linux-scsi.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* increase SD_STATS ?
@ 2005-08-19  8:02 Frederic TEMPORELLI
  2005-08-19  9:27 ` Christoph Hellwig
  0 siblings, 1 reply; 3+ messages in thread
From: Frederic TEMPORELLI @ 2005-08-19  8:02 UTC (permalink / raw)
  To: linux-scsi

Hi,


Due to SD_STATS=256  (/drivers/scsi/sd.c)
there's a limit on the max /dev/sd devices.

Now, we have a large FC fabric, with more than 3500 LUNs...
=> Of course, we have an Oops when an HBA is connected to the fabric
(still sd.c, BUG in function sd_iostats_bump when sd>SD_STATS)

What (bad) side-effects can be expected if we increase SD_STATS (4096 should be 
nice, may be 8192) ?

Best regards.

-- 
Frederic TEMPORELLI

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: increase SD_STATS ?
  2005-08-19  8:02 increase SD_STATS ? Frederic TEMPORELLI
@ 2005-08-19  9:27 ` Christoph Hellwig
  2005-08-19 10:20   ` Frederic TEMPORELLI
  0 siblings, 1 reply; 3+ messages in thread
From: Christoph Hellwig @ 2005-08-19  9:27 UTC (permalink / raw)
  To: Frederic TEMPORELLI; +Cc: linux-scsi

On Fri, Aug 19, 2005 at 10:02:43AM +0200, Frederic TEMPORELLI wrote:
> Hi,
> 
> 
> Due to SD_STATS=256  (/drivers/scsi/sd.c)
> there's a limit on the max /dev/sd devices.
> 
> Now, we have a large FC fabric, with more than 3500 LUNs...
> => Of course, we have an Oops when an HBA is connected to the fabric
> (still sd.c, BUG in function sd_iostats_bump when sd>SD_STATS)
> 
> What (bad) side-effects can be expected if we increase SD_STATS (4096 
> should be nice, may be 8192) ?

>From you description it looks like you are using some odd 2.4.x vendor
tree.  Just upgrade to latest 2.6 and these limitations will go away.


^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: increase SD_STATS ?
  2005-08-19  9:27 ` Christoph Hellwig
@ 2005-08-19 10:20   ` Frederic TEMPORELLI
  0 siblings, 0 replies; 3+ messages in thread
From: Frederic TEMPORELLI @ 2005-08-19 10:20 UTC (permalink / raw)
  Cc: linux-scsi

Hello Christoph,

Thanks for your answer.

Yes, I investigate about my kernel source (2.6.12) and I find that patches based 
on a RHEL kernel sources have been applied. This explains the odd 2.4 sd.c.

Best regards.


Christoph Hellwig wrote:
> On Fri, Aug 19, 2005 at 10:02:43AM +0200, Frederic TEMPORELLI wrote:
> 
>>Hi,
>>
>>
>>Due to SD_STATS=256  (/drivers/scsi/sd.c)
>>there's a limit on the max /dev/sd devices.
>>
>>Now, we have a large FC fabric, with more than 3500 LUNs...
>>=> Of course, we have an Oops when an HBA is connected to the fabric
>>(still sd.c, BUG in function sd_iostats_bump when sd>SD_STATS)
>>
>>What (bad) side-effects can be expected if we increase SD_STATS (4096 
>>should be nice, may be 8192) ?
> 
> 
>>From you description it looks like you are using some odd 2.4.x vendor
> tree.  Just upgrade to latest 2.6 and these limitations will go away.
> 
> -
> To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 


-- 
Frederic TEMPORELLI

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2005-08-19 10:20 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2005-08-19  8:02 increase SD_STATS ? Frederic TEMPORELLI
2005-08-19  9:27 ` Christoph Hellwig
2005-08-19 10:20   ` Frederic TEMPORELLI

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).