Bug 234605 - math/libpgmath: Fails to build: error: argument value 85 is outside the valid range [0, 15]
Summary: math/libpgmath: Fails to build: error: argument value 85 is outside the valid...
Status: Closed Overcome By Events
Alias: None
Product: Ports & Packages
Classification: Unclassified
Component: Individual Port(s) (show other bugs)
Version: Latest
Hardware: Any Any
: --- Affects Only Me
Assignee: Johannes M Dieterich
URL:
Keywords: needs-qa
Depends on:
Blocks:
 
Reported: 2019-01-04 07:19 UTC by Rainer Hurling
Modified: 2019-02-07 15:02 UTC (History)
2 users (show)

See Also:
bugzilla: maintainer-feedback? (jmd)
koobs: merge-quarterly?

Attachments
Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Rainer Hurling 2019-01-04 07:19:25 UTC 
When I try to build math/libpgmath on a recent FreeBSD !3.0-CURRENT amd64 with up to date ports tree, I get the following error:

[..snip..]
[1/31] /usr/bin/c++  -DLINUX -DLINUX86 -DLINUX8664 -DMAXCPUS=256 -DMAXCPUSL=8 -DMAXCPUSR=8 -DPG_PIC -DTARGET_LINUX -DTARGET_LINUX_X86 -DTARGET_LINUX_X8664 -DTARGET_X86 -DTARGET_X8664 -D__gnu_linux__ -I/usr/ports/math/libpgmath/work/flang-53e368b/runtime/libpgmath/lib/common -I/usr/ports/math/libpgmath/work/flang-53e368b/runtime/libpgmath/lib/x86_64 -I/usr/ports/math/libpgmath/work/flang-53e368b/runtime/libpgmath/lib/common/log/fma3 -O2 -pipe -fstack-protector -fno-strict-aliasing -O2 -pipe -fstack-protector -fno-strict-aliasing -fPIC   -m64 -O3 -mtune=core-avx2 -march=core-avx2 -MD -MT lib/common/log/fma3/CMakeFiles/common_log_fma3.dir/fdlog1.cpp.o -MF lib/common/log/fma3/CMakeFiles/common_log_fma3.dir/fdlog1.cpp.o.d -o lib/common/log/fma3/CMakeFiles/common_log_fma3.dir/fdlog1.cpp.o -c /usr/ports/math/libpgmath/work/flang-53e368b/runtime/libpgmath/lib/common/log/fma3/fdlog1.cpp
FAILED: lib/common/log/fma3/CMakeFiles/common_log_fma3.dir/fdlog1.cpp.o 
/usr/bin/c++  -DLINUX -DLINUX86 -DLINUX8664 -DMAXCPUS=256 -DMAXCPUSL=8 -DMAXCPUSR=8 -DPG_PIC -DTARGET_LINUX -DTARGET_LINUX_X86 -DTARGET_LINUX_X8664 -DTARGET_X86 -DTARGET_X8664 -D__gnu_linux__ -I/usr/ports/math/libpgmath/work/flang-53e368b/runtime/libpgmath/lib/common -I/usr/ports/math/libpgmath/work/flang-53e368b/runtime/libpgmath/lib/x86_64 -I/usr/ports/math/libpgmath/work/flang-53e368b/runtime/libpgmath/lib/common/log/fma3 -O2 -pipe -fstack-protector -fno-strict-aliasing -O2 -pipe -fstack-protector -fno-strict-aliasing -fPIC   -m64 -O3 -mtune=core-avx2 -march=core-avx2 -MD -MT lib/common/log/fma3/CMakeFiles/common_log_fma3.dir/fdlog1.cpp.o -MF lib/common/log/fma3/CMakeFiles/common_log_fma3.dir/fdlog1.cpp.o.d -o lib/common/log/fma3/CMakeFiles/common_log_fma3.dir/fdlog1.cpp.o -c /usr/ports/math/libpgmath/work/flang-53e368b/runtime/libpgmath/lib/common/log/fma3/fdlog1.cpp
/usr/ports/math/libpgmath/work/flang-53e368b/runtime/libpgmath/lib/common/log/fma3/fdlog1.cpp:39:9: error: argument value 85 is outside the valid range [0, 15] [-Wargument-outside-range]
    t = _mm_blend_epi32(INT2DBL_HI, t, H55_D); 
        ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
/usr/lib/clang/7.0.1/include/avx2intrin.h:766:12: note: expanded from macro '_mm_blend_epi32'
  (__m128i)__builtin_ia32_pblendd128((__v4si)(__m128i)(V1), \
           ^


I wasn't able to check, but possibly this is the reason for two newer patches[1][2] in libpgmath sources?


[1] https://github.com/flang-compiler/flang/commit/3660d02c4290abdeac1dca8b2ede71554033482d
[2] https://github.com/flang-compiler/flang/commit/76894366d6435780a7d1a1ae76382bb40c651ded
Comment 1 Johannes M Dieterich freebsd_committer 2019-02-06 02:43:33 UTC 
Can you confirm that this is still an issue w/ the latest port update? Thank you!
Comment 2 Rainer Hurling 2019-02-07 14:55:09 UTC 
(In reply to Johannes M Dieterich from comment #1)

I am very sorry I forgot to mention, that the latest update of math/libpgmath builds fine again on 13.0-CURRENT. Many thanks for the update and from my point of view, the PR can be closed now.
Comment 3 Walter Schwarzenfeld freebsd_triage 2019-02-07 15:02:22 UTC 
See Comment2. Close here, overcome by events.