From: drepper@gmail.com
To: Wu Fengguang <fengguang.wu@intel.com>
Cc: Taras Glek <tglek@mozilla.com>,
Johannes Weiner <hannes@cmpxchg.org>,
"linux-mm@kvack.org" <linux-mm@kvack.org>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: Re: Downsides to madvise/fadvise(willneed) for application startup
Date: Sun, 11 Apr 2010 21:43:00 -0700 (PST) [thread overview]
Message-ID: <g7wssj9j6ukus9yti3UYAxe124vaj_firegpg@mail.gmail.com> (raw)
In-Reply-To: <20100412022704.GB5151@localhost>
[-- Attachment #1: Type: text/plain, Size: 936 bytes --]
On Sun, Apr 11, 2010 at 19:27, Wu Fengguang <fengguang.wu@intel.com> wrote:
> Yes, every binary/library starts with this 512b read. It is requested
> by ld.so/ld-linux.so, and will trigger a 4-page readahead. This is not
> good readahead. I wonder if ld.so can switch to mmap read for the
> first read, in order to trigger a larger 128kb readahead.
We first need to know the sizes of the segments and their location in the binary. The binaries we use now are somewhat well laid out. The read-only segment starts at offset 0 etc. But this doesn't have to be the case. The dynamic linker has to be generic. Also, even if we start mapping at offset zero, now much to map? The file might contain debug info which must not be mapped. Therefore the first read loads enough of the headers to make all of the decisions. Yes, we could do a mmap of one page instead of the read. But that's more expansive in general, isn't it?
[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 272 bytes --]
next prev parent reply other threads:[~2010-04-12 4:43 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <4BBA6776.5060804@mozilla.com>
2010-04-06 9:51 ` Downsides to madvise/fadvise(willneed) for application startup Johannes Weiner
2010-04-06 21:57 ` Taras Glek
2010-04-06 22:26 ` Johannes Weiner
2010-04-06 22:39 ` Taras Glek
2010-04-07 2:24 ` Wu Fengguang
2010-04-07 2:54 ` Taras Glek
2010-04-07 4:06 ` Minchan Kim
2010-04-07 7:14 ` Wu Fengguang
2010-04-07 7:33 ` Minchan Kim
2010-04-07 7:47 ` Wu Fengguang
2010-04-07 8:06 ` Minchan Kim
2010-04-07 8:13 ` Wu Fengguang
2010-04-07 7:38 ` Wu Fengguang
2010-04-08 17:44 ` Taras Glek
2010-04-12 2:27 ` Wu Fengguang
2010-04-12 3:25 ` Minchan Kim
2010-04-12 4:58 ` Wu Fengguang
2010-04-12 4:43 ` drepper [this message]
2010-04-12 4:46 ` Taras Glek
2010-04-12 4:50 ` Wu Fengguang
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=g7wssj9j6ukus9yti3UYAxe124vaj_firegpg@mail.gmail.com \
--to=drepper@gmail.com \
--cc=fengguang.wu@intel.com \
--cc=hannes@cmpxchg.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=tglek@mozilla.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).