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

Fixes in xxxxx for IEEE_DIVIDE_BY_ZERO FPE; separate cpu/gpu namespaces and fix runtest segfault #723

Merged
merged 155 commits into from
Jul 21, 2023
Merged
Changes from 1 commit
Commits
Show all changes
155 commits
Select commit Hold shift + click to select a range
8775b2d
[fpe] in ggtt.sa tests, add comments about how to run a single test i…
valassi Jul 14, 2023
ea07a20
[fpe] in ggtt.sa, add copyright header when dumping new reference fil…
valassi Jul 14, 2023
d75e426
[fpe] in ggttsa cudacpp.mk, try to debug #701 IEEE_DIVIDE_BY_ZERO (se…
valassi Jul 14, 2023
6fc09d8
[fpe] revert addition of -ffpe flags in ggttsa cudacpp.mk
valassi Jul 14, 2023
8efb726
[fpe] in ggtt.sa testxxx.cc, enable FPE floating point exception sign…
valassi Jul 14, 2023
22bd10d
[fpe] in ggtt.sa testxxx.cc, add a very simple signal handler for FPE…
valassi Jul 14, 2023
f64590a
[fpe] in ggtt.sa testxxx.cc, add some context information to the FPE …
valassi Jul 14, 2023
b35b772
[fpe] in ggtt.sa testxxx.cc, disable FPE if environment variable CUDA…
valassi Jul 17, 2023
f2598e0
[fpe] in ggtt.sa testxxx.cc, turn on event by event debugging
valassi Jul 14, 2023
46b6a0f
[fpe] in ggtt.sa testxxx.cc, cleanup (remove an unnecessary reinterpr…
valassi Jul 14, 2023
5c1470a
[fpe] in ggtt.sa testxxx.cc, add prepareTest (and improve FPE signal …
valassi Jul 14, 2023
7674824
[fpe] in ggtt.sa HelAmps_sm.h, first (OLD!) attempt of BUG FIX FOR #7…
valassi Jul 14, 2023
fdacc5e
[fpe] in ggtt.sa HelAmps_sm.h, add some debugging printouts for ixxxxx
valassi Jul 17, 2023
0372e2e
[fpe] revert the last two changes in ggtt.sa HelAmps_sm.h ixxxxx, wil…
valassi Jul 17, 2023
f607906
[fpe] in ggtt.sa mgOnGpuVectors.h, add maskand function
valassi Jul 17, 2023
7d1336b
[fpe] in ggtt.sa testmisc.ss, remove maskand function as it exists in…
valassi Jul 17, 2023
b0fca94
[fpe] in ggtt.sa testxxx.cc, add more debugging printouts
valassi Jul 14, 2023
8745a98
[fpe] in ggtt.sa mgOnGpuVectors.h, add constructor "cxtype_v( const f…
valassi Jul 17, 2023
68d787f
[fpe] in ggtt.sa HelAmps_sm.h, new BUG FIX FOR #701 in function ixxxxx
valassi Jul 17, 2023
7e91a0e
[fpe] in ggtt.sa testxxx.cc, reenable resetHstMomentaToPar0 at the be…
valassi Jul 17, 2023
de6492c
[fpe] in ggtt.sa testxxx.cc, ensure that ipzxxx handles SIMD vectors …
valassi Jul 14, 2023
18dd262
[fpe] in ggtt.sa HelAmps_sm.h, new BUG FIX FOR #701 in function vxxxxx
valassi Jul 17, 2023
709ec5d
[fpe] in ggtt.sa cudacpp makefiles, add gcov target
valassi Jul 17, 2023
eb5594d
[fpe] in ggt.sa .gitignore, add gcov suffixes to gitignore
valassi Jul 17, 2023
e4957d5
[fpe] revert the previous change: will instead remove gcov files in '…
valassi Jul 17, 2023
fc120fa
[fpe] in ggtt.sa cudacpp makefiles, remove files with gcov suffixes i…
valassi Jul 17, 2023
e372403
[fpe] in gggtt.sa cudaccp makefiles, remove all gcov attempts - what …
valassi Jul 17, 2023
4380121
[fpe] in ggtt.sa HelAmps_sm.h, new BUG FIX FOR #701 in function oxxxxx
valassi Jul 17, 2023
0bd0980
[fpe] in ggtt.sa HelAmps_sm.h, partial fix for my new bug in oxxxxx
valassi Jul 17, 2023
4cc392d
[fpe] in ggtt.sa HelAmps_sa.h, remove old code with old comments
valassi Jul 17, 2023
6ec612c
[fpe] in ggtt.sa HelAmps_sm.h, partial fix for my new bug in oxxxxx: …
valassi Jul 17, 2023
06aac19
[fpe] in ggtt.sa HelAmps_sm.h, partial fix for my new bug in oxxxxx: …
valassi Jul 17, 2023
0240749
[fpe] in ggtt.sa HelAmps_sm.h, add a comment about ip/im in the new i…
valassi Jul 17, 2023
940c0c4
[fpe] in ggtt.sa testxxx.cc, ensure that opzxxx handles SIMD vectors …
valassi Jul 17, 2023
e025512
[fpe] in ggtt.sa testxxx.cc, use the same trick as for ipz/opzxxx als…
valassi Jul 17, 2023
c658021
[fpe] in ggtt.sa HelAmps_sm.h, make a minor cosmetic change in the ne…
valassi Jul 17, 2023
278190d
[fpe] in ggtt.sa testxxx.cc, use the same trick as for ipz/opzxxx for…
valassi Jul 17, 2023
2d653b2
[fpe] in ggtt.sa testmisc.cc, install a simpler FPE handler
valassi Jul 17, 2023
8d31349
[fpe] revert the previous change - will disable the FPE signals inste…
valassi Jul 17, 2023
8c468a3
[fpe] in ggtt.sa testxxx.cc, disable FPE handlers at the end of the test
valassi Jul 17, 2023
550f3af
[fpe] in ggtt.sa testxxx.cc, reshuffle tests to have 16 with m==0 and…
valassi Jul 17, 2023
8ca5144
[fpe] in ggtt.sa, replace newly dumped testxxx_cc_ref.txt (and disabl…
valassi Jul 17, 2023
a683a5c
[fpe] in ggtt.sa testxxx.cc, disable the trick for all functions but …
valassi Jul 17, 2023
f154998
[fpe] in ggtt.sa HelAmps_sm.h, DISABLE AUTO-VECTORIZATION on the whol…
valassi Jul 17, 2023
ff3b0b9
[fpe] in ggtt.sa HelAmps_sm.h, add a first fix for my fix of #701 (em…
valassi Jul 17, 2023
93a4912
[fpe] in ggtt.sa HelAmps_sm.h, DISABLE AUTO-VECTORIZATION on the whol…
valassi Jul 17, 2023
8680f94
[fpe] in ggtt.sa HelAmps_sm.h, DISABLE AUTO-VECTORIZATION on the whol…
valassi Jul 17, 2023
71e6f01
[fpe] in ggtt.sa HelAmps_sm.h, fix builds for AVX=none... the tests s…
valassi Jul 17, 2023
f7f0252
[fpe] in ggtt.sa mgOnGpuVectors.h, add maskand function also for CUDA
valassi Jul 17, 2023
13df781
[fpe] in ggtt.sa HelAmps_sm.h, fix builds for FPTYPE=f with SIMD
valassi Jul 17, 2023
c83a836
[fpe] in ggtt.sa cudacpp makefiles, remove -DDEBUG2 and add back -lin…
valassi Jul 18, 2023
377baa1
[fpe] in ggtt.sa cudacpp makefiles, remove -lineinfo again from debug…
valassi Jul 18, 2023
10ef191
[fpe] in ggtt.sa cudacpp_src makefile, add -march=x86-64 to AVX=none …
valassi Jul 18, 2023
5a6a297
[fpe] rerun tput ggtt.mad test as a reference now
valassi Jul 18, 2023
9f24960
[namespace] in ggtt.sa, start separation of cpu/gpu namespaces: first…
valassi Jul 19, 2023
cb90519
[namespace] in ggtt.sa, start separation of cpu/gpu namespaces: first…
valassi Jul 19, 2023
7de23d5
[namespace] in ggtt.sa, start separation of cpu/gpu namespaces: furth…
valassi Jul 19, 2023
5d4de00
[namespace] in ggtt.sa, start separation of cpu/gpu namespaces: furth…
valassi Jul 19, 2023
3844169
[namespace] in ggtt.sa, start separation of cpu/gpu namespaces: furth…
valassi Jul 19, 2023
811da80
[namespace] in ggtt.sa, start separation of cpu/gpu namespaces: furth…
valassi Jul 19, 2023
6d7994f
[namespace] in ggtt.sa, start separation of cpu/gpu namespaces: furth…
valassi Jul 19, 2023
103e7a8
[namespace] in ggtt.sa, start separation of cpu/gpu namespaces: furth…
valassi Jul 19, 2023
b714142
[namespace] in ggtt.sa, first fixes for cpu/gpu namespace in Paramete…
valassi Jul 19, 2023
0ae9b56
[namespace] in ggtt.sa mgOnGpuFptypes.h, move fpsqrt and other functi…
valassi Jul 19, 2023
edaf3da
[namespace] in ggtt.sa mgOnGpuCxtypes.h, move nx2 back to the common …
valassi Jul 19, 2023
52431dc
[namespace] in ggtt.sa MemoryAccessMomenta.h, move the class into the…
valassi Jul 19, 2023
4db9f23
[namespace] in ggtt.sa Bridge.h, remove cpu/gpu namespace qualifiers …
valassi Jul 19, 2023
4643f37
[namespace] in six ggtt.sa MemoryXXX.h, move the class into the cpu/g…
valassi Jul 19, 2023
c7d0349
[namespace] in ggtt.sa BridgeKernels.cc, fix namespaces for CPPProcess
valassi Jul 19, 2023
337cc8e
[namespace] in ggtt.sa, further fix for cpu/gpu namespace in Paramete…
valassi Jul 19, 2023
7e44ef6
[namespace] in ggtt.sa, add gParameters_sm.cu and buiold it with CUDA…
valassi Jul 19, 2023
97a28da
[namespace] in ggtt.sa, further fix for cpu/gpu namespace in mgOnGpuC…
valassi Jul 19, 2023
a219873
[fpe] cherry-pick the last commit from the namespace branch and fix a…
valassi Jul 19, 2023
8a5d3b1
[fpe] rerun tput for ggtt.sa and copy the log: there is a big perform…
valassi Jul 19, 2023
be4a0cb
[fpe] in ggtt.sa HelAmps_sm.h, REENABLE AUTO-VECTORIZATION on the who…
valassi Jul 19, 2023
6e9e253
[fpe] in ggtt.sa HelAmps_sm.h, add a printout that clearly shows the …
valassi Jul 19, 2023
88e4303
[fpe] in ggtt.sa HelAmps_sm.h, try to prevent vectorization in ixxxxx…
valassi Jul 19, 2023
5a005cc
[fpe] in ggtt.sa HelAmps_sm.h, try to prevent vectorization in ixxxxx…
valassi Jul 19, 2023
f3c331e
[fpe] in ggtt.sa HelAmps_sm.h, try to prevent vectorization in ixxxxx…
valassi Jul 19, 2023
b2e71e5
[fpe] in ggtt.sa HelAmps_sm.h, try to prevent FPEs in ixxxxx by using…
valassi Jul 19, 2023
98e664a
[fpe] in ggtt.sa HelAmps_sm.h, finally succeed in preventing the FPE …
valassi Jul 19, 2023
a7e952c
[fpe] in ggtt.sa HelAmps_sm.h, clean up ixxxx implementation by remov…
valassi Jul 19, 2023
106789b
[fpe] in ggtt.sa HelAmps_sm.h, add a printout that clearly shows the …
valassi Jul 19, 2023
3ae5087
[fpe] in ggtt.sa HelAmps_sm.h, try to avoid the FPE in vxxxxx using t…
valassi Jul 19, 2023
8194f15
[fpe] in ggtt.sa HelAmps_sm.h, printing out denom for each ieppV woul…
valassi Jul 19, 2023
6ed2c0d
[fpe] in ggtt.sa HelAmps_sm.h, avoid the FPE in vxxxxx using the same…
valassi Jul 19, 2023
dbd4e91
[fpe] in ggtt.sa HelAmps_sm.h, clean up vxxxx implementation by remov…
valassi Jul 19, 2023
e70f2ae
[fpe] in ggtt.sa HelAmps_sm.h, add a printout that clearly shows the …
valassi Jul 19, 2023
5437ca8
[fpe] in ggtt.sa HelAmps_sm.h, avoid the FPE in oxxxxx using the same…
valassi Jul 19, 2023
57ee968
[fpe] in ggtt.sa HelAmps_sm.h, add a printout that clearly shows the …
valassi Jul 19, 2023
39232c5
[fpe] == MAJOR CHANGE OF STRATEGY!!! == in ggtt.sa HelAmps_sm.h, move…
valassi Jul 19, 2023
04e0085
[fpe] == COMPLETE MAJOR CHANGE OF STRATEGY!!! == in ggtt.sa HelAmps_s…
valassi Jul 19, 2023
c7f194c
[fpe] rerun tput for ggtt.sa and copy the log: recover the previous p…
valassi Jul 19, 2023
b677c21
[fpe] revert to the ggtt tput performance log from upstream/master fo…
valassi Jul 19, 2023
a7a1d35
[namespace] in ggtt.sa, remove gParameters_sm.cu symlink in src, inst…
valassi Jul 19, 2023
6cb7578
[fpe] progress in backport from ggtt.sa to CODEGEN - test classes
valassi Jul 19, 2023
c0d4fa2
[fpe] clang-format CODEGEN for ggtt.sa test*.cc (also in ggtt.sa)
valassi Jul 19, 2023
f5b0cdb
[fpe] progress in backport from ggtt.sa to CODEGEN - Memory* files
valassi Jul 19, 2023
ea860ce
[fpe] clang-format CODEGEN for Memory* (also in ggtt.sa)
valassi Jul 19, 2023
85c8295
[fpe] progress in backport from ggtt.sa to CODEGEN - src/mgOnGpu*
valassi Jul 19, 2023
863f302
[fpe] clang-format CODEGEN for mgOnGpu* (also in ggtt.sa)
valassi Jul 19, 2023
5b19df7
[fpe] progress in backport from ggtt.sa to CODEGEN - Bridge*
valassi Jul 19, 2023
26eb6b6
[fpe] clang-format CODEGEN for Bridge* (also in ggtt.sa)
valassi Jul 19, 2023
0ef9448
[fpe] progress in backport from ggtt.sa to CODEGEN - makefiles
valassi Jul 19, 2023
9a95a74
[fpe] progress in backport from ggtt.sa to CODEGEN - Parameters.cc
valassi Jul 19, 2023
d9d4902
[fpe] progress in backport from ggtt.sa to CODEGEN - Parameters.h
valassi Jul 19, 2023
4b9bc3f
[fpe] in ggtt.sa, fix cudacpp.mk to remove clang14 warning on alma9 (…
valassi Jul 20, 2023
b02dd23
[fpe] in ggtt.sa HelAmps_sm.h, add a printout that clearly shows the …
valassi Jul 20, 2023
81de612
[fpe] in ggtt.sa HelAmps_sm.h, adding more printouts in ixxxxx bypass…
valassi Jul 20, 2023
bf505c6
[fpe] in ggtt.sa HelAmps_sm.h, remove some printouts in ixxxxx, the F…
valassi Jul 20, 2023
240a080
[fpe] in ggtt.sa HelAmps_sm.h, first attempt to use the 'volatile' ke…
valassi Jul 20, 2023
4be246b
[fpe] in ggtt.sa HelAmps_sm.h, finally(?) fix the FPE issue using the…
valassi Jul 20, 2023
163d71b
[fpe] in ggtt.sa HelAmps_sm.h, clean up comments about FPE and volati…
valassi Jul 20, 2023
60a5d9a
[fpe] in ggtt.sa HelAmps_sm.h, COMPLETE the addition of volatile to a…
valassi Jul 20, 2023
8ac0041
[fpe] in ggtt.sa cudacpp.mk, fix build flags for nvcc using icpx in d…
valassi Jul 20, 2023
3b91f0b
[fpe] in ggtt.sa testxxx.cc, turn off debug printouts at last! the FP…
valassi Jul 20, 2023
4eed3c3
[fpe] progress in backport from ggtt.sa to CODEGEN - new versions of …
valassi Jul 20, 2023
aba2ee2
[fpe] progress in backport from ggtt.sa to CODEGEN - fix spacing of w…
valassi Jul 20, 2023
8894774
[fpe] progress in backport from ggtt.sa to CODEGEN - fix spacing of w…
valassi Jul 20, 2023
85f560f
[fpe] in ggtt.sa HelAmps_sm.h, fix formatting with clang-format befor…
valassi Jul 20, 2023
43f7dfb
[fpe] in ggtt.sa Parameters_sm.h, improve structure and fix formattin…
valassi Jul 20, 2023
46ec689
[fpe] in CODEGEN, adapt cpp_model_parameters_h.inc template from ggtt…
valassi Jul 20, 2023
d51fc2f
[fpe] progress in backport from ggtt.sa to CODEGEN - fix spacing of a…
valassi Jul 20, 2023
9115b1d
[fpe] in CODEGEN, BACKPORT THE FPE FIX FOR #701! This completes the b…
valassi Jul 20, 2023
e34f94b
[fpe] regenerate ggtt.sa - all ok, stable
valassi Jul 20, 2023
690a2ce
[fpe] in CODEGEN, add the ALOHA copyright line where appropriate
valassi Jul 20, 2023
1bed1ba
[fpe] regenerate ggtt.sa - all ok, stable
valassi Jul 20, 2023
2269181
[fpe] in CODEGEN, fix spacing (clang formatting) in EFT Parameters.h
valassi Jul 20, 2023
89ea5d3
[fpe] regenerate heft_gg_h.sa - all ok, stable (and formatting is now…
valassi Jul 20, 2023
d32b85b
[fpe] regenerate ggtt.sa again - all ok, stable
valassi Jul 20, 2023
9c76fc1
[fpe] in CODEGEN, fix parameter handling for SUSY (move code from pri…
valassi Jul 20, 2023
1b31e83
[fpe] in CODEGEN, fix spacing (clang formatting) for SUSY in Paramete…
valassi Jul 20, 2023
32365c9
[fpe] in CODEGEN, fix spacing (clang formatting) for nobm_pp_ttW
valassi Jul 20, 2023
62c5347
[fpe] in CODEGEN, fix spacing (clang formatting) for smeft_gg_tttt
valassi Jul 20, 2023
2aec131
[fpe] regenerate ggtt and heft_h_gg sa again - all ok, stable
valassi Jul 20, 2023
671ff9c
[fpe] in CODEGEN, fix spacing (clang formatting) for ee_mumu
valassi Jul 20, 2023
6901bb6
[fpe] regenerate ggtt and heft_h_gg sa again - all ok, stable
valassi Jul 20, 2023
3cd4e17
[fpe] regenerate the other 5 processes SA - clang formatting is ok
valassi Jul 20, 2023
a5588bf
[fpe] regenerate all seven processes mad, at last
valassi Jul 20, 2023
223adc2
[namespace/fpe] in ggtt.mad, fix cpu/gpu namespaces in MemoryAccessNu…
valassi Jul 20, 2023
dcb6610
[namespace/fpe] backport MemoryAccessNumerators/Denominators changes …
valassi Jul 20, 2023
dd264f5
[fpe] regenerate ggtt.mad, all ok
valassi Jul 20, 2023
d01ba9d
[fpe] manually copy the two fixed files to the other 6 mad and 7 sa p…
valassi Jul 20, 2023
4bf1160
[fpe] rerun tput test for eemumu and ggtt - all looks ok
valassi Jul 20, 2023
41c6a6d
[fpe] rerun tput test also for ggttg* - all looks ok
valassi Jul 20, 2023
9a68868
[fpe] rerun 78 tput alltees, all ok
valassi Jul 21, 2023
9212960
[fpe] rerun 15 tmad - ggttgg tests fail again #655 as expected
valassi Jul 21, 2023
3658f3f
[fpe] Revert to upstream/master performance logs in tput and tmad for…
valassi Jul 21, 2023
bf5727b
[fpe] in ggtt.sa, remove feenableexcept on MacOS where it is not defi…
valassi Jul 21, 2023
e93ba8a
[fpe] backport workaround for #730 on MacOS to CODEGEN from ggtt.sa
valassi Jul 21, 2023
324581d
[namespace/fpe] in ggtt.sa makefiles, add 'export CUFLAGS' in SubProc…
valassi Jul 21, 2023
a1d5983
[namespace/fpe] backport fix for #731 (HRDCOD=1 builds in cuda of non…
valassi Jul 21, 2023
66b8cfe
[fpe] regenerate gg_tt and heft_gg_h sa - all ok, differences as expe…
valassi Jul 21, 2023
838e59a
[fpe] regenerate the other 5 processes sa with fixes for #730 and #731
valassi Jul 21, 2023
49f9d3f
[fpe] ** COMPLETE FPE ** regenerate all 7 processes mad with fixes fo…
valassi Jul 21, 2023
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
Next Next commit
[fpe] in ggtt.sa mgOnGpuVectors.h, add constructor "cxtype_v( const f…
…ptype& r )" to create cx vectors from fp scalars
valassi committed Jul 17, 2023

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature. The key has expired.
commit 8745a986e56f06a95761258b925a032e58b0cb8d
1 change: 1 addition & 0 deletions epochX/cudacpp/gg_tt.sa/src/mgOnGpuVectors.h
Original file line number Diff line number Diff line change
@@ -77,6 +77,7 @@ namespace mgOnGpu /* clang-format off */
cxtype_v( cxtype_v&& ) = default;
cxtype_v( const fptype_v& r, const fptype_v& i ) : m_real( r ), m_imag( i ) {}
cxtype_v( const fptype_v& r ) : m_real( r ), m_imag{ 0 } {} // IIII=0000
cxtype_v( const fptype& r ) : m_real( fptype_v{} + r ), m_imag{ 0 } {} // IIII=0000
cxtype_v& operator=( const cxtype_v& ) = default;
cxtype_v& operator=( cxtype_v&& ) = default;
cxtype_v& operator+=( const cxtype_v& c ) { m_real += c.real(); m_imag += c.imag(); return *this; }