From: Jeff King <peff@peff.net>
To: Jonathan Nieder <jrnieder@gmail.com>
Cc: git@vger.kernel.org
Subject: Re: [PATCH v4 08/23] ewah: compressed bitmap implementation
Date: Thu, 23 Jan 2014 15:29:40 -0500 [thread overview]
Message-ID: <20140123202940.GB329@sigill.intra.peff.net> (raw)
In-Reply-To: <20140123202342.GZ18964@google.com>
On Thu, Jan 23, 2014 at 12:23:42PM -0800, Jonathan Nieder wrote:
> > The memcpy solution is taken from read-cache.c, but as we noted, it
> > probably hasn't been used a lot. The blk_sha1 get_be may be faster, as
> > it converts as it reads.
>
> I doubt there's much difference either way, especially after an
> optimizer gets its hands on it. According to [1] ARM has no fast
> byte swap instruction so with -O0 the byte-at-a-time implementation is
> probably faster there. I can try a performance test if you like.
If you're curious and have time, go ahead and benchmark what I posted
against what you posted (with your fix). But you'll probably need a big
repo like the kernel to notice anything.
But I don't mind that much if we just use the memcpy trick for now. It's
nice and obvious, and we can always change it later if somebody has
numbers (I doubt it will be all that noticeable anyway; this isn't
nearly as tight a loop as the BLK_SHA1 code).
-Peff
next prev parent reply other threads:[~2014-01-23 20:29 UTC|newest]
Thread overview: 68+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-12-21 13:56 [PATCH v4 0/22] pack bitmaps Jeff King
2013-12-21 13:59 ` [PATCH v4 01/23] sha1write: make buffer const-correct Jeff King
2013-12-22 9:06 ` Christian Couder
2013-12-21 13:59 ` [PATCH v4 02/23] revindex: Export new APIs Jeff King
2013-12-21 13:59 ` [PATCH v4 03/23] pack-objects: Refactor the packing list Jeff King
2013-12-21 13:59 ` [PATCH v4 04/23] pack-objects: factor out name_hash Jeff King
2013-12-21 13:59 ` [PATCH v4 05/23] revision: allow setting custom limiter function Jeff King
2013-12-21 13:59 ` [PATCH v4 06/23] sha1_file: export `git_open_noatime` Jeff King
2013-12-21 13:59 ` [PATCH v4 07/23] compat: add endianness helpers Jeff King
2013-12-21 13:59 ` [PATCH v4 08/23] ewah: compressed bitmap implementation Jeff King
2014-01-23 2:05 ` Jonathan Nieder
2014-01-23 18:33 ` Jeff King
2014-01-23 18:35 ` [PATCH 1/2] compat: move unaligned helpers to bswap.h Jeff King
2014-01-23 19:41 ` Jonathan Nieder
2014-01-23 19:44 ` Jeff King
2014-01-23 19:56 ` Jonathan Nieder
2014-01-23 20:04 ` Jeff King
2014-01-23 20:08 ` Jonathan Nieder
2014-01-23 20:09 ` Jeff King
2014-01-23 18:35 ` [PATCH 2/2] ewah: support platforms that require aligned reads Jeff King
2014-01-23 19:52 ` [PATCH v4 08/23] ewah: compressed bitmap implementation Jonathan Nieder
2014-01-23 20:03 ` Jeff King
2014-01-23 20:12 ` Jonathan Nieder
2014-01-23 20:13 ` Jeff King
2014-01-23 20:23 ` Jonathan Nieder
2014-01-23 20:29 ` Jeff King [this message]
2014-01-23 20:38 ` Jeff King
2014-01-23 20:14 ` Shawn Pearce
2014-01-23 20:26 ` Jeff King
2014-01-23 21:53 ` brian m. carlson
2014-01-23 22:07 ` Jeff King
2014-01-23 22:17 ` Jonathan Nieder
2014-01-23 22:26 ` Jeff King
2014-01-23 22:33 ` Jonathan Nieder
2014-01-23 20:18 ` Jonathan Nieder
2014-01-23 21:20 ` [PATCH v2 0/3] unaligned reads from .bitmap files Jeff King
2014-01-23 21:23 ` [PATCH 1/3] block-sha1: factor out get_be and put_be wrappers Jeff King
2014-01-23 23:19 ` Jonathan Nieder
2014-01-23 21:26 ` [PATCH 2/3] read-cache: use get_be32 instead of hand-rolled ntoh_l Jeff King
2014-01-23 23:34 ` Jonathan Nieder
2014-01-24 2:22 ` Jeff King
2014-01-23 21:27 ` [PATCH 3/3] ewah: support platforms that require aligned reads Jeff King
2014-01-23 23:44 ` Jonathan Nieder
2014-01-23 23:49 ` Vicent Martí
2014-01-24 0:15 ` Jonathan Nieder
2014-01-23 23:17 ` [PATCH v2 0/3] unaligned reads from .bitmap files Jonathan Nieder
2013-12-21 13:59 ` [PATCH v4 09/23] documentation: add documentation for the bitmap format Jeff King
2013-12-21 14:00 ` [PATCH v4 10/23] pack-bitmap: add support for bitmap indexes Jeff King
2013-12-21 14:00 ` [PATCH v4 11/23] pack-objects: split add_object_entry Jeff King
2013-12-21 14:00 ` [PATCH v4 12/23] pack-objects: use bitmaps when packing objects Jeff King
2013-12-21 14:00 ` [PATCH v4 13/23] rev-list: add bitmap mode to speed up object lists Jeff King
2013-12-21 14:00 ` [PATCH v4 14/23] pack-objects: implement bitmap writing Jeff King
2013-12-21 14:00 ` [PATCH v4 15/23] repack: stop using magic number for ARRAY_SIZE(exts) Jeff King
2013-12-21 14:00 ` [PATCH v4 16/23] repack: turn exts array into array-of-struct Jeff King
2013-12-21 14:00 ` [PATCH v4 17/23] repack: handle optional files created by pack-objects Jeff King
2013-12-21 14:00 ` [PATCH v4 18/23] repack: consider bitmaps when performing repacks Jeff King
2013-12-21 14:00 ` [PATCH v4 19/23] count-objects: recognize .bitmap in garbage-checking Jeff King
2013-12-21 14:00 ` [PATCH v4 20/23] t: add basic bitmap functionality tests Jeff King
2013-12-21 14:00 ` [PATCH v4 21/23] t/perf: add tests for pack bitmaps Jeff King
2013-12-21 14:00 ` [PATCH v4 22/23] pack-bitmap: implement optional name_hash cache Jeff King
2013-12-21 14:00 ` [PATCH v4 23/23] compat/mingw.h: Fix the MinGW and msvc builds Jeff King
2013-12-25 22:08 ` Erik Faye-Lund
2013-12-28 10:00 ` Jeff King
2013-12-28 10:06 ` Vicent Martí
2013-12-28 15:58 ` Ramsay Jones
2013-12-21 14:03 ` [PATCH v4 0/22] pack bitmaps Jeff King
2013-12-21 14:05 ` Jeff King
2013-12-21 18:34 ` Thomas Rast
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=20140123202940.GB329@sigill.intra.peff.net \
--to=peff@peff.net \
--cc=git@vger.kernel.org \
--cc=jrnieder@gmail.com \
/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).