From: David Laight <David.Laight@ACULAB.COM>
To: 'Keith Busch' <kbusch@kernel.org>,
"linux-nvme@lists.infradead.org" <linux-nvme@lists.infradead.org>,
"linux-block@vger.kernel.org" <linux-block@vger.kernel.org>,
"linux-crypto@vger.kernel.org" <linux-crypto@vger.kernel.org>,
"x86@kernel.org" <x86@kernel.org>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Cc: "axboe@kernel.dk" <axboe@kernel.dk>, "hch@lst.de" <hch@lst.de>,
"martin.petersen@oracle.com" <martin.petersen@oracle.com>,
"colyli@suse.de" <colyli@suse.de>
Subject: RE: [PATCHv3 10/10] x86/crypto: add pclmul acceleration for crc64
Date: Tue, 22 Feb 2022 17:02:16 +0000 [thread overview]
Message-ID: <a7e806ed3c074534a24b74f827bcc914@AcuMS.aculab.com> (raw)
In-Reply-To: <20220222163144.1782447-11-kbusch@kernel.org>
From: Keith Busch
> Sent: 22 February 2022 16:32
>
> The crc64 table lookup method is inefficient, using a significant number
> of CPU cycles in the block stack per IO. If available on x86, use a
> PCLMULQDQ implementation to accelerate the calculation.
>
> The assembly from this patch was mostly generated by gcc from a C
> program using library functions provided by x86 intrinsics, and measures
> ~20x faster than the table lookup.
I think I'd like to see the C code and compiler options used to
generate the assembler as comments in the committed source file.
Either that or reasonable comments in the assembler.
It is also quite a lot of code.
What is the break-even length for 'cold cache' including the FPU saves.
...
> +.section .rodata
> +.align 32
> +.type shuffleMasks, @object
> +.size shuffleMasks, 32
> +shuffleMasks:
> + .string ""
> + .ascii "\001\002\003\004\005\006\007\b\t\n\013\f\r\016\017\217\216\215"
> + .ascii "\214\213\212\211\210\207\206\205\204\203\202\201\200"
That has to be the worst way to define 32 bytes.
> +.section .rodata.cst16,"aM",@progbits,16
> +.align 16
> +.LC0:
> + .quad -1523270018343381984
> + .quad 2443614144669557164
> + .align 16
> +.LC1:
> + .quad 2876949357237608311
> + .quad 3808117099328934763
Not sure what those are, but I bet there are better ways to
define/describe them.
David
-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK
Registration No: 1397386 (Wales)
next prev parent reply other threads:[~2022-02-22 17:02 UTC|newest]
Thread overview: 47+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-02-22 16:31 [PATCHv3 00/10] 64-bit data integrity field support Keith Busch
2022-02-22 16:31 ` [PATCHv3 01/10] block: support pi with extended metadata Keith Busch
2022-02-25 16:01 ` Christoph Hellwig
2022-02-22 16:31 ` [PATCHv3 02/10] nvme: allow integrity on extended metadata formats Keith Busch
2022-02-25 16:02 ` Christoph Hellwig
2022-02-22 16:31 ` [PATCHv3 03/10] asm-generic: introduce be48 unaligned accessors Keith Busch
2022-02-22 16:52 ` Chaitanya Kulkarni
2022-02-25 16:03 ` Christoph Hellwig
2022-02-25 17:53 ` Joe Perches
2022-02-25 17:59 ` Keith Busch
2022-02-22 16:31 ` [PATCHv3 04/10] linux/kernel: introduce lower_48_bits macro Keith Busch
2022-02-22 16:45 ` Joe Perches
2022-02-22 16:50 ` Christoph Hellwig
2022-02-22 16:56 ` Keith Busch
2022-02-22 18:43 ` Joe Perches
2022-02-22 20:09 ` David Laight
2022-02-22 20:31 ` Joe Perches
2022-02-22 21:12 ` Keith Busch
2022-02-22 21:17 ` Joe Perches
2022-02-22 16:58 ` Joe Perches
2022-02-22 17:09 ` David Laight
2022-02-22 17:14 ` Chaitanya Kulkarni
2022-02-22 16:48 ` Chaitanya Kulkarni
2022-02-22 16:31 ` [PATCHv3 05/10] lib: add rocksoft model crc64 Keith Busch
2022-02-25 16:04 ` Christoph Hellwig
2022-02-22 16:31 ` [PATCHv3 06/10] crypto: add rocksoft 64b crc framework Keith Busch
2022-02-22 19:50 ` Eric Biggers
2022-02-22 19:54 ` Eric Biggers
2022-02-22 20:09 ` Keith Busch
2022-02-25 16:11 ` Christoph Hellwig
2022-02-22 19:56 ` Eric Biggers
2022-02-22 16:31 ` [PATCHv3 07/10] lib: add crc64 tests Keith Busch
2022-02-22 16:50 ` Chaitanya Kulkarni
2022-02-25 16:05 ` Christoph Hellwig
2022-02-25 16:12 ` Keith Busch
2022-02-25 16:19 ` Christoph Hellwig
2022-02-22 16:31 ` [PATCHv3 08/10] block: add pi for nvme enhanced integrity Keith Busch
2022-02-25 16:14 ` Christoph Hellwig
2022-03-02 3:15 ` Martin K. Petersen
2022-02-22 16:31 ` [PATCHv3 09/10] nvme: add support for enhanced metadata Keith Busch
2022-02-25 16:17 ` Christoph Hellwig
2022-03-02 3:18 ` Martin K. Petersen
2022-02-22 16:31 ` [PATCHv3 10/10] x86/crypto: add pclmul acceleration for crc64 Keith Busch
2022-02-22 17:02 ` David Laight [this message]
2022-02-22 17:14 ` Keith Busch
2022-02-22 20:06 ` Eric Biggers
2022-02-22 20:51 ` Keith Busch
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=a7e806ed3c074534a24b74f827bcc914@AcuMS.aculab.com \
--to=david.laight@aculab.com \
--cc=axboe@kernel.dk \
--cc=colyli@suse.de \
--cc=hch@lst.de \
--cc=kbusch@kernel.org \
--cc=linux-block@vger.kernel.org \
--cc=linux-crypto@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-nvme@lists.infradead.org \
--cc=martin.petersen@oracle.com \
--cc=x86@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).