From: Doug Evans <dje@google.com>
To: Tom Tromey <tromey@redhat.com>,
Ulrich Weigand <uweigand@de.ibm.com>,
gdb-patches@sourceware.org
Subject: Re: [10/15] Basic value access routines
Date: Wed, 24 Jun 2009 16:26:00 -0000 [thread overview]
Message-ID: <e394668d0906240925t3fe4ceb7mccb792e67bb5dfbc@mail.gmail.com> (raw)
In-Reply-To: <20090624154618.GA1825@caradoc.them.org>
On Wed, Jun 24, 2009 at 8:46 AM, Daniel Jacobowitz<drow@false.org> wrote:
> On Wed, Jun 24, 2009 at 09:31:16AM -0600, Tom Tromey wrote:
>> A related idea that occurred to me is that, with Jan's type GC, we
>> could detach types from objfiles and "intern" them (something like
>> bcache-for-types). This would only be a win if there are many
>> identical types distributed across objfiles, though.
>
> There's plenty of identical types within a single objfile. Load GDB
> with -readnow, and I find 66 copies of the typedef for "FILE", for
> instance.
As a data point, I compiled gdb with -g -O2 and again with -gdwarf-4 -O2.
-rwxr-x--- 1 dje eng 16321609 Jun 24 08:52 gdb
-rwxr-x--- 1 dje eng 13019406 Jun 24 08:50 gdb-dw4
Starting gdb with -readnow and then examining space usage:
-g -O2:
gdb$ ps l 24343
F UID PID PPID PRI NI VSZ RSS WCHAN STAT TTY TIME COMMAND
0 67641 24343 24342 20 0 117356 97764 - S+ pts/13 0:00 ../../../..
-gdwarf-4 -O2:
gdb$ ps l 24349
F UID PID PPID PRI NI VSZ RSS WCHAN STAT TTY TIME COMMAND
0 67641 24349 16588 20 0 72088 53280 - S+ pts/13 0:00 ../../../..
Until my comdat types patch is vetted more, these numbers are to be
taken with a grain of salt, but they are promising.
next prev parent reply other threads:[~2009-06-24 16:26 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-06-09 15:20 Ulrich Weigand
2009-06-15 16:35 ` Doug Evans
2009-06-15 16:59 ` Ulrich Weigand
2009-06-22 20:32 ` Per-type architecture (Re: [10/15] Basic value access routines) Ulrich Weigand
2009-06-22 20:59 ` Daniel Jacobowitz
2009-06-23 0:41 ` Ulrich Weigand
2009-06-23 13:49 ` Daniel Jacobowitz
2009-06-23 14:02 ` Pedro Alves
2009-06-23 17:06 ` Ulrich Weigand
2009-06-23 17:30 ` Daniel Jacobowitz
2009-06-23 15:55 ` Doug Evans
2009-06-23 16:42 ` Mark Kettenis
2009-06-23 16:53 ` Daniel Jacobowitz
2009-06-23 16:57 ` Ulrich Weigand
2009-06-23 17:36 ` Doug Evans
2009-06-24 15:31 ` [10/15] Basic value access routines Tom Tromey
2009-06-24 15:46 ` Daniel Jacobowitz
2009-06-24 16:26 ` Doug Evans [this message]
2009-06-24 19:41 ` Tom Tromey
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=e394668d0906240925t3fe4ceb7mccb792e67bb5dfbc@mail.gmail.com \
--to=dje@google.com \
--cc=gdb-patches@sourceware.org \
--cc=tromey@redhat.com \
--cc=uweigand@de.ibm.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