* [PATCH 1/3] cciss: bug fix for hpacucli
@ 2005-11-18 16:33 mikem
2005-11-18 20:49 ` Jens Axboe
2005-11-18 21:45 ` Philippe Pegon
0 siblings, 2 replies; 8+ messages in thread
From: mikem @ 2005-11-18 16:33 UTC (permalink / raw)
To: akpm, axboe; +Cc: linux-kernel, linux-scsi, hpa, sitniko
Patch 1 of 3
This patch fixes a bug that breaks hpacucli, a command line interface
for the HP Array Config Utility. Without this fix the utility will
not detect any controllers in the system. I thought I had already fixed
this, but I guess not.
Thanks to all who reported the issue. Please consider this this inclusion.
Signed-off-by: Mike Miller
--------------------------------------------------------------------------------
include/linux/cciss_ioctl.h | 2 +-
1 files changed, 1 insertion(+), 1 deletion(-)
diff -puN include/linux/cciss_ioctl.h~cciss_fix_for_cli include/linux/cciss_ioctl.h
--- linux-2.6.14.2/include/linux/cciss_ioctl.h~cciss_fix_for_cli 2005-11-18 10:20:38.692304096 -0600
+++ linux-2.6.14.2-mikem/include/linux/cciss_ioctl.h 2005-11-18 10:21:14.112919344 -0600
@@ -10,8 +10,8 @@
typedef struct _cciss_pci_info_struct
{
unsigned char bus;
- unsigned short domain;
unsigned char dev_fn;
+ unsigned short domain;
__u32 board_id;
} cciss_pci_info_struct;
_
^ permalink raw reply [flat|nested] 8+ messages in thread* Re: [PATCH 1/3] cciss: bug fix for hpacucli
2005-11-18 16:33 [PATCH 1/3] cciss: bug fix for hpacucli mikem
@ 2005-11-18 20:49 ` Jens Axboe
2005-11-21 8:28 ` Jens Axboe
2005-11-18 21:45 ` Philippe Pegon
1 sibling, 1 reply; 8+ messages in thread
From: Jens Axboe @ 2005-11-18 20:49 UTC (permalink / raw)
To: mikem; +Cc: akpm, linux-kernel, linux-scsi, hpa, sitniko
On Fri, Nov 18 2005, mikem wrote:
> Patch 1 of 3
>
> This patch fixes a bug that breaks hpacucli, a command line interface
> for the HP Array Config Utility. Without this fix the utility will
> not detect any controllers in the system. I thought I had already fixed
> this, but I guess not.
>
> Thanks to all who reported the issue. Please consider this this inclusion.
Lovely, hope this makes it able to configure my drives on the tiger now
:).
Applied.
--
Jens Axboe
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [PATCH 1/3] cciss: bug fix for hpacucli
2005-11-18 20:49 ` Jens Axboe
@ 2005-11-21 8:28 ` Jens Axboe
2005-11-21 16:46 ` mikem
0 siblings, 1 reply; 8+ messages in thread
From: Jens Axboe @ 2005-11-21 8:28 UTC (permalink / raw)
To: mikem; +Cc: akpm, linux-kernel, linux-scsi, hpa, sitniko
On Fri, Nov 18 2005, Jens Axboe wrote:
> On Fri, Nov 18 2005, mikem wrote:
> > Patch 1 of 3
> >
> > This patch fixes a bug that breaks hpacucli, a command line interface
> > for the HP Array Config Utility. Without this fix the utility will
> > not detect any controllers in the system. I thought I had already fixed
> > this, but I guess not.
> >
> > Thanks to all who reported the issue. Please consider this this inclusion.
>
> Lovely, hope this makes it able to configure my drives on the tiger now
> :).
This sort-of makes it work. I get some complaints about unaligned access
when setting up a test array:
=> controller slot=0 create type=logicaldrive drives=all raid=1 drivetype=sas
.hpacucli(12458): unaligned access to 0x60000fffffcb370e, ip=0x40000000003c8550
=> controller slot=0 create type=logicaldrive drives=all raid=1 drivetype=sata
.hpacucli(12458): unaligned access to 0x60000fffffcb4aee, ip=0x40000000003c8550
.hpacucli(12458): unaligned access to 0x60000fffffcb370e, ip=0x40000000003c8550
.hpacucli(12458): unaligned access to 0x60000fffffcb370e, ip=0x40000000003c8550
Invoking hpacucli again later on makes it trigger a kobj warning:
Badness in kref_get at lib/kref.c:32
Call Trace:
[<a000000100013560>] show_stack+0x80/0xa0
sp=e0000001f5547b20 bsp=e0000001f5541250
[<a000000100013e30>] dump_stack+0x30/0x60
sp=e0000001f5547cf0 bsp=e0000001f5541238
[<a00000010034fec0>] kref_get+0xa0/0xc0
sp=e0000001f5547cf0 bsp=e0000001f5541218
[<a00000010034e470>] kobject_get+0x30/0x60
sp=e0000001f5547cf0 bsp=e0000001f55411f0
[<a00000010034eb80>] kobject_add+0x20/0x380
sp=e0000001f5547cf0 bsp=e0000001f55411b8
[<a0000001001b7ff0>] register_disk+0x70/0x200
sp=e0000001f5547cf0 bsp=e0000001f5541188
[<a0000001003385d0>] add_disk+0x90/0xc0
sp=e0000001f5547cf0 bsp=e0000001f5541168
[<a000000202e1aa40>] rebuild_lun_table+0xaa0/0xb20 [cciss]
sp=e0000001f5547cf0 bsp=e0000001f55410a8
[<a000000202e1bbc0>] cciss_ioctl+0xa20/0x3be0 [cciss]
sp=e0000001f5547d00 bsp=e0000001f5540fc0
[<a0000001003346b0>] blkdev_driver_ioctl+0xd0/0x160
sp=e0000001f5547e10 bsp=e0000001f5540f78
[<a000000100335050>] blkdev_ioctl+0x170/0xb40
sp=e0000001f5547e10 bsp=e0000001f5540f20
[<a000000100147cc0>] block_ioctl+0x40/0x60
sp=e0000001f5547e10 bsp=e0000001f5540ef0
[<a000000100160740>] do_ioctl+0x160/0x1a0
sp=e0000001f5547e10 bsp=e0000001f5540eb0
[<a000000100160880>] vfs_ioctl+0x100/0x8e0
sp=e0000001f5547e10 bsp=e0000001f5540e68
[<a0000001001610c0>] sys_ioctl+0x60/0xc0
sp=e0000001f5547e20 bsp=e0000001f5540de8
[<a00000010000bc80>] ia64_ret_from_syscall+0x0/0x20
sp=e0000001f5547e30 bsp=e0000001f5540de8
[<a000000000010640>] __kernel_syscall_via_break+0x0/0x20
sp=e0000001f5548000 bsp=e0000001f5540de8
--
Jens Axboe
^ permalink raw reply [flat|nested] 8+ messages in thread* Re: [PATCH 1/3] cciss: bug fix for hpacucli
2005-11-21 8:28 ` Jens Axboe
@ 2005-11-21 16:46 ` mikem
2005-11-21 17:23 ` Jens Axboe
2005-11-21 17:37 ` H. Peter Anvin
0 siblings, 2 replies; 8+ messages in thread
From: mikem @ 2005-11-21 16:46 UTC (permalink / raw)
To: Jens Axboe; +Cc: akpm, linux-kernel, linux-scsi, hpa, sitniko
On Mon, Nov 21, 2005 at 09:28:10AM +0100, Jens Axboe wrote:
> On Fri, Nov 18 2005, Jens Axboe wrote:
> > On Fri, Nov 18 2005, mikem wrote:
> > > Patch 1 of 3
> > >
> > > This patch fixes a bug that breaks hpacucli, a command line interface
> > > for the HP Array Config Utility. Without this fix the utility will
> > > not detect any controllers in the system. I thought I had already fixed
> > > this, but I guess not.
> > >
> > > Thanks to all who reported the issue. Please consider this this inclusion.
> >
> > Lovely, hope this makes it able to configure my drives on the tiger now
> > :).
>
> This sort-of makes it work. I get some complaints about unaligned access
> when setting up a test array:
>
> => controller slot=0 create type=logicaldrive drives=all raid=1 drivetype=sas
> .hpacucli(12458): unaligned access to 0x60000fffffcb370e, ip=0x40000000003c8550
> => controller slot=0 create type=logicaldrive drives=all raid=1 drivetype=sata
> .hpacucli(12458): unaligned access to 0x60000fffffcb4aee, ip=0x40000000003c8550
> .hpacucli(12458): unaligned access to 0x60000fffffcb370e, ip=0x40000000003c8550
> .hpacucli(12458): unaligned access to 0x60000fffffcb370e, ip=0x40000000003c8550
This seems to be coming out of user space. We'll work with the application
folks to investigate. There is a library called infomanager that's used
by the app. There may be an issue there. Call you strace the program and
send me the results? I haven't seen this in my lab with a vendor kernel.
>
> Invoking hpacucli again later on makes it trigger a kobj warning:
>
> Badness in kref_get at lib/kref.c:32
Hmmm, is this with my put_disk patch installed?
mikem
>
> Call Trace:
> [<a000000100013560>] show_stack+0x80/0xa0
> sp=e0000001f5547b20 bsp=e0000001f5541250
> [<a000000100013e30>] dump_stack+0x30/0x60
> sp=e0000001f5547cf0 bsp=e0000001f5541238
> [<a00000010034fec0>] kref_get+0xa0/0xc0
> sp=e0000001f5547cf0 bsp=e0000001f5541218
> [<a00000010034e470>] kobject_get+0x30/0x60
> sp=e0000001f5547cf0 bsp=e0000001f55411f0
> [<a00000010034eb80>] kobject_add+0x20/0x380
> sp=e0000001f5547cf0 bsp=e0000001f55411b8
> [<a0000001001b7ff0>] register_disk+0x70/0x200
> sp=e0000001f5547cf0 bsp=e0000001f5541188
> [<a0000001003385d0>] add_disk+0x90/0xc0
> sp=e0000001f5547cf0 bsp=e0000001f5541168
> [<a000000202e1aa40>] rebuild_lun_table+0xaa0/0xb20 [cciss]
> sp=e0000001f5547cf0 bsp=e0000001f55410a8
> [<a000000202e1bbc0>] cciss_ioctl+0xa20/0x3be0 [cciss]
> sp=e0000001f5547d00 bsp=e0000001f5540fc0
> [<a0000001003346b0>] blkdev_driver_ioctl+0xd0/0x160
> sp=e0000001f5547e10 bsp=e0000001f5540f78
> [<a000000100335050>] blkdev_ioctl+0x170/0xb40
> sp=e0000001f5547e10 bsp=e0000001f5540f20
> [<a000000100147cc0>] block_ioctl+0x40/0x60
> sp=e0000001f5547e10 bsp=e0000001f5540ef0
> [<a000000100160740>] do_ioctl+0x160/0x1a0
> sp=e0000001f5547e10 bsp=e0000001f5540eb0
> [<a000000100160880>] vfs_ioctl+0x100/0x8e0
> sp=e0000001f5547e10 bsp=e0000001f5540e68
> [<a0000001001610c0>] sys_ioctl+0x60/0xc0
> sp=e0000001f5547e20 bsp=e0000001f5540de8
> [<a00000010000bc80>] ia64_ret_from_syscall+0x0/0x20
> sp=e0000001f5547e30 bsp=e0000001f5540de8
> [<a000000000010640>] __kernel_syscall_via_break+0x0/0x20
> sp=e0000001f5548000 bsp=e0000001f5540de8
>
> --
> Jens Axboe
>
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [PATCH 1/3] cciss: bug fix for hpacucli
2005-11-21 16:46 ` mikem
@ 2005-11-21 17:23 ` Jens Axboe
2005-11-21 17:37 ` H. Peter Anvin
1 sibling, 0 replies; 8+ messages in thread
From: Jens Axboe @ 2005-11-21 17:23 UTC (permalink / raw)
To: mikem; +Cc: akpm, linux-kernel, linux-scsi, hpa, sitniko
On Mon, Nov 21 2005, mikem wrote:
> On Mon, Nov 21, 2005 at 09:28:10AM +0100, Jens Axboe wrote:
> > On Fri, Nov 18 2005, Jens Axboe wrote:
> > > On Fri, Nov 18 2005, mikem wrote:
> > > > Patch 1 of 3
> > > >
> > > > This patch fixes a bug that breaks hpacucli, a command line interface
> > > > for the HP Array Config Utility. Without this fix the utility will
> > > > not detect any controllers in the system. I thought I had already fixed
> > > > this, but I guess not.
> > > >
> > > > Thanks to all who reported the issue. Please consider this this inclusion.
> > >
> > > Lovely, hope this makes it able to configure my drives on the tiger now
> > > :).
> >
> > This sort-of makes it work. I get some complaints about unaligned access
> > when setting up a test array:
> >
> > => controller slot=0 create type=logicaldrive drives=all raid=1 drivetype=sas
> > .hpacucli(12458): unaligned access to 0x60000fffffcb370e, ip=0x40000000003c8550
> > => controller slot=0 create type=logicaldrive drives=all raid=1 drivetype=sata
> > .hpacucli(12458): unaligned access to 0x60000fffffcb4aee, ip=0x40000000003c8550
> > .hpacucli(12458): unaligned access to 0x60000fffffcb370e, ip=0x40000000003c8550
> > .hpacucli(12458): unaligned access to 0x60000fffffcb370e, ip=0x40000000003c8550
>
> This seems to be coming out of user space. We'll work with the application
> folks to investigate. There is a library called infomanager that's used
> by the app. There may be an issue there. Call you strace the program and
> send me the results? I haven't seen this in my lab with a vendor kernel.
Sure I'll try, I'll boot the box again tomorrow.
> > Invoking hpacucli again later on makes it trigger a kobj warning:
> >
> > Badness in kref_get at lib/kref.c:32
>
> Hmmm, is this with my put_disk patch installed?
Yep, it's with current git HEAD (so all 3 patches you sent).
--
Jens Axboe
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [PATCH 1/3] cciss: bug fix for hpacucli
2005-11-21 16:46 ` mikem
2005-11-21 17:23 ` Jens Axboe
@ 2005-11-21 17:37 ` H. Peter Anvin
2005-11-21 17:45 ` Jens Axboe
1 sibling, 1 reply; 8+ messages in thread
From: H. Peter Anvin @ 2005-11-21 17:37 UTC (permalink / raw)
To: mikem; +Cc: Jens Axboe, akpm, linux-kernel, linux-scsi, sitniko
mikem wrote:
>>
>>This sort-of makes it work. I get some complaints about unaligned access
>>when setting up a test array:
>>
>>=> controller slot=0 create type=logicaldrive drives=all raid=1 drivetype=sas
>>.hpacucli(12458): unaligned access to 0x60000fffffcb370e, ip=0x40000000003c8550
>>=> controller slot=0 create type=logicaldrive drives=all raid=1 drivetype=sata
>>.hpacucli(12458): unaligned access to 0x60000fffffcb4aee, ip=0x40000000003c8550
>>.hpacucli(12458): unaligned access to 0x60000fffffcb370e, ip=0x40000000003c8550
>>.hpacucli(12458): unaligned access to 0x60000fffffcb370e, ip=0x40000000003c8550
>
> This seems to be coming out of user space. We'll work with the application
> folks to investigate. There is a library called infomanager that's used
> by the app. There may be an issue there. Call you strace the program and
> send me the results? I haven't seen this in my lab with a vendor kernel.
>
What machine is this for? For x86-64, the above are non-canonical
addresses. IA64?
-hpa
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [PATCH 1/3] cciss: bug fix for hpacucli
2005-11-21 17:37 ` H. Peter Anvin
@ 2005-11-21 17:45 ` Jens Axboe
0 siblings, 0 replies; 8+ messages in thread
From: Jens Axboe @ 2005-11-21 17:45 UTC (permalink / raw)
To: H. Peter Anvin; +Cc: mikem, akpm, linux-kernel, linux-scsi, sitniko
On Mon, Nov 21 2005, H. Peter Anvin wrote:
> mikem wrote:
> >>
> >>This sort-of makes it work. I get some complaints about unaligned access
> >>when setting up a test array:
> >>
> >>=> controller slot=0 create type=logicaldrive drives=all raid=1
> >>drivetype=sas
> >>.hpacucli(12458): unaligned access to 0x60000fffffcb370e,
> >>ip=0x40000000003c8550
> >>=> controller slot=0 create type=logicaldrive drives=all raid=1
> >>drivetype=sata
> >>.hpacucli(12458): unaligned access to 0x60000fffffcb4aee,
> >>ip=0x40000000003c8550
> >>.hpacucli(12458): unaligned access to 0x60000fffffcb370e,
> >>ip=0x40000000003c8550
> >>.hpacucli(12458): unaligned access to 0x60000fffffcb370e,
> >>ip=0x40000000003c8550
> >
> >This seems to be coming out of user space. We'll work with the application
> >folks to investigate. There is a library called infomanager that's used
> >by the app. There may be an issue there. Call you strace the program and
> >send me the results? I haven't seen this in my lab with a vendor kernel.
> >
>
> What machine is this for? For x86-64, the above are non-canonical
> addresses. IA64?
Yup, ia64.
--
Jens Axboe
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [PATCH 1/3] cciss: bug fix for hpacucli
2005-11-18 16:33 [PATCH 1/3] cciss: bug fix for hpacucli mikem
2005-11-18 20:49 ` Jens Axboe
@ 2005-11-18 21:45 ` Philippe Pegon
1 sibling, 0 replies; 8+ messages in thread
From: Philippe Pegon @ 2005-11-18 21:45 UTC (permalink / raw)
To: mikem; +Cc: akpm, axboe, linux-kernel, linux-scsi, hpa, sitniko
mikem wrote:
> Patch 1 of 3
>
> This patch fixes a bug that breaks hpacucli, a command line interface
> for the HP Array Config Utility. Without this fix the utility will
> not detect any controllers in the system. I thought I had already fixed
> this, but I guess not.
>
> Thanks to all who reported the issue. Please consider this this inclusion.
reply too fast ;)
thanks you a lot for that one
--
Philippe Pegon
^ permalink raw reply [flat|nested] 8+ messages in thread
end of thread, other threads:[~2005-11-21 17:44 UTC | newest]
Thread overview: 8+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2005-11-18 16:33 [PATCH 1/3] cciss: bug fix for hpacucli mikem
2005-11-18 20:49 ` Jens Axboe
2005-11-21 8:28 ` Jens Axboe
2005-11-21 16:46 ` mikem
2005-11-21 17:23 ` Jens Axboe
2005-11-21 17:37 ` H. Peter Anvin
2005-11-21 17:45 ` Jens Axboe
2005-11-18 21:45 ` Philippe Pegon
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox