From owner-linux-arm-kernel@lists.arm.linux.org.uk Tue Feb 1 12:32:22 2000 Received: (from majordomo@localhost) by zen.ics.uwe.ac.uk (8.9.3/8.8.7) id MAA24480 for linux-arm-kernel-outgoing; Tue, 1 Feb 2000 12:32:22 GMT Received: from court2.aleph1.co.uk (IDENT:root@aleph1.demon.co.uk [158.152.9.127]) by zen.ics.uwe.ac.uk (8.9.3/8.8.7) with ESMTP id MAA24476 for ; Tue, 1 Feb 2000 12:32:13 GMT Received: from chewy.aleph1.co.uk ([194.70.44.163]) by court2.aleph1.co.uk with smtp (Exim 2.05 #1) id 12FcRt-0004QZ-00 for linux-arm-kernel@lists.arm.linux.org.uk; Tue, 1 Feb 2000 12:30:25 +0000 Date: Tue, 1 Feb 2000 12:30:24 +0000 (GMT) From: Wookey Subject: Re: v2.3 kernels on RPC w/ ARM710 To: ARMlinux kernel In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII X-Organization: Aleph One,Cambridge, UK. Tel +44 1223 811679 X-Mailer: ANT RISCOS Marcel [ver 1.50] Sender: owner-linux-arm-kernel@lists.arm.linux.org.uk Precedence: bulk On Mon 31 Jan, Hanske;Stefan wrote: > > Hi there, > > I've found that the problems with v2.3.xx kernels are more generic than I > thought. I've tried all v2.3.xx kernels from ftp.armlinux.org/pub/lkab - > it's still the same as with my home built kernels. > The kernels are initializing fine, but when init has been started, it is > terminated with a SIGSEGV and the kernel dies with a register dump. > > Could someone with a RiscPC and ARM[67]10 processor please verify this > problem? It doesn't work for me either (ARM710 RPC) but not the same symptoms. I just tried booting linux-2/3/35-cvs-riscpc-current (13 Jan 2000) using !linux and it uncompresses ok, then says 'done, booting kernel and the machine hangs. I haven't tried compiling my own or any other boot methods yet but will report back if I find anything useful. Wookey -- Aleph One Ltd, Bottisham, CAMBRIDGE, CB5 9BA, UK Tel (00 44) 1223 811679 work: http://www.aleph1.co.uk/ play: http://www.chaos.org.uk/~wookey/ unsubscribe: body of `unsubscribe linux-arm-kernel' to majordomo@lists.arm.linux.org.uk From owner-linux-arm-kernel@lists.arm.linux.org.uk Tue Feb 1 14:34:15 2000 Received: (from majordomo@localhost) by zen.ics.uwe.ac.uk (8.9.3/8.8.7) id OAA30588 for linux-arm-kernel-outgoing; Tue, 1 Feb 2000 14:34:15 GMT Received: from mail.uni-greifswald.de (mail.uni-greifswald.de [141.53.8.33]) by zen.ics.uwe.ac.uk (8.9.3/8.8.7) with ESMTP id OAA30584 for ; Tue, 1 Feb 2000 14:34:12 GMT Received: from localhost (sh990154@localhost) by mail.uni-greifswald.de (8.9.2/8.9.2) with SMTP id PAA08328; Tue, 1 Feb 2000 15:34:04 +0100 (MET) Date: Tue, 1 Feb 2000 15:34:04 +0100 (MET) From: "Hanske;Stefan" To: Wookey cc: ARMlinux kernel Subject: Re: v2.3 kernels on RPC w/ ARM710 In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-linux-arm-kernel@lists.arm.linux.org.uk Precedence: bulk On Tue, 1 Feb 2000, Wookey wrote: > On Mon 31 Jan, Hanske;Stefan wrote: > > > > Hi there, > > > > I've found that the problems with v2.3.xx kernels are more generic than I > > thought. I've tried all v2.3.xx kernels from ftp.armlinux.org/pub/lkab - > > it's still the same as with my home built kernels. > > The kernels are initializing fine, but when init has been started, it is > > terminated with a SIGSEGV and the kernel dies with a register dump. > > > > Could someone with a RiscPC and ARM[67]10 processor please verify this > > problem? > > It doesn't work for me either (ARM710 RPC) but not the same symptoms. > > I just tried booting linux-2/3/35-cvs-riscpc-current (13 Jan 2000) using > !linux and it uncompresses ok, then says 'done, booting kernel and the > machine hangs. > > I haven't tried compiling my own or any other boot methods yet but will > report back if I find anything useful. All the other v2.3 versions work (more or less). As for the register dump... I've sent it already some day. I'll see if I still find it. unsubscribe: body of `unsubscribe linux-arm-kernel' to majordomo@lists.arm.linux.org.uk From owner-linux-arm-kernel@lists.arm.linux.org.uk Tue Feb 1 17:55:14 2000 Received: (from majordomo@localhost) by zen.ics.uwe.ac.uk (8.9.3/8.8.7) id RAA08958 for linux-arm-kernel-outgoing; Tue, 1 Feb 2000 17:55:14 GMT Received: from earth.light.com (c657338-a.plstn1.sfba.home.com [24.1.83.42]) by zen.ics.uwe.ac.uk (8.9.3/8.8.7) with ESMTP id RAA08950 for ; Tue, 1 Feb 2000 17:55:09 GMT Received: from localhost (rscott@localhost) by earth.light.com (8.9.3/8.8.7) with ESMTP id JAA24411 for ; Tue, 1 Feb 2000 09:54:58 -0800 X-Authentication-Warning: earth.light.com: rscott owned process doing -bs Date: Tue, 1 Feb 2000 09:54:58 -0800 (PST) From: Rob Scott To: linux-arm-kernel@lists.arm.linux.org.uk Subject: Status of non-zero physical memory offset kernel? Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-linux-arm-kernel@lists.arm.linux.org.uk Precedence: bulk I'd like to use 2.3.39 as a starting point for my porting efforts, but Russell's release note states that it doesn't support architectures with non-zero phyical memory offsets. Any hope we'll see this fixed soon? If not, which kernel should I use? Thanks, Rob Scott unsubscribe: body of `unsubscribe linux-arm-kernel' to majordomo@lists.arm.linux.org.uk From owner-linux-arm-kernel@lists.arm.linux.org.uk Wed Feb 2 00:58:57 2000 Received: (from majordomo@localhost) by zen.ics.uwe.ac.uk (8.9.3/8.8.7) id AAA29420 for linux-arm-kernel-outgoing; Wed, 2 Feb 2000 00:58:57 GMT Received: from mail.uni-greifswald.de (mail.uni-greifswald.de [141.53.8.33]) by zen.ics.uwe.ac.uk (8.9.3/8.8.7) with ESMTP id AAA29416 for ; Wed, 2 Feb 2000 00:58:55 GMT Received: from localhost (sh990154@localhost) by mail.uni-greifswald.de (8.9.2/8.9.2) with SMTP id BAA02449 for ; Wed, 2 Feb 2000 01:58:54 +0100 (MET) Date: Wed, 2 Feb 2000 01:58:54 +0100 (MET) From: "Hanske;Stefan" To: linux-arm-kernel@lists.arm.linux.org.uk Subject: Fwd: Kernel compiling problems (fwd) Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-linux-arm-kernel@lists.arm.linux.org.uk Precedence: bulk ---------- Forwarded message ---------- Date: Wed, 2 Feb 2000 01:01:51 +0100 (MET) From: Stefan Hanske To: sh990154@mail.uni-greifswald.de Subject: Fwd: Kernel compiling problems > INIT: version 2.74 booting > INIT: PANIC: segmentation violation! giving up... > Unable to handle kernel paging request at virtual address c100052c > pgd=c04a8000 > *pgd=00000000, *pmd=00000000 > Internal error: Oops: 2 > CPU: 0 > pc: [] lr: [] > sp: c0479f98 ip: bffff98c fp: 00000000 > r10: 0205b454 r9: 0205b430 r8: bffff998 > r7: c0fffc3c r6: 4d9abb30 r5: 000d0000 r4: e92dd8f0 > r3: c0479fd4 r2: 00001d25 r1: 8f000000 r0: c100052c > Flags: nZcv IRQs off FIQs on Mode SVC_32 Segment user > Control: EE133F10 Table: EB003774 DAC: EA00000B > Process init (pid:1, stackpage=c0479000) > Stack: > c0479f80: c000f7c4 c0019254 40000093 ffffffff 00000002 > bffff998 > c0479fa0: 00000000 00000008 00000002 bffffa18 00000006 0205b428 bffff998 > 0205b430 > c0479fc0: 0205b454 bffffc00 bffff98c c100052c 02001548 02004408 20000010 > ffffffff > c0479fe0: C000d6f8 c01f2ccc c01e5a84 c01e5ad0 c0179fe4 c047a000 c0011444 > c000d708 > Backtrace: no frame pointer > code: 70253d70 0000000a (e5904000) e1a029a4 e2022002 where pc is in "cpu_arm6_data_abort" and lr comes from "dabt_usr" > However, I can boot into a shell via init=/bin/bash at the kernel > command > line. > When I try to execute any program from there by entering it's path/name > at the > command line, nothing happens. I can type in characters, but I can't > stop > the process (not even via SYSREQ-I or similar). > I've added some code to fork() and execve() that issue some printk()s at > different points. The result is that the shell fork()s successfully but > execve() isn't called. > Kernels tried: v2.3.19 and up (to 2.3.39-rmk1) > Binutils: 2.9.1.0.23 (patched with some redhat patches for arm) and > 2.9.5.0.22 (unpatched) > gcc: 2.95.2 with patch from ftp.netwinder.org/users/p/philb) > System: RiscPC with ARM710, various kernel configurations > Build method: native and cross-compiled > > Has anyone some hints what might me wrong??? Has anybody running > v2.3.xx- > kernels on RiscPC with non-SA processor??? > > Tia. > > Stefan ------------------------------------------------------------------- Just because you're paranoid doesn't mean they're not really after you! ------------------------------------------------------------------- unsubscribe: body of `unsubscribe linux-arm-kernel' to majordomo@lists.arm.linux.org.uk From owner-linux-arm-kernel@lists.arm.linux.org.uk Tue Feb 8 00:54:37 2000 Received: (from majordomo@localhost) by zen.ics.uwe.ac.uk (8.9.3/8.8.7) id AAA09351 for linux-arm-kernel-outgoing; Tue, 8 Feb 2000 00:54:37 GMT Received: from mail2.digital.com (mail2.digital.com [204.123.2.56]) by zen.ics.uwe.ac.uk (8.9.3/8.8.7) with ESMTP id AAA09347 for ; Tue, 8 Feb 2000 00:54:34 GMT Received: from pobox1.pa.dec.com (pobox1.pa.dec.com [16.1.240.19]) by mail2.digital.com (8.9.2/8.9.3/WV2.0h) with SMTP id QAA04620 for ; Mon, 7 Feb 2000 16:54:22 -0800 (PST) Received: from stowe.pa.dec.com by pobox1.pa.dec.com (5.65v3.2/1.1.10.5/07Nov97-1157AM) id AA09887; Mon, 7 Feb 2000 16:53:52 -0800 Received: by stowe.pa.dec.com; (5.65v4.0/1.1.8.2/06Jun96-0357PM) id AA15407; Mon, 7 Feb 2000 16:53:48 -0800 Date: Mon, 7 Feb 2000 16:53:48 -0800 Message-Id: <200002080053.AA15407@stowe.pa.dec.com> From: Deborah Wallach To: sa1100-linux@pa.dec.com, linux-arm-kernel@lists.arm.linux.org.uk Subject: Itsy schematics and updated Linux software available Address: Compaq Computer Corporation; Western Research Laboratory; 250 University Avenue; Palo Alto, CA 94301; USA Phone: 650-617-3315 Sender: owner-linux-arm-kernel@lists.arm.linux.org.uk Precedence: bulk In order to promote collaboration and to advance the state of pocket computing research, Compaq Computer Corporation's Research Laboratories are making available the schematics and manufacturing information for the Itsy V1.5 Pocket Computer. The Itsy is a flexible research platform, based on the StrongARM 1100, intended to enable hardware and software research in pocket computing. It is a small, low-power handheld device with a highly flexible interface, designed to encourage the development of innovative research projects, such as novel user interfaces, new applications, power management techniques, hardware extensions, etc. The information we are making available includes the hardware specifications, a low level monitor, and the Linux source code for the Itsy. Information about the project is available at: www.research.compaq.com/wrl/itsy More information on the distribution and how to download it is available at: www.research.compaq.com/wrl/itsy/download.html Deborah Wallach Western Research Laboratory Compaq Computer Corporation unsubscribe: body of `unsubscribe linux-arm-kernel' to majordomo@lists.arm.linux.org.uk From owner-linux-arm-kernel@lists.arm.linux.org.uk Tue Feb 8 20:28:44 2000 Received: (from majordomo@localhost) by zen.ics.uwe.ac.uk (8.9.3/8.8.7) id UAA05073 for linux-arm-kernel-outgoing; Tue, 8 Feb 2000 20:28:44 GMT Received: from caramon.arm.linux.org.uk (root@dyn-33.linux.theplanet.co.uk [195.92.244.33]) by zen.ics.uwe.ac.uk (8.9.3/8.8.7) with ESMTP id UAA05067 for ; Tue, 8 Feb 2000 20:28:38 GMT Received: from raistlin.arm.linux.org.uk (root@raistlin [195.92.244.35]) by caramon.arm.linux.org.uk (8.9.3/8.9.3) with ESMTP id UAA18940; Tue, 8 Feb 2000 20:14:07 GMT From: Russell King - ARM Linux Admin Received: (from linux@localhost) by raistlin.arm.linux.org.uk (8.7.4/8.7.3) id UAA00553; Tue, 8 Feb 2000 20:10:31 GMT Message-Id: <200002082010.UAA00553@raistlin.arm.linux.org.uk> Subject: Re: Fwd: Kernel compiling problems (fwd) To: sh990154@mail.uni-greifswald.de (Hanske;Stefan) Date: Tue, 8 Feb 2000 20:10:31 +0000 (GMT) Cc: linux-arm-kernel@lists.arm.linux.org.uk In-Reply-To: from "Hanske;Stefan" at Feb 02, 2000 01:58:54 AM X-Mailer: ELM [version 2.5 PL1] MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-linux-arm-kernel@lists.arm.linux.org.uk Precedence: bulk Hanske;Stefan writes: > > Unable to handle kernel paging request at virtual address c100052c > > pgd=c04a8000 > > *pgd=00000000, *pmd=00000000 > > Internal error: Oops: 2 > > CPU: 0 > > pc: [] lr: [] > > sp: c0479f98 ip: bffff98c fp: 00000000 > > r10: 0205b454 r9: 0205b430 r8: bffff998 > > r7: c0fffc3c r6: 4d9abb30 r5: 000d0000 r4: e92dd8f0 > > r3: c0479fd4 r2: 00001d25 r1: 8f000000 r0: c100052c > > Flags: nZcv IRQs off FIQs on Mode SVC_32 Segment user > > Control: EE133F10 Table: EB003774 DAC: EA00000B > > Process init (pid:1, stackpage=c0479000) > > Stack: > > c0479f80: c000f7c4 c0019254 40000093 ffffffff 00000002 > > bffff998 > > c0479fa0: 00000000 00000008 00000002 bffffa18 00000006 0205b428 bffff998 > > 0205b430 > > c0479fc0: 0205b454 bffffc00 bffff98c c100052c 02001548 02004408 20000010 > > ffffffff > > c0479fe0: C000d6f8 c01f2ccc c01e5a84 c01e5ad0 c0179fe4 c047a000 c0011444 > > c000d708 > > Backtrace: no frame pointer > > code: 70253d70 0000000a (e5904000) e1a029a4 e2022002 Ok, found this one! The following patch should sort this out: --- linux-orig/arch/arm/mm/proc-arm6,7.S Mon Dec 27 17:16:05 1999 +++ linux/arch/arm/mm/proc-arm6,7.S Tue Feb 8 20:09:03 2000 @@ -103,9 +103,9 @@ .align ENTRY(cpu_arm6_data_abort) -Ldata_simple: ldr r4, [r0] @ read instruction causing problem mov r2, r4, lsr #19 @ r2 b1 = L +Ldata_simple: and r2, r2, #2 @ check read/write bit mrc p15, 0, r0, c6, c0, 0 @ get FAR mrc p15, 0, r1, c5, c0, 0 @ get FSR _____ |_____| ------------------------------------------------- ---+---+- | | Russell King linux@arm.linux.org.uk --- --- | | | | http://www.arm.linux.org.uk/~rmk/armlinux.html / / | | +-+-+ --- -+- / | THE developer of ARM Linux |+| /|\ / | | | --- | +-+-+ ------------------------------------------------- /\\\ | unsubscribe: body of `unsubscribe linux-arm-kernel' to majordomo@lists.arm.linux.org.uk From owner-linux-arm-kernel@lists.arm.linux.org.uk Tue Feb 15 06:50:06 2000 Received: (from majordomo@localhost) by zen.ics.uwe.ac.uk (8.9.3/8.8.7) id GAA09832 for linux-arm-kernel-outgoing; Tue, 15 Feb 2000 06:50:06 GMT Received: from fire.light.com (IDENT:root@c657338-a.plstn1.sfba.home.com [24.19.133.127]) by zen.ics.uwe.ac.uk (8.9.3/8.8.7) with ESMTP id GAA09797 for ; Tue, 15 Feb 2000 06:50:02 GMT Received: from localhost (rscott@localhost) by fire.light.com (8.9.3/8.9.3) with ESMTP id WAA22622 for ; Mon, 14 Feb 2000 22:55:24 -0800 Date: Mon, 14 Feb 2000 22:55:24 -0800 (PST) From: Rob Scott To: linux-arm-kernel@lists.arm.linux.org.uk Subject: printk in init/main.c Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-linux-arm-kernel@lists.arm.linux.org.uk Precedence: bulk Hi, I'm trying to get my first printk(linux_banner) in init/main.c to work. I can't figure out how console_drivers gets initialized. It's used in the following fragment from printk (where the actual printing gets done): if (msg_level < console_loglevel && console_drivers) { struct console *c = console_drivers; while(c) { if ((c->flags & CON_ENABLED) && c->write) c->write(c, msg, p - msg + line_feed); c = c->next; } } The printk(linux_banner) is called before console_init in the function start_kernel as the following fragment from init/main.c shows: /* * Interrupts are still disabled. Do necessary setups, then * enable them */ printk(linux_banner); setup_arch(&command_line, &memory_start, &memory_end); memory_start = paging_init(memory_start,memory_end); trap_init(); init_IRQ(); sched_init(); time_init(); parse_options(command_line); /* * HACK ALERT! This is early. We're enabling the console before * we've done PCI setups etc, and console_init() must be aware of * this. But we do want output early, in case something goes wrong. */ memory_start = console_init(memory_start,memory_end); My guess is that there's some "interesting" compile time behavour that I'm missing... Thanks in advance for your help! Rob Scott unsubscribe: body of `unsubscribe linux-arm-kernel' to majordomo@lists.arm.linux.org.uk From owner-linux-arm-kernel@lists.arm.linux.org.uk Tue Feb 15 16:37:06 2000 Received: (from majordomo@localhost) by zen.ics.uwe.ac.uk (8.9.3/8.8.7) id QAA22295 for linux-arm-kernel-outgoing; Tue, 15 Feb 2000 16:37:06 GMT Received: from xanadu.vipswitch.com (generic199.197.205.205.in-addr.arpa [205.205.197.199] (may be forged)) by zen.ics.uwe.ac.uk (8.9.3/8.8.7) with ESMTP id QAA22287 for ; Tue, 15 Feb 2000 16:37:01 GMT Date: Tue, 15 Feb 2000 11:31:56 -0500 (EST) From: Nicolas Pitre To: Rob Scott cc: linux-arm-kernel@lists.arm.linux.org.uk Subject: Re: printk in init/main.c In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-linux-arm-kernel@lists.arm.linux.org.uk Precedence: bulk On Mon, 14 Feb 2000, Rob Scott wrote: > > Hi, > > I'm trying to get my first printk(linux_banner) in init/main.c to work. I > can't figure out how console_drivers gets initialized. It's used in the > following fragment from printk (where the actual printing gets done): [...] > The printk(linux_banner) is called before console_init in the function > start_kernel as the following fragment from init/main.c shows: [...] All printk's before console_init() are stored in a buffer. The buffer is purged only when the console is initialized, but your kernel may have crashed before that point. You may consider using printascii() and friends. See arch/arm/kernel/debug-*.S, or arch/arm/kernel/head_*.s depending on the version you are using. Nicolas unsubscribe: body of `unsubscribe linux-arm-kernel' to majordomo@lists.arm.linux.org.uk From owner-linux-arm-kernel@lists.arm.linux.org.uk Tue Feb 15 22:22:13 2000 Received: (from majordomo@localhost) by zen.ics.uwe.ac.uk (8.9.3/8.8.7) id WAA17598 for linux-arm-kernel-outgoing; Tue, 15 Feb 2000 22:22:13 GMT Received: from caramon.arm.linux.org.uk (root@p17-robin-gui.tch.enablis.net [194.168.180.77]) by zen.ics.uwe.ac.uk (8.9.3/8.8.7) with ESMTP id WAA17594 for ; Tue, 15 Feb 2000 22:22:10 GMT Received: from raistlin.arm.linux.org.uk (root@raistlin [192.168.0.3]) by caramon.arm.linux.org.uk (8.9.3/8.9.3) with ESMTP id WAA16146 for ; Tue, 15 Feb 2000 22:22:08 GMT Received: (from rmk@localhost) by raistlin.arm.linux.org.uk (8.7.4/8.7.3) id WAA03340 for linux-arm-kernel@lists.arm.linux.org.uk; Tue, 15 Feb 2000 22:18:09 GMT Received: from caramon.arm.linux.org.uk (root@p17-robin-gui.tch.enablis.net [194.168.180.77]) by zen.ics.uwe.ac.uk (8.9.3/8.8.7) with ESMTP id WAA16884; Tue, 15 Feb 2000 22:12:30 GMT Received: from raistlin.arm.linux.org.uk (root@raistlin [192.168.0.3]) by caramon.arm.linux.org.uk (8.9.3/8.9.3) with ESMTP id WAA16047; Tue, 15 Feb 2000 22:12:28 GMT From: Russell King - ARM Linux Admin Received: (from linux@localhost) by raistlin.arm.linux.org.uk (8.7.4/8.7.3) id WAA03229; Tue, 15 Feb 2000 22:08:28 GMT Message-Id: <200002152208.WAA03229@raistlin.arm.linux.org.uk> Subject: Mailing lists moving... To: linux-arm@lists.arm.linux.org.uk, linux-arm-kernel@lists.arm.linux.org.uk, linux-arm-announce@lists.arm.linux.org.uk Date: Tue, 15 Feb 2000 22:08:27 +0000 (GMT) X-Mailer: ELM [version 2.5 PL1] MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-linux-arm-kernel@lists.arm.linux.org.uk Precedence: bulk To all subscribers of the following mailing lists: linux-arm@lists.arm.linux.org.uk linux-arm-announce@lists.arm.linux.org.uk linux-arm-kernel@lists.arm.linux.org.uk Over the next few days, the mailing lists will be migrated to a new machine. Please note that there is the possibility that any new subscriptions/ any unsubscriptions sent over the next day or so will be lost. If you receive a majordomo reply for your request, please check the mail headers to find out which machine it has come from. If it was sent out by 'zen.ics.uwe.ac.uk' then only the old lists have handled it. Please resend the request in a couple of days. There is no need to re-subscribe to the lists - at the time of writing, both mailing list sites contain exactly the same subscriber database. _____ |_____| ------------------------------------------------- ---+---+- | | Russell King linux@arm.linux.org.uk --- --- | | | | http://www.arm.linux.org.uk/~rmk/armlinux.html / / | | +-+-+ --- -+- / | THE developer of ARM Linux |+| /|\ / | | | --- | +-+-+ ------------------------------------------------- /\\\ | unsubscribe: body of `unsubscribe linux-arm-kernel' to majordomo@lists.arm.linux.org.uk From owner-linux-arm-kernel@lists.arm.linux.org.uk Sun Feb 20 22:54:51 2000 Received: (from majordomo@localhost) by parcelfarce.linux.theplanet.co.uk (8.9.3/8.9.3) id WAA02804 for linux-arm-kernel-outgoing; Sun, 20 Feb 2000 22:54:51 GMT Received: from caramon.arm.linux.org.uk (root@p59-cordelia-gui.tch.enablis.net [212.250.233.59]) by parcelfarce.linux.theplanet.co.uk (8.9.3/8.9.3) with ESMTP id WAA02800 for ; Sun, 20 Feb 2000 22:54:49 GMT Received: from raistlin.arm.linux.org.uk (root@raistlin [192.168.0.3]) by caramon.arm.linux.org.uk (8.9.3/8.9.3) with ESMTP id WAA01200; Sun, 20 Feb 2000 22:53:01 GMT From: Russell King - ARM Linux Admin Received: (from linux@localhost) by raistlin.arm.linux.org.uk (8.7.4/8.7.3) id WAA04629; Sun, 20 Feb 2000 22:52:49 GMT Message-Id: <200002202252.WAA04629@raistlin.arm.linux.org.uk> Subject: Re: slow signals? To: Philip.Blundell@pobox.com (Philip Blundell) Date: Sun, 20 Feb 2000 22:52:48 +0000 (GMT) Cc: linux-arm-kernel@lists.arm.linux.org.uk In-Reply-To: from "Philip Blundell" at Feb 20, 2000 10:47:56 PM X-Mailer: ELM [version 2.5 PL1] MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-linux-arm-kernel@lists.arm.linux.org.uk Precedence: bulk Philip Blundell writes: > >> Signal handler overhead seems to be 50% up between 2.3.39 and 2.3.46. > >> Anybody have any idea what happened? > >Nothing has changed in the signal handling dept recently in the ARM tree. > > I suppose it's possible I'm suffering some kind of local lossage, though it > doesn't seem all that likely. Do you have any benchmark results for recent > kernels? Nope, haven't benchmarked the kernel this side of the millenium yet. _____ |_____| ------------------------------------------------- ---+---+- | | Russell King linux@arm.linux.org.uk --- --- | | | | http://www.arm.linux.org.uk/~rmk/armlinux.html / / | | +-+-+ --- -+- / | THE developer of ARM Linux |+| /|\ / | | | --- | +-+-+ ------------------------------------------------- /\\\ | unsubscribe: body of `unsubscribe linux-arm-kernel' to majordomo@lists.arm.linux.org.uk From owner-linux-arm-kernel@lists.arm.linux.org.uk Sun Feb 20 23:04:30 2000 Received: (from majordomo@localhost) by parcelfarce.linux.theplanet.co.uk (8.9.3/8.9.3) id XAA02890 for linux-arm-kernel-outgoing; Sun, 20 Feb 2000 23:04:30 GMT Received: from kings-cross.london.uk.eu.org (mail@tazenda.demon.co.uk [158.152.220.239]) by parcelfarce.linux.theplanet.co.uk (8.9.3/8.9.3) with ESMTP id XAA02886 for ; Sun, 20 Feb 2000 23:04:27 GMT Received: from localhost ([::ffff:127.0.0.1] helo=kings-cross.london.uk.eu.org ident=phil) by kings-cross.london.uk.eu.org with esmtp (Exim 3.11 #1) id 12MfN0-0005Kh-00; Sun, 20 Feb 2000 23:02:30 +0000 X-Mailer: exmh version 2.0.2 2/24/98 (debian) To: Russell King - ARM Linux Admin cc: linux-arm-kernel@lists.arm.linux.org.uk Subject: Re: slow signals? In-Reply-To: Message from Russell King - ARM Linux Admin of "Sun, 20 Feb 2000 22:52:48 GMT." <200002202252.WAA04629@raistlin.arm.linux.org.uk> References: <200002202252.WAA04629@raistlin.arm.linux.org.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sun, 20 Feb 2000 23:02:30 +0000 From: Philip Blundell Message-Id: Sender: owner-linux-arm-kernel@lists.arm.linux.org.uk Precedence: bulk >Nope, haven't benchmarked the kernel this side of the millenium yet. I'd be interested to see a copy of the results next time you do. p. unsubscribe: body of `unsubscribe linux-arm-kernel' to majordomo@lists.arm.linux.org.uk From owner-linux-arm-kernel@lists.arm.linux.org.uk Sun Feb 20 23:19:36 2000 Received: (from majordomo@localhost) by parcelfarce.linux.theplanet.co.uk (8.9.3/8.9.3) id XAA03011 for linux-arm-kernel-outgoing; Sun, 20 Feb 2000 23:19:36 GMT Received: from caramon.arm.linux.org.uk (root@p59-cordelia-gui.tch.enablis.net [212.250.233.59]) by parcelfarce.linux.theplanet.co.uk (8.9.3/8.9.3) with ESMTP id XAA02956; Sun, 20 Feb 2000 23:15:12 GMT Received: from raistlin.arm.linux.org.uk (root@raistlin [192.168.0.3]) by caramon.arm.linux.org.uk (8.9.3/8.9.3) with ESMTP id XAA01319; Sun, 20 Feb 2000 23:13:24 GMT From: Russell King - ARM Linux Admin Received: (from linux@localhost) by raistlin.arm.linux.org.uk (8.7.4/8.7.3) id XAA04683; Sun, 20 Feb 2000 23:13:12 GMT Message-Id: <200002202313.XAA04683@raistlin.arm.linux.org.uk> Subject: Mailing lists moved, FTP site moving To: linux-arm@lists.arm.linux.org.uk, linux-arm-kernel@lists.arm.linux.org.uk, linux-arm-announce@lists.arm.linux.org.uk Date: Sun, 20 Feb 2000 23:13:11 +0000 (GMT) X-Mailer: ELM [version 2.5 PL1] MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-linux-arm-kernel@lists.arm.linux.org.uk Precedence: bulk Hi, I am pleased to report that the mailing lists have now been moved onto a new server and are now running correctly. The old lists have been disabled, which means that the archives held on 'ftp.arm.linux.org.uk' will be out of date until the FTP site is properly transferred. Unfortunately, some of the archived messages were lost in this move, which I hope will not cause too many problems for anyone. I shall look into transferring the FTP site shortly. Please ensure that you are using the correct address, namely: "ftp.arm.linux.org.uk" or "ftp.arm.uk.linux.org" The first address is preferable. This is especially true for sites mirroring the main FTP site. _____ |_____| ------------------------------------------------- ---+---+- | | Russell King linux@arm.linux.org.uk --- --- | | | | http://www.arm.linux.org.uk/~rmk/armlinux.html / / | | +-+-+ --- -+- / | THE developer of ARM Linux |+| /|\ / | | | --- | +-+-+ ------------------------------------------------- /\\\ | unsubscribe: body of `unsubscribe linux-arm-kernel' to majordomo@lists.arm.linux.org.uk From owner-linux-arm-kernel@lists.arm.linux.org.uk Mon Feb 21 06:30:21 2000 Received: (from majordomo@localhost) by parcelfarce.linux.theplanet.co.uk (8.9.3/8.9.3) id GAA04482 for linux-arm-kernel-outgoing; Mon, 21 Feb 2000 06:30:21 GMT Received: from davinci.artisan.calpoly.edu (root@davinci.artisan.calpoly.edu [129.65.60.31]) by parcelfarce.linux.theplanet.co.uk (8.9.3/8.9.3) with ESMTP id GAA04478 for ; Mon, 21 Feb 2000 06:30:19 GMT From: bowu@calpoly.edu Received: from localhost (root@localhost) by davinci.artisan.calpoly.edu (8.8.6 (PHNE_17135)/8.8.6) with SMTP id WAA17466 for linux-arm-kernel@lists.arm.linux.org.uk; Sun, 20 Feb 2000 22:28:19 -0800 (PST) X-OpenMail-Hops: 1 Date: Sun, 20 Feb 2000 22:28:05 -0800 Message-Id: Subject: How to reset the remote_debug mode under SDT for EBSA MIME-Version: 1.0 TO: linux-arm-kernel@lists.arm.linux.org.uk Content-Type: text/plain; charset=US-ASCII Content-Disposition: inline Content-Transfer-Encoding: 7bit Sender: owner-linux-arm-kernel@lists.arm.linux.org.uk Precedence: bulk Hi, I am doing development on the ebsa board under SDT (strongarm development tools) now. I encountered a problem: after I use remote_debug mode to download the image to the flash ram on the board, if I make some change on the code and rebuild the image, I can not download it to the board unless I reboot the computer. Does anyone know how to reset the remote_debug mode, then I can download the rebuild code without rebooting the machine? Thanks a lot. Bo unsubscribe: body of `unsubscribe linux-arm-kernel' to majordomo@lists.arm.linux.org.uk From owner-linux-arm-kernel@lists.arm.linux.org.uk Tue Feb 22 14:21:15 2000 Received: (from majordomo@localhost) by parcelfarce.linux.theplanet.co.uk (8.9.3/8.9.3) id OAA11053 for linux-arm-kernel-outgoing; Tue, 22 Feb 2000 14:21:15 GMT Received: from c003.sfo.cp.net (c003-h004.c003.sfo.cp.net [209.228.14.112]) by parcelfarce.linux.theplanet.co.uk (8.9.3/8.9.3) with SMTP id OAA11049 for ; Tue, 22 Feb 2000 14:21:10 GMT From: cm_lee@sina.com.tw Received: (cpmta 22674 invoked from network); 22 Feb 2000 06:14:09 -0800 Date: 22 Feb 2000 06:14:09 -0800 Message-ID: <20000222141409.22673.cpmta@c003.sfo.cp.net> X-Sent: 22 Feb 2000 14:14:09 GMT Received: from [140.109.20.119] by mail.sina.com.tw with HTTP; 22 Feb 2000 06:14:09 PST Content-Type: text/plain; charset=big5 Content-Disposition: inline Mime-Version: 1.0 To: linux-arm-kernel@lists.arm.linux.org.uk X-Mailer: Web Mail 3.0 Subject: Building arm toolchain problem Sender: owner-linux-arm-kernel@lists.arm.linux.org.uk Precedence: bulk Dear All: I have problem on building arm tool chain!! I use the following source, kernel 2.2.14, gcc-2.95.2, glibc-2.1.2, binutils-2.9.5.0.27. I followed the processed proposed by Chris Rutter (Thank you, Chris), except that I can not find the suitable patch for gcc 2.95.2 and binutils-2.9.5.0.27. The gcc and binutils are normally built. However, when I build the glibc, the make process is out with this error "/usr/local/arm/bin/ld: unrecognized emulation mode: elf32arm Supported emulations: armelf_linux armelf_linux26 armelf" Could anyone tell me why? Thanks kirk _______________________________________________________________ http://www.SINA.com - #1 Destination Site for Chinese Worldwide unsubscribe: body of `unsubscribe linux-arm-kernel' to majordomo@lists.arm.linux.org.uk From owner-linux-arm-kernel@lists.arm.linux.org.uk Wed Feb 23 07:12:09 2000 Received: (from majordomo@localhost) by parcelfarce.linux.theplanet.co.uk (8.9.3/8.9.3) id HAA14027 for linux-arm-kernel-outgoing; Wed, 23 Feb 2000 07:12:09 GMT Received: from c003.sfo.cp.net (c003-h002.c003.sfo.cp.net [209.228.14.110]) by parcelfarce.linux.theplanet.co.uk (8.9.3/8.9.3) with SMTP id HAA14023 for ; Wed, 23 Feb 2000 07:12:07 GMT From: cm_lee@sina.com.tw Received: (cpmta 23494 invoked from network); 22 Feb 2000 23:09:39 -0800 Date: 22 Feb 2000 23:09:39 -0800 Message-ID: <20000223070939.23493.cpmta@c003.sfo.cp.net> X-Sent: 23 Feb 2000 07:09:39 GMT Received: from [140.109.20.119] by mail.sina.com.tw with HTTP; 22 Feb 2000 23:09:39 PST Content-Type: text/plain; charset=big5 Content-Disposition: inline Mime-Version: 1.0 To: jsutherland@accelent.com;, linux-arm-kernel@lists.arm.linux.org.uk X-Mailer: Web Mail 3.0 Subject: RE: Building arm toolchain problem Sender: owner-linux-arm-kernel@lists.arm.linux.org.uk Precedence: bulk > I use the following source, kernel 2.2.14, gcc-2.95.2, > glibc-2.1.2, binutils-2.9.5.0.27. > > However, when I build the glibc, the make process is out with this error > "/usr/local/arm/bin/ld: unrecognized emulation mode: elf32arm > Supported emulations: armelf_linux armelf_linux26 armelf" > > George France from the cross-gcc mailing list sent me this hack after I > stumbled onto this problem: > > Use your favorite editor to modify the gcc file 'specs', to get the correct > behaviour from gcc, amend the *link section of your specs file as follows: > > *link: > %{h*} %{version:-v} %{b} %{Wl,*:%*} %{static:-Bstatic} > %{shared:-shared > } %{symbolic:-Bsymbolic} %{rdynamic:-export-dynamic} > %{!dynamic-linker: > -dynamic-linker /lib/ld-linux.so.2} -X %{mbig-endian:-EB} > %{mapcs-26:-m ar > 2) melf_linux26} %{!mapcs-26:-m armelf_linux} -p It works!! Thanks a lot! However, now I encountered another error on building glibc. When it starts to build libdb1.so, arm-linux-gcc just lists many "multiple definition of __(variable name like __dummy , __gcc_bcmp ...etc)" messages. And says that these variables are first defined in libgcc.a !! What happened!!?? How shoud I do to avoid these errors?? Thanks! kirk _______________________________________________________________ http://www.SINA.com - #1 Destination Site for Chinese Worldwide unsubscribe: body of `unsubscribe linux-arm-kernel' to majordomo@lists.arm.linux.org.uk From owner-linux-arm-kernel@lists.arm.linux.org.uk Fri Feb 25 20:44:38 2000 Received: (from majordomo@localhost) by parcelfarce.linux.theplanet.co.uk (8.9.3/8.9.3) id UAA27366 for linux-arm-kernel-outgoing; Fri, 25 Feb 2000 20:44:38 GMT Received: from crl.dec.com (crl.dec.com [192.58.206.2]) by parcelfarce.linux.theplanet.co.uk (8.9.3/8.9.3) with ESMTP id UAA27362 for ; Fri, 25 Feb 2000 20:44:37 GMT Received: from crlwebmail.crl.dec.com (crlwebmail.crl.dec.com [16.11.0.2]) by crl.dec.com (8.8.8/RWD-1.2) with ESMTP id PAA07142; Fri, 25 Feb 2000 15:42:58 -0500 (EST) Received: by crlwebmail.crl.dec.com with Internet Mail Service (5.5.2650.21) id ; Fri, 25 Feb 2000 15:46:34 -0500 Message-ID: From: George France To: linux-arm@vger.rutgers.edu, linux-arm-kernel@lists.arm.linux.org.uk Subject: Binary tar ball of i386 to armv4l (Strong-ARM) Cross Tool Chain a vailable for download. Date: Fri, 25 Feb 2000 15:43:14 -0500 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2650.21) Content-Type: text/plain; charset="iso-8859-1" Sender: owner-linux-arm-kernel@lists.arm.linux.org.uk Precedence: bulk There have been so many questions and problems on the mailing list about building a Cross Tool Chain. We have decided to post a 38 Mbyte binary (executable) tar-ball here: http://crl.research.compaq.com/projects/personalserver/sw_download.html it is an arm-linux cross Tool Chain. The Tool Chain is made up of: binutils-2.9.5.0.22, gcc-2.95.2 - C, C++, f77, CHILL, java and objc. glibc-2.1.2. with the International crypt library. (THIS IS NOT FOR EXPORT) The Tool Chain is compiled for a i386 host with an armv4l target. Armv4 = the instructions set for the SA-110, SA-1100 and SA-1110 processors. l = little edian. Installation notes: The Tool Chain must be installed in /skiff/local. It will not work from any other path. The only other problem that you may have with the include files is that the tar ball was setup for Linux 2.2.14. You may need to setup a symbolic links for: ln -s /usr/src/linux/include/asm /skiff/local/arm-linux/include/asm ln -s /usr/src/linux/include/linux /skiff/local/arm-linux/include/linux or you could copy the /usr/src/linux/include/asm and /usr/src/linux/include/linux directories to the /skiff/local/arm-linux/include. Be certain that you have run 'make menuconfig' or equivalent, then 'make dep'. This will verify that your kernel tree is up-to-date and the correct symbolic links are up-to-date. This tool chain has glibc symbol versioning. If you are using a netwinder, you may have to compile your code with static libs. If there is interest, we may make a binary i386 to Strong-ARM cross gdb available for remote debugging. --George George France, france@crl.dec.com Cambridge Research Laboratory, Compaq Computer Corporation One Kendall Square, Building 700 MS: CRL Cambridge, MA 02139 USA unsubscribe: body of `unsubscribe linux-arm-kernel' to majordomo@lists.arm.linux.org.uk From owner-linux-arm-kernel@lists.arm.linux.org.uk Fri Feb 25 21:35:34 2000 Received: (from majordomo@localhost) by parcelfarce.linux.theplanet.co.uk (8.9.3/8.9.3) id VAA27569 for linux-arm-kernel-outgoing; Fri, 25 Feb 2000 21:35:34 GMT Received: from duteinh.et.tudelft.nl (duteinh.et.tudelft.nl [130.161.42.1]) by parcelfarce.linux.theplanet.co.uk (8.9.3/8.9.3) with ESMTP id VAA27565 for ; Fri, 25 Feb 2000 21:35:34 GMT Received: from kalman by duteinh.et.tudelft.nl (8.9.3/BB-2.0) id WAA13191; Fri, 25 Feb 2000 22:33:50 +0100 (MET) Date: Fri, 25 Feb 2000 22:33:50 +0100 (MET) Message-Id: <200002252133.WAA13191@duteinh.et.tudelft.nl> From: Erik Mouw To: france@crl.dec.com Cc: linux-arm@vger.rutgers.edu, linux-arm-kernel@lists.arm.linux.org.uk Subject: Re: Binary tar ball of i386 to armv4l (Strong-ARM) Cross Tool Chain a vailable for download. In-Reply-To: Organization: Eric conspiracy secret labs X-Face: "!Z|1XjFg!3Iw)BwM^M\k)J5KHiPddz}#/Zw%?$&Q!xqob"OA|})3_@&BTTiQERo?/^g[Z<9P4R8:5C:aqq.cw!hotLc=+u-"+1rq@9$V^8PbOUlodyi_sejt+D`iqkLpx$>.aXl%vnfPIt"LK|/%>94_Zrm8jy?`_[RKt'TuoxUxfk(gMhY(n|DBBl~c$dhO>G%6}R4+DGX3|idBm4oU&UM/XUP4I9aoN{]`ax%B?rgD(I3OJ<62Y6L X-Mailer: XCmail 1.0.0 - with PGP support, PGP engine version 0.5 (Linux) X-Mailerorigin: http://www.fsai.fh-trier.de/~schmitzj/Xclasses/XCmail/ X-Eric-Conspiracy: There is no conspiracy! X-Loop: erik@it.et.tudelft.nl MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Sender: owner-linux-arm-kernel@lists.arm.linux.org.uk Precedence: bulk On Fri, 25 Feb 2000 15:43:14 -0500, George France wrote: > There have been so many questions and problems on the mailing list about > building a Cross Tool Chain. We have decided to post a 38 Mbyte binary > (executable) tar-ball here: > > > http://crl.research.compaq.com/projects/personalserver/sw_download.html > > it is an arm-linux cross Tool Chain. The Tool Chain is made up of: > > binutils-2.9.5.0.22, > gcc-2.95.2 - C, C++, f77, CHILL, java and objc. > glibc-2.1.2. with the International crypt library. (THIS IS NOT FOR > EXPORT) > > The Tool Chain is compiled for a i386 host with an armv4l target. Armv4 = > the instructions set for the SA-110, SA-1100 and SA-1110 processors. l = > little edian. Hmm, it's a little embarrasing for me to tell you that you have been doing double work. I announced my tarball with almost the same specifications (only C and C++ as languages) two weeks ago. My toolchain was built from: binutils-2.9.5.0.22.tar.bz2 linux-2.2.13.tar.gz + patch-2.2.13-rmk2.gz + diff-2.2.13-rmk2-np14.gz gcc-2.95.2.tar.gz + gcc-2.95.2-diff-991022.gz + chris-rutter-patch-for-gcc-2.95.2 glibc-2.1.2.tar.bz2 + glibc-crypt-2.1.tar.gz + glibc-linuxthreads-2.1.2.tar.bz2 I only announced it at the linux-arm and sa1100-linux mailing lists, maybe I should have send the announcement to the linux-arm-kernel mailing list, too. My toolchain can be freely downloaded by non-US users (but also by US users, of course ;-), because we don't have cryptographic export rules over here in The Netherlands. The URL: http://www-ict.its.tudelft.nl/~erik/open-source/arm-linux-cross/ Erik -- J.A.K. (Erik) Mouw, Information and Communication Theory Group, Department of Electrical Engineering, Faculty of Information Technology and Systems, Delft University of Technology, PO BOX 5031, 2600 GA Delft, The Netherlands Phone: +31-15-2785859 Fax: +31-15-2781843 Email J.A.K.Mouw@its.tudelft.nl WWW: http://www-ict.its.tudelft.nl/~erik/ -- LART. 250 MIPS under one Watt. http://www-ict.its.tudelft.nl/~erik/open-source/LART/ unsubscribe: body of `unsubscribe linux-arm-kernel' to majordomo@lists.arm.linux.org.uk From owner-linux-arm-kernel@lists.arm.linux.org.uk Sun Feb 27 07:01:15 2000 Received: (from majordomo@localhost) by parcelfarce.linux.theplanet.co.uk (8.9.3/8.9.3) id HAA02669 for linux-arm-kernel-outgoing; Sun, 27 Feb 2000 07:01:15 GMT Received: from fire.light.com (IDENT:root@c657338-a.plstn1.sfba.home.com [24.19.133.127]) by parcelfarce.linux.theplanet.co.uk (8.9.3/8.9.3) with ESMTP id HAA02665 for ; Sun, 27 Feb 2000 07:01:13 GMT Received: from localhost (rscott@localhost) by fire.light.com (8.9.3/8.9.3) with ESMTP id XAA02337 for ; Sat, 26 Feb 2000 23:06:56 -0800 Date: Sat, 26 Feb 2000 23:06:55 -0800 (PST) From: Rob Scott To: linux-arm-kernel@lists.arm.linux.org.uk Subject: Domain specifications Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-linux-arm-kernel@lists.arm.linux.org.uk Precedence: bulk I need a clue about how the domain field in a process structure is generated and/or how page/section tables are checked for permission. The problem I'm trying to solve is when switching tasks via proc-arm6,7.S, the domain register is written with the value 1d, and the system eventually hangs solid when an timer interrupt occurs. If the domain register write is commented out, leaving the value 1f, then the kernel finishes booting (until the point it can't find a root device, since I don't have one defined yet). I'm not sure if the 1d value for the domain register is correct. If it is, then that means that "Page and section permission bits are checked", according to "ARM System Architecture". If I'm reading the source correctly, that means that the user memory is being checked. So, my guess is that the page permissions aren't being set correctly, and I don't know where to look to check this. Perhaps I'm missing something in my processor specific includes?? Thanks in advance for any hints! Rob Scott unsubscribe: body of `unsubscribe linux-arm-kernel' to majordomo@lists.arm.linux.org.uk From owner-linux-arm-kernel@lists.arm.linux.org.uk Sun Feb 27 09:25:51 2000 Received: (from majordomo@localhost) by parcelfarce.linux.theplanet.co.uk (8.9.3/8.9.3) id JAA03106 for linux-arm-kernel-outgoing; Sun, 27 Feb 2000 09:25:51 GMT Received: from caramon.arm.linux.org.uk (root@p32-magpie-gui.tch.enablis.net [194.168.180.32]) by parcelfarce.linux.theplanet.co.uk (8.9.3/8.9.3) with ESMTP id JAA03102 for ; Sun, 27 Feb 2000 09:25:49 GMT Received: from flint.arm.linux.org.uk (root@flint [192.168.0.4]) by caramon.arm.linux.org.uk (8.9.3/8.9.3) with ESMTP id JAA00703; Sun, 27 Feb 2000 09:22:41 GMT Received: (from linux@localhost) by flint.arm.linux.org.uk (8.9.3/8.9.3) id JAA05589; Sun, 27 Feb 2000 09:16:57 GMT From: Russell King - ARM Linux Admin Message-Id: <200002270916.JAA05589@flint.arm.linux.org.uk> Subject: Re: Domain specifications To: rscott@mtrob.fdns.net (Rob Scott) Date: Sun, 27 Feb 2000 09:16:56 +0000 (GMT) Cc: linux-arm-kernel@lists.arm.linux.org.uk In-Reply-To: from "Rob Scott" at Feb 26, 2000 11:06:55 PM X-Mailer: ELM [version 2.5 PL1] MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-linux-arm-kernel@lists.arm.linux.org.uk Precedence: bulk Rob Scott writes: > The problem I'm trying to solve is when switching tasks via proc-arm6,7.S, > the domain register is written with the value 1d, and the system 1d is one of the expected values that is written to the domain register from any task that is not going to access 'kernel' data: 1 d 01 11 01 | | \- user domain (client) | \---- kernel and page table domain (manager) \------- IO domain (client) The page tables are setup such that virtual page0 (with the machine exception vectors in) is write-protected from SVC mode (since the user domain is set to client). This relies on the 'S' bit in the control register to be set, but the 'R' bit to be clear. > eventually hangs solid when an timer interrupt occurs. If the domain > register write is commented out, leaving the value 1f, then the kernel > finishes booting (until the point it can't find a root device, since I > don't have one defined yet). That sounds like virtual page0 is being unmapped by something, or the control register is getting corrupted, and causing the meaning of one combination of permission bits to mean "inaccessible" rather than "write protected". Could you supply details of which kernel version, and the exact processor type you are experiencing this problem with please? > Perhaps I'm missing something in my processor specific includes?? Have you changed them? _____ |_____| ------------------------------------------------- ---+---+- | | Russell King linux@arm.linux.org.uk --- --- | | | | http://www.arm.linux.org.uk/~rmk/armlinux.html / / | | +-+-+ --- -+- / | THE developer of ARM Linux |+| /|\ / | | | --- | +-+-+ ------------------------------------------------- /\\\ | unsubscribe: body of `unsubscribe linux-arm-kernel' to majordomo@lists.arm.linux.org.uk From owner-linux-arm-kernel@lists.arm.linux.org.uk Sun Feb 27 16:03:23 2000 Received: (from majordomo@localhost) by parcelfarce.linux.theplanet.co.uk (8.9.3/8.9.3) id QAA04514 for linux-arm-kernel-outgoing; Sun, 27 Feb 2000 16:03:23 GMT Received: from fire.light.com (IDENT:root@c657338-a.plstn1.sfba.home.com [24.19.133.127]) by parcelfarce.linux.theplanet.co.uk (8.9.3/8.9.3) with ESMTP id QAA04510 for ; Sun, 27 Feb 2000 16:03:22 GMT Received: from localhost (rscott@localhost) by fire.light.com (8.9.3/8.9.3) with ESMTP id IAA02850; Sun, 27 Feb 2000 08:08:44 -0800 Date: Sun, 27 Feb 2000 08:08:43 -0800 (PST) From: Rob Scott To: Russell King - ARM Linux Admin cc: linux-arm-kernel@lists.arm.linux.org.uk Subject: Re: Domain specifications In-Reply-To: <200002270916.JAA05589@flint.arm.linux.org.uk> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-linux-arm-kernel@lists.arm.linux.org.uk Precedence: bulk On Sun, 27 Feb 2000, Russell King - ARM Linux Admin wrote: > Rob Scott writes: > > The problem I'm trying to solve is when switching tasks via proc-arm6,7.S, > > the domain register is written with the value 1d, and the system > > 1d is one of the expected values that is written to the domain register > from any task that is not going to access 'kernel' data: > > 1 d > 01 11 01 > | | \- user domain (client) > | \---- kernel and page table domain (manager) > \------- IO domain (client) > > The page tables are setup such that virtual page0 (with the machine > exception vectors in) is write-protected from SVC mode (since the user > domain is set to client). This relies on the 'S' bit in the control > register to be set, but the 'R' bit to be clear. > Thanks for the explaination! It's good to know that 1d is correct... > > eventually hangs solid when an timer interrupt occurs. If the domain > > register write is commented out, leaving the value 1f, then the kernel > > finishes booting (until the point it can't find a root device, since I > > don't have one defined yet). > > That sounds like virtual page0 is being unmapped by something, or the control > register is getting corrupted, and causing the meaning of one combination > of permission bits to mean "inaccessible" rather than "write protected". Humm - I'm assuming that this would cause some sort of MMU fault, but no error message is generated... > > Could you supply details of which kernel version, and the exact processor > type you are experiencing this problem with please? The base kernel is 2.2.14, patched to rmk3. I've been adding code for my processor, which is an ARM720T. It's embedded in a LinkUp Systems L7200. > > > Perhaps I'm missing something in my processor specific includes?? > > Have you changed them? Actually, since this is a new processor type, I've had to create them. I've used bits from arch-rpc and arch-ebsa100. Thanks, Rob unsubscribe: body of `unsubscribe linux-arm-kernel' to majordomo@lists.arm.linux.org.uk From owner-linux-arm-kernel@lists.arm.linux.org.uk Sun Feb 27 16:53:56 2000 Received: (from majordomo@localhost) by parcelfarce.linux.theplanet.co.uk (8.9.3/8.9.3) id QAA04840 for linux-arm-kernel-outgoing; Sun, 27 Feb 2000 16:53:56 GMT Received: from caramon.arm.linux.org.uk (root@p59-robin-gui.tch.enablis.net [194.168.180.119]) by parcelfarce.linux.theplanet.co.uk (8.9.3/8.9.3) with ESMTP id QAA04836 for ; Sun, 27 Feb 2000 16:53:54 GMT Received: from raistlin.arm.linux.org.uk (root@raistlin [192.168.0.3]) by caramon.arm.linux.org.uk (8.9.3/8.9.3) with ESMTP id QAA02428; Sun, 27 Feb 2000 16:52:17 GMT From: Russell King - ARM Linux Admin Received: (from linux@localhost) by raistlin.arm.linux.org.uk (8.7.4/8.7.3) id QAA05183; Sun, 27 Feb 2000 16:52:14 GMT Message-Id: <200002271652.QAA05183@raistlin.arm.linux.org.uk> Subject: Re: Domain specifications To: rscott@mtrob.fdns.net (Rob Scott) Date: Sun, 27 Feb 2000 16:52:14 +0000 (GMT) Cc: linux-arm-kernel@lists.arm.linux.org.uk In-Reply-To: from "Rob Scott" at Feb 27, 2000 08:08:43 AM X-Mailer: ELM [version 2.5 PL1] MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-linux-arm-kernel@lists.arm.linux.org.uk Precedence: bulk Rob Scott writes: > > That sounds like virtual page0 is being unmapped by something, or the control > > register is getting corrupted, and causing the meaning of one combination > > of permission bits to mean "inaccessible" rather than "write protected". > > Humm - I'm assuming that this would cause some sort of MMU fault, but no > error message is generated... Unfortunately, if the first page goes amiss, then no error message can be generated - this page contains the machine exception vectors, so you end up getting stuck in the prefetch abort vector. (the processor tries to fetch an instruction from the exception vectors, and prefetch aborts ad infinitum). _____ |_____| ------------------------------------------------- ---+---+- | | Russell King linux@arm.linux.org.uk --- --- | | | | http://www.arm.linux.org.uk/~rmk/armlinux.html / / | | +-+-+ --- -+- / | THE developer of ARM Linux |+| /|\ / | | | --- | +-+-+ ------------------------------------------------- /\\\ | unsubscribe: body of `unsubscribe linux-arm-kernel' to majordomo@lists.arm.linux.org.uk From owner-linux-arm-kernel@lists.arm.linux.org.uk Sun Feb 27 18:12:16 2000 Received: (from majordomo@localhost) by parcelfarce.linux.theplanet.co.uk (8.9.3/8.9.3) id SAA05200 for linux-arm-kernel-outgoing; Sun, 27 Feb 2000 18:12:16 GMT Received: from kings-cross.london.uk.eu.org (mail@tazenda.demon.co.uk [158.152.220.239]) by parcelfarce.linux.theplanet.co.uk (8.9.3/8.9.3) with ESMTP id SAA05196 for ; Sun, 27 Feb 2000 18:12:14 GMT Received: from localhost ([::ffff:127.0.0.1] helo=kings-cross.london.uk.eu.org ident=phil) by kings-cross.london.uk.eu.org with esmtp (Exim 3.11 #1) id 12P829-00006U-00; Sun, 27 Feb 2000 18:03:09 +0000 X-Mailer: exmh version 2.0.2 2/24/98 (debian) To: patches@arm.linux.org.uk, linux-arm-kernel@lists.arm.linux.org.uk Subject: assembler tweaks Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sun, 27 Feb 2000 18:03:09 +0000 From: Philip Blundell Message-Id: Sender: owner-linux-arm-kernel@lists.arm.linux.org.uk Precedence: bulk Here is a small patch containing some assembler optimizations: 1. Replace STM with STR where appropriate 2. No point including the WFS-ignorer if NWFPE is built in. 3. No point zeroing R10 on entry to the kernel if -fomit-frame-pointer is in use. PATCH FOLLOWS diff -udpNr rmk/linux/arch/arm/lib/findbit.S linux-000227/arch/arm/lib/findbit.S --- rmk/linux/arch/arm/lib/findbit.S Mon Feb 7 21:22:51 2000 +++ linux-000227/arch/arm/lib/findbit.S Fri Jan 14 22:28:51 2000 @@ -53,10 +53,10 @@ ENTRY(find_next_zero_bit) beq Lfindzbit1lp @ If new byte, goto old routine ldrb r3, [r0, r2, lsr#3] orr r3, r3, #0xFF00 @ Set top bits so we wont get confused - stmfd sp!, {r4} + str r4, [sp, #-4]! and r4, r2, #7 mov r3, r3, lsr r4 @ Shift right by no. of bits - ldmfd sp!, {r4} + ldr r4, [sp], #4 and r3, r3, #0xFF teq r3, #0xFF orreq r2, r2, #7 diff -udpNr rmk/linux/arch/arm/lib/strrchr.S linux-000227/arch/arm/lib/strrchr.S --- rmk/linux/arch/arm/lib/strrchr.S Mon Feb 7 21:22:51 2000 +++ linux-000227/arch/arm/lib/strrchr.S Sun Feb 27 17:24:32 2000 @@ -12,7 +12,7 @@ .text ENTRY(strrchr) - stmfd sp!, {lr} + str lr, [sp, #-4]! mov r3, #0 1: ldrb r2, [r0], #1 teq r2, r1 diff -udpNr rmk/linux/arch/arm/kernel/entry-armo.S linux-000227/arch/arm/kernel/entry-armo.S --- rmk/linux/arch/arm/kernel/entry-armo.S Mon Feb 7 21:22:51 2000 +++ linux-000227/arch/arm/kernel/entry-armo.S Sun Jan 9 16:16:18 2000 @@ -22,6 +22,7 @@ * adhering to the above criteria. */ #include +#include #include #include @@ -79,6 +80,12 @@ #define S_R1 4 #define S_R0 0 +#ifdef CONFIG_FRAME_POINTER +#define zero_frame mov fp, #0 +#else +#define zero_frame +#endif + #ifdef IOC_BASE /* IOC / IOMD based hardware */ .equ ioc_base_high, IOC_BASE & 0xff000000 @@ -213,7 +220,7 @@ irq_prio_h: .byte 0, 8, 9, 8,10,10,10,1 stmfd sp!, {r0 - r12} ;\ mov r0, #-1 ;\ str r0, [sp, #S_OLD_R0] ;\ - mov fp, #0 + zero_frame #define SVC_IRQ_SAVE_ALL \ str sp, [sp, #-16]! ;\ @@ -224,7 +231,7 @@ irq_prio_h: .byte 0, 8, 9, 8,10,10,10,1 stmfd sp!, {r0 - r12} ;\ mov r0, #-1 ;\ str r0, [sp, #S_OLD_R0] ;\ - mov fp, #0 + zero_frame #define SVC_RESTORE_ALL \ ldmfd sp, {r0 - pc}^ @@ -261,7 +268,7 @@ vector_undefinstr: tst lr,#3 bne __und_svc save_user_regs - mov fp, #0 + zero_frame teqp pc, #I_BIT | MODE_SVC .Lbug_undef: ldr r4, .LC2 @@ -283,6 +290,10 @@ __und_svc: SVC_SAVE_ALL @ Non-user mo bl SYMBOL_NAME(do_undefinstr) SVC_RESTORE_ALL +#ifdef CONFIG_NWFPE + /* The FPE is always present */ + .equ fpe_not_present, 0 +#else /* We get here if an undefined instruction happens and the floating * point emulator is not present. If the offending instruction was * a WFS, we just perform a normal return as if we had emulated the @@ -319,6 +330,7 @@ wfs_mask_data: .word 0x0e200110 @ WFS/ .word 0x0d0d0100 @ LDF [sp]/STF [sp] .word 0x0d0b0100 @ LDF [fp]/STF [fp] .word 0x0f0f0f00 +#endif .LC2: .word SYMBOL_NAME(fp_enter) @@ -686,7 +698,7 @@ ENTRY(__switch_to) /* * initialise the trap system */ -ENTRY(trap_init) +ENTRY(__trap_init) stmfd sp!, {r4 - r7, lr} adr r1, .Ljump_addresses ldmia r1, {r1 - r7, ip, lr} diff -udpNr rmk/linux/arch/arm/kernel/entry-armv.S linux-000227/arch/arm/kernel/entry-armv.S --- rmk/linux/arch/arm/kernel/entry-armv.S Mon Feb 7 21:22:51 2000 +++ linux-000227/arch/arm/kernel/entry-armv.S Sat Jan 15 10:56:26 2000 @@ -68,6 +68,12 @@ #define OFF_CR_ALIGNMENT(x) cr_alignment - x +#ifdef CONFIG_FRAME_POINTER +#define zero_frame mov fp, #0 +#else +#define zero_frame +#endif + #ifdef IOC_BASE /* IOC / IOMD based hardware */ #include @@ -421,7 +546,7 @@ __und_invalid: sub sp, sp, #S_FRAME_SIZE ldr r4, .LCund mov r1, #BAD_UNDEFINSTR @ int reason -1: mov fp, #0 +1: zero_frame ldmia r4, {r5 - r7} @ Get XXX pc, cpsr, old_r0 add r4, sp, #S_PC stmia r4, {r5 - r7} @ Save XXX pc, cpsr, old_r0 @@ -688,11 +821,11 @@ t: .ascii "Prefetch -> undefined instru ENTRY(__switch_to) stmfd sp!, {r4 - sl, fp, lr} @ Store most regs on stack mrs ip, cpsr - stmfd sp!, {ip} @ Save cpsr_SVC + str ip, [sp, #-4]! @ Save cpsr_SVC str sp, [r0, #TSS_SAVE] @ Save sp_SVC ldr sp, [r1, #TSS_SAVE] @ Get saved sp_SVC ldr r2, [r1, #TSS_DOMAIN] - ldmfd sp!, {ip} + ldr ip, [sp], #4 mcr p15, 0, r2, c3, c0 @ Set domain register msr spsr, ip @ Save tasks CPSR into SPSR for this return ldmfd sp!, {r4 - sl, fp, pc}^ @ Load all regs saved previously diff -udpNr rmk/linux/arch/arm/kernel/entry-common.S linux-000227/arch/arm/kernel/entry-common.S --- rmk/linux/arch/arm/kernel/entry-common.S Sun Feb 20 14:10:43 2000 +++ linux-000227/arch/arm/kernel/entry-common.S Sun Feb 27 15:29:10 2000 @@ -85,7 +76,7 @@ tip .req r7 @ temporary IP .align 5 vector_swi: save_user_regs mask_pc lr, lr - mov fp, #0 + zero_frame ldr scno, [lr, #-4] @ get SWI instruction arm700_bug_check scno, ip #ifdef CONFIG_ALIGNMENT_TRAP unsubscribe: body of `unsubscribe linux-arm-kernel' to majordomo@lists.arm.linux.org.uk From owner-linux-arm-kernel@lists.arm.linux.org.uk Sun Feb 27 18:21:53 2000 Received: (from majordomo@localhost) by parcelfarce.linux.theplanet.co.uk (8.9.3/8.9.3) id SAA05301 for linux-arm-kernel-outgoing; Sun, 27 Feb 2000 18:21:53 GMT Received: from kings-cross.london.uk.eu.org (mail@tazenda.demon.co.uk [158.152.220.239]) by parcelfarce.linux.theplanet.co.uk (8.9.3/8.9.3) with ESMTP id SAA05297 for ; Sun, 27 Feb 2000 18:21:51 GMT Received: from localhost ([::ffff:127.0.0.1] helo=kings-cross.london.uk.eu.org ident=phil) by kings-cross.london.uk.eu.org with esmtp (Exim 3.11 #1) id 12P8I9-0000G7-00; Sun, 27 Feb 2000 18:19:41 +0000 X-Mailer: exmh version 2.0.2 2/24/98 (debian) To: patches@arm.linux.org.uk, linux-arm-kernel@lists.arm.linux.org.uk Subject: system call changes Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sun, 27 Feb 2000 18:19:41 +0000 From: Philip Blundell Message-Id: Sender: owner-linux-arm-kernel@lists.arm.linux.org.uk Precedence: bulk Currently the system call entry path is set up to directly handle functions with up to six arguments. This is wasteful because there is only one, comparatively rare, system call that actually uses all six arguments; all the others are forced to push a useless word onto the stack. This patch reworks things so that only a single word of arguments is pushed in the standard entry veneer; sys_mmap2 is implemented in assembler so it can sort its own arguments out. The other place that has to change is sys_syscall, which fortuitously ends up a cycle or two faster as well. I'd be interested to hear thoughts on this. PATCH FOLLOWS --- rmk/linux/arch/arm/kernel/entry-common.S Sun Feb 20 14:10:43 2000 +++ armlinux/arch/arm/kernel/entry-common.S Sun Feb 27 18:03:26 2000 @@ -4,16 +4,6 @@ */ #define S_OFF 8 -#define SYSCALL_REGS r4, r5 - -/* - * Define to favour ARM8, ARM9 and StrongARM cpus. This says that it is - * cheaper to use two LDR instructions than a two-register LDM, if the - * latter would entail calculating an address specially. - */ -#if defined(CONFIG_CPU_SA110) -#define HARVARD_CACHE -#endif .macro get_softirq, rd #ifdef __SMP__ @@ -95,9 +85,7 @@ #endif enable_irqs ip - stmdb sp!, {SYSCALL_REGS} @ new style: (r0 = arg1, r4 = arg5, r5 = arg6) - @ Note that we dont have to handle - @ sys_syscalls arg7 here + str r4, [sp, #-S_OFF]! @ push fifth arg adrsvc al, lr, fast_syscall_return bic scno, scno, #0xff000000 @ mask off SWI op-code @@ -141,10 +129,11 @@ b SYMBOL_NAME(deferred) .align 5 - + .type __softirq_state, %object __softirq_state: .word SYMBOL_NAME(softirq_state) + .type sys_call_table, %object ENTRY(sys_call_table) #include "calls.S" @@ -153,15 +142,17 @@ */ @ r0 = syscall number @ r5 = syscall table + .type sys_syscall, %function SYMBOL_NAME(sys_syscall): eor scno, r0, #OS_NUMBER << 20 cmp scno, #NR_syscalls @ check range - add ip, sp, #S_OFF - ldmleib ip, {r0 - r3, SYSCALL_REGS} @ get our args - stmleia sp, {SYSCALL_REGS} @ Put our arg on the stack + stmleia sp, {r5, r6} @ shuffle args + movle r0, r1 + movle r1, r2 + movle r2, r3 + movle r3, r4 ldrle pc, [tbl, scno, lsl #2] - mov r0, #-ENOSYS - RETINSTR(mov,pc,lr) + b sys_ni_syscall sys_fork_wrapper: add r0, sp, #S_OFF @@ -198,6 +189,23 @@ sys_sigaltstack_wrapper: ldr r2, [sp, #S_OFF + S_SP] b do_sigaltstack + +/* + * Note: off_4k is always units of 4K. If we can't do the requested + * offset, we return EINVAL. + */ +sys_mmap2: +#if PAGE_SHIFT > 12 + tst r5, #PGOFF_MASK + moveq r5, r5, lsr #PGOFF_SHIFT + streq r5, [sp, #4] + beq do_mmap2 + mov r0, #-EINVAL + mov pc, lr +#else + str r5, [sp, #4] + b do_mmap2 +#endif .data diff -udpNr rmk/linux/arch/arm/kernel/sys_arm.c linux-000227/arch/arm/kernel/sys_arm.c --- rmk/linux/arch/arm/kernel/sys_arm.c Mon Feb 7 21:23:42 2000 +++ linux-000227/arch/arm/kernel/sys_arm.c Mon Feb 7 22:02:55 2000 @@ -50,7 +50,7 @@ asmlinkage int sys_pipe(unsigned long * } /* common code for old and new mmaps */ -static inline long do_mmap2( +long int do_mmap2( unsigned long addr, unsigned long len, unsigned long prot, unsigned long flags, unsigned long fd, unsigned long pgoff) @@ -77,27 +77,6 @@ static inline long do_mmap2( fput(file); out: return error; -} - -#define PGOFF_SHIFT (PAGE_SHIFT - 12) -#define PGOFF_MASK (~((1 << PGOFF_SHIFT) - 1)) - -/* - * Note: off_4k is always units of 4K. If we can't do the requested - * offset, we return EINVAL. - */ -asmlinkage long -sys_mmap2(unsigned long addr, unsigned long len, unsigned long prot, - unsigned long flags, unsigned long fd, unsigned long off_4k) -{ - unsigned long pgoff; - - if (off_4k & ~PGOFF_MASK) - return -EINVAL; - - pgoff = off_4k >> PGOFF_SHIFT; - - return do_mmap2(addr, len, prot, flags, fd, pgoff); } struct mmap_arg_struct { unsubscribe: body of `unsubscribe linux-arm-kernel' to majordomo@lists.arm.linux.org.uk From owner-linux-arm-kernel@lists.arm.linux.org.uk Sun Feb 27 20:47:09 2000 Received: (from majordomo@localhost) by parcelfarce.linux.theplanet.co.uk (8.9.3/8.9.3) id UAA05870 for linux-arm-kernel-outgoing; Sun, 27 Feb 2000 20:47:09 GMT Received: from kings-cross.london.uk.eu.org (mail@tazenda.demon.co.uk [158.152.220.239]) by parcelfarce.linux.theplanet.co.uk (8.9.3/8.9.3) with ESMTP id UAA05865 for ; Sun, 27 Feb 2000 20:47:07 GMT Received: from localhost ([::ffff:127.0.0.1] helo=kings-cross.london.uk.eu.org ident=phil) by kings-cross.london.uk.eu.org with esmtp (Exim 3.11 #1) id 12PAZ4-0000Hq-00; Sun, 27 Feb 2000 20:45:18 +0000 X-Mailer: exmh version 2.0.2 2/24/98 (debian) To: Russell King Cc: linux-arm-kernel@lists.arm.linux.org.uk Subject: Re: Patch received: system call changes In-Reply-To: Message from Russell King of "Sun, 27 Feb 2000 20:40:33 GMT." <200002272040.UAA08130@raistlin.arm.linux.org.uk> References: <200002272040.UAA08130@raistlin.arm.linux.org.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sun, 27 Feb 2000 20:45:18 +0000 From: Philip Blundell Message-Id: Sender: owner-linux-arm-kernel@lists.arm.linux.org.uk Precedence: bulk >You're missing definitions for PGOFF_MASK and PGOFF_SHIFT Oh yes, sorry. p. diff -udpNr rmk/linux/arch/arm/lib/getconsdata.c linux-000227/arch/arm/lib/getconsdata.c --- rmk/linux/arch/arm/lib/getconsdata.c Mon Feb 7 21:22:51 2000 +++ linux-000227/arch/arm/lib/getconsdata.c Sat Jan 15 22:53:34 2000 @@ -91,3 +91,6 @@ unsigned long PAGE_SZ = PAGE_SIZE; unsigned long KSWI_BASE = 0x900000; unsigned long KSWI_SYS_BASE = 0x9f0000; unsigned long SYS_ERROR0 = 0x9f0000; + +unsigned long PGOFF_SHIFT = (PAGE_SHIFT - 12); +unsigned long PGOFF_MASK = ((1 << (PAGE_SHIFT - 12)) - 1); unsubscribe: body of `unsubscribe linux-arm-kernel' to majordomo@lists.arm.linux.org.uk