Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

rebase? #1

Merged
merged 348 commits into from
Jul 7, 2019
Merged

rebase? #1

merged 348 commits into from
Jul 7, 2019

Conversation

TiborGY
Copy link
Owner

@TiborGY TiborGY commented Jul 7, 2019

No description provided.

martin-frbg and others added 30 commits February 14, 2019 22:57
icc has known problems with mixed-language builds that ifort can handle just fine. Fixes #1956
With the Intel compiler on Linux, prefer ifort for the final link step
Fix unannounced modification of input operand 8 (lda4) in Haswell GEMVN microkernel
Fix wrong constraints in inline assembly of Haswell DTRSM kernel
rework indices to allow marking argument lda4 as input and output. For #2009
rework indices to allow marking argument lda as input and output.
rework indices to allow marking i,as and bs as both input and output (marked operand n1 as well for simplicity). For #2009
Reword/expand comments in Makefile.rule
Fix inline assembly constraints in various x86_64 GEMVN kernels
Fix inline assembly constraints in Bulldozer TRSM kernels
same bug as seen in #1388, somehow missed in corresponding PR #1389
revert workaround introduced for issue #1332 as the actual cause appears to be my incorrect fix from #1262 (see #1388)
to make compilation complete reliably within the 1h limit again
Move one of clobber fixes to right place
as xcode10 deprecated i386
Make x86_32 imply NO_AVX2, NO_AVX512 in addition to NO_AVX
Make sure that AVX512 is disabled in 32bit builds
to avoid surprises from defining either as zero. Fixes #2035 by addressing some concerns from #1422
martin-frbg and others added 23 commits June 10, 2019 19:12
Avoid unintentional activation of TLS codepath via USE_TLS=0
fixes compile failure with pgi 18.10 as reported on OpenBLAS-users
PGI assembler does not like the initialization in the constraints.
Fix DTRMMKERNEL register save for power8 64-bit mode (Fix for #2166)
Fix build on FreeBSD/powerpc64.
FreeBSD needs DCBT_ARG=0 as well.
FreeBSD needs those macros too.
Fix build on PPC970 for FreeBSD
power9 cgemm/ctrmm. new sgemm 8x16
(and other small fixes to match Makefile behaviour)
needs to be unset in the cache for the change to have any effect
(also copy BUNDERSCORE settings for FORTRAN from the corresponding Makefile)
Fix CMAKE compilation with MinGW32 and add it to Appveyor
@TiborGY TiborGY closed this Jul 7, 2019
@TiborGY TiborGY reopened this Jul 7, 2019
@TiborGY TiborGY merged commit 2a669e7 into TiborGY:patch-1 Jul 7, 2019
TiborGY added a commit that referenced this pull request Jul 7, 2019
This reverts commit 2a669e7.
@TiborGY TiborGY mentioned this pull request Jul 7, 2019
TiborGY added a commit that referenced this pull request Jul 7, 2019
This reverts commit 2a669e7.
TiborGY pushed a commit that referenced this pull request Jul 7, 2019
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.