On 10/17/2014 06:46 PM, Martin K. Petersen wrote:
>> "Petr" == Petr Vandrovec writes:
>
> Petr> is there any reason to do blacklisting? Why not let first request
> Petr> fail, and switch to non-write-same code path once that happens?
>
> Well, we do. But we try to avoid confusing users with
> "Petr" == Petr Vandrovec writes:
Petr> is there any reason to do blacklisting? Why not let first request
Petr> fail, and switch to non-write-same code path once that happens?
Well, we do. But we try to avoid confusing users with error messages if
we can avoid it. And if we know it's not g
On 10/14/2014 5:57 PM, Martin K. Petersen wrote:
"Petr" == Petr Vandrovec writes:
Petr,
Petr> Logic (from 2011, commit 8af1954d172a46a63e5e79dae523a6d74715e458)
Petr> says that EOPNOTSUPP is returned when DISCARD request failed, as
Petr> discarding is optional, and failures can be safely igno
> "Petr" == Petr Vandrovec writes:
Petr,
Petr> Logic (from 2011, commit 8af1954d172a46a63e5e79dae523a6d74715e458)
Petr> says that EOPNOTSUPP is returned when DISCARD request failed, as
Petr> discarding is optional, and failures can be safely ignored. That
Petr> is definitely not true for WR
2001
From: Petr Vandrovec
Date: Fri, 10 Oct 2014 23:10:25 -0700
Subject: [PATCH] Do not silently discard WRITE_SAME requests
When device does not support WRITE_SAME, after first failure
block layer starts throwing away WRITE_SAME requests without
warning anybody, leading to the data corruption.
L
> "Petr" == Petr Vandrovec writes:
Petr> After investigating, problem seems to be in a way completion
Petr> handler for WRITE_SAME handles EOPNOTSUPP error, causing
Petr> all-but-first WRITE_SAME request on the LVM device to be silently
Petr> ignored - command is never issued, but success is
Thanks,
Petr Vandrovec
From: Petr Vandrovec
Subject: [PATCH] Do not silently discard WRITE_SAME requests
When device does not support WRITE_SAME, after first failure
block layer starts throwing away WRITE_SAME requests without
warning anybody,
7 matches
Mail list logo
