lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <yq1zk43zftc.fsf@sermon.lab.mkp.net>
Date:	Wed, 03 Oct 2012 12:22:07 -0400
From:	"Martin K. Petersen" <martin.petersen@...cle.com>
To:	Kent Overstreet <koverstreet@...gle.com>
Cc:	NeilBrown <neilb@...e.de>, Jens Axboe <axboe@...nel.dk>,
	Shaohua Li <shli@...ionio.com>,
	lkml <linux-kernel@...r.kernel.org>, martin.petersen@...cle.com
Subject: Re: [PATCH] block: makes bio_split support bio without data

>>>>> "Kent" == Kent Overstreet <koverstreet@...gle.com> writes:

Kent> I think we should be able to split REQ_DISCARD bios that have a
Kent> payload or REQ_WRITE_SAME bios just fine though - for both of
Kent> those cases, the payload doesn't correspond to a particular
Kent> sector, so just copy the original bvec to the two splits and don't
Kent> do anything else to it.

DISCARD bios come down with a single bvec that is later used in the SCSI
disk driver to describe a memory page that can then be mapped into a
scatter-gather list. The reason for this is that both ATA TRIM and SCSI
UNMAP put the block range descriptors in the payload rather than in the
command itself. By the time MD calls bio_split there will be an empty
bvec in the bio.

For WRITE SAME the parent payload contains a bvec describing a single
logical block of data (i.e. typically 512 bytes). The same bvec is used
for both bios in the pair.

For neither DISCARD, nor WRITE SAME do we need to muck with bv_offset
and bv_len. As a result, my patch uses the bio_is_rw() conditional to
wrap the the bvec munging code.

-- 
Martin K. Petersen	Oracle Linux Engineering
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ