This bugzilla service is closed. All entries have been migrated to https://gitlab.com/libeigen/eigen

Bug 1637

Summary:	Register spilling with recent gcc & clang
Product:	Eigen	Reporter:	Gael Guennebaud <gael.guennebaud>
Component:	Core - matrix products	Assignee:	Nobody <eigen.nobody>
Status:	RESOLVED FIXED
Severity:	Performance Problem	CC:	chtz, gael.guennebaud
Priority:	Normal
Version:	unspecified
Hardware:	All
OS:	All
Whiteboard:
Bug Depends on:
Bug Blocks:	814

Description Gael Guennebaud 2018-12-06 17:08:57 UTC

We have some performance regressions with recent compiler versions of both clang and gcc. The problem is bad register allocation leading to spilling. More precisely, here is what I found:

With AVX (no FMA):
- gcc 5:    OK
- gcc >= 6: spilling in 2pX4 kernel
- clang:    OK

With AVX+FMA:
- gcc: OK
- clang 5: OK
- clang >=6: spilling in 3pX4 kernel

With AVX512f(+FMA)
- gcc: ok
- clang: ok

Comment 1 Gael Guennebaud 2018-12-06 22:34:31 UTC

For gcc, I managed to trick it with more aggressive asm comments to isolate each "EIGEN_GEBP_ONESTEP":

asm("" : [a0] "+x" (A0), [a1] "+x" (A1) ); \


Such a trick does not work for clang.

Comment 2 Gael Guennebaud 2018-12-07 08:26:57 UTC

For gcc:
https://bitbucket.org/eigen/eigen/commits/9f52fde0348

Comment 3 Gael Guennebaud 2018-12-07 09:06:16 UTC

for clang:
https://bitbucket.org/eigen/eigen/commits/40e26d3f60fb/

I backported the two workaround to 3.3

Comment 4 Nobody 2019-12-04 18:13:53 UTC

-- GitLab Migration Automatic Message --

This bug has been migrated to gitlab.com's GitLab instance and has been closed from further activity.

You can subscribe and participate further through the new bug through this link to our GitLab instance: https://gitlab.com/libeigen/eigen/issues/1637.