[issue17781] optimize compilation options

Zachary Ware added the comment: After 2.5 years without response, I think the answer is probably "no" :) -- resolution: -> fixed stage: -> resolved status: pending -> closed ___ Python tracker

[issue17781] optimize compilation options

Changes by R. David Murray : -- status: open -> pending ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe: https:/

[issue17781] optimize compilation options

Changes by R. David Murray : -- nosy: +r.david.murray status: pending -> open ___ Python tracker ___ ___ Python-bugs-list mailing list

[issue17781] optimize compilation options

Zachary Ware added the comment: PGO is available as `make profile-opt`, LTO has a patch in issue25702. Are there any other interesting options worth investigating? -- nosy: +zach.ware status: open -> pending ___ Python tracker

[issue17781] optimize compilation options

Alecsandru Patrascu added the comment: Hi, I added a dedicated issue just for LTO only when using GCC and CLANG (http://bugs.python.org/issue25702), that works well with PGO also. -- nosy: +alecsandru.patrascu ___ Python tracker

[issue17781] optimize compilation options

Terry J. Reedy added the comment: #24915 is about adding pgo and has a slew of patches. -- nosy: +terry.reedy ___ Python tracker ___ _

[issue17781] optimize compilation options

Matthias Klose added the comment: On 22.09.2015 12:31, Antoine Pitrou wrote: > Also note LTO can make compilation times much longer (it's the linking step > actually, which can take minutes). use -flto=jobserver -- ___ Python tracker

[issue17781] optimize compilation options

Antoine Pitrou added the comment: > Hum, does it make sense to enable LTO without PGO? Probably not. By the way, I now have a small ARM system to play with, and there the gain of LTO+PGO over PGO alone is around 10%. Also note LTO can make compilation times much longer (it's the linking step

[issue17781] optimize compilation options

STINNER Victor added the comment: > For the record, the gain for LTO+PGO (with "-flto -O3") over PGO alone seems > to be between 0% and 10% for most benchmarks. Hum, does it make sense to enable LTO without PGO? -- nosy: +haypo ___ Python tracker <

[issue17781] optimize compilation options

Antoine Pitrou added the comment: You can test for yourself by passing `CFLAGS="-flto -O3" LDFLAGS="-flto -O3"` to ./configure (using gcc). -- ___ Python tracker ___ ___

[issue17781] optimize compilation options

Antoine Pitrou added the comment: For the record, the gain for LTO+PGO (with "-flto -O3") over PGO alone seems to be between 0% and 10% for most benchmarks. -- ___ Python tracker __

[issue17781] optimize compilation options

Antoine Pitrou added the comment: Note this patch is likely wrong, as it doesn't add the optimization options to the linker invocation. According to the gcc does, """To use the link-time optimizer, -flto and optimization options should be specified at compile time and during the final link""".

[issue17781] optimize compilation options

Raymond Hettinger added the comment: I would like the see LTO enabled. The intermodule calls to code in abstract.c would become less expensive. -- nosy: +rhettinger ___ Python tracker

[issue17781] optimize compilation options

Antoine Pitrou added the comment: LTO (Link-Time Optimization) is not the same as PGO, though I guess it can take advantage of PGO for its heuristics. -- ___ Python tracker ___

[issue17781] optimize compilation options

Stefan Behnel added the comment: Issue 24915 suggests PGO and comes with an actual patch. I suggest rejecting this ticket as too broad. -- nosy: +scoder ___ Python tracker ___ _

[issue17781] optimize compilation options

Changes by Stefan Krah : -- nosy: -skrah ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe: https://mail.python.o

[issue17781] optimize compilation options

Changes by Stefan Krah : -- nosy: +skrah ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe: https://mail.python.or

[issue17781] optimize compilation options

Matthias Klose added the comment: the proposed patch is wrong. when linking with -flto, you should pass all the relevant CFLAGS to the linker as well. Also pass -fuse-linker-plugin. and this should be an opt-in, not the default. Depending on the architecture and the compiler version, -flto is

[issue17781] optimize compilation options

Christian Heimes added the comment: Here is a patch for -flto. You need to run autoconf to re-generate configure, too. -- keywords: +patch nosy: +christian.heimes Added file: http://bugs.python.org/file29920/lto_flag.patch ___ Python tracker

[issue17781] optimize compilation options

Matthias Klose added the comment: most of that can be attributed to the pgo build, which is upstream for a long time. the second thing to do is to build with lto, and see what speedups you get in addition. and it certainly helps to build the interpreter statically (without --enable-shared). b

[issue17781] optimize compilation options

New submission from Antoine Pitrou: Ubuntu's system Python 3.3 shows consistently better performance than a vanilla Python 3.3: around 10-15% faster in general (see attached benchmark numbers). If this can be attributed to different compilation options, it would be nice to backport those optio