Re: cvs commit: src/lib/libc/stdlib malloc.c rb.h

Bruce M. Simpson wrote: Jason Evans wrote: Log: Implement red-black trees without using parent pointers, and store the color bit in the least significant bit of the right child pointer, in order to reduce red-black tree linkage overhead by ~2X as compared to sys/tree.h. Interesting.

Re: cvs commit: src/lib/libc/stdlib malloc.c rb.h

Jason Evans wrote: Log: Implement red-black trees without using parent pointers, and store the color bit in the least significant bit of the right child pointer, in order to reduce red-black tree linkage overhead by ~2X as compared to sys/tree.h. Interesting. Will this work in the ker

Re: cvs commit: src/lib/libc/stdlib malloc.c

Jason Evans wrote: jasone 2007-12-28 07:24:20 UTC FreeBSD src repository Modified files: lib/libc/stdlib malloc.c Log: Maintain two trees instead of one (old_chunks --> old_chunks_{ad,szad}) in order to support re-use of multi-chunk unused regions within the DSS for

Re: cvs commit: src/lib/libc/stdlib malloc.c

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Jason Evans wrote: > jasone 2007-11-27 03:17:30 UTC > > FreeBSD src repository > > Modified files: > lib/libc/stdlib malloc.c > Log: > Implement dynamic load balancing of thread-->arena mapping, based on lock > contention. T

Re: cvs commit: src/lib/libc/stdlib malloc.c

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 > > Proper tuning parameter configuration for this change is a finicky business, > and it is very much machine-dependent. One seemingly promising solution > would be to run a tuning program during operating system installation that > computes

Re: cvs commit: src/lib/libc/stdlib malloc.c

On Sun, Mar 25, 2007 at 06:08:37PM +0300, Andrew Pantyukhin wrote: > On 3/23/07, Jason Evans <[EMAIL PROTECTED]> wrote: > >Matteo Riondato wrote: > >> On Fri, Mar 23, 2007 at 05:05:48AM +, Jason Evans wrote: > >>> jasone 2007-03-23 05:05:48 UTC > >>> > >>> FreeBSD src repository > >>> >

Re: cvs commit: src/lib/libc/stdlib malloc.c

On 3/23/07, Jason Evans <[EMAIL PROTECTED]> wrote: Matteo Riondato wrote: > On Fri, Mar 23, 2007 at 05:05:48AM +, Jason Evans wrote: >> jasone 2007-03-23 05:05:48 UTC >> >> FreeBSD src repository >> >> Modified files: >> lib/libc/stdlib malloc.c > >> Revision ChangesP

Re: cvs commit: src/lib/libc/stdlib malloc.c

On Fri, Mar 23, 2007 at 10:58:15PM +, Jason Evans wrote: > jasone 2007-03-23 22:58:15 UTC > > FreeBSD src repository > > Modified files: > lib/libc/stdlib malloc.c > Revision ChangesPath > 1.143 +297 -151 src/lib/libc/stdlib/malloc.c I can confirm this commi

Re: cvs commit: src/lib/libc/stdlib malloc.c

Matteo Riondato wrote: On Fri, Mar 23, 2007 at 05:05:48AM +, Jason Evans wrote: jasone 2007-03-23 05:05:48 UTC FreeBSD src repository Modified files: lib/libc/stdlib malloc.c Revision ChangesPath 1.142 +337 -328 src/lib/libc/stdlib/malloc.c Can this co

Re: cvs commit: src/lib/libc/stdlib malloc.c

On Fri, Mar 23, 2007 at 05:05:48AM +, Jason Evans wrote: > jasone 2007-03-23 05:05:48 UTC > > FreeBSD src repository > > Modified files: > lib/libc/stdlib malloc.c > Revision ChangesPath > 1.142 +337 -328 src/lib/libc/stdlib/malloc.c Can this commit be the c

Re: cvs commit: src/lib/libc/stdlib malloc.c

On 2/1/07, John Baldwin <[EMAIL PROTECTED]> wrote: On Wednesday 31 January 2007 19:16, Jason Evans wrote: > Coleman Kane wrote: > > Modified files: > > lib/libc/stdlib malloc.c > > Log: > > Fix a utrace(2)-related bug in calloc(3). > > > > Integrate various p

Re: cvs commit: src/lib/libc/stdlib malloc.c

On Wednesday 31 January 2007 19:16, Jason Evans wrote: > Coleman Kane wrote: > > Modified files: > > lib/libc/stdlib malloc.c > > Log: > > Fix a utrace(2)-related bug in calloc(3). > > > > Integrate various pedantic cleanups. > > > > Submitted by: Andr

Re: cvs commit: src/lib/libc/stdlib malloc.c

Coleman Kane wrote: Modified files: lib/libc/stdlib malloc.c Log: Fix a utrace(2)-related bug in calloc(3). Integrate various pedantic cleanups. Submitted by: Andrew Doran <[EMAIL PROTECTED] > Revision ChangesPa

Re: cvs commit: src/lib/libc/stdlib malloc.c

On 1/31/07, Jason Evans <[EMAIL PROTECTED]> wrote: jasone 2007-01-31 22:54:19 UTC FreeBSD src repository Modified files: lib/libc/stdlib malloc.c Log: Fix a utrace(2)-related bug in calloc(3). Integrate various pedantic cleanups. Submitted by: Andrew Doran <[EMAIL

Re: cvs commit: src/lib/libc/stdlib malloc.c

Olivier Houchard wrote this message on Thu, Jul 27, 2006 at 14:36 +: > cognet 2006-07-27 14:36:28 UTC > > FreeBSD src repository > > Modified files: > lib/libc/stdlib malloc.c > Log: > Use 4 as QUANTUM_2POW_MIN on arm as it is on any other architecture, to > avoid >

Re: cvs commit: src/lib/libc/stdlib malloc.c

Peter Jeremy wrote: On Wed, 2006-Apr-05 18:46:24 +, Jason Evans wrote: jasone 2006-04-05 18:46:24 UTC FreeBSD src repository Modified files: lib/libc/stdlib malloc.c Log: Add an unreachable return statement, in order to avoid a compiler warning for non-standard optimization

Re: cvs commit: src/lib/libc/stdlib malloc.c

On Wed, 2006-Apr-05 18:46:24 +, Jason Evans wrote: >jasone 2006-04-05 18:46:24 UTC > > FreeBSD src repository > > Modified files: >lib/libc/stdlib malloc.c > Log: > Add an unreachable return statement, in order to avoid a compiler warning > for non-standard optimization leve

Re: cvs commit: src/lib/libc/stdlib malloc.c

Jason Evans wrote: jasone 2006-03-24 22:13:49 UTC FreeBSD src repository Modified files: lib/libc/stdlib malloc.c Log: Convert TINY_MIN_2POW from a cpp macro to tiny_min_2pow (a variable), and determine its value at run time according to other relevant values. This

Re: cvs commit: src/lib/libc/stdlib malloc.c

On Thursday 26 January 2006 22:02, Jason Evans wrote: > On Jan 26, 2006, at 6:51 PM, Andrey Chernov wrote: > > I worry about my system performance. Could you please make all > > statistics > > and debug code conditional, debending on some flags which could be > > set as > > malloc options? Even in

Re: cvs commit: src/lib/libc/stdlib malloc.c

On Fri, Jan 27, 2006 at 09:45:46AM +0100, Poul-Henning Kamp wrote: > In message <[EMAIL PROTECTED]>, Andrey Chernov writes: > >> Don't use -current if you can't tolerate the minor performance > >> hit associated with making FreeBSD better. > > > >Don't ever try to tell anybody what to do if you don

Re: cvs commit: src/lib/libc/stdlib malloc.c

Ulrich Spörlein wrote: > What about introducing the right make(1) magic to have > NO_MALLOC_DEBUG=true work when put into make.conf? Ah, forget about it. I just read the follow-up commit. Ulrich Spoerlein -- PGP Key ID: 20FEE9DD Encrypted mail welcome! Fingerprint: AEC

Re: cvs commit: src/lib/libc/stdlib malloc.c

Gleb Smirnoff wrote: > On Fri, Jan 27, 2006 at 05:51:50AM +0300, Andrey Chernov wrote: > > It is much easy to turn some tests off via malloc options than to > > comment out manually corresponding malloc.c defines after each > > cvsup. > > Just use CVS instead of cvsup, and you won't need to do any

Re: cvs commit: src/lib/libc/stdlib malloc.c

In message <[EMAIL PROTECTED]>, Andrey Chernov writes: >> Don't use -current if you can't tolerate the minor performance >> hit associated with making FreeBSD better. > >Don't ever try to tell anybody what to do if you don't know all his >circumstacnes. Especially when being not asked. Why not,

Re: cvs commit: src/lib/libc/stdlib malloc.c

On Fri, Jan 27, 2006 at 05:51:50AM +0300, Andrey Chernov wrote: A> On Fri, Jan 27, 2006 at 02:36:44AM +, Jason Evans wrote: A> > jasone 2006-01-27 02:36:44 UTC A> > A> > FreeBSD src repository A> > A> > Modified files: A> > lib/libc/stdlib malloc.c A> > Log: A> > Clean

Re: cvs commit: src/lib/libc/stdlib malloc.c

In message: <[EMAIL PROTECTED]> Steve Kargl <[EMAIL PROTECTED]> writes: : On Fri, Jan 27, 2006 at 07:38:43AM +0300, Andrey Chernov wrote: : > On Thu, Jan 26, 2006 at 08:14:12PM -0800, Steve Kargl wrote: : > > On Fri, Jan 27, 2006 at 06:24:03AM +0300, Andrey Chernov wrote: : > > > On Thu

Re: cvs commit: src/lib/libc/stdlib malloc.c

On Jan 26, 2006, at 9:09 PM, Andrey Chernov wrote: On Fri, Jan 27, 2006 at 04:42:10AM +, Jason Evans wrote: Add NO_MALLOC_EXTRAS, so that various extra features that can cause performance degradation can be disabled via something like the following in /etc/malloc.conf: C

Re: cvs commit: src/lib/libc/stdlib malloc.c

On Fri, Jan 27, 2006 at 04:42:10AM +, Jason Evans wrote: > Add NO_MALLOC_EXTRAS, so that various extra features that can cause > performance degradation can be disabled via something like the following > in /etc/malloc.conf: > > CFLAGS+=-DNO_MALLOC_EXTRAS Thanx. You probably

Re: cvs commit: src/lib/libc/stdlib malloc.c

On Fri, Jan 27, 2006 at 07:38:43AM +0300, Andrey Chernov wrote: > On Thu, Jan 26, 2006 at 08:14:12PM -0800, Steve Kargl wrote: > > On Fri, Jan 27, 2006 at 06:24:03AM +0300, Andrey Chernov wrote: > > > On Thu, Jan 26, 2006 at 09:15:02PM -0600, Christian S.J. Peron wrote: > > > > These malloc debuggi

Re: cvs commit: src/lib/libc/stdlib malloc.c

On Thu, Jan 26, 2006 at 08:14:12PM -0800, Steve Kargl wrote: > On Fri, Jan 27, 2006 at 06:24:03AM +0300, Andrey Chernov wrote: > > On Thu, Jan 26, 2006 at 09:15:02PM -0600, Christian S.J. Peron wrote: > > > These malloc debugging/stats features have been invaluable. I vote to > > > accept the perf

Re: cvs commit: src/lib/libc/stdlib malloc.c

On Fri, Jan 27, 2006 at 06:24:03AM +0300, Andrey Chernov wrote: > On Thu, Jan 26, 2006 at 09:15:02PM -0600, Christian S.J. Peron wrote: > > These malloc debugging/stats features have been invaluable. I vote to > > accept the performance hit in exchange for useful debugging/stats > > features in -

Re: cvs commit: src/lib/libc/stdlib malloc.c

Jason Evans wrote: On Jan 26, 2006, at 6:51 PM, Andrey Chernov wrote: I worry about my system performance. Could you please make all statistics and debug code conditional, debending on some flags which could be set as malloc options? Even in case this is a test period now, some people perha

Re: cvs commit: src/lib/libc/stdlib malloc.c

On Thu, Jan 26, 2006 at 09:15:02PM -0600, Christian S.J. Peron wrote: > These malloc debugging/stats features have been invaluable. I vote to > accept the performance hit in exchange for useful debugging/stats > features in -CURRENT, and disable them in -STABLE/-RELEASE when they are > ready. Y

Re: cvs commit: src/lib/libc/stdlib malloc.c

On Thu, Jan 26, 2006 at 07:02:17PM -0800, Jason Evans wrote: > Unfortunately, run-time checks for the debugging features are > expensive, since the debugging code is scattered throughout > malloc.c. This pretty much mandates compile-time configuration, for > performance reasons. Checking on

Re: cvs commit: src/lib/libc/stdlib malloc.c

Jason Evans wrote: On Jan 26, 2006, at 6:51 PM, Andrey Chernov wrote: I worry about my system performance. Could you please make all statistics and debug code conditional, debending on some flags which could be set as malloc options? Even in case this is a test period now, some people perh

Re: cvs commit: src/lib/libc/stdlib malloc.c

On Jan 26, 2006, at 6:51 PM, Andrey Chernov wrote: I worry about my system performance. Could you please make all statistics and debug code conditional, debending on some flags which could be set as malloc options? Even in case this is a test period now, some people perhaps don't want some te

Re: cvs commit: src/lib/libc/stdlib malloc.c

On Fri, Jan 27, 2006 at 02:36:44AM +, Jason Evans wrote: > jasone 2006-01-27 02:36:44 UTC > > FreeBSD src repository > > Modified files: > lib/libc/stdlib malloc.c > Log: > Clean up statistics gathering and printing. I worry about my system performance. Could you pleas

Re: cvs commit: src/lib/libc/stdlib malloc.c

* Jason Evans <[EMAIL PROTECTED]> [ Date: 2006-01-19 ] [ w.r.t. cvs commit: src/lib/libc/stdlib malloc.c ] > jasone 2006-01-20 03:11:11 UTC > > FreeBSD src repository > > Modified files: > lib/libc/stdlib malloc.c > Log: > Use uintptr_t rather than size_t when casti

Re: cvs commit: src/lib/libc/stdlib malloc.c

On Jan 19, 2006, at 11:20 AM, Jason Evans wrote: jasone 2006-01-19 19:20:43 UTC FreeBSD src repository Modified files: lib/libc/stdlib malloc.c Log: Revert addtion of assertions in revision 1.99. These assertions cause problems in cases where regions are faked up for

Re: cvs commit: src/lib/libc/stdlib malloc.c

On 2006-01-15 21:04, Jason Evans <[EMAIL PROTECTED]> wrote: > On Jan 14, 2006, at 1:32 PM, Giorgos Keramidas wrote: > >On 2006-01-13 11:07, Jason Evans <[EMAIL PROTECTED]> wrote: > >>On Jan 12, 2006, at 5:29 PM, Giorgos Keramidas wrote: > >>> > >>>[...] > >>> > >>>Does this look like an off-by-one

Re: cvs commit: src/lib/libc/stdlib malloc.c

On Jan 15, 2006, at 9:13 PM, Jason Evans wrote: jasone 2006-01-16 05:13:49 UTC FreeBSD src repository Modified files: lib/libc/stdlib malloc.c Log: [...] Allow chunks to be as small as the page size. This change allows us to reduce the chunk size to the point that a

Re: cvs commit: src/lib/libc/stdlib malloc.c

On Jan 14, 2006, at 1:32 PM, Giorgos Keramidas wrote: On 2006-01-13 11:07, Jason Evans <[EMAIL PROTECTED]> wrote: On Jan 12, 2006, at 5:29 PM, Giorgos Keramidas wrote: [...] Does this look like an off-by-one error to you too Jason? Apparently, the allocated size of s->data is s->size, which i

Re: cvs commit: src/lib/libc/stdlib malloc.c

On 2006-01-15 01:05, Jason Evans <[EMAIL PROTECTED]> wrote: >On Jan 14, 2006, at 7:28 PM, Giorgos Keramidas wrote: >> >> Jason, >> >> is this related to the malloc changes in any way. I'm curious >> why the default return type of `int' wasn't a problem so far. >> Has the definition of userland poi

Re: cvs commit: src/lib/libc/stdlib malloc.c

On Sun, 15 Jan 2006, Jason Evans wrote: On amd64, jemalloc uses mmap() to get chunks of memory to carve up. It's possible that these chunks are above 4 GB, which means that the high bits are important, but sizeof(int) is 4, not large enough to store such a pointer. With sbrk(), the addresses

Re: cvs commit: src/lib/libc/stdlib malloc.c

On Jan 14, 2006, at 7:28 PM, Giorgos Keramidas wrote: Jason, is this related to the malloc changes in any way. I'm curious why the default return type of `int' wasn't a problem so far. Has the definition of userland pointers changed recently from a type that could fit in an `int' to someth

Re: cvs commit: src/lib/libc/stdlib malloc.c

On 2006-01-14 19:13, Pascal Hofstee <[EMAIL PROTECTED]> wrote: > Giorgos Keramidas wrote: > >On 2006-01-14 14:31, Pascal Hofstee <[EMAIL PROTECTED]> wrote: > > > >Hmmm, this is very interesting. I see that irssi links to glib-2.8.5 > >here, but where did you get a port for glib of a newer version ?

Re: cvs commit: src/lib/libc/stdlib malloc.c

Giorgos Keramidas wrote: On 2006-01-14 14:31, Pascal Hofstee <[EMAIL PROTECTED]> wrote: Hmmm, this is very interesting. I see that irssi links to glib-2.8.5 here, but where did you get a port for glib of a newer version ? By using the marcusmerge script available at http://www.marcuscom.com:

Re: cvs commit: src/lib/libc/stdlib malloc.c

On 2006-01-15 04:46, Giorgos Keramidas <[EMAIL PROTECTED]> wrote: > It seems strange that most of the addresses are in the 0x8+ > range and the adress that triggers the core dump is much smaller: > > (gdb) x/s 0x30e3100 > 0x30e3100: > > But the real string I typed when irssi crashed

Re: cvs commit: src/lib/libc/stdlib malloc.c

On 2006-01-14 14:13, Jason Evans <[EMAIL PROTECTED]> wrote: > >>If updating to either before or after the broken > >>posix_memalign() revision, and need help figuring out the > >>issue, please let me know. > > > >More issues come up after updating to today's CURRENT version of > >malloc. In partic

Re: cvs commit: src/lib/libc/stdlib malloc.c

On 2006-01-14 14:13, Jason Evans <[EMAIL PROTECTED]> wrote: > >Apparently it does seem related to posix_memalign() changes. > >When I bootstrap Emacs without posix_memalign(), by manually > >tweaking src/config.h after configure runs, and #undef > >POSIX_MEMALIGN, then it passes the bootstrap stage

Re: cvs commit: src/lib/libc/stdlib malloc.c

On 2006-01-14 14:31, Pascal Hofstee <[EMAIL PROTECTED]> wrote: > Jason Evans wrote: > >Hmm, that's interesting. I'll mess around with emacs later on today > >(though I only have i386 hardware at the moment). Right now, I'm > >still building xorg on my machine in order to try to reproduce the gtk

Re: cvs commit: src/lib/libc/stdlib malloc.c

Jason Evans wrote: On Jan 14, 2006, at 1:32 PM, Giorgos Keramidas wrote: On 2006-01-13 11:07, Jason Evans <[EMAIL PROTECTED]> wrote: On Jan 12, 2006, at 5:29 PM, Giorgos Keramidas wrote: Does it look like I'm right in thinking that this is a bug in Emacs? This looks like a bug in emacs, as y

Re: cvs commit: src/lib/libc/stdlib malloc.c

On Jan 14, 2006, at 1:32 PM, Giorgos Keramidas wrote: On 2006-01-13 11:07, Jason Evans <[EMAIL PROTECTED]> wrote: On Jan 12, 2006, at 5:29 PM, Giorgos Keramidas wrote: Does it look like I'm right in thinking that this is a bug in Emacs? This looks like a bug in emacs, as you say, but I don't

Re: cvs commit: src/lib/libc/stdlib malloc.c

On 2006-01-13 11:07, Jason Evans <[EMAIL PROTECTED]> wrote: >On Jan 12, 2006, at 5:29 PM, Giorgos Keramidas wrote: >> >> [...] >> >> Does this look like an off-by-one error to you too Jason? >> Apparently, the allocated size of s->data is s->size, which is 873 >> bytes, but then Emacs tries to acce

Re: cvs commit: src/lib/libc/stdlib malloc.c

On 2006-01-13 11:07, Jason Evans <[EMAIL PROTECTED]> wrote: >On Jan 12, 2006, at 5:29 PM, Giorgos Keramidas wrote: >> Does this look like an off-by-one error to you too Jason? >> Apparently, the >> allocated size of s->data is s->size, which is 873 bytes, but then >> Emacs tries >> to access s->dat

Re: cvs commit: src/lib/libc/stdlib malloc.c

On Jan 12, 2006, at 5:29 PM, Giorgos Keramidas wrote: [...] Does this look like an off-by-one error to you too Jason? Apparently, the allocated size of s->data is s->size, which is 873 bytes, but then Emacs tries to access s->data[873]. Does it look like I'm right in thinking that this i

Re: cvs commit: src/lib/libc/stdlib malloc.c

On 2006-01-12 20:28, Giorgos Keramidas <[EMAIL PROTECTED]> wrote: >On 2006-01-12 10:17, Jason Evans <[EMAIL PROTECTED]> wrote: >>> Fix a bitwise logic error in posix_memalign(). >>> >>> Reported by:glebius >>> >>> Revision ChangesPath >>> 1.92 +2 -2 src/lib/libc/stdlib/malloc

Re: cvs commit: src/lib/libc/stdlib malloc.c

On 2006-01-12 10:17, Jason Evans <[EMAIL PROTECTED]> wrote: >On Jan 12, 2006, at 10:09 AM, Jason Evans wrote: >>jasone 2006-01-12 18:09:26 UTC >> >> FreeBSD src repository >> >> Modified files: >>lib/libc/stdlib malloc.c >> Log: >> Fix a bitwise logic error in posix_memalign(). >>

Re: cvs commit: src/lib/libc/stdlib malloc.c

On Jan 12, 2006, at 10:09 AM, Jason Evans wrote: jasone 2006-01-12 18:09:26 UTC FreeBSD src repository Modified files: lib/libc/stdlib malloc.c Log: Fix a bitwise logic error in posix_memalign(). Reported by:glebius Revision ChangesPath 1.92 +2 -2