kernel for testing

[Arjan van de Ven]

Hi,

we're looking at doing a kernel update in about a week or so, a
candidate kernel for testing just has been uploaded to the test area for
fc2; kernel 2.6.8-1.520

Greetings,
    Arjan van de Ven

[Brian Millett]

>  Hi,
>
>  we're looking at doing a kernel update in about a week or so, a
>  candidate kernel for testing just has been uploaded to the test area for
>  fc2; kernel 2.6.8-1.520
>
>  Greetings,
>      Arjan van de Ven
>
>
>  --
>  fedora-test-list mailing list
>  fedora-test-list(a)redhat.com
>  To unsubscribe:
>  http://www.redhat.com/mailman/listinfo/fedora-test-list 
Arjan, I use the latest orinoco software for the prism2.5 card I use
because of the need to have monitoring.  So I compile the module for every
kernel.  Works just great, except for this am with the 2.6.8-1.520 kernel.
 I am using the rawhide updates, so the gcc is version 3.4.1 20040714 (Red
Hat 3.4.1-7).  When I do  modprobe orinoco_pci, I get the following error:

Aug 14 08:46:18 shaka kernel: hermes: version magic '2.6.8-1.520 686
REGPARM 4KSTACKS gcc-3.4' should be '2.6.8-1.520 686 REGPARM 4KSTACKS
gcc-3.3'
Aug 14 08:46:18 shaka kernel: orinoco: version magic '2.6.8-1.520 686
REGPARM 4KSTACKS gcc-3.4' should be '2.6.8-1.520 686 REGPARM 4KSTACKS
gcc-3.3'

-- 
Brian Millett
Enterprise Consulting Group  "Shifts in paradigms
[phone]           often cause nose bleeds."
bpmATec-groupDOTcom                     Greg Glenn

[Arjan van de Ven]

On Sat, Aug 14, 2004 at 09:27:02AM -0500, Brian Millett wrote:
>  Aug 14 08:46:18 shaka kernel: hermes: version magic '2.6.8-1.520
> 686
>  REGPARM 4KSTACKS gcc-3.4' should be '2.6.8-1.520 686 REGPARM 4KSTACKS
>  gcc-3.3'
>  Aug 14 08:46:18 shaka kernel: orinoco: version magic '2.6.8-1.520 686
>  REGPARM 4KSTACKS gcc-3.4' should be '2.6.8-1.520 686 REGPARM 4KSTACKS
>  gcc-3.3' 
yeah well FC2 has gcc 3.3 so this candidate fc2 kernel is also compiled with
gcc 3.3 ;)

[Bart Kalita]

Silly Q but whats the address?

TIA


On Sat, 2004-08-14 at 14:25 +0200, Arjan van de Ven wrote:
>  Hi,
>  
>  we're looking at doing a kernel update in about a week or so, a
>  candidate kernel for testing just has been uploaded to the test area for
>  fc2; kernel 2.6.8-1.520
>  
>  Greetings,
>      Arjan van de Ven
>  
>

[Janina Sajka]

Bart Kalita writes:
>  Silly Q but whats the address? 
ftp://download.fedora.redhat.com/pub/fedora/linux/core/updates/testing/2/
and then your arch.

>  
>  TIA
>  
>  
>  On Sat, 2004-08-14 at 14:25 +0200, Arjan van de Ven wrote:
>  > Hi,
>  > 
>  > we're looking at doing a kernel update in about a week or so, a
>  > candidate kernel for testing just has been uploaded to the test area for
>  > fc2; kernel 2.6.8-1.520
>  > 
>  > Greetings,
>  >     Arjan van de Ven
>  > 
>  > 
>  
>  
>  -- 
>  fedora-test-list mailing list
>  fedora-test-list(a)redhat.com
>  To unsubscribe: 
>  http://www.redhat.com/mailman/listinfo/fedora-test-list 
-- 
	
				Janina Sajka, Chair
				Accessibility Workgroup
				Free Standards Group (FSG)

janina(a)freestandards.org	Phone: [phone]

[Clyde E. Kunkel]

On Sat, 2004-08-14 at 14:25 +0200, Arjan van de Ven wrote:
>  Hi,
>  
>  we're looking at doing a kernel update in about a week or so, a
>  candidate kernel for testing just has been uploaded to the test area for
>  fc2; kernel 2.6.8-1.520
>  
>  Greetings,
>      Arjan van de Ven
>  
>   
Intalled on FedoraCore 3 T1 system, ASUS P4c800-E.  Looking ok so
far....

[Janina Sajka]

Clyde E. Kunkel writes:
>  On Sat, 2004-08-14 at 14:25 +0200, Arjan van de Ven wrote:
>  > Hi,
>  > 
>  > we're looking at doing a kernel update in about a week or so, a
>  > candidate kernel for testing just has been uploaded to the test area for
>  > fc2; kernel 2.6.8-1.520
>  > 
>  > Greetings,
>  >     Arjan van de Ven
>  >  
So far so good on a dual Opteron.

Still fussing with my RME Multiface ALSA interface, though. Not sure if
it's a kernel issue, though.

[Féliciano Matias]

Le sam 14/08/2004 à 14:25, Arjan van de Ven a écrit :
>  Hi,
>  
>  we're looking at doing a kernel update in about a week or so, a
>  candidate kernel for testing just has been uploaded to the test area for
>  fc2; kernel 2.6.8-1.520
>   
2.6.8 or 2.6.8.1 ?
http://marc.theaimsgroup.com/?l=linux-kernel&m=109248077922701&w=2

>  Greetings,
>      Arjan van de Ven
>

[Alan Cox]

On Sat, Aug 14, 2004 at 10:16:42PM +0200, Matias Feliciano wrote:
>  > fc2; kernel 2.6.8-1.520
>  > 
>  
>  2.6.8 or 2.6.8.1 ?
>  http://marc.theaimsgroup.com/?l=linux-kernel&m=109248077922701&w=2 
And hopefully with the additional SG_IO fix 8)

[Sean Bruno]

R u folks looking for some FC2 users to test this kernel release prior
to its actual release?  I have a couple of different machines I can test
it on, if that is your request.

On Sat, 2004-08-14 at 13:22, Alan Cox wrote:
>  On Sat, Aug 14, 2004 at 10:16:42PM +0200, Matias Feliciano wrote:
>  > > fc2; kernel 2.6.8-1.520
>  > > 
>  > 
>  > 2.6.8 or 2.6.8.1 ?
>  > http://marc.theaimsgroup.com/?l=linux-kernel&m=109248077922701&w=2
>  
>  And hopefully with the additional SG_IO fix 8) 
--

[Jef Spaleta]

On Sat, 14 Aug 2004 13:34:16 -0700, Sean Bruno <sean.bruno(a)dsl-only.net&gt; wrote:
>  R u folks looking for some FC2 users to test this kernel release
> prior
>  to its actual release?  I have a couple of different machines I can test
>  it on, if that is your request. 
That is sort of the point of the testing-updates branch for each release of fc.

Quoting http://fedora.redhat.com/download/updates.html :

Proposed Fedora Updates

Before updated packages are considered Fedora Updates, they may be
placed in a testing directory for public testing. After they have been
tested more throughly by the community, they can be moved into the
Fedora Updates category.

http://download.fedora.redhat.com/pub/fedora/linux/core/updates/testing/ 


-jef

[Jim Cornette]

Arjan van de Ven wrote:
>  Hi,
>  
>  we're looking at doing a kernel update in about a week or so, a
>  candidate kernel for testing just has been uploaded to the test area for
>  fc2; kernel 2.6.8-1.520
>  
>  Greetings,
>      Arjan van de Ven
>  
>   
There were no errors reported during kernel installation. Also, the 
kernel boots into the proper runlevel again. inittab is set back to 
runlevel 5 and 3 is added in grub.conf. No surprises!

The only error in boot log is below, which has been recorded for a long 
time.

Aug 14 20:47:02 cornette-fc2 mdmpd: mdmpd startup failed
Aug 14 20:47:02 cornette-fc2 mdmpd: mdmpd failed

Are there any particular items to test? USB, Firewire, etc?

Jim

-- 
New screensaver released: Curtains for Windows.

[Ralf Corsepius]

On Sat, 2004-08-14 at 14:25, Arjan van de Ven wrote:
>  Hi,
>  
>  we're looking at doing a kernel update in about a week or so, a
>  candidate kernel for testing just has been uploaded to the test area for
>  fc2; kernel 2.6.8-1.520During an "su -l someuser" with
> "someuser"'s home on a remote nfs-server: 
Aug 15 13:00:47 mccallum kernel: Unable to handle kernel NULL pointer dereference at
virtual address 00000014
Aug 15 13:00:47 mccallum kernel:  printing eip:
Aug 15 13:00:47 mccallum kernel: 22d7fb71
Aug 15 13:00:47 mccallum kernel: *pde = 00000000
Aug 15 13:00:47 mccallum kernel: Oops: 0002 [#5]
Aug 15 13:00:47 mccallum kernel: Modules linked in: nfs snd_mixer_oss snd_ens1371
snd_rawmidi snd_seq_device snd_pcm snd_page_alloc snd_timer snd_ac97_codec gameport snd
soundcore nvidia(U) usbserial parport_pc lp parport nfsd exportfs lockd md5 ipv6 autofs4
ipt_TOS ipt_MASQUERADE ipt_REJECT ipt_LOG ipt_state ip_nat_irc ip_nat_tftp ip_nat_ftp
ip_conntrack_irc ip_conntrack_tftp ip_conntrack_ftp ipt_multiport ipt_conntrack
iptable_filter iptable_mangle iptable_nat ip_conntrack ip_tables sunrpc 8139too mii hisax
crc_ccitt isdn slhc floppy sg scsi_mod microcode dm_mod button battery asus_acpi ac ext3
jbd
Aug 15 13:00:47 mccallum kernel: CPU:    0
Aug 15 13:00:47 mccallum kernel: EIP:    0060:[<22d7fb71>]    Tainted: P
Aug 15 13:00:47 mccallum kernel: EFLAGS: 00210246   (2.6.8-1.520)
Aug 15 13:00:47 mccallum kernel: EIP is at nfs3_request_init+0xb/0x13 [nfs]
Aug 15 13:00:47 mccallum kernel: eax: 00000000   ebx: 05551960   ecx: 0520fd5c   edx:
06b81ea0
Aug 15 13:00:47 mccallum kernel: esi: 0570ed74   edi: 00000000   ebp: 0311a080   esp:
0570ed44
Aug 15 13:00:47 mccallum kernel: ds: 007b   es: 007b   ss: 0068
Aug 15 13:00:47 mccallum kernel: Process xauth (pid: 5744, threadinfo=0570e000
task=056453a0)
Aug 15 13:00:47 mccallum kernel: Stack: 0570ed5c 22d7861e 05551960 1d725800 0520fd5c
06b81ea0 1d244b3c 00000000
Aug 15 13:00:47 mccallum kernel:        0000000a 22d8f283 00000000 00000000 0570ed70
009fd5a2 0520fc40 00000000
Aug 15 13:00:47 mccallum kernel:        0520fd18 00000000 22d7cb35 00000000 00000035
00000035 0311a080 0520fd5c
Aug 15 13:00:47 mccallum kernel: Call Trace:
Aug 15 13:00:47 mccallum kernel:  [<22d7861e>] nfs_create_request+0x106/0x113 [nfs]
Aug 15 13:00:47 mccallum kernel:  [<22d7cb35>] nfs_update_request+0x430/0x503 [nfs]
Aug 15 13:00:47 mccallum kernel:  [<22d7cd97>] nfs_updatepage+0x113/0x1dd [nfs]
Aug 15 13:00:47 mccallum kernel:  [<22d746eb>] nfs_commit_write+0xcf/0x15f [nfs]
Aug 15 13:00:47 mccallum kernel:  [<0213fadc>]
generic_file_aio_write_nolock+0x6c8/0x855
Aug 15 13:00:47 mccallum kernel:  [<02153be3>] anon_vma_prepare+0x1f/0x215
Aug 15 13:00:47 mccallum kernel:  [<0214efb9>] do_anonymous_page+0x24b/0x2ba
Aug 15 13:00:47 mccallum kernel:  [<0213fd49>] generic_file_aio_write+0x76/0x8c
Aug 15 13:00:47 mccallum kernel:  [<22d74862>] nfs_file_write+0xe7/0x104 [nfs]
Aug 15 13:00:47 mccallum kernel:  [<02160b0f>] do_sync_write+0x6a/0x99
Aug 15 13:00:47 mccallum kernel:  [<0214d7e2>] follow_page_pfn+0xec/0xfd
Aug 15 13:00:47 mccallum kernel:  [<02160bf6>] vfs_write+0xb8/0xe4
Aug 15 13:00:47 mccallum kernel:  [<02160cc0>] sys_write+0x3c/0x62
Aug 15 13:00:47 mccallum kernel: Code: ff 40 14 89 43 18 5b c3 56 31 f6 53 89 c3 8b 40 0c
39 d0 75

Ralf

[Mike Chambers]

On Sun, 2004-08-15 at 13:06 +0200, Ralf Corsepius wrote:
>  During an "su -l someuser" with "someuser"'s
> home on a remote nfs-server:
>  
>  Aug 15 13:00:47 mccallum kernel: Unable to handle kernel NULL pointer dereference at
> virtual address 00000014
>  Aug 15 13:00:47 mccallum kernel:  printing eip:
>  Aug 15 13:00:47 mccallum kernel: 22d7fb71
>  Aug 15 13:00:47 mccallum kernel: *pde = 00000000
>  Aug 15 13:00:47 mccallum kernel: Oops: 0002 [#5]
>  Aug 15 13:00:47 mccallum kernel: Modules linked in: nfs snd_mixer_oss snd_ens1371
> snd_rawmidi snd_seq_device snd_pcm snd_page_alloc snd_timer snd_ac97_codec gameport snd
> soundcore nvidia(U) usbserial parport_pc lp parport nfsd exportfs lockd md5 ipv6 autofs4
> ipt_TOS ipt_MASQUERADE ipt_REJECT ipt_LOG ipt_state ip_nat_irc ip_nat_tftp ip_nat_ftp
> ip_conntrack_irc ip_conntrack_tftp ip_conntrack_ftp ipt_multiport ipt_conntrack
> iptable_filter iptable_mangle iptable_nat ip_conntrack ip_tables sunrpc 8139too mii hisax
> crc_ccitt isdn slhc floppy sg scsi_mod microcode dm_mod button battery asus_acpi ac ext3
> jbd
>  Aug 15 13:00:47 mccallum kernel: CPU:    0
>  Aug 15 13:00:47 mccallum kernel: EIP:    0060:[<22d7fb71>]    Tainted: P
>  Aug 15 13:00:47 mccallum kernel: EFLAGS: 00210246   (2.6.8-1.520)
>  Aug 15 13:00:47 mccallum kernel: EIP is at nfs3_request_init+0xb/0x13 [nfs]
>  Aug 15 13:00:47 mccallum kernel: eax: 00000000   ebx: 05551960   ecx: 0520fd5c   edx:
> 06b81ea0
>  Aug 15 13:00:47 mccallum kernel: esi: 0570ed74   edi: 00000000   ebp: 0311a080   esp:
> 0570ed44
>  Aug 15 13:00:47 mccallum kernel: ds: 007b   es: 007b   ss: 0068
>  Aug 15 13:00:47 mccallum kernel: Process xauth (pid: 5744, threadinfo=0570e000
> task=056453a0)
>  Aug 15 13:00:47 mccallum kernel: Stack: 0570ed5c 22d7861e 05551960 1d725800 0520fd5c
> 06b81ea0 1d244b3c 00000000
>  Aug 15 13:00:47 mccallum kernel:        0000000a 22d8f283 00000000 00000000 0570ed70
> 009fd5a2 0520fc40 00000000
>  Aug 15 13:00:47 mccallum kernel:        0520fd18 00000000 22d7cb35 00000000 00000035
> 00000035 0311a080 0520fd5c
>  Aug 15 13:00:47 mccallum kernel: Call Trace:
>  Aug 15 13:00:47 mccallum kernel:  [<22d7861e>] nfs_create_request+0x106/0x113
> [nfs]
>  Aug 15 13:00:47 mccallum kernel:  [<22d7cb35>] nfs_update_request+0x430/0x503
> [nfs]
>  Aug 15 13:00:47 mccallum kernel:  [<22d7cd97>] nfs_updatepage+0x113/0x1dd [nfs]
>  Aug 15 13:00:47 mccallum kernel:  [<22d746eb>] nfs_commit_write+0xcf/0x15f [nfs]
>  Aug 15 13:00:47 mccallum kernel:  [<0213fadc>]
> generic_file_aio_write_nolock+0x6c8/0x855
>  Aug 15 13:00:47 mccallum kernel:  [<02153be3>] anon_vma_prepare+0x1f/0x215
>  Aug 15 13:00:47 mccallum kernel:  [<0214efb9>] do_anonymous_page+0x24b/0x2ba
>  Aug 15 13:00:47 mccallum kernel:  [<0213fd49>] generic_file_aio_write+0x76/0x8c
>  Aug 15 13:00:47 mccallum kernel:  [<22d74862>] nfs_file_write+0xe7/0x104 [nfs]
>  Aug 15 13:00:47 mccallum kernel:  [<02160b0f>] do_sync_write+0x6a/0x99
>  Aug 15 13:00:47 mccallum kernel:  [<0214d7e2>] follow_page_pfn+0xec/0xfd
>  Aug 15 13:00:47 mccallum kernel:  [<02160bf6>] vfs_write+0xb8/0xe4
>  Aug 15 13:00:47 mccallum kernel:  [<02160cc0>] sys_write+0x3c/0x62
>  Aug 15 13:00:47 mccallum kernel: Code: ff 40 14 89 43 18 5b c3 56 31 f6 53 89 c3 8b 40 0c
> 39 d0 75 
Yep, got the same thing here.  So far the below kernels give me the
problems above with nfs...

kernel-2.6.7-1.515
kernel-2.6.7-1.517
kernel-2.6.8-1.520

Running FC3t1 + rawhide

[Arjan van de Ven]

>  Yep, got the same thing here.  So far the below kernels give me the
>  problems above with nfs... 
are your kernels tainted too?

[Ralf Corsepius]

On Sun, 2004-08-15 at 16:21, Arjan van de Ven wrote:
>  > Yep, got the same thing here.  So far the below kernels give me
> the
>  > problems above with nfs...
>  
>  are your kernels tainted too? 
It doesn't matter.

The same happens with an untainted kernel:

Aug 15 18:31:04 mccallum kernel: Unable to handle kernel NULL pointer dereference at
virtual address 00000014
Aug 15 18:31:04 mccallum kernel:  printing eip:
Aug 15 18:31:04 mccallum kernel: 22becb71
Aug 15 18:31:04 mccallum kernel: *pde = 00000000
Aug 15 18:31:04 mccallum kernel: Oops: 0002 [#1]
Aug 15 18:31:04 mccallum kernel: Modules linked in: nfs usbserial parport_pc lp parport
nfsd exportfs lockd md5 ipv6 autofs4 ipt_TOS ipt_MASQUERADE ipt_REJECT ipt_LOG ipt_state
ip_nat_irc ip_nat_tftp ip_nat_ftp ip_conntrack_irc ip_conntrack_tftp ip_conntrack_ftp
ipt_multiport ipt_conntrack iptable_filter iptable_mangle iptable_nat ip_conntrack
ip_tables sunrpc 8139too mii hisax crc_ccitt isdn slhc floppy sg scsi_mod microcode dm_mod
button battery asus_acpi ac ext3 jbd
Aug 15 18:31:04 mccallum kernel: CPU:    0
Aug 15 18:31:04 mccallum kernel: EIP:    0060:[<22becb71>]    Not tainted
Aug 15 18:31:04 mccallum kernel: EFLAGS: 00010246   (2.6.8-1.520)
Aug 15 18:31:04 mccallum kernel: EIP is at nfs3_request_init+0xb/0x13 [nfs]
Aug 15 18:31:04 mccallum kernel: eax: 00000000   ebx: 1b45b7e0   ecx: 1b4809fc   edx:
1b24d780
Aug 15 18:31:04 mccallum kernel: esi: 1aa98c80   edi: 00000000   ebp: 03178b40   esp:
1aa98c50
Aug 15 18:31:04 mccallum kernel: ds: 007b   es: 007b   ss: 0068
Aug 15 18:31:04 mccallum su(pam_unix)[3485]: session closed for user pictures
Aug 15 18:31:04 mccallum kernel: Process bash (pid: 3486, threadinfo=1aa98000
task=1b23a090)
Aug 15 18:31:04 mccallum kernel: Stack: 1aa98c68 22be561e 1b45b7e0 035b3a00 1b4809fc
1b24d780 1d244b3c 00000000
Aug 15 18:31:04 mccallum kernel:        0000000a 22bfc283 00000000 00000000 22beb228
00000004 1b4809fc 1aa98ddc
Aug 15 18:31:04 mccallum kernel:        03178b40 000000bf 22be7d45 00000000 000000bf
00000000 0213da36 1aa98d30
Aug 15 18:31:04 mccallum kernel: Call Trace:
Aug 15 18:31:04 mccallum kernel:  [<22be561e>] nfs_create_request+0x106/0x113 [nfs]
Aug 15 18:31:04 mccallum kernel:  [<22beb228>] nfs_sync_inode+0x4c/0x57 [nfs]
Aug 15 18:31:04 mccallum kernel:  [<22be7d45>] readpage_async_filler+0x54/0xfe
[nfs]
Aug 15 18:31:04 mccallum kernel:  [<0213da36>] add_to_page_cache+0x9f/0x12f
Aug 15 18:31:04 mccallum kernel:  [<22be7cf1>] readpage_async_filler+0x0/0xfe [nfs]
Aug 15 18:31:04 mccallum kernel:  [<02144df6>] read_cache_pages+0x6e/0xdf
Aug 15 18:31:04 mccallum kernel:  [<0211be05>] autoremove_wake_function+0x0/0x2d
Aug 15 18:31:04 mccallum kernel:  [<22a608a1>] rpc_call_sync+0x7a/0x87 [sunrpc]
Aug 15 18:31:04 mccallum kernel:  [<22be7e5e>] nfs_readpages+0x6f/0x91 [nfs]
Aug 15 18:31:04 mccallum kernel:  [<02144e9a>] read_pages+0x33/0xdd
Aug 15 18:31:04 mccallum kernel:  [<021420c4>] buffered_rmqueue+0x1e9/0x20c
Aug 15 18:31:04 mccallum kernel:  [<0214239b>] __alloc_pages+0x2b4/0x2be
Aug 15 18:31:04 mccallum kernel:  [<02145500>] do_page_cache_readahead+0x29f/0x2bf
Aug 15 18:31:04 mccallum kernel:  [<02145663>] page_cache_readahead+0x143/0x1b0
Aug 15 18:31:04 mccallum kernel:  [<0213e516>] do_generic_mapping_read+0x94/0x305
Aug 15 18:31:04 mccallum kernel:  [<0213e9e5>] __generic_file_aio_read+0x15d/0x177
Aug 15 18:31:04 mccallum kernel:  [<0213e787>] file_read_actor+0x0/0x101
Aug 15 18:31:04 mccallum kernel:  [<0213ea3f>] generic_file_aio_read+0x40/0x47
Aug 15 18:31:04 mccallum kernel:  [<22be1349>] nfs_file_read+0xcc/0xd6 [nfs]
Aug 15 18:31:04 mccallum kernel:  [<02160992>] do_sync_read+0x6a/0x99
Aug 15 18:31:04 mccallum kernel:  [<02160a79>] vfs_read+0xb8/0xe4
Aug 15 18:31:04 mccallum kernel:  [<02160c5e>] sys_read+0x3c/0x62
Aug 15 18:31:04 mccallum kernel: Code: ff 40 14 89 43 18 5b c3 56 31 f6 53 89 c3 8b 40 0c
39 d0 75

[FYI: This is with FC2 and yout original FC2 updates/testing rpm, using
automounted nfs homes]

Ralf

[Mike Chambers]

On Sun, 2004-08-15 at 16:21 +0200, Arjan van de Ven wrote:
>  > Yep, got the same thing here.  So far the below kernels give me
> the
>  > problems above with nfs...
>  
>  are your kernels tainted too? 
Nope, all Fedora built kernels, nothing adding/taken away, ran just as
they were installed.

Mike

[Michal Jaegermann]

On Sun, Aug 15, 2004 at 06:59:43AM -0500, Mike Chambers wrote:
>  
>  Yep, got the same thing here.  So far the below kernels give me the
>  problems above with nfs...
>  
>  kernel-2.6.7-1.515
>  kernel-2.6.7-1.517
>  kernel-2.6.8-1.520 
It looks like that the following patch

===== fs/nfs/file.c 1.40 vs edited =====
--- 1.40/fs/nfs/file.c	2004-08-09 14:58:00 -04:00
+++ edited/fs/nfs/file.c	2004-08-13 22:54:01 -04:00
@@ -72,7 +72,7 @@
 
 static int nfs_check_flags(int flags)
 {
-	if (flags & (O_APPEND | O_DIRECT))
+	if ((flags & (O_APPEND | O_DIRECT)) == (O_APPEND | O_DIRECT))
 		return -EINVAL;
 
 	return 0;
@@ -89,7 +89,7 @@
 	int res;
 
 	res = nfs_check_flags(filp->f_flags);
-	if (!res)
+	if (res)
 		return res;
 
 	lock_kernel();

was not included.  At least for kernel-2.6.8-1.520.

The above was posted by Jeff Garzik on lkml and that piece missing
was the reason for a "paper bag" 2.6.8.1 release.  Look for
a posting "[PATCH] Re: [2.6.8-rc4-bk] NFS oops on x86-64" from
13 Aug 2004 23:23:34 -0400 and a Linus comment "Ahh. Jeff posted the
right one, obviously".

Is a bugzilla ticket opened for this bug?  Curiously enough
I was not hit by that even when using NFS.

   Michal

[Michal Jaegermann]

On Sun, Aug 15, 2004 at 06:59:43AM -0500, Mike Chambers wrote:
>  
>  Yep, got the same thing here.  So far the below kernels give me the
>  problems above with nfs...
>  
>  kernel-2.6.7-1.515
>  kernel-2.6.7-1.517
>  kernel-2.6.8-1.520 
I have my NFS oops with 2.6.8-1.520 on x86_64; which is not a big
surprise. :-)  The issue does not seem to have a bugzilla entry but
I wonder if there is a point in opening one. Arjan?

   Michal

[Sean Bruno]

Does the oops look like bugzilla 128036?

On Sun, 2004-08-15 at 15:49, Michal Jaegermann wrote:
>  On Sun, Aug 15, 2004 at 06:59:43AM -0500, Mike Chambers wrote:
>  > 
>  > Yep, got the same thing here.  So far the below kernels give me the
>  > problems above with nfs...
>  > 
>  > kernel-2.6.7-1.515
>  > kernel-2.6.7-1.517
>  > kernel-2.6.8-1.520
>  
>  I have my NFS oops with 2.6.8-1.520 on x86_64; which is not a big
>  surprise. :-)  The issue does not seem to have a bugzilla entry but
>  I wonder if there is a point in opening one. Arjan?
>  
>     Michal 
--

[Michal Jaegermann]

On Sun, Aug 15, 2004 at 03:52:13PM -0700, Sean Bruno wrote:
>  Does the oops look like bugzilla 128036? 
No.  That one is quite different.  This is really an issue
with 2.6.8 and some "pre" kernels which Arjan, clearly, used
as a base for some late 2.6.7=1.... variants and kernel-2.6.8-1.520.
You need that patch I posted earlier in this thread.

   Michal

ppp + radius plugin

[marek cervenka]

hi,

why isn't pppd compiled with radius plugin?


-- 
-------------------------------------
Marek Cervenka
Centrum Vypocetni Techniky
CVT - http://cvt.fpf.slu.cz
FPF SLU OPAVA - http://www.fpf.slu.cz
=====================================

[Charles Curley]

On Sat, Aug 14, 2004 at 02:25:04PM +0200, Arjan van de Ven wrote:
>  Hi,
>  
>  we're looking at doing a kernel update in about a week or so, a
>  candidate kernel for testing just has been uploaded to the test area for
>  fc2; kernel 2.6.8-1.520 
Test article 1: Compaq Pressario 4160, Pentium 75 - 200, 132.793 MHz.

* Eth0 (00:05.0 Ethernet controller: Lite-On Communications Inc
  LNE100TX (rev 20), Linux Tulip driver version 1.1.13 (May 11, 2002))
  failed to initialize correctly. I could ping other hosts from the
  target, but couldn't ping (or ssh) to the target. At shutdown, the
  system reported that it had a different MAC address than
  expected. This is new.

Aug 15 09:58:18 issola kernel: FDC 0 is a post-1991 82077
Aug 15 09:58:18 issola kernel: Linux Tulip driver version 1.1.13 (May 11, 2002)
Aug 15 09:58:18 issola kernel: tulip0:  MII transceiver #1 config 3000 status 7829
advertising 01e1.
Aug 15 09:58:18 issola kernel: eth0: Lite-On 82c168 PNIC rev 32 at 0xc500f000,
00:02:E3:09:73:AB, IRQ 11.
Aug 15 09:58:19 issola kernel: ip_tables: (C) 2000-2002 Netfilter core team
Aug 15 09:58:19 issola kernel: ip_conntrack version 2.1 (576 buckets, 4608 max) - 320
bytes per conntrack
Aug 15 09:58:19 issola kernel: Linux Tulip driver version 1.1.13 (May 11, 2002)
Aug 15 09:58:19 issola kernel: tulip0:  MII transceiver #1 config 1000 status 782d
advertising 01e1.
Aug 15 09:58:19 issola kernel: eth0: Lite-On 82c168 PNIC rev 32 at 0xc5014000,
00:02:E3:09:73:AB, IRQ 11.
Aug 15 09:58:19 issola kernel: eth0: Setting full-duplex based on MII#1 link partner
capability of 45e1.
Aug 15 09:58:19 issola kernel: eth0: Setting full-duplex based on MII#1 link partner
capability of 45e1.
Aug 15 09:58:44 issola rpcidmapd: rpc.idmapd startup succeeded

The second line about setting full duplex, the penultimate line above,
is not in the logs from booting to other kernels.

* USB hub and floppy drive (Vendor: Y-E DATA Model: USB-FDU, Rev:
  5.01) worked fine.

* Kingston 256MB USB key
  (http://www.qbik.ch/usb/devices/showdev.php?id=2032) failed at
  fdisk. This device is known to work on FC1 and FC2 kernel
  2.6.6-1.435.2.3.

Test article 2: Toshiba Tecra 8000, Pentium II (Deschutes), 266.621
MHz.

* Floppy drive: No access to the floppy diskette is performed (an
  access is normal). Instead, I get a message: "usb 1-1: control
  timeout on ep0out". At least the machine does not lock up, as it has
  in the past.

* Kingston USB key drive: Similar response:

Aug 15 11:15:03 teckla kernel: usb 1-1: new full speed USB device using address 2
Aug 15 11:15:08 teckla kernel: usb 1-1: control timeout on ep0out

Both machines run very slowly after the Kingston device is inserted.

On both test machines, the installation left the default boot kernel
in grub.conf to point to the prior default kernel (was "default=0",
changed it to "default=1"). As these are test kernels, this is
probably a good idea.

-- 

Charles Curley                  /"\    ASCII Ribbon Campaign
Looking for fine software       \ /    Respect for open standards
and/or writing?                  X     No HTML/RTF in email
http://www.charlescurley.com    / \    No M$ Word docs in email

Key fingerprint = CE5C 6645 A45A 64E4 94C0  809C FFF6 4C48 4ECD DFDB

[Charles Curley]

On Sun, Aug 15, 2004 at 11:49:48AM -0600, Charles Curley wrote:
>  On Sat, Aug 14, 2004 at 02:25:04PM +0200, Arjan van de Ven wrote:
>  > Hi,
>  > 
>  > we're looking at doing a kernel update in about a week or so, a
>  > candidate kernel for testing just has been uploaded to the test area for
>  > fc2; kernel 2.6.8-1.520
>  
>  Test article 1: Compaq Pressario 4160, Pentium 75 - 200, 132.793 MHz.
>   

>  
>  * USB hub and floppy drive (Vendor: Y-E DATA Model: USB-FDU, Rev:
>    5.01) worked fine.
>  
>  * Kingston 256MB USB key
>    (http://www.qbik.ch/usb/devices/showdev.php?id=2032) failed at
>    fdisk. This device is known to work on FC1 and FC2 kernel
>    2.6.6-1.435.2.3.
>  
>  Test article 2: Toshiba Tecra 8000, Pentium II (Deschutes), 266.621
>  MHz. 
Addendum

Test article 3: Walmart special, 1.1GHz Athlon.

Both devices work fine and dandy with kernels kernel-2.6.5-1.358,
kernel-2.6.7-1.494.2.2, and kernel-2.6.8-1.520.

I conjecture that the problem with the other test systems is that they
are older, slower, and the USB code can't move fast enough.

-- 

Charles Curley                  /"\    ASCII Ribbon Campaign
Looking for fine software       \ /    Respect for open standards
and/or writing?                  X     No HTML/RTF in email
http://www.charlescurley.com    / \    No M$ Word docs in email

Key fingerprint = CE5C 6645 A45A 64E4 94C0  809C FFF6 4C48 4ECD DFDB

[Féliciano Matias]

Le sam 14/08/2004 à 14:25, Arjan van de Ven a écrit :
>  Hi,
>  
>  we're looking at doing a kernel update in about a week or so, a
>  candidate kernel for testing just has been uploaded to the test area for
>  fc2; kernel 2.6.8-1.520
>   
Some times I have a kernel panic.
I use a custom configuration (see attachment) because the adsl driver
(bewan proprietary) doesn't work with CONFIG_REGPARAM.

After ppp0 is up, I launch :
/sbin/tc qdisc add dev ppp0 root tbf rate 250kbit latency 50ms burst 1540

After 1 or 2 second I got (copy by hand) :
tbf_enqueue  [sch_tbf]
dev_queue_xmit
ip_finish_output
ip_push_pending_frames
udp_push_pending_frames
udp_sendmsg
memcpy_fromiovec
inet_sendmsg
sock_sendmsg
save_i387_fxsave
setup_sigcontext
sys_socketcall
syscall_call
Code : Bad EIP value.
<O> Kernel panic : Fatal exception in interrupt
 In interrupt handler - not syncing


>  Greetings,
>      Arjan van de Ven

[Féliciano Matias]

Le lun 16/08/2004 à 17:22, Matias Feliciano a écrit :
>  [...]
>  /sbin/tc qdisc add dev ppp0 root tbf rate 250kbit latency 50ms burst 1540
>  [...]
>  <O> Kernel panic : Fatal exception in interrupt
>   In interrupt handler - not syncing
>   
Fix in 2.6.8-1.521 !

[Miguel Pereira Leite]

Hi.

Maybe it's late but, i've searched arround and did not find any report
about this.

I'm using FC2 full updated, now with 2.6.8-1.520. Looking good. But
today i've inserted a USB pen drive (my usb is 1.1), the machine crashed
after 2 or 3 seconds.
Talking about it in IRC some told me to unlock se writing protection on
the pen drive. With this unlocked, i can insert ir safely.
i dont know if this is a previous issue, because i don't remember if
i've ever inserted it with protection ON.
This is repetable.
Sorry for my english.

A Sáb, 2004-08-14 às 13:25, Arjan van de Ven escreveu:
>  Hi,
>  
>  we're looking at doing a kernel update in about a week or so, a
>  candidate kernel for testing just has been uploaded to the test area for
>  fc2; kernel 2.6.8-1.520
>  
>  Greetings,
>      Arjan van de Ven 
-- 
Miguel Pereira Leite <mpleite(a)ppl-ti.com&gt;
Proença & Pereira Leite - Tecnologias de Informação, Lda

[Miguel Pereira Leite]

Hi again.

Sorry for my last post.

It's not apenning anymore. it might have something to do with my
bluetooth dongle that it was on that usb connector. i'll check it.


Regards,,
A Qua, 2004-08-18 às 11:25, Miguel Pereira Leite escreveu:
>  Hi.
>  
>  Maybe it's late but, i've searched arround and did not find any report
>  about this.
>  
>  I'm using FC2 full updated, now with 2.6.8-1.520. Looking good. But
>  today i've inserted a USB pen drive (my usb is 1.1), the machine crashed
>  after 2 or 3 seconds.
>  Talking about it in IRC some told me to unlock se writing protection on
>  the pen drive. With this unlocked, i can insert ir safely.
>  i dont know if this is a previous issue, because i don't remember if
>  i've ever inserted it with protection ON.
>  This is repetable.
>  Sorry for my english.
>  
>  A Sáb, 2004-08-14 às 13:25, Arjan van de Ven escreveu:
>  > Hi,
>  > 
>  > we're looking at doing a kernel update in about a week or so, a
>  > candidate kernel for testing just has been uploaded to the test area for
>  > fc2; kernel 2.6.8-1.520
>  > 
>  > Greetings,
>  >     Arjan van de Ven 
-- 
Miguel Pereira Leite <mpleite(a)ppl-ti.com&gt;
Proença & Pereira Leite - Tecnologias de Informação, Lda

[Alan Cox]

On Wed, Aug 18, 2004 at 11:25:51AM +0100, Miguel Pereira Leite wrote:
>  I'm using FC2 full updated, now with 2.6.8-1.520. Looking good.
> But
>  today i've inserted a USB pen drive (my usb is 1.1), the machine crashed
>  after 2 or 3 seconds.
>  Talking about it in IRC some told me to unlock se writing protection on
>  the pen drive. With this unlocked, i can insert ir safely.
>  i dont know if this is a previous issue, because i don't remember if
>  i've ever inserted it with protection ON. 
Please file this in bugzilla. Its very good debug info