 Apr 1998 07:41:17 BST." <199804210641.HAA22434@odie.barnet.ac.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 21 Apr 1998 09:36:56 +0100 From: Philip Blundell Message-Id: Status: RO >Assuming (1) I've understood the code right; (2) K&R haven't misdocumented >the behaviour of strrchr and (3) Linux strrchr is supposed to be the >same as ANSI C strrchr, then the current implementation of strrchr has >an out by one error, which leads me to ask if anything actually uses it? Possibly not. I certainly can't think of anything vital off the top of my head. Do the versions for other architectures behave the same? p. From owner-linux-arm-outgoing@vger.rutgers.edu Tue Apr 21 09:53:08 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id JAA22815 for ; Tue, 21 Apr 1998 09:53:06 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:29512 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390589-15032>; Tue, 21 Apr 1998 11:53:03 +0300 Received: by vger.rutgers.edu id <970791-319>; Tue, 21 Apr 1998 04:45:03 -0400 Received: from lenzie.cent.gla.ac.uk ([130.209.30.11]:44355 "EHLO lenzie.cent.gla.ac.uk" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <970790-319>; Tue, 21 Apr 1998 04:44:35 -0400 Received: from localhost (9606585c@localhost) by lenzie.cent.gla.ac.uk (8.8.4/8.8.8) with SMTP id JAA09815; Tue, 21 Apr 1998 09:51:53 +0100 (BST) Date: Tue, 21 Apr 1998 09:51:52 +0100 (BST) From: James Craig <9606585c@udcf.gla.ac.uk> X-Sender: 9606585c@lenzie.cent.gla.ac.uk To: Alan Cox cc: linux-arm@vger.rutgers.edu Subject: Re: Installing on an 4 MB A5000t In-Reply-To: <199804202251.XAA05456@snowcrash.cymru.net> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO On Mon, 20 Apr 1998, Alan Cox wrote: > > 2) The RedHat installer cannot be used on floppy to install on machines > > with 4MB of ram. With a future PartMan, it will be possible to copy > > a small partition image into the drive, thus allowing non-floppy > > based installation. > There is also the SGI and MacLinux installer work which uses libext2fs > under the native OS to unpack a base cpio for the install system > (making it do RPM is on the job list) I must admit, I'd thought about using a RISC OS application to do something along the same lines. This would fix both the RH and Slackware problems to the point that we could use either distribution without any great fuss. In fact, while we're at it, .deb support would be nice too. :) > > a) The MEMC is quite unlike any other memory manager in the world, > > and doesn't fit any sensible memory management model that Linux > > could provide. Therefore, the MM is less than optimal. > > DaveM is having a think on the diffs btw Yeah, it's seriously weird. 32K pages are a nuisance and a half. > > I'm beginning to wonder if the old machines are really worth the > > effort, and whether we're going to be able to continue integrating > > the architecture-specifics for these machines with obsolete > > processors into Linus' kernel source tree. > Rephrase that to 'who with an old ARM cares enough to fix it' ;) Well, I don't *have* a new ARM, so I don't have a choice in the matter - if I want to use ARM linux at all, I need old ARM support. Guess I get to help. Roll on the summer holidays. :) -- James Craig <9606585c@udcf.gla.ac.uk> unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From 9606585c@udcf.gla.ac.uk Tue Apr 21 09:56:48 1998 Return-Path: <9606585c@udcf.gla.ac.uk> Received: from lenzie.cent.gla.ac.uk (lenzie.cent.gla.ac.uk [130.209.30.11]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id JAA22835 for ; Tue, 21 Apr 1998 09:56:47 +0100 Received: from localhost (9606585c@localhost) by lenzie.cent.gla.ac.uk (8.8.4/8.8.8) with SMTP id JAA12376; Tue, 21 Apr 1998 09:56:58 +0100 (BST) Date: Tue, 21 Apr 1998 09:56:56 +0100 (BST) From: James Craig <9606585c@udcf.gla.ac.uk> X-Sender: 9606585c@lenzie.cent.gla.ac.uk To: Matthew Wilcox cc: linux-arm@vger.rutgers.edu Subject: Re: fd0* device for ADFS shape discs under i386 Linux In-Reply-To: <199804210436.FAA21957@odie.barnet.ac.uk> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Status: RO On Tue, 21 Apr 1998, Matthew Wilcox wrote: > James Craig > > Yeah, I can quite definately see the problem. However, could we store a > > hash table of recently-accessed-inodes and just cache their directory > > info? It's not like you can't search and ADFS filesystem for a particular > > ID, it just takes a while. For small enough numbers of files, (say a > > floppy) you could do it at mount/umount time. > > Truly. I've written ADFS file readers and writers before, but I've never > > tried integrating one into a totally different OS's idea of a filing > > system. In the meantime, we could probably fairly quickly code up an ADFS > > equivalent of the mtools for DOS - The old IscaFS that Phil Norman wrote > > would do the job just fine if we could manage to fiddle it to run under > > Linux. :) > > Uhh.. IscaFS reads ext2 filesystems from RISC OS. So does ExeterFS > (IscaFS is an image filesystem and is extremely sexy. ExeterFS is more > like Spark and has a number of bugs.) Yep, sorry all, I was seriously confused when I wrote that bit. I'll just go shoot myself. ;) I think the next best thing I have is a set of old ADFS E-format readers/writers I wrote for the BBC B+ in BASIC. Time to get to work on them again. :) BTW, anyone know - ARM linux OK with 4 floppy drives on an A410/1? > -- > Set Alias$Case Set Alias$[ |||| |MSet Alias$Otherwise Set Alias$[ \ Matthew > "" |MSet Alias$When If %0=%%0 Then Set Alias$[ "" ||MIf %0=%%0 \ Wilcox > Then Set Alias$Otherwise Set Alias$[ |||||||||||||||| ||MIf \ > %0=%%0 Then Set Alias$When Set Alias$[ |||||||||||||||| Must try tying that in sometime, and seeing what happens. :) -- James Craig <9606585c@udcf.gla.ac.uk> From 9606585c@udcf.gla.ac.uk Tue Apr 21 10:04:25 1998 Return-Path: <9606585c@udcf.gla.ac.uk> Received: from lenzie.cent.gla.ac.uk (lenzie.cent.gla.ac.uk [130.209.30.11]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id KAA22884 for ; Tue, 21 Apr 1998 10:04:23 +0100 Received: from localhost (9606585c@localhost) by lenzie.cent.gla.ac.uk (8.8.4/8.8.8) with SMTP id KAA16097; Tue, 21 Apr 1998 10:04:37 +0100 (BST) Date: Tue, 21 Apr 1998 10:04:34 +0100 (BST) From: James Craig <9606585c@udcf.gla.ac.uk> X-Sender: 9606585c@lenzie.cent.gla.ac.uk To: Matthew Wilcox cc: linux-arm@vger.rutgers.edu Subject: Re: Installing on an 4 MB A5000 In-Reply-To: <199804210330.EAA21711@odie.barnet.ac.uk> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Status: RO On Tue, 21 Apr 1998, Matthew Wilcox wrote: > Russell King - ARM Linux Admin > > > > Hi. > > > > Maybe I ought to make my position quite clear on this subject: > > > > 1) I believe that RedHat offers the best compromise for ARM Linux. > > That's your belief and you're entitled to hold it. Since you created the > distribution, others should accept that. I still like the idea of creating > a Debian distribution, but I think that time spent working on that would > be wasted before we have libc 6. Possibly, although I'm not sure. I can't think of any reason why slackware or Red Hat should be any different in that respect. :) > > 2) The RedHat installer cannot be used on floppy to install on machines > > with 4MB of ram. With a future PartMan, it will be possible to copy > > a small partition image into the drive, thus allowing non-floppy > > based installation. > > That will be really sexy. Yep, very nice idea. That would solve a *lot* of problems with the setup. > > 4) The 2.1.xx kernels currently are going to require a lot of changes > > to get them to run properly on the older architectures: > > a) The MEMC is quite unlike any other memory manager in the world, > > and doesn't fit any sensible memory management model that Linux > > could provide. Therefore, the MM is less than optimal. > > You can say that again! Yep. :) > > b) With the advent of ELF, and it's requirement to map the same > > data in two logical locations, this will require more > > modifications to the kernel. > > I *thought* it had been agreed that that wasn't actually necessary, it was > merely the way the x86 decided to do it? Likewise, but hey, I don't understand ELF all that well anyway. :) > > c) The large page size requires numerous modifications to the > > way memory is allocated in several parts of Linux. > > Yes, I noticed that in the patches. Are there not also issues on Alpha's > with their 8k page sizes? I guess they tend to have so much more memory > that this is a moot point. Truly. > > d) The kernel's malloc() routines need a hack to reduce the > > inherent overheads that it produces on large page machines. > > I'm not saying that it'll be impossible to do, but more that > > I'm beginning to wonder if the old machines are really worth the > > effort, and whether we're going to be able to continue integrating > > the architecture-specifics for these machines with obsolete > > processors into Linus' kernel source tree. Leaving support for large page sizes in the kernel would be a good move, since it increases the range of different CPUs that linux *could* be ported to with relatively little work. > > 5) It is my intention to continue integrating the sources as is or as > > patches allow for these old machines. I shall be working on this > > area, but since my resources are required for other purposes, this > > will not happen with any great speed. If someone wishes to take > > over the kernel admin/hacking for the A5000, they are welcome. If > > you want to know more details about this, then please mail me. I'm in no position to help at present, sadly. :( > IMO, it would make more sense to separate out the port into an arm26 and > an arm32. Yeah, I know this is a really bad time to mention it :-) Of > course, there's nothing to stop arm32 machines executing arm26 binaries. > This might encourage someone to take responsibility for the arm26 tree > off your hands. Interesting idea, but I suspect that it would lead to a complete halt on progress on the arm26 architecture. At least with the two together, we still get progress - if arm26 and arm32 were seperate, we'd have to update the arm26 code seperately from the arm32 stuff, even if the changes were applicable to both architectures. -- James Craig <9606585c@udcf.gla.ac.uk> From alan@snowcrash.cymru.net Tue Apr 21 10:32:13 1998 Return-Path: Received: from snowcrash.cymru.net (snowcrash.cymru.net [163.164.160.3]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id KAA22952 for ; Tue, 21 Apr 1998 10:32:10 +0100 Received: (from alan@localhost) by snowcrash.cymru.net (8.8.7/8.7.1) id KAA14726; Tue, 21 Apr 1998 10:32:27 +0100 From: Alan Cox Message-Id: <199804210932.KAA14726@snowcrash.cymru.net> Subject: Re: fd0* device for ADFS shape discs under i386 Linux To: willy@odie.barnet.ac.uk (Matthew Wilcox) Date: Tue, 21 Apr 1998 10:32:25 +0100 (BST) Cc: alan@cymru.net, ajb85@cam.ac.uk, linux-arm@vger.rutgers.edu In-Reply-To: <199804210423.FAA21878@odie.barnet.ac.uk> from "Matthew Wilcox" at Apr 21, 98 05:23:35 am Content-Type: text Status: RO > after all 2.2 will be soon enough. However, would this scheme not screw > NFS rather badly? I thought inodes had to persist for the lifetime of the > file. NFS has the rather nasty requirement of having to preserve them > over reboots. [I've implemented an NFS server for RISC OS. It is nasty. > You may see it released rsn if certain other things happen.] So you cant export them back out via NFS (ditto for SMB back out via NFS already). Is that a big issue ? From owner-linux-arm-outgoing@vger.rutgers.edu Tue Apr 21 12:05:54 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id MAA23271 for ; Tue, 21 Apr 1998 12:05:53 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:28787 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390843-15032>; Tue, 21 Apr 1998 14:04:10 +0300 Received: by vger.rutgers.edu id <970798-319>; Tue, 21 Apr 1998 06:56:00 -0400 Received: from lenzie.cent.gla.ac.uk ([130.209.30.11]:39442 "EHLO lenzie.cent.gla.ac.uk" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <970886-319>; Tue, 21 Apr 1998 06:52:38 -0400 Received: from localhost (9606585c@localhost) by lenzie.cent.gla.ac.uk (8.8.4/8.8.8) with SMTP id KAA00881; Tue, 21 Apr 1998 10:31:40 +0100 (BST) Date: Tue, 21 Apr 1998 10:31:35 +0100 (BST) From: James Craig <9606585c@udcf.gla.ac.uk> X-Sender: 9606585c@lenzie.cent.gla.ac.uk To: Russell King - ARM Linux Admin cc: linux-arm@vger.rutgers.edu Subject: Re: Installing on an 4 MB A5000 In-Reply-To: <199804202013.VAA00529@raistlin.armlinux.org> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO On Mon, 20 Apr 1998, Russell King - ARM Linux Admin wrote: > James Craig writes: > > Yep, I know about this problem. It's a bit of a problem (understatement of > > the century) but we're more likely to be able to manage when the installer > > is just a tar+gunzip, instead of some weird format that stores huge heaps > > of completely useless info. > > That's where you're wrong. You're talking about tar + gunzip + sh + installer. > With RedHat, you've just got the installer and cpio. There's nothing more > to it. Try ps -ax. I wish it were that simple. RH tends to run a lot of other junk too. I admit, slackware uses a lot more seperate processes though. > > The slackware packages include the scripts too - If you look closely at a > > slackware package, it contains a install/script file usually which handles > > installation. Most slackware packages uninstall fairly cleanly too. The > > other *BIG*, in fact *HUGE* advantage is that you can get away with no > > installer whatsoever. The good old fashioned sash shell (stand-alone > > shell) includes tar in it - and compiles perfectly for the ARM leaving > > bucketloads of RAM free even on a 4M machine. You can just use it's > > built-in tar command to unpack the slackware packages - it works great. > > And if you want to teach Joe Bloggs how to use it, then go ahead. But that's > not the end of it - it's all well untaring, but what about these installation > scripts you mentioned. You have mentioned that you can unpack the sources, > but what about the next stage. The total effort required to run the scripts is: cd /mnt .. install/install Doesn't seem very difficult, even for Joe Bloggs. Not to forget, as I pointed out, getting ARM Linux on a 4M machine right now is so damn difficult, anyone even attempting it is probably going to manage that, even if I have to write a HOWTO on it. > > I don't really think people who are smart enough to get it to even *boot* > > are going to have much trouble figuring out that the need to install perl > > before they can use a perl script. Not to forget, of course, RPM is truly > > hideous to use anyway. Trying to do anything other than one of the > > predefined action is seriously fiddly, and you don't stand much change > > of getting an RPM port to other OSs to let you modify packages before > > installing them - e.g. seperating individual files out to fit them onto > > 720K floppies.. Also - removing a package which other programs are using > > isn't smart usually - but you need to do it to make a decent job of upgrading > > versions on something. I'd much rather it just shut up and got on with it. :) > > Well then, you should just make sure you use the --nodeps option everytime. > The dependencies are provided for people who may not realise the ramifications > of the action that they are about to perform. > I'd like to ask you a couple of questions at this point. Have you tried > creating a distribution? Have you tried making a RedHat package? Yes to the first. I've recompiled slackware at various points, created slack boot disks from scratch etc. (i.e. recompiling everything that goes on the disk, figuring out what I need on it, and putting it all together with my own install scripts) I've never created an RH package because I don't use RH. At all. Whatsoever. I tried it, really didn't like it, abandoned it. I've created debian packages in the past, although I've never actually had much reason to use it. I know one thing - creating a slackware .tar.gz is a lot easier than creating a RH .RPM. tar cvfz . isn't a big chore. > If the answer to these questions is no, then I would suggest that you try > this *before* commenting on it, since you're trying to talk about a subject > that you have only half if not less of the total facts about. I do know about doing this, in fact, I've even installed slackware on a 2 meg machine in the past. Having no RAM free is something I'm well used to, and I have to distribute modified IRC servers and bots around QuakeNet (www.quakenet.eu.org) where I'm an IRCop. Most of the other IRCops aren't too hot at linux installation, so I need to package them in a way they can *all* handle. RPM, they can't. .tar.gz causes no problems. > You forget, I *have* tried the tar method. I *have* tried the RPM method. > I know what a pain the two are. I know the problems that each one cause. > I know the benefits of each. I know this because I have had *experience* > of them both. I've tried the tar method, and looked at the RPM method for about 3 hours getting nothing but crashes, and RPM complaining about it's databases. I gave in at that point since there's no reason for me to use a packaging method that excludes anyone who *isn't* using Red Hat (I've tried installing RPM on a slackware machine, and frankly, it's a pain in the butt.) > As someone who has spent half a year creating the ARM Linux distribution, > I really *KNOW* how difficult it is to create a distribution. It is not > a trivial matter, and without the RPM system, it would not be here today. > I *KNOW* where the problems lie. I congratulate you on doing as well as you have. The current distribution seems to work extremely well on machines with more than 4 meg of RAM, and I'm quite seriously impressed with the amount of work you've put in. We all owe you a lot for that. > Now, stop arguing about something that is IRRELLEVENT and discuss SENSIBLY > THE REAL PROBLEM AT HAND - how to get AN installer to work in 4MB, not > I DON'T LIKE REDHAT AND THATS WHY IT DOESN'T WORK. I *have* been discussing this seriously. The bottom line is that on an ix86, slackware can install on a 2 meg machine. (I can show an i386sx40 with 2 meg running Linux as proof if anyone doesn't believe me.) On an ix86, RH can't install (easily) on less than an 8 meg machine. I think it's fairly clear which one uses less RAM on the ix86 architecture. I know the ARM memory management is a bit (OK, very) different, and the overheads on processes are rather heavier, but I'd have to say the difference between the two memory footprints is rather large. In summary, the reason I think that a slackware distribution might be a good thing (TM) are: i) Slackware has a much smaller memory footprint on the ix86, which implies that it *may* also have a smaller footprint on the ARM machines, helping with the 4M problem. ii) Slackware's packages can be installed manually more easily, and can be modified on OSs other than Linux if the need arises. They can also be split up and otherwise modified a little more readily than RPMs. iii) A minimum slackware installation is rather smaller than a minimum Red Hat installation. A slackware install including GCC can fit in about 20 meg (assuming you don't have the current rather huge kernel sources in there). The last time I tried Red Hat (correct me if this figure is wrong, I haven't used RH in quite some time) the smallest with that setup was around the 80 meg mark. > > My commiserations to your friend, losing his soul to the dark side of > > Linux. He's fallen in with evil companions. ;) > > Not at all - he was going to use a tar file, until I explained RPMs to > him. He immediately saw the enormous advantages that a RPM gives you > over a simple tar.gz file. It *was* a joke, Russ, there's no need to be so irritable over it. > -- James Craig <9606585c@udcf.gla.ac.uk> unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Tue Apr 21 12:13:12 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id MAA23288 for ; Tue, 21 Apr 1998 12:13:11 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:4219 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390791-15032>; Tue, 21 Apr 1998 14:13:20 +0300 Received: by vger.rutgers.edu id <970790-319>; Tue, 21 Apr 1998 07:05:06 -0400 Received: from snowcrash.cymru.net ([163.164.160.3]:1554 "EHLO snowcrash.cymru.net" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <970791-319>; Tue, 21 Apr 1998 07:04:52 -0400 Received: (from alan@localhost) by snowcrash.cymru.net (8.8.7/8.7.1) id MAA16513; Tue, 21 Apr 1998 12:12:28 +0100 From: Alan Cox Message-Id: <199804211112.MAA16513@snowcrash.cymru.net> Subject: Re: Installing on an 4 MB A5000 To: 9606585c@udcf.gla.ac.uk (James Craig) Date: Tue, 21 Apr 1998 12:12:25 +0100 (BST) Cc: linux@arm.uk.linux.org, linux-arm@vger.rutgers.edu In-Reply-To: from "James Craig" at Apr 21, 98 10:31:35 am Content-Type: text X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO > created debian packages in the past, although I've never actually had much > reason to use it. I know one thing - creating a slackware .tar.gz is a lot > easier than creating a RH .RPM. tar cvfz . isn't a big > chore. cd /usr/src/redhat rpm -ba SPECS/* builds a full Red Hat 8) unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Tue Apr 21 13:00:38 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id NAA23781 for ; Tue, 21 Apr 1998 13:00:36 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:4219 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390947-15032>; Tue, 21 Apr 1998 14:57:13 +0300 Received: by vger.rutgers.edu id <970798-319>; Tue, 21 Apr 1998 07:45:01 -0400 Received: from red.csi.cam.ac.uk ([131.111.8.70]:34663 "HELO red.csi.cam.ac.uk" ident: "exim") by vger.rutgers.edu with SMTP id <970923-319>; Tue, 21 Apr 1998 07:43:08 -0400 Received: from dmf20 by red.csi.cam.ac.uk with smtp (Exim 1.73 #1) id 0yRb8t-0000uU-00; Tue, 21 Apr 1998 12:23:15 +0100 Date: Tue, 21 Apr 1998 12:23:15 +0100 (BST) From: David X-Sender: dmf20@red.csi.cam.ac.uk To: Alan Cox cc: James Craig <9606585c@udcf.gla.ac.uk>, linux@arm.uk.linux.org, linux-arm@vger.rutgers.edu Subject: Re: Installing on an 4 MB A5000 In-Reply-To: <199804211112.MAA16513@snowcrash.cymru.net> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO > cd /usr/src/redhat > rpm -ba SPECS/* > > builds a full Red Hat 8) If you happen to have "rpm" in your path. Which I'm fairly sure I don't. I know *exactly* where to find and how to use tar, though. David. unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From dmf20@hermes.cam.ac.uk Tue Apr 21 13:26:53 1998 Return-Path: Received: from red.csi.cam.ac.uk (exim@red.csi.cam.ac.uk [131.111.8.70]) by odie.barnet.ac.uk (8.8.6/8.8.6) with SMTP id NAA23913 for ; Tue, 21 Apr 1998 13:26:51 +0100 Received: from dmf20 by red.csi.cam.ac.uk with smtp (Exim 1.73 #1) id 0yRc8k-0001js-00; Tue, 21 Apr 1998 13:27:10 +0100 Date: Tue, 21 Apr 1998 13:27:10 +0100 (BST) From: David X-Sender: dmf20@red.csi.cam.ac.uk To: Matthew Wilcox cc: linux-arm@vger.rutgers.edu Subject: Re: Installing on an 4 MB A5000 In-Reply-To: <199804211212.NAA23845@odie.barnet.ac.uk> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Status: RO > > If you happen to have "rpm" in your path. Which I'm fairly sure I don't. > > I know *exactly* where to find and how to use tar, though. > > I'm sorry, what is your point? If you have a redhat setup then it is > significantly easier to maintain and manage than a slackware setup. > How many slackware-based machines do you manage? Looking after more > than 2 is a real hassle. My point is: I don't really care what the name of the distribution is. I like to know what's going onto my system. I don't have ARMLinux - it won't support my hardware. All I know is that if I wish to examine, and/or use the contents of part of distribution, then I know with absolute certainty (without even thinking about it) how to extract a .tgz and how to control which directory it goes in and so on and that I can trivially create them on any UNIX system. I'm not saying that there's anything wrong with any particular distribution or install system. What I am saying is that a particular benefit of .tgzs is that it is *obvious* what you can do with it. I might also suggest that the system used by NetBSD, IMHO: (a) couldn't be easier (b) quite definitely uses .tar.gzs. (And we note, doesn't require a boot floppy of any sort, although it probably won't work as a solution to the 4Mb problem.) David. PS - Please can we start a comp.dist.tar.advocacy newsgroup?? ;-) From owner-linux-arm-outgoing@vger.rutgers.edu Tue Apr 21 13:42:24 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id NAA24044 for ; Tue, 21 Apr 1998 13:42:20 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:22680 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391184-15032>; Tue, 21 Apr 1998 15:40:05 +0300 Received: by vger.rutgers.edu id <970799-319>; Tue, 21 Apr 1998 08:31:38 -0400 Received: from odie.barnet.ac.uk ([194.82.202.98]:26171 "EHLO odie.barnet.ac.uk" ident: "willy") by vger.rutgers.edu with ESMTP id <970935-319>; Tue, 21 Apr 1998 08:23:45 -0400 Received: (from willy@localhost) by odie.barnet.ac.uk (8.8.6/8.8.6) id NAA23939 for linux-arm@vger.rutgers.edu; Tue, 21 Apr 1998 13:30:25 +0100 From: Matthew Wilcox Message-Id: <199804211230.NAA23939@odie.barnet.ac.uk> Subject: improved strlen.S To: linux-arm@vger.rutgers.edu Date: Tue, 21 Apr 1998 13:30:20 +0100 (BST) X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Thanks to Mark Wooding for providing me with the improved detect-zero-byte code. I found a bug in the pre-word case. This should be suitable for linux but is /not/ suitable for libc as it does not have the correct behaviour when the character being searched for is the zero byte. 44 instructions, but 2 fewer in the inner loop, and a couple removed from the direct path in. Anyway: strchr: stmfd r13!, {r14} tst r0, #3 beq 2f 1: ldrb r2, [r0], #1 cmp r2, r1 cmpne r2, #0 tstne r0, #3 bne 1b bcs 6f 2: orr r1, r1, r1, LSL #8 orr r1, r1, r1, LSL #16 mov r14, #1 orr r14, r14, r14, LSL #8 orr r14, r14, r14, LSL #16 3: ldr r2, [r0], #4 eor r3, r2, r1 sub ip, r3, r14 bic ip, ip, r3 tst ip, r14, lsl #7 beq 5f sub ip, r2, r14 bic ip, ip, r2 tst ip, r14, lsl #7 bne 3b b 7f 5: tst r2, #&000000ff beq 7f tst r3, #&000000ff subeq r0, r0, #1 tstne r3, #&0000ff00 subeq r0, r0, #2 ldmeqfd r13!, {pc} tst r2, #&0000ff00 tstne r2, #&00ff0000 beq 7f tst r3, #&00ff0000 subeq r0, r0, #1 tstne r3, #&ff000000 ldmeqfd r13!, {pc} b 7f 6: cmp r2, #0 subne r0, r0, #1 7: moveq r0, #0 ldmfd r13!, {pc} -- Set Alias$Case Set Alias$[ |||| |MSet Alias$Otherwise Set Alias$[ \ Matthew "" |MSet Alias$When If %0=%%0 Then Set Alias$[ "" ||MIf %0=%%0 \ Wilcox Then Set Alias$Otherwise Set Alias$[ |||||||||||||||| ||MIf \ %0=%%0 Then Set Alias$When Set Alias$[ |||||||||||||||| unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Tue Apr 21 13:46:29 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id NAA24073 for ; Tue, 21 Apr 1998 13:46:27 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:22680 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391125-15032>; Tue, 21 Apr 1998 15:45:50 +0300 Received: by vger.rutgers.edu id <970798-319>; Tue, 21 Apr 1998 08:37:45 -0400 Received: from beech.sucs.soton.ac.uk ([152.78.129.138]:64085 "EHLO beech.sucs.soton.ac.uk" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <970791-319>; Tue, 21 Apr 1998 08:37:21 -0400 Received: from bright.ecs.soton.ac.uk (bright.ecs.soton.ac.uk [152.78.64.201]) by beech.sucs.soton.ac.uk (8.8.8/relay-02) with SMTP id NAA13655; Tue, 21 Apr 1998 13:44:33 +0100 (BST) Received: from penelope.ecs.soton.ac.uk by bright.ecs.soton.ac.uk; Tue, 21 Apr 98 13:41:42 BST From: Russ King Message-Id: <8938.199804211249@penelope.ecs.soton.ac.uk> Subject: Re: Installing on an 4 MB A5000 To: 9606585c@udcf.gla.ac.uk (James Craig) Date: Tue, 21 Apr 1998 13:49:06 +0100 (BST) Cc: linux-arm@vger.rutgers.edu In-Reply-To: from "James Craig" at Apr 21, 98 10:31:35 am X-Mailer: ELM [version 2.4 PL24] Content-Type: text X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO James Craig writes: > i) Slackware has a much smaller memory footprint on the ix86, which > implies that it *may* also have a smaller footprint on the ARM machines, > helping with the 4M problem. This I'd like to see - the RedHat install disk can be as small as 720k. With the slackware install disks, I believe that they require all sorts of binaries, which I found was extremely difficult to fit on a 1.44MB floppy on the ARM systems, even with trying shared or static libraries. This is more a problem on the older Archimedes machines, where you can *only* have 720k disks. This is where the RedHat installer has it's advantage - it *is* smaller than the total space required by the slackware installer. > ii) Slackware's packages can be installed manually more easily, and can be > modified on OSs other than Linux if the need arises. They can also be > split up and otherwise modified a little more readily than RPMs. They can't be split up properly under RiscOS - as I'm trying to point out elsewhere, although you *can* read tars, and *can* write tars, any attempt to do so under a non-unix operating system will cause valuable information to be lost. As such, I don't see this as an issue. > iii) A minimum slackware installation is rather smaller than a minimum Red > Hat installation. A slackware install including GCC can fit in about 20 > meg (assuming you don't have the current rather huge kernel sources in > there). The last time I tried Red Hat (correct me if this figure is wrong, > I haven't used RH in quite some time) the smallest with that setup was > around the 80 meg mark. The minimun RedHat install is actually 40MB, not your grossly overrated 80MB mark. However, rather than comparing the minimum size, we ought to be qualifing it with what is *in* that installation. After all, we're comparing apples and oranges here. > It *was* a joke, Russ, there's no need to be so irritable over it. ;) PS, I've removed myself from the CC: line *because* I'm on the mailing list! unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Tue Apr 21 13:52:01 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id NAA24131 for ; Tue, 21 Apr 1998 13:52:00 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:22680 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390640-15032>; Tue, 21 Apr 1998 15:50:11 +0300 Received: by vger.rutgers.edu id <970798-319>; Tue, 21 Apr 1998 08:41:59 -0400 Received: from beech.sucs.soton.ac.uk ([152.78.129.138]:64500 "EHLO beech.sucs.soton.ac.uk" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <970796-319>; Tue, 21 Apr 1998 08:41:41 -0400 Received: from bright.ecs.soton.ac.uk (bright.ecs.soton.ac.uk [152.78.64.201]) by beech.sucs.soton.ac.uk (8.8.8/relay-02) with SMTP id NAA14618 for ; Tue, 21 Apr 1998 13:49:08 +0100 (BST) Received: from penelope.ecs.soton.ac.uk by bright.ecs.soton.ac.uk; Tue, 21 Apr 98 13:46:18 BST From: Russ King Message-Id: <9194.199804211253@penelope.ecs.soton.ac.uk> Subject: *** RE-READ THIS *** To: linux-arm@vger.rutgers.edu Date: Tue, 21 Apr 1998 13:53:42 +0100 (BST) X-Mailer: ELM [version 2.4 PL24] Content-Type: text X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Look 'ere you lot. I *am* definitely subscribed to the list. Do not include me in the CC: on *any* mailings whatsoever, *even* if you're replying to one already on the mailing list if you're sending your mail to the list. This goes for *all* mails to the list. I *don't* want to download twice as much email each night. The amount of redundant mail I'm getting is proving to be more a problem than Internet Spam!!! Please STOP IT NOW, or else I shall cause these CC:'d mails to bounce straight back to you. I've tried asking nicely, but I'm afraid that didn't work. Thank you. (Very hacked off). unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Tue Apr 21 13:56:42 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id NAA24153 for ; Tue, 21 Apr 1998 13:56:41 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:22680 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390970-15032>; Tue, 21 Apr 1998 15:55:44 +0300 Received: by vger.rutgers.edu id <970800-319>; Tue, 21 Apr 1998 08:47:32 -0400 Received: from odie.barnet.ac.uk ([194.82.202.98]:26436 "EHLO odie.barnet.ac.uk" ident: "willy") by vger.rutgers.edu with ESMTP id <970790-319>; Tue, 21 Apr 1998 08:47:15 -0400 Received: (from willy@localhost) by odie.barnet.ac.uk (8.8.6/8.8.6) id NAA24057; Tue, 21 Apr 1998 13:43:32 +0100 From: Matthew Wilcox Message-Id: <199804211243.NAA24057@odie.barnet.ac.uk> Subject: Re: Installing on an 4 MB A5000 To: dmf20@hermes.cam.ac.uk (David) Date: Tue, 21 Apr 1998 13:43:30 +0100 (BST) Cc: linux-arm@vger.rutgers.edu In-Reply-To: from "David" at Apr 21, 98 01:27:10 pm X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO David > My point is: I don't really care what the name of the distribution is. I > like to know what's going onto my system. I don't have ARMLinux - it > won't support my hardware. All I know is that if I wish to examine, > and/or use the contents of part of distribution, then I know with > absolute certainty (without even thinking about it) how to extract a .tgz > and how to control which directory it goes in and so on and that I can > trivially create them on any UNIX system. rpm is available in source form. I believe it compiles on NetBSD for one. You know what's in the rpm because there's an option to rpm to tell you. Even better, it has a description, and a pgp signature from the author. But all this is missing your fundamentally wrong assumption which is that a distribution is prepared in order for you to be able to pick it apart and use it on an alien system. > I'm not saying that there's anything wrong with any particular > distribution or install system. What I am saying is that a particular > benefit of .tgzs is that it is *obvious* what you can do with it. And I'm saying this is not an advantage. > PS - Please can we start a comp.dist.tar.advocacy newsgroup?? ;-) This discussion should certainly not be happening on linux-arm. Maybe one of the linux advocacy groups would be better. -- Set Alias$Case Set Alias$[ |||| |MSet Alias$Otherwise Set Alias$[ \ Matthew "" |MSet Alias$When If %0=%%0 Then Set Alias$[ "" ||MIf %0=%%0 \ Wilcox Then Set Alias$Otherwise Set Alias$[ |||||||||||||||| ||MIf \ %0=%%0 Then Set Alias$When Set Alias$[ |||||||||||||||| unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Tue Apr 21 14:55:53 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id OAA24390 for ; Tue, 21 Apr 1998 14:55:51 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:25165 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390965-15032>; Tue, 21 Apr 1998 16:52:46 +0300 Received: by vger.rutgers.edu id <970803-319>; Tue, 21 Apr 1998 09:38:02 -0400 Received: from odie.barnet.ac.uk ([194.82.202.98]:26734 "EHLO odie.barnet.ac.uk" ident: "willy") by vger.rutgers.edu with ESMTP id <970928-319>; Tue, 21 Apr 1998 09:33:49 -0400 Received: (from willy@localhost) by odie.barnet.ac.uk (8.8.6/8.8.6) id OAA24310; Tue, 21 Apr 1998 14:40:58 +0100 From: Matthew Wilcox Message-Id: <199804211340.OAA24310@odie.barnet.ac.uk> Subject: first cut at strrchr.S To: linux-arm@vger.rutgers.edu Date: Tue, 21 Apr 1998 14:40:56 +0100 (BST) Cc: mdw@ebi.ac.uk X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO This includes the improvement that mdw told me about with respect to strchr. It's 53 instructions, a little more complicated than strchr. But if it's worth optimising strchr, it's worth optimising strrchr since it's likely to have to churn through more data. Again, this is untested and I'd welcome testers. Does anyone object to me posting these bits of assembler? Would you prefer I placed them on ftp.barnet.ac.uk and only announced them here? -- Set Alias$Case Set Alias$[ |||| |MSet Alias$Otherwise Set Alias$[ \ Matthew "" |MSet Alias$When If %0=%%0 Then Set Alias$[ "" ||MIf %0=%%0 \ Wilcox Then Set Alias$Otherwise Set Alias$[ |||||||||||||||| ||MIf \ %0=%%0 Then Set Alias$When Set Alias$[ |||||||||||||||| strrchr: stmfd r13!, {r4, r14} mov ip, #0 tst r0, #3 beq 3f 1: ldrb r2, [r0], #1 teq r2, r1 subeq ip, r0, #1 cmp r2, #0 tstne r0, #3 bne 1b bcs 5f 3: orr r1, r1, r1, LSL #8 orr r1, r1, r1, LSL #16 mov r4, #1 orr r4, r4, r4, LSL #8 orr r4, r4, r4, LSL #16 4: ldr r2, [r0], #4 eor r3, r2, r1 sub r14, r3, r4 bic r14, r14, r3 tst r14, r4, lsl #7 moveq r3, ip subeq ip, r0, #4 sub r14, r2, r4 bic r14, ip, r2 tst r14, r4, lsl #7 bne 4b 5: teq ip, #0 beq 6f tst ip, #3 bne 6f ldr r2, [ip] eor r0, r2, r1 tst r2, #&000000ff tstne r2, #&0000ff00 orreq r0, r0, #&000ff000 tstne r2, #&00ff0000 orreq r0, r0, #&ff000000 mvn r2, #0 tst r0, #&ff000000 addeq r2, r2, #1 tstne r0, #&00ff0000 addeq r2, r2, #1 tstne r0, #&0000ff00 addeq r2, r2, #1 tstne r0, #&000000ff addeqs r2, r2, #1 addge r0, ip, r2 ldmgefd r13!, {r4, pc} mov ip, r3 b 5b 6: mov r0, ip mov pc, r14 unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Tue Apr 21 21:41:35 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id VAA26139 for ; Tue, 21 Apr 1998 21:41:34 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:18545 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391106-15032>; Tue, 21 Apr 1998 23:38:54 +0300 Received: by vger.rutgers.edu id <970791-319>; Tue, 21 Apr 1998 16:27:00 -0400 Received: from odie.barnet.ac.uk ([194.82.202.98]:31431 "EHLO odie.barnet.ac.uk" ident: "willy") by vger.rutgers.edu with ESMTP id <970929-319>; Tue, 21 Apr 1998 16:26:34 -0400 Received: (from willy@localhost) by odie.barnet.ac.uk (8.8.6/8.8.6) id VAA26095 for linux-arm@vger.rutgers.edu; Tue, 21 Apr 1998 21:34:11 +0100 From: Matthew Wilcox Message-Id: <199804212034.VAA26095@odie.barnet.ac.uk> Subject: StrongARM + Ether3 To: linux-arm@vger.rutgers.edu Date: Tue, 21 Apr 1998 21:34:10 +0100 (BST) X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO I've just upgraded my RiscPC to a StrongARM (hooray ;-) However, the Ether3 module no longer works. It complains about not being able to determine the width of the podule bus. So I recompiled the kernel, telling the configure script to compile for a StrongARM instead of an ARM6/7. This had the same problem. Then I put the ARM610 back in, the original kernel worked fine, and the strongarm one died horribly with the arm610 ;-) This is 2.0.33-19980120. Should 2.1.97 work out of the box? -- Set Alias$Case Set Alias$[ |||| |MSet Alias$Otherwise Set Alias$[ \ Matthew "" |MSet Alias$When If %0=%%0 Then Set Alias$[ "" ||MIf %0=%%0 \ Wilcox Then Set Alias$Otherwise Set Alias$[ |||||||||||||||| ||MIf \ %0=%%0 Then Set Alias$When Set Alias$[ |||||||||||||||| unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Wed Apr 22 18:14:41 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id SAA32097 for ; Wed, 22 Apr 1998 18:14:35 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:28713 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391244-2280>; Wed, 22 Apr 1998 20:14:35 +0300 Received: by vger.rutgers.edu id <971106-9428>; Wed, 22 Apr 1998 12:58:29 -0400 Received: from cptsg4.univ-mrs.fr ([139.124.7.104]:22766 "HELO cptsg4" ident: "TIMEDOUT") by vger.rutgers.edu with SMTP id <971120-9428>; Wed, 22 Apr 1998 12:55:03 -0400 Received: from cptsg4 (localhost [127.0.0.1]) by cptsg4 (950413.SGI.8.6.12/950213.SGI.AUTOCF) via SMTP id TAA11788 for ; Wed, 22 Apr 1998 19:02:05 +0200 Message-ID: <353E228D.41C6@cpt.univ-mrs.fr> Date: Wed, 22 Apr 1998 19:02:05 +0200 From: Vincent PENNE Organization: Centre de Physique Theorique de Marseille X-Mailer: Mozilla 2.01S (X11; I; IRIX64 6.2 IP28) MIME-Version: 1.0 To: linux-arm@vger.rutgers.edu Subject: libstdc++ References: <9194.199804211253@penelope.ecs.soton.ac.uk> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO I've tried to compile libg++-2.7. It works fine with small programm (like hello world, or a little bigger : playsid (which works very fine !)) But when I try more complex programm, I have big problem : for example wxWindows library. Actually, it works as far as I do not need the standard C library, which make me think that there is a conflict between the two. Actually it would be very understandable since libc is a version less than 2.7 (2.6 I think ?) Does anybody have already compiled a working c++ library ? Thank Vincent. /==========================================================\ | Vincent PENNE, Centre de Physique Theorique | |Tel. (0033)4 91 26 95 04 CNRS - LUMINY - Case 907 | |Fax. (0033)4 91 26 95 53 F-13288 MARSEILLE CEDEX 9 | |Email: penne@cpt.univ-mrs.fr FRANCE | |http://www.cpt.univ-mrs.fr | \==========================================================/ unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From linux@arm.uk.linux.org Wed Apr 22 22:07:43 1998 Return-Path: Received: from snowcrash.cymru.net (snowcrash.cymru.net [163.164.160.3]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id WAA00796 for ; Wed, 22 Apr 1998 22:07:41 +0100 Received: from caramon.armlinux.org (dialup1.cymru.net [163.164.160.65]) by snowcrash.cymru.net (8.8.7/8.7.1) with ESMTP id WAA17379; Wed, 22 Apr 1998 22:08:11 +0100 Received: from raistlin.armlinux.org (raistlin [192.168.0.3]) by caramon.armlinux.org (8.7.4/8.7.3) with ESMTP id UAA28972; Wed, 22 Apr 1998 20:21:51 +0100 From: Russell King - ARM Linux Admin Received: (from linux@localhost) by raistlin.armlinux.org (8.7.4/8.7.3) id TAA00403; Wed, 22 Apr 1998 19:54:38 +0100 Message-Id: <199804221854.TAA00403@raistlin.armlinux.org> Subject: Re: StrongARM + Ether3 To: willy@odie.barnet.ac.uk (Matthew Wilcox) Date: Wed, 22 Apr 1998 19:54:37 +0100 (BST) Cc: linux-arm@vger.rutgers.edu In-Reply-To: <199804212034.VAA26095@odie.barnet.ac.uk> from "Matthew Wilcox" at Apr 21, 98 09:34:10 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Status: RO Matthew Wilcox writes: > I've just upgraded my RiscPC to a StrongARM (hooray ;-) However, the > Ether3 module no longer works. It complains about not being able to > determine the width of the podule bus. Aha, we may be getting somewhere with those Ether3/EtherB cards that don't work here. This is telling me that there is something different when a StrongARM is used. The StrongARM doesn't replicate a byte write across all byte lanes, and as such, the test in the Ether3/EtherB driver is failing to detect the podule bus width correctly. As such, it is reporting an error which prevents the driver from recognising the ether chip. Next weekend's patches should allow many more people to use their Ether3/EtherB cards... _____ |_____| ------------------------------------------------- ---+---+- | | Russell King linux@arm.uk.linux.org --- --- | | | | http://www.arm.uk.linux.org/~rmk/armlinux.html / / | | +-+-+ --- -+- / | THE developer of ARM Linux |+| /|\ / | | | --- | +-+-+ ------------------------------------------------- /\\\ | From owner-linux-arm-outgoing@vger.rutgers.edu Sat Apr 25 14:09:40 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id OAA15383 for ; Sat, 25 Apr 1998 14:09:38 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:17311 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390517-23961>; Sat, 25 Apr 1998 16:10:24 +0300 Received: by vger.rutgers.edu id <971016-29806>; Sat, 25 Apr 1998 08:50:11 -0400 Received: from methan.in-berlin.de ([160.45.10.13]:4103 "HELO methan.in-berlin.de" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with SMTP id <971453-29806>; Sat, 25 Apr 1998 08:49:54 -0400 Received: by methan.in-berlin.de (Smail3.2.0.98) from bolzen.in-berlin.de (194.94.235.1) with esmtp id ; Sat, 25 Apr 1998 15:09:20 +0200 (MET DST) Received: by bolzen.in-berlin.de (Smail3.2) from yoda with uucp id ; Sat, 25 Apr 1998 15:09:04 +0200 (MET DST) Received: by yoda.in-berlin.de (Smail3.1.28.1 #0) id m0ySrrb-000000C; Fri, 24 Apr 98 21:26 GMT Date: Fri, 24 Apr 1998 23:17:03 +0200 From: Stefan Voss To: linux-arm@vger.rutgers.edu Subject: Re: PowerTec initialization problem Message-ID: <1cedfd3b48%voss@yoda.in-berlin.de> In-Reply-To: <199804192200.XAA04139@raistlin.armlinux.org> Organization: Individual Network (IN), Berlin, Germany X-Mailer: Messenger v1.34c for RISC OS MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Posting-Agent: RISC OS Newsbase 0.60d X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO In message <199804192200.XAA04139@raistlin.armlinux.org>, Russell King wrote: > Stefan Voss writes: > > I compiled a new kernel with PowerTec SCSI support from the latest release > > of the kernel patches (April 13th). > > > > When booting this kernel my PowerTec SCSI II card is recognized ok but > > shortly after that the following error message is displayed: > > > > scsi: aborting command due to timeout: pid 0, scsi0, channel 0, id 0, Test > > Unit Ready 00 00 00 00 > > scsi0: fas126_abort: command executing > > SCSI host 0 abort (pid 0) timed out resetting > > SCSI bus is being reset for host 0 channel 0 > > scsi0: fas126_reset > > > > The above messages are repeated in an endless loop thus hanging the boot > > process. > > I have this week been working on the powertec code, and have just released > new patches. Please wait a couple of days before downloading them since I > haven't actually tested these ones (time ran out...). If you see nothing > on the mailing list saying they're bad by Tuesday, then try them out.... OK, I waited until Friday before getting the patches from April 19th :) Unfortunately a kernel compiled with these patches show the same result (see above) just a little more verbosely. > | | Russell King linux@arm.uk.linux.org --- --- Regards, Stefan -- Stefan Voss (voss@yoda.in-berlin.de) unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Mon Apr 27 22:45:05 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id WAA29282 for ; Mon, 27 Apr 1998 22:45:04 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:17122 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390945-23948>; Tue, 28 Apr 1998 00:44:13 +0300 Received: by vger.rutgers.edu id <970985-4553>; Mon, 27 Apr 1998 17:25:45 -0400 Received: from later.CS.Berkeley.EDU ([128.32.37.29]:4502 "EHLO later.CS.Berkeley.EDU" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <971030-4553>; Mon, 27 Apr 1998 17:25:12 -0400 Received: from localhost (rmartin@localhost) by later.CS.Berkeley.EDU (8.8.3/8.6.9) with SMTP id OAA10605 for ; Mon, 27 Apr 1998 14:38:53 -0700 (PDT) X-Authentication-Warning: later.CS.Berkeley.EDU: rmartin owned process doing -bs Date: Mon, 27 Apr 1998 14:38:52 -0700 (PDT) From: Rich Martin To: linux-arm@vger.rutgers.edu Subject: Bootp for the EBSA 110? Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO We have a few of the EBSA 110 boards that we would like to boot Linux on. Using the bootp that came with DEC's SDT we can get 3 packets returned to the board via tftp before bootp croaks. How are others with the EBSA getting the kernel on? I've also gotten wind of alternate versions of bootp for the EBSA. Can someone point me to them? Thanks -- -Rich unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Tue Apr 28 15:35:03 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id PAA32745 for ; Tue, 28 Apr 1998 15:35:02 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:11475 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391822-23948>; Tue, 28 Apr 1998 17:32:54 +0300 Received: by vger.rutgers.edu id <970941-8838>; Tue, 28 Apr 1998 10:13:48 -0400 Received: from alushta.NL.net ([193.78.240.22]:4455 "EHLO alushta.NL.net" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <971150-8838>; Tue, 28 Apr 1998 10:10:52 -0400 Received: from ichfsnl by alushta.NL.net with UUCP id <14392-15555>; Tue, 28 Apr 1998 16:27:31 +0200 Received: from ichsun5.noname (ichsun5 [192.3.2.14]) by ichfsnl.ich.nl (8.6.12/8.6.12) with SMTP id QAA17165 for ; Tue, 28 Apr 1998 16:34:16 +0200 Received: by ichsun5.noname (4.1/SMI-4.1) id AA29962; Tue, 28 Apr 98 16:24:09 +0200 Message-Id: <19980428162409.A29953@ich.nl> Date: Tue, 28 Apr 1998 16:24:09 +0200 From: Hans Bogaards To: ARM Linux Mailing List Subject: Installing with 2 floppy drives Mail-Followup-To: ARM Linux Mailing List Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.92.1i X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Hello, I'm stilll trying to get Linux installed on my A5000 (Yes, I'm stubborn) I now have two floppy drives and they work o.k. under Risc OS. My idea was to make linux mount a floppy-disk in one drive as 'root' and a floppy in the other drive as 'supplemental' (with the correct images of course) But when I try this by giving the kernel the followoing options: root=/dev/fd1 ramdisk_size=0 The kernel fails to mount the root-disk and gives the following error (approx, because I don't have my arch at hand) end_request: I/O error. failure to read device 02:01 My question is now, why can't linux find the second floppy drive and is it possible to do it this way (maybe with a changed kernel or root-disk or with other kernel options) -- Tot mails! "You can't help respecting someone Hans who can spell Tuesday, even if he spells it wrong" Winnie the Pooh unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Tue Apr 28 21:07:46 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id VAA01494 for ; Tue, 28 Apr 1998 21:07:45 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:28950 "EHLO vger.rutgers.edu" ident: "TIMEDOUT") by orava.funet.fi with ESMTP id <390844-23948>; Tue, 28 Apr 1998 23:05:18 +0300 Received: by vger.rutgers.edu id <970941-8838>; Tue, 28 Apr 1998 15:46:31 -0400 Received: from snowcrash.cymru.net ([163.164.160.3]:1516 "EHLO snowcrash.cymru.net" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <971186-8838>; Tue, 28 Apr 1998 15:45:21 -0400 Received: from caramon.armlinux.org (dialup4.cymru.net [163.164.160.68]) by snowcrash.cymru.net (8.8.7/8.7.1) with ESMTP id VAA05656; Tue, 28 Apr 1998 21:03:23 +0100 Received: from raistlin.armlinux.org (raistlin [192.168.0.3]) by caramon.armlinux.org (8.7.4/8.7.3) with ESMTP id UAA01606; Tue, 28 Apr 1998 20:10:54 +0100 From: Russell King - ARM Linux Admin Received: (from linux@localhost) by raistlin.armlinux.org (8.7.4/8.7.3) id UAA00344; Tue, 28 Apr 1998 20:10:40 +0100 Message-Id: <199804281910.UAA00344@raistlin.armlinux.org> Subject: Re: Installing with 2 floppy drives To: hansb@ich.nl (Hans Bogaards) Date: Tue, 28 Apr 1998 20:10:39 +0100 (BST) Cc: linux-arm@vger.rutgers.edu In-Reply-To: <19980428162409.A29953@ich.nl> from "Hans Bogaards" at Apr 28, 98 04:24:09 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Hans Bogaards writes: > My question is now, why can't linux find the second floppy drive and is it > possible to do it this way (maybe with a changed kernel or root-disk or with > other kernel options) RiscOS is a bit sneeky when it boots up - if you have 1 drive configured, it checks each physical drive position starting at zero to find the drive. If you have two, then they can be in positions 1 and 3 for example. However, Linux *expects* then to be configured so that drive 0 is position 0, drive 1 is position 1 etc, where: position 0 = slot 0 drive 0 position 1 = slot 0 drive 1 position 2 = slot 1 drive 0 position 3 = slot 1 drive 1 Unfortunately, the Linux floppy driver as it stands (it's getting to be rather unmaintainable and is due for a re-write soon) is not able to perform the detection of the drive position, and as such, they must be configured correctly. _____ |_____| ------------------------------------------------- ---+---+- | | Russell King linux@arm.uk.linux.org --- --- | | | | http://www.arm.uk.linux.org/~rmk/armlinux.html / / | | +-+-+ --- -+- / | THE developer of ARM Linux |+| /|\ / | | | --- | +-+-+ ------------------------------------------------- /\\\ | unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Tue Apr 28 21:41:21 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id VAA01851 for ; Tue, 28 Apr 1998 21:41:20 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:2223 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391813-23948>; Tue, 28 Apr 1998 23:41:14 +0300 Received: by vger.rutgers.edu id <971203-8838>; Tue, 28 Apr 1998 16:22:02 -0400 Received: from snowcrash.cymru.net ([163.164.160.3]:1427 "EHLO snowcrash.cymru.net" ident: "TIMEDOUT") by vger.rutgers.edu with ESMTP id <971214-8838>; Tue, 28 Apr 1998 16:21:38 -0400 Received: from caramon.armlinux.org (dialup4.cymru.net [163.164.160.68]) by snowcrash.cymru.net (8.8.7/8.7.1) with ESMTP id VAA06110; Tue, 28 Apr 1998 21:39:41 +0100 Received: from raistlin.armlinux.org (raistlin [192.168.0.3]) by caramon.armlinux.org (8.7.4/8.7.3) with ESMTP id VAA01887; Tue, 28 Apr 1998 21:39:55 +0100 From: Russell King - ARM Linux Admin Received: (from linux@localhost) by raistlin.armlinux.org (8.7.4/8.7.3) id VAA02379; Tue, 28 Apr 1998 21:39:36 +0100 Message-Id: <199804282039.VAA02379@raistlin.armlinux.org> Subject: Re: Armlinux 2.0.33 compile fails To: glennr@es.co.nz Date: Tue, 28 Apr 1998 21:39:36 +0100 (BST) Cc: linux-arm@vger.rutgers.edu In-Reply-To: <19980428.210433.59@flipper.penguin-eng.co.nz> from "Glenn Ramsey" at Apr 28, 98 09:04:33 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Glenn Ramsey writes: > > > I have been unable to compile the latest 2.0.33 kernel with > > > the 19980426 patch or the 19980419 patch > > If this doesn't work, let me know, and please include a copy of > > the .config file (in the linux/ directory). > > It didn't work. Well done ;) You've found a problem in Linus' source code that apparantly no one else has found yet. Here is the patch that I'm forwarding for 2.0.34 to fix this problem: --- orig/linux/net/netsyms.c Sun Jan 18 21:58:25 1998 +++ linux/net/netsyms.c Tue Apr 28 20:37:55 1998 @@ -13,6 +13,7 @@ #include #include #include +#include #ifdef CONFIG_AX25 #include Also, there appear to be link errors in this configuration, which are fixed by: --- orig/linux/net/core/sock.c Sun Jan 18 21:57:37 1998 +++ linux/net/core/sock.c Tue Apr 28 20:55:05 1998 @@ -72,6 +72,7 @@ * easier (suggested by Craig Metz). * Michael Pall : SO_ERROR returns positive errno again * Elliot Poger : Added support for SO_BINDTODEVICE. + * Russell King : Add #ifdef CONFIG_INET to SO_BINDTODEVICE * * To Fix: * @@ -247,13 +248,13 @@ if(err) return err; memcpy_fromfs(&req,optval,sizeof(req)); - +#ifdef CONFIG_INET /* Remove any cached route for this socket. */ if (sk->ip_route_cache) { ip_rt_put(sk->ip_route_cache); sk->ip_route_cache=NULL; } - +#endif if (*(req.ifr_name) == '\0') { sk->bound_device = NULL; } else { --- orig/linux/net/ethernet/eth.c Mon Aug 5 08:13:55 1996 +++ linux/net/ethernet/eth.c Tue Apr 28 20:56:16 1998 @@ -30,6 +30,8 @@ * Alan Cox : Protect against forwarding explosions with * older network drivers and IFF_ALLMULTI. * Christer Weinigel : Better rebuild header message. + * Russell King : eth_header_cache_bind and eth_header_cache_update + * only compiled if CONFIG_INET is selected * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License @@ -218,6 +220,7 @@ return htons(ETH_P_802_2); } +#ifdef CONFIG_INET /* * Upper level calls this function to bind hardware header cache entry. * If the call is successful, then corresponding Address Resolution Protocol @@ -258,6 +261,7 @@ memcpy(hh->hh_data, haddr, ETH_ALEN); hh->hh_uptodate = 1; } +#endif /* * Copy from an ethernet device memory space to an sk_buff while checksumming if IP You will also have to edit arch/arm/lib/Makefile to include checksum.o unconditionally. These will, naturally, be in my next 2.0 kernel patches on Sunday (and hopefully the 2.0.34 kernel release from Linus when it happens). _____ |_____| ------------------------------------------------- ---+---+- | | Russell King linux@arm.uk.linux.org --- --- | | | | http://www.arm.uk.linux.org/~rmk/armlinux.html / / | | +-+-+ --- -+- / | THE developer of ARM Linux |+| /|\ / | | | --- | +-+-+ ------------------------------------------------- /\\\ | unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Wed Apr 29 18:25:19 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id SAA06208 for ; Wed, 29 Apr 1998 18:25:18 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:24984 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391813-6839>; Wed, 29 Apr 1998 20:22:31 +0300 Received: by vger.rutgers.edu id <971482-8838>; Wed, 29 Apr 1998 11:47:29 -0400 Received: from post-12.mail.demon.net ([194.217.242.41]:44556 "HELO post.mail.demon.net" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with SMTP id <971728-8838>; Wed, 29 Apr 1998 11:18:56 -0400 Received: from (globe) [158.152.30.195] by post.mail.demon.net with smtp (Exim 1.82 #2) id 0yUU5a-0001w2-00; Wed, 29 Apr 1998 11:27:46 +0100 Received: from (spring.nexus.co.uk) [192.0.0.3] (root) by globe with smtp (Exim 1.82 #1) id 0yUTwn-0001jj-00; Wed, 29 Apr 1998 11:18:41 +0100 Received: from localhost (spring.nexus.co.uk) [127.0.0.1] (pb) by spring.nexus.co.uk with esmtp (Exim 1.82 #1) id 0yUTwm-0001Q6-00; Wed, 29 Apr 1998 11:18:40 +0100 X-Mailer: exmh version 2.0zeta 7/24/97 To: linux-arm@vger.rutgers.edu Subject: uaccess ramblings Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 29 Apr 1998 11:18:39 +0100 From: Philip Blundell Message-Id: X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Hi, I was thinking about the code we use for user accesses on ARM2/3 machines just now (can you tell I'm not having a particularly busy day?) At the moment all the get_user calls translate into a function call to one of two addresses depending on whether we are in `kernel' or `user' mode (indirected through current->tss). I was wondering whether we could actually press the MEMC hardware into service to avoid the need for this. If we always use LDRT/STRT instructions for get_user() and put_user() then the ARM will ask for translation every time. If I remember right, though I don't have a MEMC datasheet here to check, you can set the privilege level that will be used for translated accesses by writing to a MEMC register. All we need to do is arrange for set_fs(KERNEL_DS) to set the privilege level to 1, and set_fs(USER_DS) to set it to 0, and arrange the page tables such that level 1 allows access to anything. This is closer to the ARM6 implementation, and I suspect it should be faster. Incidentally, I think the `wp_works_ok' macros in processor.h are probably completely redundant and can be removed -- the armv one is also set to the wrong value. :-) p. unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Thu Apr 30 00:31:54 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id AAA08388 for ; Thu, 30 Apr 1998 00:31:53 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:9582 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390544-18338>; Thu, 30 Apr 1998 02:31:54 +0300 Received: by vger.rutgers.edu id <971499-8838>; Wed, 29 Apr 1998 19:02:18 -0400 Received: from snowcrash.cymru.net ([163.164.160.3]:1306 "EHLO snowcrash.cymru.net" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <971566-8838>; Wed, 29 Apr 1998 18:40:15 -0400 Received: from caramon.armlinux.org (dialup2.cymru.net [163.164.160.66]) by snowcrash.cymru.net (8.8.7/8.7.1) with ESMTP id AAA29871; Thu, 30 Apr 1998 00:01:26 +0100 Received: from raistlin.armlinux.org (raistlin [192.168.0.3]) by caramon.armlinux.org (8.7.4/8.7.3) with ESMTP id XAA03278; Wed, 29 Apr 1998 23:56:57 +0100 From: Russell King Received: (from rmk@localhost) by raistlin.armlinux.org (8.7.4/8.7.3) id WAA00536; Wed, 29 Apr 1998 22:37:50 +0100 Message-Id: <199804292137.WAA00536@raistlin.armlinux.org> Subject: Re: cache flush ramblings To: pb@nexus.co.uk (Philip Blundell) Date: Wed, 29 Apr 1998 22:37:49 +0100 (BST) Cc: linux-arm@vger.rutgers.edu In-Reply-To: from "Philip Blundell" at Apr 29, 98 11:05:10 am X-Location: london.england.earth.mulky-way.universe Reply-To: rmk@arm.uk.linux.org X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Philip Blundell writes: > I was looking at the code in proc-sa110.S to flush the cache just now. It > seems that we maintain a flag (clean_switch) to say whether the Dcache might > be polluted by data from the flushing area and use this to decide whether or > not to read the whole 32k of data. It only ever reads 16k. Please read the code again. _____ |_____| ------------------------------------------------- ---+---+- | | Russell King rmk@arm.uk.linux.org --- --- | | | | http://www.arm.uk.linux.org/~rmk/home.html / / | | +-+-+ --- -+- / | THE developer of ARM Linux |+| /|\ / | | | --- | +-+-+ ------------------------------------------------- /\\\ | unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Thu Apr 30 00:50:09 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id AAA08449 for ; Thu, 30 Apr 1998 00:50:08 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:31446 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390594-18338>; Thu, 30 Apr 1998 02:47:39 +0300 Received: by vger.rutgers.edu id <971539-8838>; Wed, 29 Apr 1998 19:09:48 -0400 Received: from snowcrash.cymru.net ([163.164.160.3]:1339 "EHLO snowcrash.cymru.net" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <971596-8838>; Wed, 29 Apr 1998 18:40:42 -0400 Received: from caramon.armlinux.org (dialup2.cymru.net [163.164.160.66]) by snowcrash.cymru.net (8.8.7/8.7.1) with ESMTP id AAA29881; Thu, 30 Apr 1998 00:01:46 +0100 Received: from raistlin.armlinux.org (raistlin [192.168.0.3]) by caramon.armlinux.org (8.7.4/8.7.3) with ESMTP id XAA03282; Wed, 29 Apr 1998 23:56:58 +0100 From: Russell King - ARM Linux Admin Received: (from linux@localhost) by raistlin.armlinux.org (8.7.4/8.7.3) id XAA00790; Wed, 29 Apr 1998 23:52:14 +0100 Message-Id: <199804292252.XAA00790@raistlin.armlinux.org> Subject: Re: uaccess ramblings To: pb@nexus.co.uk (Philip Blundell) Date: Wed, 29 Apr 1998 23:52:14 +0100 (BST) Cc: linux-arm@vger.rutgers.edu In-Reply-To: from "Philip Blundell" at Apr 29, 98 11:18:39 am X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Philip Blundell writes: > I was thinking about the code we use for user accesses on ARM2/3 machines just > now (can you tell I'm not having a particularly busy day?) Hmm, how did I tell? > At the moment all the get_user calls translate into a function call to one of > two addresses depending on whether we are in `kernel' or `user' mode > (indirected through current->tss). I was wondering whether we could actually > press the MEMC hardware into service to avoid the need for this. Unfortunately, the MEMC hardware is very restrictive in this respect. > If we always use LDRT/STRT instructions for get_user() and put_user() then the > ARM will ask for translation every time. If I remember right, though I don't > have a MEMC datasheet here to check, you can set the privilege level that will > be used for translated accesses by writing to a MEMC register. This will not work. LDRT does not enable logical translation - it just turns on the user protection (by asserting the TRAN output). The control register does not allow the checks to be turned off. I quote: Supervisor Mode - ... is selected while the SPMD input is held high. (and SPMD is generally connected to -TRAN). Operating System Mode (OS) - OS mode is selected by setting a control bit in the MEMC Control Register (which may only be done from SVC mode). OS mode is more privileged than User mode <<>> Since the kernel accesses physical RAM from time to time using the get_user and put_user macros, replacing them with a LDRT/STRT instruction will cause a fault. However, one way to improve things would be for the compiler to allow you to control what registers were used to pass data into the inline assembly fragment. Without this, it produces around 3 redundant mov instructions per get_user/put_user macro. > Incidentally, I think the `wp_works_ok' macros in processor.h are probably > completely redundant and can be removed -- the armv one is also set to the > wrong value. :-) They are redundant and have now been removed. _____ |_____| ------------------------------------------------- ---+---+- | | Russell King linux@arm.uk.linux.org --- --- | | | | http://www.arm.uk.linux.org/~rmk/armlinux.html / / | | +-+-+ --- -+- / | THE developer of ARM Linux |+| /|\ / | | | --- | +-+-+ ------------------------------------------------- /\\\ | unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Thu Apr 30 09:04:26 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id JAA09678 for ; Thu, 30 Apr 1998 09:04:25 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:17249 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390623-18338>; Thu, 30 Apr 1998 11:02:07 +0300 Received: by vger.rutgers.edu id <971629-8838>; Thu, 30 Apr 1998 03:35:14 -0400 Received: from tazenda.demon.co.uk ([158.152.220.239]:3179 "HELO kings-cross.london.uk.eu.org" ident: "exim") by vger.rutgers.edu with SMTP id <971621-8838>; Thu, 30 Apr 1998 03:35:00 -0400 Received: from kings-cross.london.uk.eu.org [127.0.0.1] (phil) by kings-cross.london.uk.eu.org with esmtp (Exim 1.71 #1) id 0yUo9e-0006SV-00; Thu, 30 Apr 1998 08:53:18 +0100 X-Mailer: exmh version 2.0zeta 7/24/97 To: linux-arm@vger.rutgers.edu Subject: Re: uaccess ramblings In-reply-to: Your message of "Wed, 29 Apr 1998 23:52:14 BST." <199804292252.XAA00790@raistlin.armlinux.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 30 Apr 1998 08:53:18 +0100 From: Philip Blundell Message-Id: X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Russell King wrote: >Since the kernel accesses physical RAM from time to time using the get_user >and put_user macros, replacing them with a LDRT/STRT instruction will cause >a fault. That's a shame. In fact I think I have a plan that will still allow quite an improvement. I'll try to implement it today and make a patch. Access to physical RAM is never allowed in USER_DS (I assume) and the access_ok() check will catch that. So in put_user() we can check whether the address being accessed is above 0x02000000 - if so it must be a kernel access to physical RAM and can never fault so we can just do a plain STR (don't even have to stack the link register in fact). If it's below 0x02000000 it's an access to logical RAM; we can use STRT and let MEMC signal an exception if it's a bad access. This check will cost us a TST/BNE but that's still less than the current scheme by quite a way since it avoids all this fiddling with registers and getting values out of current->tss.uaccess->* (which requires two pointer dereferences). Does that sound reasonable or am I still overlooking something? I suppose there's a slight danger of taking an address exception in the "physical RAM" case I described but I imagine we can allow that to go untrapped and cause an Oops to be honest since only a kernel bug could cause it. Out of interest, where does the kernel access physical RAM with get_user()? >They are redundant and have now been removed. Thanks. p. unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Thu Apr 30 09:04:27 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id JAA09680 for ; Thu, 30 Apr 1998 09:04:26 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:17249 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390641-18338>; Thu, 30 Apr 1998 11:02:27 +0300 Received: by vger.rutgers.edu id <971622-8838>; Thu, 30 Apr 1998 03:37:59 -0400 Received: from tazenda.demon.co.uk ([158.152.220.239]:3183 "HELO kings-cross.london.uk.eu.org" ident: "exim") by vger.rutgers.edu with SMTP id <971608-8838>; Thu, 30 Apr 1998 03:35:11 -0400 Received: from kings-cross.london.uk.eu.org [127.0.0.1] (phil) by kings-cross.london.uk.eu.org with esmtp (Exim 1.71 #1) id 0yUnyI-0006SE-00; Thu, 30 Apr 1998 08:41:34 +0100 X-Mailer: exmh version 2.0zeta 7/24/97 To: rmk@arm.uk.linux.org cc: linux-arm@vger.rutgers.edu Subject: Re: cache flush ramblings In-reply-to: Your message of "Wed, 29 Apr 1998 22:37:49 BST." <199804292137.WAA00536@raistlin.armlinux.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 30 Apr 1998 08:41:34 +0100 From: Philip Blundell Message-Id: X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO >It only ever reads 16k. Please read the code again. Oops, you're right. In fact it already does exactly what I was suggesting. p. unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Thu Apr 30 14:09:59 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id OAA11040 for ; Thu, 30 Apr 1998 14:09:57 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:15213 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391158-18338>; Thu, 30 Apr 1998 16:08:13 +0300 Received: by vger.rutgers.edu id <970922-8838>; Thu, 30 Apr 1998 08:39:27 -0400 Received: from send1d.yahoomail.com ([205.180.60.48]:56326 "HELO send1d.yahoomail.com" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with SMTP id <971664-8838>; Thu, 30 Apr 1998 08:36:00 -0400 Message-ID: <19980430125541.19913.rocketmail@send1d.yahoomail.com> Received: from [199.33.81.223] by send1d; Thu, 30 Apr 1998 05:55:41 PDT Date: Thu, 30 Apr 1998 05:55:41 -0700 (PDT) From: klein fabien Subject: download a binary code in the kernel space To: linux-arm@vger.rutgers.edu MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO hi I m currently writting an linux driver for a multi modem PCI card I need to download a binary file into the card to drive an arm csm processor. So my question is :is it possible to access this file via my module driver and transfert it in the kernel space before downloading it trough PCI bus?? Is there another of doing this ? (using mmap ??..) In advance Thanks a lot KLEIN FABIEN Sophia Antipolis France _________________________________________________________ DO YOU YAHOO!? Get your free @yahoo.com address at http://mail.yahoo.com unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Thu Apr 30 16:52:16 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id QAA11722 for ; Thu, 30 Apr 1998 16:52:14 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:8622 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391246-18338>; Thu, 30 Apr 1998 18:50:23 +0300 Received: by vger.rutgers.edu id <971705-8838>; Thu, 30 Apr 1998 11:21:46 -0400 Received: from snowcrash.cymru.net ([163.164.160.3]:2036 "EHLO snowcrash.cymru.net" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <971707-8838>; Thu, 30 Apr 1998 11:13:52 -0400 Received: (from alan@localhost) by snowcrash.cymru.net (8.8.7/8.7.1) id QAA10530; Thu, 30 Apr 1998 16:38:25 +0100 From: Alan Cox Message-Id: <199804301538.QAA10530@snowcrash.cymru.net> Subject: Re: download a binary code in the kernel space To: fabklein@yahoo.com (klein fabien) Date: Thu, 30 Apr 1998 16:38:23 +0100 (BST) Cc: linux-arm@vger.rutgers.edu In-Reply-To: <19980430125541.19913.rocketmail@send1d.yahoomail.com> from "klein fabien" at Apr 30, 98 05:55:41 am Content-Type: text X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO > I m currently writting an linux driver for a multi modem PCI card > I need to download a binary file into the card to drive an arm csm > processor. > So my question is :is it possible to access this file via my module > driver and transfert it in the kernel space before downloading it > trough PCI bus?? Yes. Its far far neater to have a user program that configures the unit however and does an ioctl() passing you the data unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Thu Apr 30 17:13:42 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id RAA11797 for ; Thu, 30 Apr 1998 17:13:38 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:14981 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391250-18338>; Thu, 30 Apr 1998 19:12:27 +0300 Received: by vger.rutgers.edu id <971714-8838>; Thu, 30 Apr 1998 11:38:01 -0400 Received: from ferret.lmh.ox.ac.uk ([163.1.138.204]:31040 "HELO ferret.lmh.ox.ac.uk" ident: "qmailr") by vger.rutgers.edu with SMTP id <971722-8838>; Thu, 30 Apr 1998 11:37:07 -0400 Received: (qmail 1278 invoked by uid 504); 30 Apr 1998 16:01:50 -0000 Date: Thu, 30 Apr 1998 17:01:50 +0100 (BST) From: Matthew Kirkwood To: klein fabien cc: ARM Linux Subject: Re: download a binary code in the kernel space In-Reply-To: <199804301538.QAA10530@snowcrash.cymru.net> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO > I m currently writting an linux driver for a multi modem PCI card > I need to download a binary file into the card to drive an arm csm > processor. > So my question is :is it possible to access this file via my module > driver and transfert it in the kernel space before downloading it > trough PCI bus?? Probably the best thing to do is to have your module register a misc device and you can send it data by write()ing to or ioctl()ing it. The alternative is to do what at least one other driver does (a SCSI one, isn't it?) and just compile the code into the module/kernel. Matthew. unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Thu Apr 30 19:15:05 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id TAA12550 for ; Thu, 30 Apr 1998 19:15:01 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:23615 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <390406-18338>; Thu, 30 Apr 1998 21:15:20 +0300 Received: by vger.rutgers.edu id <970876-8838>; Thu, 30 Apr 1998 13:49:35 -0400 Received: from mail.denic.net ([195.35.6.2]:1709 "EHLO mail.all.de" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <970872-8838>; Thu, 30 Apr 1998 13:49:21 -0400 Received: by mail.all.de (Smail3.2) from bolzen.denic.net (194.94.235.1) with esmtp id ; Thu, 30 Apr 1998 20:14:31 +0200 (MET DST) Received: by bolzen.denic.net (Smail3.2) from yoda with uucp id ; Thu, 30 Apr 1998 20:13:17 +0200 (MET DST) Received: by yoda.in-berlin.de (Smail3.1.28.1 #0) id m0yUzmD-000000C; Thu, 30 Apr 98 18:17 GMT Date: Thu, 30 Apr 1998 20:15:27 +0200 From: Stefan Voss To: linux-arm@vger.rutgers.edu Subject: Re: PowerTec initialization problem Message-ID: In-Reply-To: <199804251424.PAA01476@raistlin.armlinux.org> Organization: Individual Network (IN), Berlin, Germany X-Mailer: Messenger v1.34c for RISC OS MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Posting-Agent: RISC OS Newsbase 0.60d X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO In message <199804251424.PAA01476@raistlin.armlinux.org>, Russell King wrote: > [...] > > Since you reported the problem, I have seen it a couple of times. It appears > that the FAS chip disappears from memory under certain circumstances (which > I'm not sure of yet). What chip does the driver believe you have? Does it > believe you have a 53C9x-type or 53C90-type? The boot messages say it's a 53C90-type > | | Russell King linux@arm.uk.linux.org --- --- Regards, Stefan -- Stefan Voss (voss@yoda.in-berlin.de) unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Fri May 1 00:44:27 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id AAA13542 for ; Fri, 1 May 1998 00:44:26 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:5396 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391535-3135>; Fri, 1 May 1998 02:43:25 +0300 Received: by vger.rutgers.edu id <971773-8838>; Thu, 30 Apr 1998 18:52:16 -0400 Received: from snowcrash.cymru.net ([163.164.160.3]:1115 "EHLO snowcrash.cymru.net" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <971817-8838>; Thu, 30 Apr 1998 18:29:19 -0400 Received: from caramon.armlinux.org (dialup1.cymru.net [163.164.160.65]) by snowcrash.cymru.net (8.8.7/8.7.1) with ESMTP id XAA18350; Thu, 30 Apr 1998 23:55:17 +0100 Received: from raistlin.armlinux.org (raistlin [192.168.0.3]) by caramon.armlinux.org (8.7.4/8.7.3) with ESMTP id XAA06213; Thu, 30 Apr 1998 23:54:12 +0100 From: Russell King - ARM Linux Admin Received: (from linux@localhost) by raistlin.armlinux.org (8.7.4/8.7.3) id XAA01574; Thu, 30 Apr 1998 23:29:55 +0100 Message-Id: <199804302229.XAA01574@raistlin.armlinux.org> Subject: Re: PowerTec initialization problem To: voss@yoda.in-berlin.de (Stefan Voss) Date: Thu, 30 Apr 1998 23:29:54 +0100 (BST) Cc: linux-arm@vger.rutgers.edu In-Reply-To: from "Stefan Voss" at Apr 30, 98 08:15:27 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Stefan Voss writes: > The boot messages say it's a 53C90-type Ok, it's not a new problem then, and it's one that I've seen when trying to turn the terminators on... Try inserting the module using: insmod powertec_scsi term=0 to see if that helps. If it does, then I think I know where the problem lies. _____ |_____| ------------------------------------------------- ---+---+- | | Russell King linux@arm.uk.linux.org --- --- | | | | http://www.arm.uk.linux.org/~rmk/armlinux.html / / | | +-+-+ --- -+- / | THE developer of ARM Linux |+| /|\ / | | | --- | +-+-+ ------------------------------------------------- /\\\ | unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Fri May 1 00:48:42 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id AAA13558 for ; Fri, 1 May 1998 00:48:41 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:5396 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391127-3135>; Fri, 1 May 1998 02:47:47 +0300 Received: by vger.rutgers.edu id <971807-8838>; Thu, 30 Apr 1998 18:52:29 -0400 Received: from snowcrash.cymru.net ([163.164.160.3]:1113 "EHLO snowcrash.cymru.net" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <971818-8838>; Thu, 30 Apr 1998 18:29:20 -0400 Received: from caramon.armlinux.org (dialup1.cymru.net [163.164.160.65]) by snowcrash.cymru.net (8.8.7/8.7.1) with ESMTP id XAA18353; Thu, 30 Apr 1998 23:55:19 +0100 Received: from raistlin.armlinux.org (raistlin [192.168.0.3]) by caramon.armlinux.org (8.7.4/8.7.3) with ESMTP id XAA06215; Thu, 30 Apr 1998 23:54:13 +0100 From: Russell King Received: (from rmk@localhost) by raistlin.armlinux.org (8.7.4/8.7.3) id WAA01441; Thu, 30 Apr 1998 22:53:02 +0100 Message-Id: <199804302153.WAA01441@raistlin.armlinux.org> Subject: EBSA-110 instructions (fwd) To: zskiraly@stanford.edu Date: Thu, 30 Apr 1998 22:53:02 +0100 (BST) Cc: linux-arm@vger.rutgers.edu In-Reply-To: <001101bd7470$53679b60$23b040ab@zskiraly> from "Zsolt Kiraly" at Apr 30, 98 12:44:01 pm X-Location: london.england.earth.mulky-way.universe Reply-To: rmk@arm.uk.linux.org X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Zsolt Kiraly writes: > I would like to run ARM Linux on an EBSA-110. You wrote on your web page > that this is possible. Could you send some installation instructions and > info on your setup so I can do this ? The reason that there aren't any on the web pages is that I don't have the time to write any. However, this reply is also going to the mailing list, so someone there may be able to suggest something. If someone does manage something, then I'll add it to the web site. _____ |_____| ------------------------------------------------- ---+---+- | | Russell King rmk@arm.uk.linux.org --- --- | | | | http://www.arm.uk.linux.org/~rmk/home.html / / | | +-+-+ --- -+- / | THE developer of ARM Linux |+| /|\ / | | | --- | +-+-+ ------------------------------------------------- /\\\ | unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Fri May 1 00:52:30 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id AAA13582 for ; Fri, 1 May 1998 00:52:28 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:5396 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391136-3135>; Fri, 1 May 1998 02:51:25 +0300 Received: by vger.rutgers.edu id <971813-8838>; Thu, 30 Apr 1998 18:52:43 -0400 Received: from snowcrash.cymru.net ([163.164.160.3]:1041 "EHLO snowcrash.cymru.net" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <971816-8838>; Thu, 30 Apr 1998 18:28:58 -0400 Received: from caramon.armlinux.org (dialup1.cymru.net [163.164.160.65]) by snowcrash.cymru.net (8.8.7/8.7.1) with ESMTP id XAA18330 for ; Thu, 30 Apr 1998 23:55:02 +0100 Received: from raistlin.armlinux.org (raistlin [192.168.0.3]) by caramon.armlinux.org (8.7.4/8.7.3) with ESMTP id XAA06201 for ; Thu, 30 Apr 1998 23:54:09 +0100 From: Russell King - ARM Linux Admin Received: (from linux@localhost) by raistlin.armlinux.org (8.7.4/8.7.3) id XAA01624 for linux-arm@vger.rutgers.edu; Thu, 30 Apr 1998 23:43:39 +0100 Message-Id: <199804302243.XAA01624@raistlin.armlinux.org> Subject: Web pages... To: linux-arm@vger.rutgers.edu Date: Thu, 30 Apr 1998 23:43:38 +0100 (BST) X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO Hi All, If anyone has any useful *small* modifications / information / web page links that they would like to be added to the ARM Linux web site, please let me know in the next week or so. Thanks. _____ |_____| ------------------------------------------------- ---+---+- | | Russell King linux@arm.uk.linux.org --- --- | | | | http://www.arm.uk.linux.org/~rmk/armlinux.html / / | | +-+-+ --- -+- / | THE developer of ARM Linux |+| /|\ / | | | --- | +-+-+ ------------------------------------------------- /\\\ | unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu From owner-linux-arm-outgoing@vger.rutgers.edu Fri May 1 02:29:26 1998 Return-Path: Received: from orava.funet.fi (orava.funet.fi [128.214.248.46]) by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id CAA13941 for ; Fri, 1 May 1998 02:29:25 +0100 Received: from vger.rutgers.edu ([128.6.190.2]:16590 "EHLO vger.rutgers.edu" ident: "root") by orava.funet.fi with ESMTP id <391710-3135>; Fri, 1 May 1998 04:29:52 +0300 Received: by vger.rutgers.edu id <971797-8838>; Thu, 30 Apr 1998 21:02:15 -0400 Received: from later.CS.Berkeley.EDU ([128.32.37.29]:1736 "EHLO later.CS.Berkeley.EDU" ident: "NO-IDENT-SERVICE") by vger.rutgers.edu with ESMTP id <970973-8838>; Thu, 30 Apr 1998 21:01:46 -0400 Received: from localhost (rmartin@localhost) by later.CS.Berkeley.EDU (8.8.3/8.6.9) with SMTP id SAA14507; Thu, 30 Apr 1998 18:28:43 -0700 (PDT) X-Authentication-Warning: later.CS.Berkeley.EDU: rmartin owned process doing -bs Date: Thu, 30 Apr 1998 18:28:43 -0700 (PDT) From: Rich Martin To: linux-arm@vger.rutgers.edu cc: zskiraly@stanford.edu Subject: Re: EBSA-110 instructions (fwd) In-Reply-To: <199804302153.WAA01441@raistlin.armlinux.org> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Orcpt: rfc822;linux-arm@vger.rutgers.edu Sender: owner-linux-arm@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu Status: RO I'm willing to write a set of pages up, but I still can't get our little old EBSA up :( It's currenty dying in tftp_load() function of the bootp supplied in DEC's SDT 2.0. If/when I get the system running I'll type up a set of pages ... (tips appreciated :). Also, we had a 2 bad boards out of 3, so be careful. One board was DOA and the other's ethernet controller just doesn't work. The LEDs are all dark. I don't think it was a software problem since the other board gets packets across fine. -Rich --------------------------------------------------------------------- On Thu, 30 Apr 1998, Russell King wrote: > > I would like to run ARM Linux on an EBSA-110. You wrote on your web page > > that this is possible. Could you send some installation instructions and > > info on your setup so I can do this ? > > The reason that there aren't any on the web pages is that I don't have the > time to write any. However, this reply is also going to the mailing list, > so someone there may be able to suggest something. > If someone does manage something, then I'll add it to the web site. unsubscribe: body of `unsubscribe linux-arm' to majordomo@vger.rutgers.edu