On Fedora 31 with GCC 9.2.1, compiling qemuxml2argvtest takes
about 36 seconds since
commit 30c6d99209cd36bba8e14795622f658827698522
Author: Jiri Denemark <jdenemar@redhat.com>
Date: Thu Oct 24 17:51:42 2019 +0200
qemuxml2argvtest: Update host arch for DO_TEST*ARCH* tests
The optimizer is hitting some pathelogical performance behaviour due to
the high number of branches in the mymain() method.
Pushing the branch tests down into the testCompareXMLToArgv method
brings the compile time down to 3 seconds.
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
---
tests/qemuxml2argvtest.c | 10 ++++++----
1 file changed, 6 insertions(+), 4 deletions(-)
diff --git a/tests/qemuxml2argvtest.c b/tests/qemuxml2argvtest.c
index a2791d0460..86b8899921 100644
--- a/tests/qemuxml2argvtest.c
+++ b/tests/qemuxml2argvtest.c
@@ -412,6 +412,9 @@ testCompareXMLToArgv(const void *data)
size_t i;
qemuDomainObjPrivatePtr priv = NULL;
+ if (info->arch != VIR_ARCH_NONE && info->arch != VIR_ARCH_X86_64)
+ qemuTestSetHostArch(driver.caps, info->arch);
+
memset(&monitor_chr, 0, sizeof(monitor_chr));
if (!(conn = virGetConnect()))
@@ -580,6 +583,9 @@ testCompareXMLToArgv(const void *data)
virSetConnectStorage(NULL);
virObjectUnref(conn);
VIR_FREE(migrateURI);
+ if (info->arch != VIR_ARCH_NONE && info->arch != VIR_ARCH_X86_64)
+ qemuTestSetHostArch(driver.caps, VIR_ARCH_NONE);
+
return ret;
}
@@ -699,13 +705,9 @@ mymain(void)
__VA_ARGS__, ARG_END) < 0) \
return EXIT_FAILURE; \
testInfoSetPaths(&info, _suffix); \
- if (info.arch != VIR_ARCH_NONE && info.arch != VIR_ARCH_X86_64) \
- qemuTestSetHostArch(driver.caps, info.arch); \
if (virTestRun("QEMU XML-2-ARGV " _name _suffix, \
testCompareXMLToArgv, &info) < 0) \
ret = -1; \
- if (info.arch != VIR_ARCH_NONE && info.arch != VIR_ARCH_X86_64) \
- qemuTestSetHostArch(driver.caps, VIR_ARCH_NONE); \
testQemuInfoClear(&info); \
} while (0)
--
2.23.0
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
On 11/27/19 1:22 PM, Daniel P. Berrangé wrote: > On Fedora 31 with GCC 9.2.1, compiling qemuxml2argvtest takes > about 36 seconds since > > commit 30c6d99209cd36bba8e14795622f658827698522 > Author: Jiri Denemark <jdenemar@redhat.com> > Date: Thu Oct 24 17:51:42 2019 +0200 > > qemuxml2argvtest: Update host arch for DO_TEST*ARCH* tests > > The optimizer is hitting some pathelogical performance behaviour due to > the high number of branches in the mymain() method. > > Pushing the branch tests down into the testCompareXMLToArgv method > brings the compile time down to 3 seconds. > > Signed-off-by: Daniel P. Berrangé <berrange@redhat.com> > --- In my Fedora 31 box, 'make' time of qemuxml2argvtest.c with this patch: real 0m4.113s user 0m3.838s sys 0m0.360s Without it: real 0m35.099s user 0m34.573s sys 0m0.544s Reviewed-by: Daniel Henrique Barboza <danielhb413@gmail.com> -- libvir-list mailing list libvir-list@redhat.com https://www.redhat.com/mailman/listinfo/libvir-list
On Wed, Nov 27, 2019 at 04:22:42PM +0000, Daniel P. Berrangé wrote: >On Fedora 31 with GCC 9.2.1, compiling qemuxml2argvtest takes >about 36 seconds since > > commit 30c6d99209cd36bba8e14795622f658827698522 > Author: Jiri Denemark <jdenemar@redhat.com> > Date: Thu Oct 24 17:51:42 2019 +0200 > > qemuxml2argvtest: Update host arch for DO_TEST*ARCH* tests > >The optimizer is hitting some pathelogical performance behaviour due to *pathological >the high number of branches in the mymain() method. > Missing bugzilla link: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=58479 >Pushing the branch tests down into the testCompareXMLToArgv method >brings the compile time down to 3 seconds. > -fno-var-tracking-assignments brings it down even more for me Can we turn the flag off for tests? Jano >Signed-off-by: Daniel P. Berrangé <berrange@redhat.com> >--- > tests/qemuxml2argvtest.c | 10 ++++++---- > 1 file changed, 6 insertions(+), 4 deletions(-) > -- libvir-list mailing list libvir-list@redhat.com https://www.redhat.com/mailman/listinfo/libvir-list
On 11/27/19 8:38 PM, Ján Tomko wrote: > On Wed, Nov 27, 2019 at 04:22:42PM +0000, Daniel P. Berrangé wrote: >> On Fedora 31 with GCC 9.2.1, compiling qemuxml2argvtest takes >> about 36 seconds since >> >> commit 30c6d99209cd36bba8e14795622f658827698522 >> Author: Jiri Denemark <jdenemar@redhat.com> >> Date: Thu Oct 24 17:51:42 2019 +0200 >> >> qemuxml2argvtest: Update host arch for DO_TEST*ARCH* tests >> >> The optimizer is hitting some pathelogical performance behaviour due to > > *pathological > >> the high number of branches in the mymain() method. >> > > Missing bugzilla link: > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=58479 > >> Pushing the branch tests down into the testCompareXMLToArgv method >> brings the compile time down to 3 seconds. >> > > -fno-var-tracking-assignments brings it down even more for me > > Can we turn the flag off for tests? Don't that hurt us when trying to debug tests? Michal -- libvir-list mailing list libvir-list@redhat.com https://www.redhat.com/mailman/listinfo/libvir-list
On Thu, Nov 28, 2019 at 11:06:12AM +0100, Michal Privoznik wrote: > On 11/27/19 8:38 PM, Ján Tomko wrote: > > On Wed, Nov 27, 2019 at 04:22:42PM +0000, Daniel P. Berrangé wrote: > > > On Fedora 31 with GCC 9.2.1, compiling qemuxml2argvtest takes > > > about 36 seconds since > > > > > > commit 30c6d99209cd36bba8e14795622f658827698522 > > > Author: Jiri Denemark <jdenemar@redhat.com> > > > Date: Thu Oct 24 17:51:42 2019 +0200 > > > > > > qemuxml2argvtest: Update host arch for DO_TEST*ARCH* tests > > > > > > The optimizer is hitting some pathelogical performance behaviour due to > > > > *pathological > > > > > the high number of branches in the mymain() method. > > > > > > > Missing bugzilla link: > > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=58479 > > > > > Pushing the branch tests down into the testCompareXMLToArgv method > > > brings the compile time down to 3 seconds. > > > > > > > -fno-var-tracking-assignments brings it down even more for me > > > > Can we turn the flag off for tests? > > Don't that hurt us when trying to debug tests? That seems to be the implication from the info pages on this option Annotate assignments to user variables early in the compilation and attempt to carry the annotations over throughout the compilation all the way to the end, in an attempt to improve debug information while optimizing. Use of '-gdwarf-4' is recommended along with it. Even with this enabled by default it is quite painful debugging due to variables being optimized out, so I'm loathe to make it worse just to shave off a few seconds. Regards, Daniel -- |: https://berrange.com -o- https://www.flickr.com/photos/dberrange :| |: https://libvirt.org -o- https://fstop138.berrange.com :| |: https://entangle-photo.org -o- https://www.instagram.com/dberrange :| -- libvir-list mailing list libvir-list@redhat.com https://www.redhat.com/mailman/listinfo/libvir-list
© 2016 - 2024 Red Hat, Inc.