zhangyiss/spack
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Zlib gets a 50-60% performance win when compiled with -O2 (#5838)

Browse Source 
Chasing a performance regression has lead me to this change, going from default optimization gives a significant performance win. The sweet spot for zlib is apparently `-O2`, both `-Ofast` and `-O3` are slightly worse (regression is about 3% compared with `-O2) in my testing. 

Happy to share my methodology with people so we can benchmark on a wider variety of systems.
This commit is contained in:
David Hows 2017-10-25 22:53:35 +11:00 committed by Todd Gamblin
parent ce8a698b48
commit 05b3178ca4
1 changed files with 4 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
var/spack/repos/builtin/packages/zlib/package.py
View File

@ -45,6 +45,8 @@ class Zlib(Package):
description='Produce position-independent code (for shared libs)') description='Produce position-independent code (for shared libs)')
variant('shared', default=True, variant('shared', default=True,
description='Enables the build of shared libraries.') description='Enables the build of shared libraries.')
variant('optimize', default=True,
description='Enable -O2 for a more optimized lib')
patch('w_patch.patch', when="@1.2.11%cce") patch('w_patch.patch', when="@1.2.11%cce")
@ -58,6 +60,8 @@ def libs(self):
def setup_environment(self, spack_env, run_env): def setup_environment(self, spack_env, run_env):
if '+pic' in self.spec: if '+pic' in self.spec:
spack_env.append_flags('CFLAGS', self.compiler.pic_flag) spack_env.append_flags('CFLAGS', self.compiler.pic_flag)
if '+optimize' in self.spec:
spack_env.append_flags('CFLAGS', '-O2')
def install(self, spec, prefix): def install(self, spec, prefix):
config_args = [] config_args = []