On Thu, 9 Nov 2000, Jens Axboe wrote:
>
> > The second is more elegant in that it side steps the problem by
> > giving rd.c a make_request function instead of using the default
> > _make_request. This means that io_request_lock is simply never
> > claimed my rd.
>
> And this solution is
On Thu, Nov 09 2000, Neil Brown wrote:
[snip]
> DEADLOCK
> I have two patches which address this problem.
> The first is simple and simply drops ui_request_lock before calling
> getblk. This may be the appropriate one to use given the code
> freeze.
rd st
Linus,
There is a problem with drivers/block/rd.c which can lead to a
deadlock an SMP hardware.
The scenario goes:
Processor A Processor B
enter _make_request in getblk (or elsewhere)
(or generic_unplug_device)
3 matches
Mail list logo
