Re: [dm-devel] mirror patches - device failure detection/race fix/deadlock fix

[Jonathan E Brassow <jbrassow redhat com>]

On Aug 2, 2005, at 11:00 AM, Jonathan E Brassow wrote:

> The first 4 patches (0000[1234].patch) are the device failure  
> detection patches.  They've been cleaned up a bit from the last time  
> they were posted.
>
> The 'race.patch' fixes a couple race conditions that were a result of  
> the __rh_alloc() function releasing and regrabbing a lock when called.
>
> The 'deadlock.patch' switches all spin_lock calls on the region_lock  
> to spin_lock_irqsave.  Without this patch you machine will hang.
>
> The patches apply in the order described, and should apply to kernels  
> 2.6.12 - 2.6.13-rc5.
>
> I've made them all attachments to this e-mail.  If people don't like  
> it this way, let me know, and I'll do it lkml style.
>
>  brassow
>
> <00001.patch><00002.patch><00003.patch><00004.patch><race.patch><deadlo 
> ck.patch>--
> dm-devel mailing list
> dm-devel redhat com
> https://www.redhat.com/mailman/listinfo/dm-devel


I'm an idiot.  :)

The 0000[1234].patch files remain unchanged, but you can forget about  
the race & deadlock patch files.  For fixes to those problems, try this  
new race patch.

QA has confirmed that this patch has at least taken care of the  
problems I was trying to address with the other patches...

 brassow

Attachment: race2.patch
Description: Binary data