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

Builds should ignore/filter march and mtune flags? #2143

Closed
gdevenyi opened this issue May 29, 2019 · 14 comments
Closed

Builds should ignore/filter march and mtune flags? #2143

gdevenyi opened this issue May 29, 2019 · 14 comments
Milestone

Comments

@gdevenyi
Copy link

If I'm building a DYNAMIC_ARCH build, many different files are compiled with different arch flags based on their targets, however if I specify march in CFLAGS during configure, I end up with that in the flags, and when I get to a kernel that uses features my hardware doesn't support, the compile fails.

I ran into this deep inside a cmake superbuild where I want the rest of the code to use -march but cmake passes that onwards into the built-in openblas integration and breaks the build.

@brada4
Copy link
Contributor

brada4 commented May 29, 2019

The final CFLAGS/FFLAGS parameters are combined per-makefile from CCOMMON_OPT FCOMMON_OPT and COMMON_OPT flags. OpenBLAS will not build with CFLAGS set.

Might be that OpenBLAS is first in your build to do some testing, showing invalid CFLAGS, and rest of superbuild does not work either.

@martin-frbg
Copy link
Collaborator

I doubt there is an "any size fits all" solution to this - in a DYNAMIC_ARCH build, the CFLAGS will (mainly) determine what the common code will be built with, and there are probably several legitimate reasons for wanting to influence that.

@gdevenyi
Copy link
Author

@brada4 Our superbuild uses the CMAKE configuration, which accepts, and passes through CFLAGS/CXXFLAGS, maybe this is not as intended?

The CFLAGS are not invalid, this is simply -march=native, which breaks the DYNAMIC_ARCH build for CPUs newer than the local system CPU. I build the superbuild in parallel, and everything else works fine (up until OpenBLAS dependent things anyways).

Maybe this is specific to the cmake configuration which I should've mentioned to start...

@martin-frbg
Copy link
Collaborator

Can you be a bit more specific about the instruction subset that your build host does not support, and that fail to compile only with the -march=native ? (Is it AVX512 ?)

@brada4
Copy link
Contributor

brada4 commented May 29, 2019

OpenBLAS build uses variable names CFLAGS and FFLAGS internally. What you set before it interferes with normal operation. In addition it alters those to build included reference LAPACK.
Just unset them mmmmOK?

@gdevenyi
Copy link
Author

Yes @martin-frbg in my case its the AVX512 versions failing on my system because -march=native doesn't support it. Purging this setting allows those to build.

...thanks @brada4. That's the bug I'm reporting.

@martin-frbg
Copy link
Collaborator

cmake/system.cmake should already add -march=skylake-avx512 as needed - however no attempt is made at removing any pre-existing -march flags (is there a CMAKE command for this ?) and I am not sure how the compiler would react to seeing both "-march=native -march=skylake-avx512"

@gdevenyi
Copy link
Author

I guess based on my experience here, it looks like -march=native overrides the other arch settings:

> gcc -march=native -march=skylake-avx512 -E -v - </dev/null 2>&1 | grep cc1
 /usr/lib/gcc/x86_64-linux-gnu/8/cc1 -E -quiet -v -imultiarch x86_64-linux-gnu - -march=broadwell -mmmx -mno-3dnow -msse -msse2 -msse3 -mssse3 -mno-sse4a -mcx16 -msahf -mmovbe -maes -mno-sha -mpclmul -mpopcnt -mabm -mno-lwp -mfma -mno-fma4 -mno-xop -mbmi -mno-sgx -mbmi2 -mno-pconfig -mno-wbnoinvd -mno-tbm -mavx -mavx2 -msse4.2 -msse4.1 -mlzcnt -mrtm -mhle -mrdrnd -mf16c -mfsgsbase -mrdseed -mprfchw -madx -mfxsr -mxsave -mxsaveopt -mno-avx512f -mno-avx512er -mno-avx512cd -mno-avx512pf -mno-prefetchwt1 -mno-clflushopt -mno-xsavec -mno-xsaves -mno-avx512dq -mno-avx512bw -mno-avx512vl -mno-avx512ifma -mno-avx512vbmi -mno-avx5124fmaps -mno-avx5124vnniw -mno-clwb -mno-mwaitx -mno-clzero -mno-pku -mno-rdpid -mno-gfni -mno-shstk -mno-avx512vbmi2 -mno-avx512vnni -mno-vaes -mno-vpclmulqdq -mno-avx512bitalg -mno-movdiri -mno-movdir64b --param l1-cache-size=32 --param l1-cache-line-size=64 --param l2-cache-size=25600 -mtune=broadwell -fstack-protector-strong -Wformat -Wformat-security

^^ I see a bunch of "no AVX" settings which is where I'm running into the compiler failure.

cmake doesn't have a standardized way to remove items from a variable but maybe a regex works. I'm looking for examples.

@gdevenyi
Copy link
Author

gdevenyi commented May 30, 2019

Looks like a REGEX replace on the C/CXX flags can remove, something like:
https://cmake.org/cmake/help/latest/command/string.html#regular-expressions

string(REGEX REPLACE "-march=[A-Za-z0-9_]+"
       "" CMAKE_C_FLAGS_RELEASE
       "${CMAKE_C_FLAGS_RELEASE}")

^Not tested yet.

@martin-frbg
Copy link
Collaborator

From what I saw it does not appear to be that easy from "within" a cmake project. The general recommendation appears to be to either append a corresponding -no-something flag if available (which is obviously not the case here) or to move the "special" files to a separate project and be sure to adjust the CFLAGS before building it (which trivially circumvents the issue).

@martin-frbg
Copy link
Collaborator

Hmm. Stripping out the "-march=native" like you suggested does "sort of" work, but only if I replace it very early on, e.g. in the "if (NOT NO_AVX512)" part of the DYNAMIC_ARCH option handling in arch.cmake where it will affect all parts of the build. When I try to limit it to only the SkylakeX kernel part, no matter if I try to do it in system.cmake or in kernel/CMakeLists.txt, I still end up with the original "C_FLAGS=-march=native" in kernel/CMakeFiles/kernel.SKYLAKEX.dir/flags.make

@gdevenyi
Copy link
Author

gdevenyi commented Jun 9, 2019

My understanding of dynamic arch is that it should be stripped from all parts of the build, otherwise the build isn't really dynamic anyway. So this sounds good to me.

TiborGY added a commit to TiborGY/OpenBLAS that referenced this issue Jul 7, 2019
* With the Intel compiler on Linux, prefer ifort for the final link step 

icc has known problems with mixed-language builds that ifort can handle just fine. Fixes OpenMathLib#1956

* Rename operands to put lda on the input/output constraint list

* Fix wrong constraints in inline assembly

for OpenMathLib#2009

* Fix inline assembly constraints

rework indices to allow marking argument lda4 as input and output. For OpenMathLib#2009

* Fix inline assembly constraints

rework indices to allow marking argument lda as input and output.

* Fix inline assembly constraints

* Fix inline assembly constraints

* Fix inline assembly constraints in Bulldozer TRSM kernels

rework indices to allow marking i,as and bs as both input and output (marked operand n1 as well for simplicity). For OpenMathLib#2009

* Correct range_n limiting

same bug as seen in OpenMathLib#1388, somehow missed in corresponding PR OpenMathLib#1389

* Allow multithreading TRMV again

revert workaround introduced for issue OpenMathLib#1332 as the actual cause appears to be my incorrect fix from OpenMathLib#1262 (see OpenMathLib#1388)

* Fix error introduced during cleanup

* Reduce list of kernels in the dynamic arch build

to make compilation complete reliably within the 1h limit again

* init

* move fix to right place

* Fix missing -c option in AVX512 test

* Fix AVX512 test always returning false due to missing compiler option

* Make x86_32 imply NO_AVX2, NO_AVX512 in addition to NO_AVX

fixes OpenMathLib#2033

* Keep xcode8.3 for osx BINARY=32 build

as xcode10 deprecated i386

* Make sure that AVX512 is disabled in 32bit builds

for OpenMathLib#2033

* Improve handling of NO_STATIC and NO_SHARED

to avoid surprises from defining either as zero. Fixes OpenMathLib#2035 by addressing some concerns from OpenMathLib#1422

* init

* address warning introed with OpenMathLib#1814 et al

* Restore locking optimizations for OpenMP case

restore another accidentally dropped part of OpenMathLib#1468 that was missed in OpenMathLib#2004 to address performance regression reported in OpenMathLib#1461

* HiSilicon tsv110 CPUs optimization branch

add HiSilicon tsv110 CPUs  optimization branch

* add TARGET support for  HiSilicon tsv110 CPUs

* add TARGET support for HiSilicon tsv110 CPUs

* add TARGET support for HiSilicon tsv110 CPUs

* Fix module definition conflicts between LAPACK and ReLAPACK

for OpenMathLib#2043

* Do not compile in AVX512 check if AVX support is disabled

xgetbv is function depends on NO_AVX being undefined - we could change that too, but that combo is unlikely to work anyway

* ctest.c : add __POWERPC__ for PowerMac

* Fix crash in sgemm SSE/nano kernel on x86_64

Fix bug OpenMathLib#2047.

Signed-off-by: Celelibi <[email protected]>

* param.h : enable defines for PPC970 on DarwinOS

fixes:
gemm.c: In function 'sgemm_':
../common_param.h:981:18: error: 'SGEMM_DEFAULT_P' undeclared (first use in this function)
 #define SGEMM_P  SGEMM_DEFAULT_P
                  ^

* common_power.h: force DCBT_ARG 0 on PPC970 Darwin

without this, we see
../kernel/power/gemv_n.S:427:Parameter syntax error
and many more similar entries

that relates to this assembly command
dcbt 8, r24, r18

this change makes the DCBT_ARG = 0
and openblas builds through to completion on PowerMac 970
Tests pass

* Make TARGET=GENERIC compatible with DYNAMIC_ARCH=1

for issue OpenMathLib#2048

* make DYNAMIC_ARCH=1 package work on TSV110.

* make DYNAMIC_ARCH=1 package work on TSV110

* Add Intel Denverton

for OpenMathLib#2048

* Add Intel Denverton

* Change 64-bit detection as explained in OpenMathLib#2056

* Trivial typo fix

as suggested in OpenMathLib#2022

* Disable the AVX512 DGEMM kernel (again)

Due to as yet unresolved errors seen in OpenMathLib#1955 and OpenMathLib#2029

* Use POSIX getenv on Cygwin

The Windows-native GetEnvironmentVariable cannot be relied on, as
Cygwin does not always copy environment variables set through Cygwin
to the Windows environment block, particularly after fork().

* Fix for OpenMathLib#2063: The DllMain used in Cygwin did not run the thread memory
pool cleanup upon THREAD_DETACH which is needed when compiled with
USE_TLS=1.

* Also call CloseHandle on each thread, as well as on the event so as to not leak thread handles.

* AIX asm syntax changes needed for shared object creation

* power9 makefile. dgemm based on power8 kernel with following changes : 32x unrolled 16x4 kernel and 8x4 kernel using (lxv stxv butterfly rank1 update). improvement from 17 to 22-23gflops. dtrmm cases were added into dgemm itself

* Expose CBLAS interfaces for I?MIN and I?MAX

* Build CBLAS interfaces for I?MIN and I?MAX

* Add declarations for ?sum and cblas_?sum

* Add interface for ?sum (derived from ?asum)

* Add ?sum

* Add implementations of ssum/dsum and csum/zsum

as trivial copies of asum/zsasum with the fabs calls replaced by fmov to preserve code structure

* Add ARM implementations of ?sum

(trivial copies of the respective ?asum with the fabs calls removed)

* Add ARM64 implementations of ?sum

as trivial copies of the respective ?asum kernels with the fabs calls removed

* Add ia64 implementation of ?sum

as trivial copy of asum with the fabs calls removed

* Add MIPS implementation of ?sum

as trivial copy of ?asum with the fabs calls removed

* Add MIPS64 implementation of ?sum

as trivial copy of ?asum with the fabs replaced by mov to preserve code structure

* Add POWER implementation of ?sum

as trivial copy of ?asum with the fabs replaced by fmr to preserve code structure

* Add SPARC implementation of ?sum

as trivial copy of ?asum with the fabs replaced by fmov to preserve code structure

* Add x86 implementation of ?sum

as trivial copy of ?asum with the fabs calls removed

* Add x86_64 implementation of ?sum

as trivial copy of ?asum with the fabs calls removed

* Add ZARCH implementation of ?sum

as trivial copies of the respective ?asum kernels with the ABS and vflpsb calls removed

* Detect 32bit environment on 64bit ARM hardware

for OpenMathLib#2056, using same approach as OpenMathLib#2058

* Add cmake defaults for ?sum kernels

* Add ?sum

* Add ?sum definitions for generic kernel

* Add declarations for ?sum

* Add -lm and disable EXPRECISION support on *BSD

fixes OpenMathLib#2075

* Add in runtime CPU detection for POWER.

* snprintf define consolidated to common.h

* Support INTERFACE64=1

* Add support for INTERFACE64 and fix XERBLA calls

1. Replaced all instances of "int" with "blasint"
2. Added string length as "hidden" third parameter in calls to fortran XERBLA

* Correct length of name string in xerbla call

* Avoid out-of-bounds accesses in LAPACK EIG tests

see Reference-LAPACK/lapack#333

* Correct INFO=4 condition

* Disable reallocation of work array in xSYTRF

as it appears to cause memory management problems (seen in the LAPACK tests)

* Disable repeated recursion on Ab_BR in ReLAPACK xGBTRF

due to crashes in LAPACK tests

* sgemm/strmm

* Update Changelog with changes from 0.3.6

* Increment version to 0.3.7.dev

* Increment version to 0.3.7.dev

* Misc. typo fixes

Found via `codespell -q 3 -w -L ith,als,dum,nd,amin,nto,wis,ba -S ./relapack,./kernel,./lapack-netlib`

* Correct argument of CPU_ISSET for glibc <2.5

fixes OpenMathLib#2104

* conflict resolve

* Revert reference/ fixes

* Revert Changelog.txt typos

* Disable the SkyLakeX DGEMMITCOPY kernel as well

as a stopgap measure for numpy/numpy#13401 as mentioned in OpenMathLib#1955

* Disable DGEMMINCOPY as well for now

OpenMathLib#1955

* init

* Fix errors in cpu enumeration with glibc 2.6

for OpenMathLib#2114

* Change two http links to https

Closes OpenMathLib#2109

* remove redundant code OpenMathLib#2113

* Set up CI with Azure Pipelines

[skip ci]

* TST: add native POWER8 to CI

* add native POWER8 testing to
Travis CI matrix with ppc64le
os entry

* Update link to IBM MASS library, update cpu support status

* first try migrating one of the arm builds from travis

* fix tabbing in azure commands

* Update azure-pipelines.yml

take out offending lines (although stolen from https://github.com/conda-forge/opencv-feedstock azure-pipelines fiie)

* Update azure-pipelines.yml

* Update azure-pipelines.yml

* Update azure-pipelines.yml

* Update azure-pipelines.yml

* DOC: Add Azure CI status badge

* Add ARMV6 build to azure CI setup (OpenMathLib#2122)

using aytekinar's Alpine image and docker script from the Travis setup

[skip ci]

* TST: Azure manylinux1 & clean-up

* remove some of the steps & comments
from the original Azure yml template

* modify the trigger section to use
develop since OpenBLAS primarily uses
this branch; use the same batching
behavior as downstream projects NumPy/
SciPy

* remove Travis emulated ARMv6 gcc build
because this now happens in Azure

* use documented Ubuntu vmImage name for Azure
and add in a manylinux1 test run to the matrix

[skip appveyor]

* Add NO_AFFINITY to available options on Linux, and set it to ON

to match the gmake default. Fixes second part of OpenMathLib#2114

* Replace ISMIN and ISAMIN kernels on all x86_64 platforms (OpenMathLib#2125)

* Mark iamax_sse.S as unsuitable for MIN due to issue OpenMathLib#2116
* Use iamax.S rather than iamax_sse.S for ISMIN/ISAMIN on all x86_64 as workaround for OpenMathLib#2116

* Move ARMv8 gcc build from Travis to Azure

* Move ARMv8 gcc build from Travis to Azure

* Update .travis.yml

* Test drone CI

* install make

* remove sudo

* Install gcc

* Install perl

* Install gfortran and add a clang job

* gfortran->gcc-gfortran

* Switch to ubuntu and parallel jobs

* apt update

* Fix typo

* update yes

* no need of gcc in clang build

* Add a cmake build as well

* Add cmake builds and print options

* build without lapack on cmake

* parallel build

* See if ubuntu 19.04 fixes the ICE

* Remove qemu armv8 builds

* arm32 build

* Fix typo

* TST: add SkylakeX AVX512 CI test

* adapt the C-level reproducer code for some
recent SkylakeX AVX512 kernel issues, provided
by Isuru Fernando and modified by Martin Kroeker,
for usage in the utest suite

* add an Intel SDE SkylakeX emulation utest run to
the Azure CI matrix; a custom Docker build was required
because Ubuntu image provided by Azure does not support
AVX512VL instructions

* Add option USE_LOCKING for single-threaded build with locking support

for calling from concurrent threads

* Add option USE_LOCKING for single-threaded build with locking support

* Add option USE_LOCKING for SMP-like locking in USE_THREAD=0 builds

* Add option USE_LOCKING but keep default settings intact

* Remove unrelated change

* Do not try ancient PGI hacks with recent versions of that compiler

should fix OpenMathLib#2139

* Build and run utests in any case, they do their own checks for fortran availability

* Add softfp support in min/max kernels

fix for OpenMathLib#1912

* Revert "Add softfp support in min/max kernels"

* Separate implementations of AMAX and IAMAX on arm

As noted in OpenMathLib#1912 and comment on OpenMathLib#1942, the combined implementation happens to "do the right thing" on hardfp, but cannot return both value and index on softfp where they would have to share the return register

* Ensure correct output for DAMAX with softfp

* Use generic kernels for complex (I)AMAX to support softfp

* improved zgemm power9 based on power8

* upload thread safety test folder

* hook up c++ thread safety test (main Makefile)

*  add c++ thread test option to Makefile.rule

* Document NO_AVX512 

for OpenMathLib#2151

* sgemm pipeline improved, zgemm rewritten without inner packs, ABI lxvx v20 fixed with vs52

* Fix detection of AVX512 capable compilers in getarch

21eda8b introduced a check in getarch.c to test if the compiler is capable of
AVX512. This check currently fails, since the used __AVX2__ macro is only
defined if getarch itself was compiled with AVX2/AVX512 support. Make sure this
is the case by building getarch with -march=native on x86_64. It is only
supposed to run on the build host anyway.

* c_check: Unlink correct file

* power9 zgemm ztrmm optimized

* conflict resolve

* Add gfortran workaround for ABI violations in LAPACKE

for OpenMathLib#2154 (see gcc bug 90329)

* Add gfortran workaround for ABI violations

for OpenMathLib#2154 (see gcc bug 90329)

* Add gfortran workaround for potential ABI violation 

for OpenMathLib#2154

* Update fc.cmake

* Remove any inadvertent use of -march=native from DYNAMIC_ARCH builds

from OpenMathLib#2143, -march=native precludes use of more specific options like -march=skylake-avx512 in individual kernels, and defeats the purpose of dynamic arch anyway.

* Avoid unintentional activation of TLS code via USE_TLS=0

fixes OpenMathLib#2149

* Do not force gcc options on non-gcc compilers

fixes compile failure with pgi 18.10 as reported on OpenBLAS-users

* Update Makefile.x86_64

* Zero ecx with a mov instruction

PGI assembler does not like the initialization in the constraints.

* Fix mov syntax

* new sgemm 8x16

* Update dtrmm_kernel_16x4_power8.S

* PGI compiler does not like -march=native

* Fix build on FreeBSD/powerpc64.

Signed-off-by: Piotr Kubaj <[email protected]>

* Fix build for PPC970 on FreeBSD pt. 1

FreeBSD needs DCBT_ARG=0 as well.

* Fix build for PPC970 on FreeBSD pt.2

FreeBSD needs those macros too.

* cgemm/ctrmm power9

* Utest needs CBLAS but not necessarily FORTRAN

* Add mingw builds to Appveyor config

* Add getarch flags to disable AVX on x86

(and other small fixes to match Makefile behaviour)

* Make disabling DYNAMIC_ARCH on unsupported systems work

needs to be unset in the cache for the change to have any effect

* Mingw32 needs leading underscore on object names

(also copy BUNDERSCORE settings for FORTRAN from the corresponding Makefile)
@martin-frbg martin-frbg added this to the 0.3.7 milestone Aug 13, 2019
@martin-frbg
Copy link
Collaborator

Closing as fixed in 0.3.7

@gdevenyi
Copy link
Author

Wonderful thanks!

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

No branches or pull requests

3 participants