linux-kernel - Re: [PATCH 4/5] brd: Request from fdisk 4k alignment

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: <yq1389r6les.fsf@sermon.lab.mkp.net>
Date:	Mon, 10 Nov 2014 12:00:11 -0500
From:	"Martin K. Petersen" <martin.petersen@...cle.com>
To:	Boaz Harrosh <boaz@...xistor.com>
Cc:	"Martin K. Petersen" <martin.petersen@...cle.com>,
	Jens Axboe <axboe@...com>,
	Matthew Wilcox <willy@...ux.intel.com>,
	Dmitry Monakhov <dmonakhov@...nvz.org>,
	linux-kernel <linux-kernel@...r.kernel.org>,
	linux-fsdevel <linux-fsdevel@...r.kernel.org>,
	Karel Zak <kzak@...hat.com>
Subject: Re: [PATCH 4/5] brd: Request from fdisk 4k alignment

>>>>> "Boaz" == Boaz Harrosh <boaz@...xistor.com> writes:

Boaz> I do not know why I thought that only io_min does that, I can see
Boaz> now that both effect the Kernel the same way. Which scares me a
Boaz> bit.

The difference is subtle. io_min is a hint from the storage about its
preferred minimum I/O size. pbs describes the smallest unit that can be
atomically written (like a sector on a drive with 512-byte logical/4K
physical blocks). In most cases they are the same, pbs is just a
slightly harder guarantee than io_min.

What I was objecting to in your patch description was mostly the
statement you made that these values affect kernel behavior. They really
don't. Not directly, anyway. The queue limits are stacked and offsets
are adjusted based on partitions, etc. But they don't alter the kernel
runtime behavior.

The queue limits are reported to userland and will affect things like
partitioning, MD/DM tooling and mkfs.*. And therefore they only
indirectly affect the kernel's behavior.

-- 
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/