Linux 2.5.25-dj2
A few more items from the pending queue, including the large IDE patch from Jens and Andre, various updates & cleanups. Mostly clearing things out of the backlog before I jet off to Germany for a week..
As usual,..
Patch against 2.5.25 vanilla is available from:
ftp://ftp.kernel.org/pub/linux/kernel/people/davej/patches/2.5/
Merged patch archive: http://www.codemonkey.org.uk/patches/merged/
Check http://www.codemonkey.org.uk/Linux-2.5.html before reporting known bugs that are also in mainline.
-- Davej.
2.5.25-dj2
o 2.4 IDE layer forward port. (Jens Axboe, Andre Hedrick)
o Take some of the new input changes from .25
| by .26, I should be able to go with mainline
| if merging continues at current rate.
o More cleanups in the agpgart driver. (Greg-KH)
o cpufreq update. (Dominik Brodowski)
o Update DMA-mapping documentation. (Adam J. Richter)
o Fix problem with Bochs & new MTRR driver. (Manfred Spraul)
o Fix i815e agpgart support. (Nicolas Aspert)
o Fix lock hang in shm_destroy() (Hugh Dickins)
o Various SCSI driver janitor work. (Michael G. Janicki)
o Allow booting from non-zero boot CPU. (Anton Blanchard)
o ad1848 OSS compile fix. (Ryan Anderson)
o cs46xx compile fix. (Stephen Rothwell)
o 3c509 + non-isapnp compile fix. (James Morris)
o i2o proc compile fix. (Adrian Bunk)
o aironet compile fix. (Adrian Bunk)2.5.25-dj1
o Fix negative memcpy issue in orinoco driver. (Russell King) o Improvements to i386 machine check handler. (Andi Kleen) o Fix various bitops to use unsigned long's. (Geert Uytterhoeven) o Various network driver init_etherdev cleanups. (Crutcher Dunnavant) o Improved list handling in dcache.c (Matthew Wilcox) o Convert various i2c drivers to new interface. (Frank Davis, Albert Cranford) o Consistent usage of static in keyboard maps. (Keith Owens) o Various typo fixes in ipsysctl docs. (Edward J. Huff) o Region handling cleanup in boardergo isdn. (William Stinson) o Region handling cleanup in baycom hamradio driver. (William Stinson) o PCI DMA update for roadrunner HIPPI driver. (Francois Romieu) o RPC recieve cleanup. (Trond Myklebust) o Fix potential O_DIRECT oops. (Andrew Morton) o Fix module creation off-by-one's. (Peter Oberparleiter) o airo, aztcd & sonycd compile fixes. (Martin Dalecki) o Fix misleading tlb flush comment. (Manik Raina) o PCI DMA update for tlan (and misc cleanup) (Francois Romieu) o Cleanup suspend code not to need prototypes. (Pavel Machek) o Numerous pci_enable_device() cleanups. (Michael G. Janicki) o Fix garbage in /proc/partitions. (Peter Chubb) o agpgart compile fix. (Manfred Spraul) o fatfs crapectomy. (Christoph Hellwig) o Don't advise to make dep when not needed. (Christoph Hellwig)
Signature · 2 lines
| Dave Jones. http://www.codemonkey.org.uk | SuSE Labs
On Fri Jul 12 2002, Dave Jones wrote:
[....]
gcc -Wp,-MD,./.ide-scsi24.o.d -D__KERNEL__ -I/usr/src/linux/include -Wall -Wstrict-prototypes -Wno-trigraphs -O2 -fno-strict-aliasing -fno-common -fomit-frame-pointer -pipe -mpreferred-stack-boundary=2 -march=k6 -nostdinc -iwithprefix include -DKBUILD_BASENAME=ide_scsi24-c -o ide-scsi24.o ide-scsi24.c ide-scsi24.c:847: unknown field abort' specified in initializer ide-scsi24.c:847: warning: initialization from incompatible pointer type ide-scsi24.c:848: unknown field reset' specified in initializer ide-scsi24.c:848: warning: initialization from incompatible pointer type make[3]: *** [ide-scsi24.o] Error 1
make[3]: Leaving directory /usr/src/linux/drivers/scsi' make[2]: *** [scsi] Error 2
make[2]: Leaving directory /usr/src/linux/drivers'
make[1]: *** [drivers] Error 2
make[1]: Leaving directory /usr/src/linux'
make: *** [bzImage] Error 2
chiara:/usr/src/linux #
Signature · 1 line
# Heinz Diehl, 68259 Mannheim, Germany
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Quoted 1 line · Heinz Diehl
From: Heinz Diehl <[email protected]>
You beat me to it by a few minutes - I was just about to post that I am experiencing the same error.
Quoted 10 lines
gcc -Wp,-MD,./.ide-scsi24.o.d -D__KERNEL__ -I/usr/src/linux/include -Wall -Wstrict-prototypes -Wno-trigraphs -O2 -fno-strict-aliasing -fno-common -fomit-frame-pointer -pipe -mpreferred-stack-boundary=2 -march=k6 -nostdinc -iwithprefix include -DKBUILD_BASENAME=ide_scsi24-c -o ide-scsi24.o
ide-scsi24.cide-scsi24.c:847: unknown field abort' specified in initializer ide-scsi24.c:847: warning: initialization from incompatible pointer type ide-scsi24.c:848: unknown field reset' specified in initializer ide-scsi24.c:848: warning: initialization from incompatible pointer type
Quoted 10 lines
make[3]: *** [ide-scsi24.o] Error 1
make[3]: Leaving directory/usr/src/linux/drivers/scsi' make[2]: *** [scsi] Error 2
make[2]: Leaving directory/usr/src/linux/drivers' make[1]: *** [drivers] Error 2
make[1]: Leaving directory/usr/src/linux'
make: *** [bzImage] Error 2 chiara:/usr/src/linux #
- --
Ben Clifford [email protected] GPG: 30F06950
http://www.hawaga.org.uk/ben/
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iD8DBQE9MGe/sYXoezDwaVARAqnfAJ0cnls/Xi0NovRg5Cef5T0gGHyL/ACeNe1b VXmftNtyoMicwdlCow/FR58=
=7VdE
-----END PGP SIGNATURE-----
Quoted 6 lines · Ben Clifford
-----BEGIN PGP SIGNED MESSAGE-----
Quoted 4 lines
ide-scsi24.c:847: unknown field abort' specified in initializer ide-scsi24.c:847: warning: initialization from incompatible pointer type ide-scsi24.c:848: unknown field reset' specified in initializer ide-scsi24.c:848: warning: initialization from incompatible pointer type
Just kill those lines.
Dave
Signature · 2 lines
| Dave Jones. http://www.codemonkey.org.uk | SuSE Labs
Quoted 6 lines · Dave Jones
Quoted 4 lines
Quoted 4 lines
ide-scsi24.c:847: unknown field abort' specified in initializer ide-scsi24.c:847: warning: initialization from incompatible pointer type ide-scsi24.c:848: unknown field reset' specified in initializer ide-scsi24.c:848: warning: initialization from incompatible pointer type
Just kill those lines.
This leads to:
make[2]: Entering directory /usr/src/linux/init'
Generating /usr/src/linux/include/linux/compile.h (updated) gcc -Wp,-MD,./.version.o.d -D__KERNEL__ -I/usr/src/linux/include -Wall -Wstrict-prototypes -Wno-trigraphs -O2 -fno-strict-aliasing -fno-common -fomit-frame-pointer -pipe -mpreferred-stack-boundary=2
-march=k6 -nostdinc -iwithprefix include -DKBUILD_BASENAME=version
-c -o version.o version.c
ld -m elf_i386 -r -o init.o main.o version.o do_mounts.o make[2]: Leaving directory /usr/src/linux/init'
ld -m elf_i386 -T arch/i386/vmlinux.lds -e stext arch/i386/kernel/head.o arch/i386/kernel/init_task.o init/init.o --start-group arch/i386/kernel/kernel.o arch/i386/mm/mm.o kernel/kernel.o mm/mm.o fs/fs.o ipc/ipc.o /usr/src/linux/arch/i386/lib/lib.a lib/lib.a /usr/src/linux/arch/i386/lib/lib.a drivers/built-in.o sound/sound.o arch/i386/pci/pci.o net/network.o --end-group -o vmlinux fs/fs.o: In function proc_pid_stat':
fs/fs.o(.text+0x21a29): undefined reference to __udivdi3' fs/fs.o: In function kstat_read_proc':
fs/fs.o(.text+0x22a7b): undefined reference to __udivdi3' fs/fs.o(.text+0x22b11): undefined reference to __udivdi3' make[1]: *** [vmlinux] Error 1
make[1]: Leaving directory /usr/src/linux'
make: *** [bzImage] Error 2
chiara:/usr/src/linux #
Signature · 1 line
# Heinz Diehl, 68259 Mannheim, Germany
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Quoted 3 lines · Heinz Diehl
Quoted 1 line
Just kill those lines.
This leads to:
That happens even if you don't kill those lines. I have ide-scsi as a module and if I say "make bzImage modules" I get the SCSI error handling error, and if I just say "make bzImage" I get the __udivdi3 error.
The only occurence of __udvidi3 in source code seems to be in the arch-specific directories (and not the i386 one - which I am using).
Looks like it is being referenced somehow in the procfs .o files (but not procfs source)
- --
Ben Clifford [email protected] GPG: 30F06950
http://www.hawaga.org.uk/ben/
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iD8DBQE9Md9vsYXoezDwaVARAmH8AJ9Dca0CI0qTU6PkYQ5lAg6C5ZvxkQCeOXJ/ Ux2lMrRF9pRU5P2M9kr3o9c=
=bEVu
-----END PGP SIGNATURE-----
Quoted 1 line · Heinz Diehl
This leads to:
...
Quoted 9 lines
fs/fs.o: In functionproc_pid_stat':
fs/fs.o(.text+0x21a29): undefined reference to__udivdi3'fs/fs.o: In functionkstat_read_proc':
fs/fs.o(.text+0x22a7b): undefined reference to__udivdi3' fs/fs.o(.text+0x22b11): undefined reference to__udivdi3' make[1]: *** [vmlinux] Error 1
make[1]: Leaving directory/usr/src/linux'
make: *** [bzImage] Error 2 chiara:/usr/src/linux #
known problem, a fix has been posted to the list already or can be found at
http://www.physik3.uni-rostock.de/tim/kernel/2.5/jiffies64-djfix06.patch.gz
Tim
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Quoted 6 lines · Dave Jones
Quoted 4 lines
Quoted 4 lines
ide-scsi24.c:847: unknown field abort' specified in initializer ide-scsi24.c:847: warning: initialization from incompatible pointer type ide-scsi24.c:848: unknown field reset' specified in initializer ide-scsi24.c:848: warning: initialization from incompatible pointer type
Just kill those lines.
So I killed those lines (and applied the patch supplied by Tim Schmielau).
My ide-cd module was loaded automatically at boot.
I loaded the ide-scsi24 module (which didn't detect anything because the ide-cd module already had both the CD-ROM and CD-RW drives). So I rmmoded ide-cd and ide-scsi24.
Then I modprobed ide-scsi and got two oopses in quick succession which locked up the machine :-(
Anyone else have any luck with ide-scsi24?
More details to follow.
- --
Ben Clifford [email protected] GPG: 30F06950
http://www.hawaga.org.uk/ben/
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iD8DBQE9MfHxsYXoezDwaVARAj7IAJ9qZp379ALSUoZslSGyvzvPFT6HawCdHO6r GZQ1J4oBiQYKOM8JyP4SS2M=
=rKux
-----END PGP SIGNATURE-----
Quoted 8 lines · Ben Clifford
My ide-cd module was loaded automatically at boot.
I loaded the ide-scsi24 module (which didn't detect anything because the ide-cd module already had both the CD-ROM and CD-RW drives). So I rmmoded ide-cd and ide-scsi24.
Then I modprobed ide-scsi and got two oopses in quick succession which locked up the machine :-(Anyone else have any luck with ide-scsi24?
decode oops, post, cc: [email protected] and [email protected]
Dave
Signature · 2 lines
| Dave Jones. http://www.codemonkey.org.uk | SuSE Labs
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Quoted 3 lines · Dave Jones
Quoted 1 line
Anyone else have any luck with ide-scsi24?
decode oops, post, cc: [email protected] and [email protected]
Here is the first oops - I managed to get it written to disk before the second oops occurs (which locks up the machine) a few seconds later.
I captured it by saying:
modprobe ide-scsi24; sleep 3s; dmesg > a.log; sync
As I said about, a second oops occurs about 5 seconds later - in between, still appers to be responsive (I can type text into a bash prompt), but the second oops locks the machine up. I'll try and capture that now.
Ben
ksymoops 2.4.4 on i686 2.5.25-dj2+bc. Options used
-V (default)
-k /proc/ksyms (specified)
-l /proc/modules (default)
-o /lib/modules/2.5.25-dj2+bc/ (default) -m /boot/System.map-2.5.25-dj2+bc (default)
Error (regular_file): read_system_map stat /boot/System.map-2.5.25-dj2+bc failed
Unable to handle kernel paging request at virtual address d08ff420 d08ff420 *pde = 01334067 Oops: 0000 CPU: 0 EIP: 0010:[<d08ff420>] Tainted: P
Using defaults from ksymoops -t elf32-i386 -a i386
EFLAGS: 00010082
eax: d08ff420 ebx: c133f69c ecx: c02bee74 edx: c02bee74
esi: c02bee74 edi: c02bee74 ebp: 00000206 esp: cb5cfb40
ds: 0018 es: 0018 ss: 0018
Stack: c01a7014 c02bee74 c133f69c c02bee30 c02bee74 c01af5fa c02bee74 cb5ce000
c02bee74 c133f69c 00000003 00000206 c01afd75 c13402dc 00000000 c02bee74
c133f69c c02bee74 c13402dc 00000000 00000000 cb5cfba4 cb5cfba4 00000000
Call Trace: [<c01a7014>] [<c01af5fa>] [<c01afd75>] [<d08fd482>] [<d08e3611>]
[<d08e3b40>] [<d08e9e2c>] [<d08e925c>] [<d08e92da>] [<d08e3766>] [<d08e30c0>]
[<d08ea7e2>] [<c01a6d1c>] [<c01a732f>] [<d08eb182>] [<d08ea618>] [<c014e993>]
[<c014d5ce>] [<c01618d9>] [<c01619cc>] [<c0161990>] [<c014561c>] [<c0145337>]
[<c0162280>] [<c016218d>] [<c0184646>] [<c01849c1>] [<c0184c83>] [<d08fdad2>]
[<d08e4823>] [<d08f1640>] [<d08fdf84>] [<d08fd55f>] [<d08fdfa0>] [<c0118215>]
[<d08fdec4>] [<d08fc060>] [<c0106ebb>]
Code: Bad EIP value.Quoted 1 line
Quoted 1 line
EIP;
d08ff420<[ide-cd]ll_10byte_cmd_build+0/d0> <=====
Trace; c01a7014 <elv_next_request+14/a0>
Trace; c01af5fa <ide_stall_queue+2ca/360>
Trace; c01afd75 <ide_do_drive_cmd+d5/120>
Trace; d08fd482 <[ide-cd]restore_request+22/a0>
Trace; d08e3611 <[scsi_mod]__kstrtab_scsi_driverfs_bus_type+11/f0> Trace; d08e3b40 <[scsi_mod]scsi_done+0/80>
Trace; d08e9e2c <[scsi_mod]scsi_request_fn+39c/3f0> Trace; d08e925c <[scsi_mod]__scsi_insert_special+6c/a0> Trace; d08e92da <[scsi_mod]scsi_insert_special_req+1a/20> Trace; d08e3766 <[scsi_mod]scsi_wait_req+76/b0>
Trace; d08e30c0 <[scsi_mod]__ksymtab_scsi_ioctl_send_command+0/0> Trace; d08ea7e2 <[scsi_mod]scan_scsis_single+f2/740> Trace; c01a6d1c <agp_enable+b19c/b380>
Trace; c01a732f <blk_queue_make_request+5f/90>
Trace; d08eb182 <[scsi_mod]scan_scsis_target+32/170> Trace; d08ea618 <[scsi_mod]scan_scsis+2c8/3a0>
Trace; c014e993 <new_inode+63/70>
Trace; c014d5ce <d_instantiate+4e/60>
Trace; c01618d9 <wipe_partitions+a89/1260>
Trace; c01619cc <wipe_partitions+b7c/1260>
Trace; c0161990 <wipe_partitions+b40/1260>
Trace; c014561c <vfs_create+ec/7f0>
Trace; c0145337 <lookup_hash+67/90>
Trace; c0162280 <driverfs_create_file+c0/120>
Trace; c016218d <driverfs_create_dir+dd/110>
Trace; c0184646 <put_device+16/a0>
Trace; c01849c1 <device_create_file+71/80>
Trace; c0184c83 <device_remove_file+2b3/2d0>
Trace; d08fdad2 <[ide-cd]cdrom_write_intr+1e2/290> Trace; d08e4823 <[scsi_mod]scsi_register_host+213/340> Trace; d08f1640 <[scsi_mod].rodata.end+5851/8d51> Trace; d08fdf84 <[ide-cd]cdrom_lockdoor+94/f0>
Trace; d08fd55f <[ide-cd]cdrom_start_read+5f/90>
Trace; d08fdfa0 <[ide-cd]cdrom_lockdoor+b0/f0>
Trace; c0118215 <inter_module_put+7c5/8b0>
Trace; d08fdec4 <[ide-cd]cdrom_check_status+54/80> Trace; d08fc060 <[ide-cd]__module_license+0/0>
Trace; c0106ebb <__up_wakeup+107b/2e30>
1 error issued. Results may not be reliable.
- --
Ben Clifford [email protected] GPG: 30F06950
http://www.hawaga.org.uk/ben/
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iD8DBQE9MhU5sYXoezDwaVARAsMCAJ4iTjmUa9dhvj7vaDV30M152k+eHwCfdc/h be8awHV3cfoSUiKUzoVq7X4=
=u109
-----END PGP SIGNATURE-----
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Quoted 1 line · Ben Clifford
Quoted 1 line
Quoted 1 line
EIP;
d08ff420<[ide-cd]ll_10byte_cmd_build+0/d0> <=====
ide-cd.o is loaded at boot, and I have been manually unloading it before attempting to modprobe ide-scsi24.
So I rm'd the ide-cd.o from my modules directory as an easy way to stop it ever being loaded and I can modprobe ide-scsi24 without the oops happening.
I am attempting to rerecord a CD-RW at the moment to see if it works properly - it is in the blanking stage at the moment and appears to be ok...
- --
Ben Clifford [email protected] GPG: 30F06950
http://www.hawaga.org.uk/ben/
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iD8DBQE9MhrKsYXoezDwaVARAtQDAJ0RCg3YDcWVtWICxscA18KaWXr3ogCcDDpe cbnDZdbOtWSgIdVtrnBFj1U=
=KK1q
-----END PGP SIGNATURE-----