* Solaris issue with packed stat structures
@ 2014-10-30 14:34 Sébastien Bouchex Bellomié
2014-10-30 14:42 ` Jens Axboe
0 siblings, 1 reply; 5+ messages in thread
From: Sébastien Bouchex Bellomié @ 2014-10-30 14:34 UTC (permalink / raw)
To: fio@vger.kernel.org
Hi,
I'm the latest source on solaris SPARC and I have BUS errors when generating stats :
Stack looks like this :
0002bac4 sum_stat (17973e, fbc05b38, 1, 0, 0, 22) + 38
0002cf24 sum_thread_stats (179526, fbc05920, 1, 0, fbd7de7c, 4) + a8
0003113c __show_run_stats (fe40ad88, 400, 0, 0, 0, 0) + 1124
000b0b10 fio_backend (0, ffbffcf4, 0, 0, 0, 15c174) + 200
000e6a40 main (2, ffbffcf4, ffbffd00, 166130, 0, fefe0180) + c4
00017844 _start (0, 0, 0, 0, 0, 0) + 5c
It works fine on intel processors because the hardware is handling the misalignment but on Sparc, it is not handled, so a BUS error is generated
To fix this issue, I have added some padding here :
[...]
uint16_t continue_on_error;
uint16_t continue_on_error_pad; // <<<<<==== Pading here
uint64_t total_err_count;
uint32_t first_error;
[...]
Thanks
Seb
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: Solaris issue with packed stat structures
2014-10-30 14:34 Solaris issue with packed stat structures Sébastien Bouchex Bellomié
@ 2014-10-30 14:42 ` Jens Axboe
2014-10-30 14:55 ` Sébastien Bouchex Bellomié
0 siblings, 1 reply; 5+ messages in thread
From: Jens Axboe @ 2014-10-30 14:42 UTC (permalink / raw)
To: Sébastien Bouchex Bellomié, fio@vger.kernel.org
On 2014-10-30 08:34, S�bastien Bouchex Bellomi� wrote:
> Hi,
>
> I'm the latest source on solaris SPARC and I have BUS errors when generating stats :
>
> Stack looks like this :
>
> 0002bac4 sum_stat (17973e, fbc05b38, 1, 0, 0, 22) + 38
> 0002cf24 sum_thread_stats (179526, fbc05920, 1, 0, fbd7de7c, 4) + a8
> 0003113c __show_run_stats (fe40ad88, 400, 0, 0, 0, 0) + 1124
> 000b0b10 fio_backend (0, ffbffcf4, 0, 0, 0, 15c174) + 200
> 000e6a40 main (2, ffbffcf4, ffbffd00, 166130, 0, fefe0180) + c4
> 00017844 _start (0, 0, 0, 0, 0, 0) + 5c
>
> It works fine on intel processors because the hardware is handling the misalignment but on Sparc, it is not handled, so a BUS error is generated
>
> To fix this issue, I have added some padding here :
>
> [...]
> uint16_t continue_on_error;
> uint16_t continue_on_error_pad; // <<<<<==== Pading here
> uint64_t total_err_count;
> uint32_t first_error;
What -git revision are you on? The latest should have the correct padding.
--
Jens Axboe
^ permalink raw reply [flat|nested] 5+ messages in thread
* RE: Solaris issue with packed stat structures
2014-10-30 14:42 ` Jens Axboe
@ 2014-10-30 14:55 ` Sébastien Bouchex Bellomié
2014-10-30 14:56 ` Jens Axboe
0 siblings, 1 reply; 5+ messages in thread
From: Sébastien Bouchex Bellomié @ 2014-10-30 14:55 UTC (permalink / raw)
To: Jens Axboe, fio@vger.kernel.org
Hi,
The version on opencsw.org was buggy. I took the one on the git server and everything works fine.
Thanks
Seb
-----Message d'origine-----
De�: fio-owner@vger.kernel.org [mailto:fio-owner@vger.kernel.org] De la part de Jens Axboe
Envoy�: jeudi 30 octobre 2014 15:43
��: S�bastien Bouchex Bellomi�; fio@vger.kernel.org
Objet�: Re: Solaris issue with packed stat structures
On 2014-10-30 08:34, S�bastien Bouchex Bellomi� wrote:
> Hi,
>
> I'm the latest source on solaris SPARC and I have BUS errors when generating stats :
>
> Stack looks like this :
>
> 0002bac4 sum_stat (17973e, fbc05b38, 1, 0, 0, 22) + 38
> 0002cf24 sum_thread_stats (179526, fbc05920, 1, 0, fbd7de7c, 4) + a8
> 0003113c __show_run_stats (fe40ad88, 400, 0, 0, 0, 0) + 1124
> 000b0b10 fio_backend (0, ffbffcf4, 0, 0, 0, 15c174) + 200
> 000e6a40 main (2, ffbffcf4, ffbffd00, 166130, 0, fefe0180) + c4
> 00017844 _start (0, 0, 0, 0, 0, 0) + 5c
>
> It works fine on intel processors because the hardware is handling the
> misalignment but on Sparc, it is not handled, so a BUS error is
> generated
>
> To fix this issue, I have added some padding here :
>
> [...]
> uint16_t continue_on_error;
> uint16_t continue_on_error_pad; // <<<<<==== Pading here
> uint64_t total_err_count;
> uint32_t first_error;
What -git revision are you on? The latest should have the correct padding.
--
Jens Axboe
--
To unsubscribe from this list: send the line "unsubscribe fio" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: Solaris issue with packed stat structures
2014-10-30 14:55 ` Sébastien Bouchex Bellomié
@ 2014-10-30 14:56 ` Jens Axboe
2014-10-30 17:12 ` Sébastien Bouchex Bellomié
0 siblings, 1 reply; 5+ messages in thread
From: Jens Axboe @ 2014-10-30 14:56 UTC (permalink / raw)
To: Sébastien Bouchex Bellomié, fio@vger.kernel.org
On 2014-10-30 08:55, S�bastien Bouchex Bellomi� wrote:
> Hi,
>
> The version on opencsw.org was buggy. I took the one on the git server and everything works fine.
Ah I see, yes that one is a bit old. I'll tag a new release soon,
hopefully we can notify opencsw and get a new one uploaded.
--
Jens Axboe
^ permalink raw reply [flat|nested] 5+ messages in thread
* RE: Solaris issue with packed stat structures
2014-10-30 14:56 ` Jens Axboe
@ 2014-10-30 17:12 ` Sébastien Bouchex Bellomié
0 siblings, 0 replies; 5+ messages in thread
From: Sébastien Bouchex Bellomié @ 2014-10-30 17:12 UTC (permalink / raw)
To: Jens Axboe, fio@vger.kernel.org
Hi,
Yes and the latest release build is 2.0.14, so that's quiet old.
http://www.opencsw.org/qa/package/fio/
Seb
-----Message d'origine-----
De : Jens Axboe [mailto:axboe@kernel.dk]
Envoyé : jeudi 30 octobre 2014 15:57
À : Sébastien Bouchex Bellomié; fio@vger.kernel.org
Objet : Re: Solaris issue with packed stat structures
On 2014-10-30 08:55, Sébastien Bouchex Bellomié wrote:
> Hi,
>
> The version on opencsw.org was buggy. I took the one on the git server and everything works fine.
Ah I see, yes that one is a bit old. I'll tag a new release soon, hopefully we can notify opencsw and get a new one uploaded.
--
Jens Axboe
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2014-10-30 17:12 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2014-10-30 14:34 Solaris issue with packed stat structures Sébastien Bouchex Bellomié
2014-10-30 14:42 ` Jens Axboe
2014-10-30 14:55 ` Sébastien Bouchex Bellomié
2014-10-30 14:56 ` Jens Axboe
2014-10-30 17:12 ` Sébastien Bouchex Bellomié
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.