forked from llvm/llvm-project
-
Notifications
You must be signed in to change notification settings - Fork 3
[AutoBump] Merge with 046b064d (Jan 20) (3) [Only tested LLVM] #541
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
Open
jorickert
wants to merge
76
commits into
bump_to_f4943464
Choose a base branch
from
bump_to_046b064d
base: bump_to_f4943464
Could not load branches
Branch not found: {{ refName }}
Loading
Could not load tags
Nothing to show
Loading
Are you sure you want to change the base?
Some commits from the old base branch may be removed from the timeline,
and old review comments may become outdated.
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…23386) Prevents infinite loop between combineBitcastToBoolVector and hoistLogicOpWithSameOpcodeHands, which only performs the "logicop(bitcast(A),bitcast(B)) -> bitcast(logicop(A,B))" upto type legalization. combineBitcastToBoolVector doesn't care much as its mainly for AVX512 cleanup that X86DomainReassignment can't handle for us. Fixes llvm#123333
…ract/insertion nodes. NFC. (llvm#123312) Avoid the use of getIntPtrConstant for anything other than address pointer related code. Noticed while trying to use getVectorIdxConstant as a breakpoint.
This fold already exists in a couple places (DAG and CGP), where an icmps operands are swapped to allow CSE with a sub. They do not handle constants though. This patch adds an AArch64 version that can be more precise.
Model updating IV users directly in VPlan, replace fixupIVUsers. Now simple extracts are created for all phis in the exit block during initial VPlan construction. A later VPlan transform (optimizeInductionExitUsers) replaces extracts of inductions with their pre-computed values if possible. This completes the transition towards modeling all live-outs directly in VPlan. There are a few follow-ups: * emit extracts initially also for resume phis, and optimize them tougher with IV exit users * support for VPlans with multiple exits in optimizeInductionExitUsers. Depends on llvm#110004, llvm#109975 and llvm#112145.
This reverts commit c2d15ac. Causes build failures on PPC stage2 & fuchsia bots https://lab.llvm.org/buildbot/#/builders/168/builds/7650 https://lab.llvm.org/buildbot/#/builders/11/builds/11248
… to handle single bitwidth at a time. Attempt 32-bit broadcasts first, and then fallback to 64-bit broadcasts on failure. We lose an explicit assertion for matching operand numbers but X86InstrFoldTables already does something similar. Pulled out of WIP patch llvm#73509
This is to address the issue where `yield` can cause the thread to be assigned to the lowest priority. I have done lots of experiments: see the comments here: llvm#84471 (comment) And for this patch, the benchmark has been performed on a 16 core M4 MAX CPU MacBook Pro. dylib compiled with Release mode and the test compiled with optimization=speed ``` Comparing ../../../build_atomic_yield2/ref_new2.json to ../../../build_atomic_yield2/no_yield_new2.json Benchmark Time CPU Time Old Time New CPU Old CPU New ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- BM_1_atomic_1_waiter_1_notifier<KeepNotifying, NumHighPrioTasks<0>>/262144 +0.0460 +0.0392 14949926 15637503 13633314 14167327 BM_1_atomic_1_waiter_1_notifier<KeepNotifying, NumHighPrioTasks<0>>/524288 +0.0299 +0.0290 24369327 25099004 24367214 25073900 BM_1_atomic_1_waiter_1_notifier<KeepNotifying, NumHighPrioTasks<0>>/1048576 +0.0648 +0.0640 48149060 51268517 48144857 51226733 BM_1_atomic_1_waiter_1_notifier<NotifyEveryNus<50>, NumHighPrioTasks<0>>/4096 +0.0000 -0.8765 204815500 204823427 204514333 25265071 BM_1_atomic_1_waiter_1_notifier<NotifyEveryNus<50>, NumHighPrioTasks<0>>/8192 +0.0000 -0.8747 409637520 409640821 408997500 51228071 BM_1_atomic_1_waiter_1_notifier<NotifyEveryNus<50>, NumHighPrioTasks<0>>/16384 +0.0001 -0.8737 819244417 819351256 817022000 103217000 BM_1_atomic_1_waiter_1_notifier<NotifyEveryNus<100>, NumHighPrioTasks<0>>/4096 +0.0000 -0.9029 409607694 409624937 271866333 26410600 BM_1_atomic_1_waiter_1_notifier<NotifyEveryNus<100>, NumHighPrioTasks<0>>/8192 +0.0001 -0.9017 819168417 819269339 542784000 53352429 BM_1_atomic_1_waiter_1_notifier<NotifyEveryNus<100>, NumHighPrioTasks<0>>/16384 +0.0001 -0.9012 1638361750 1638522929 1089486000 107684571 BM_1_atomic_1_waiter_1_notifier<KeepNotifying, NumHighPrioTasks<4>>/262144 +0.3178 +0.3068 12777744 16838266 12764732 16681233 BM_1_atomic_1_waiter_1_notifier<KeepNotifying, NumHighPrioTasks<4>>/524288 +0.2231 +0.2225 26889415 32887842 26864138 32840550 BM_1_atomic_1_waiter_1_notifier<KeepNotifying, NumHighPrioTasks<4>>/1048576 +0.1809 +0.1799 56103004 66251660 56048000 66129583 BM_1_atomic_1_waiter_1_notifier<NotifyEveryNus<50>, NumHighPrioTasks<4>>/4096 -0.0029 -0.8708 205509986 204906011 204277333 26399538 BM_1_atomic_1_waiter_1_notifier<NotifyEveryNus<50>, NumHighPrioTasks<4>>/8192 +0.0001 -0.8711 410286709 410314199 408608000 52667692 BM_1_atomic_1_waiter_1_notifier<NotifyEveryNus<50>, NumHighPrioTasks<4>>/16384 -0.0019 -0.8713 821042916 819476441 816274000 105077000 BM_1_atomic_1_waiter_1_notifier<NotifyEveryNus<100>, NumHighPrioTasks<4>>/4096 -0.0005 -0.9015 409825792 409638429 273145333 26896400 BM_1_atomic_1_waiter_1_notifier<NotifyEveryNus<100>, NumHighPrioTasks<4>>/8192 -0.0027 -0.9014 821528125 819285433 545661000 53775308 BM_1_atomic_1_waiter_1_notifier<NotifyEveryNus<100>, NumHighPrioTasks<4>>/16384 -0.0041 -0.9014 1645204459 1638538077 1091726000 107647000 BM_1_atomic_1_waiter_1_notifier<KeepNotifying, NumHighPrioTasks<7>>/16 -0.4835 -0.4836 1609 831 1609 831 BM_1_atomic_1_waiter_1_notifier<KeepNotifying, NumHighPrioTasks<7>>/32 -0.4398 -0.4399 3167 1774 3166 1773 BM_1_atomic_1_waiter_1_notifier<KeepNotifying, NumHighPrioTasks<7>>/64 -0.4705 -0.4705 6323 3348 6323 3348 BM_1_atomic_1_waiter_1_notifier<NotifyEveryNus<50>, NumHighPrioTasks<7>>/8 +0.0005 -0.8683 400109 400314 399256 52575 BM_1_atomic_1_waiter_1_notifier<NotifyEveryNus<50>, NumHighPrioTasks<7>>/16 +0.0005 -0.8683 800055 800483 798797 105165 BM_1_atomic_1_waiter_1_notifier<NotifyEveryNus<50>, NumHighPrioTasks<7>>/32 +0.0003 -0.8680 1600058 1600585 1597266 210903 BM_1_atomic_1_waiter_1_notifier<NotifyEveryNus<100>, NumHighPrioTasks<7>>/8 +0.0004 -0.8976 800006 800365 531802 54441 BM_1_atomic_1_waiter_1_notifier<NotifyEveryNus<100>, NumHighPrioTasks<7>>/16 +0.0005 -0.8982 1599965 1600765 1064885 108429 BM_1_atomic_1_waiter_1_notifier<NotifyEveryNus<100>, NumHighPrioTasks<7>>/32 +0.0005 -0.8993 3199905 3201437 2129243 214343 BM_1_atomic_multi_waiter_1_notifier<KeepNotifying, NumWaitingThreads<3>, NumHighPrioTasks<0>>/16384 -0.0226 -0.0261 972539 950519 971198 945828 BM_1_atomic_multi_waiter_1_notifier<KeepNotifying, NumWaitingThreads<3>, NumHighPrioTasks<0>>/32768 -0.0198 -0.0221 1933294 1895054 1930720 1888094 BM_1_atomic_multi_waiter_1_notifier<KeepNotifying, NumWaitingThreads<3>, NumHighPrioTasks<0>>/65536 -0.0031 -0.0039 3835138 3823094 3827785 3812836 BM_1_atomic_multi_waiter_1_notifier<KeepNotifying, NumWaitingThreads<7>, NumHighPrioTasks<0>>/4096 +0.4380 +0.4294 571762 822185 570245 815115 BM_1_atomic_multi_waiter_1_notifier<KeepNotifying, NumWaitingThreads<7>, NumHighPrioTasks<0>>/8192 +0.0735 +0.0680 1223881 1313880 1221350 1304439 BM_1_atomic_multi_waiter_1_notifier<KeepNotifying, NumWaitingThreads<7>, NumHighPrioTasks<0>>/16384 +0.1222 +0.1205 2442071 2740519 2433105 2726274 BM_1_atomic_multi_waiter_1_notifier<KeepNotifying, NumWaitingThreads<15>, NumHighPrioTasks<0>>/1024 +0.1527 +1.2188 196081 226031 62647 139001 BM_1_atomic_multi_waiter_1_notifier<KeepNotifying, NumWaitingThreads<15>, NumHighPrioTasks<0>>/2048 +0.0757 +0.4838 387858 417228 129250 191780 BM_1_atomic_multi_waiter_1_notifier<KeepNotifying, NumWaitingThreads<15>, NumHighPrioTasks<0>>/4096 -0.0355 -0.2443 812827 784003 378109 285722 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<50>, NumWaitingThreads<3>, NumHighPrioTasks<0>>/1024 +0.0002 -0.0873 51202059 51211089 51135714 46670867 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<50>, NumWaitingThreads<3>, NumHighPrioTasks<0>>/2048 +0.0001 -0.0864 102424970 102432359 102287571 93452000 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<50>, NumWaitingThreads<3>, NumHighPrioTasks<0>>/4096 +0.0000 -0.0865 204828250 204834229 204528667 186845250 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<50>, NumWaitingThreads<7>, NumHighPrioTasks<0>>/256 +0.0003 -0.1681 12801752 12805016 12786382 10636485 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<50>, NumWaitingThreads<7>, NumHighPrioTasks<0>>/512 +0.0001 -0.1686 25601940 25604893 25565481 21254515 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<50>, NumWaitingThreads<7>, NumHighPrioTasks<0>>/1024 +0.0000 -0.1569 51210789 51211539 51150143 43122500 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<50>, NumWaitingThreads<15>, NumHighPrioTasks<0>>/64 +0.0064 -0.3503 3210430 3230869 2856780 1856063 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<50>, NumWaitingThreads<15>, NumHighPrioTasks<0>>/128 +0.0034 -0.3534 6410529 6432308 5704792 3688942 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<50>, NumWaitingThreads<15>, NumHighPrioTasks<0>>/256 +0.0011 -0.3600 12821419 12835646 11455934 7331250 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<100>, NumWaitingThreads<3>, NumHighPrioTasks<0>>/256 +0.0003 +0.0034 25600089 25608062 24375034 24457172 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<100>, NumWaitingThreads<3>, NumHighPrioTasks<0>>/512 +0.0002 -0.0000 51203798 51211795 48859857 48858500 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<100>, NumWaitingThreads<3>, NumHighPrioTasks<0>>/1024 +0.0003 +0.0008 102411321 102437524 97694429 97777286 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<100>, NumWaitingThreads<7>, NumHighPrioTasks<0>>/64 +0.0002 -0.0464 6399846 6401009 6070487 5789091 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<100>, NumWaitingThreads<7>, NumHighPrioTasks<0>>/128 +0.0002 -0.0457 12799914 12802544 12069966 11518836 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<100>, NumWaitingThreads<7>, NumHighPrioTasks<0>>/256 +0.0001 -0.0513 25599724 25602105 24202862 22962032 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<100>, NumWaitingThreads<15>, NumHighPrioTasks<0>>/16 -0.0060 +0.2575 1611779 1602148 956236 1202492 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<100>, NumWaitingThreads<15>, NumHighPrioTasks<0>>/32 -0.0064 +0.2964 3221485 3200918 1883540 2441728 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<100>, NumWaitingThreads<15>, NumHighPrioTasks<0>>/64 -0.0046 +0.3087 6432692 6403368 3701725 4844611 BM_1_atomic_multi_waiter_1_notifier<KeepNotifying, NumWaitingThreads<3>, NumHighPrioTasks<4>>/256 -0.0536 -0.0592 27458 25988 27402 25780 BM_1_atomic_multi_waiter_1_notifier<KeepNotifying, NumWaitingThreads<3>, NumHighPrioTasks<4>>/512 -0.0469 -0.0527 54745 52175 54628 51750 BM_1_atomic_multi_waiter_1_notifier<KeepNotifying, NumWaitingThreads<3>, NumHighPrioTasks<4>>/1024 -0.0297 -0.0340 108312 105095 108047 104378 BM_1_atomic_multi_waiter_1_notifier<KeepNotifying, NumWaitingThreads<7>, NumHighPrioTasks<4>>/64 -0.2445 -0.2722 15109 11414 14711 10708 BM_1_atomic_multi_waiter_1_notifier<KeepNotifying, NumWaitingThreads<7>, NumHighPrioTasks<4>>/128 -0.3132 -0.3515 32494 22317 32063 20794 BM_1_atomic_multi_waiter_1_notifier<KeepNotifying, NumWaitingThreads<7>, NumHighPrioTasks<4>>/256 -0.1397 -0.1834 52801 45424 52170 42602 BM_1_atomic_multi_waiter_1_notifier<KeepNotifying, NumWaitingThreads<15>, NumHighPrioTasks<4>>/16 +0.1679 +1.0248 28973 33837 13243 26814 BM_1_atomic_multi_waiter_1_notifier<KeepNotifying, NumWaitingThreads<15>, NumHighPrioTasks<4>>/32 -0.0481 +0.7901 39155 37273 16072 28771 BM_1_atomic_multi_waiter_1_notifier<KeepNotifying, NumWaitingThreads<15>, NumHighPrioTasks<4>>/64 -0.2075 +0.7568 57547 45606 19582 34402 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<50>, NumWaitingThreads<3>, NumHighPrioTasks<4>>/256 -0.0001 -0.0807 12802693 12800886 12775327 11744119 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<50>, NumWaitingThreads<3>, NumHighPrioTasks<4>>/512 -0.0021 -0.0867 25655056 25601315 25590407 23371667 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<50>, NumWaitingThreads<3>, NumHighPrioTasks<4>>/1024 -0.0007 -0.0832 51238801 51201975 51099071 46845733 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<50>, NumWaitingThreads<7>, NumHighPrioTasks<4>>/64 +0.0016 -0.2411 3200714 3205846 3176841 2410756 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<50>, NumWaitingThreads<7>, NumHighPrioTasks<4>>/128 +0.0008 -0.2373 6404239 6409102 6359649 4850544 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<50>, NumWaitingThreads<7>, NumHighPrioTasks<4>>/256 +0.0000 -0.2286 12805839 12806032 12713018 9806653 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<50>, NumWaitingThreads<15>, NumHighPrioTasks<4>>/16 +0.0272 +0.0563 811198 833264 482220 509345 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<50>, NumWaitingThreads<15>, NumHighPrioTasks<4>>/32 +0.0097 +0.0454 1617205 1632962 957801 1001264 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<50>, NumWaitingThreads<15>, NumHighPrioTasks<4>>/64 +0.0050 +0.0389 3217997 3234130 1927921 2002868 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<100>, NumWaitingThreads<3>, NumHighPrioTasks<4>>/256 +0.0000 -0.0009 25599763 25601039 24520071 24497071 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<100>, NumWaitingThreads<3>, NumHighPrioTasks<4>>/512 +0.0001 -0.0017 51200354 51203628 49086786 49005500 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<100>, NumWaitingThreads<3>, NumHighPrioTasks<4>>/1024 +0.0001 +0.0013 102400369 102409744 97931143 9806085 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<100>, NumWaitingThreads<7>, NumHighPrioTasks<4>>/64 -0.0017 +0.0128 6410821 6400104 5529150 5600008 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<100>, NumWaitingThreads<7>, NumHighPrioTasks<4>>/128 -0.0011 +0.0215 12817263 12803569 11025889 11263032 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<100>, NumWaitingThreads<7>, NumHighPrioTasks<4>>/256 -0.0005 +0.0193 25612704 25600332 22089065 22515677 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<100>, NumWaitingThreads<15>, NumHighPrioTasks<4>>/16 -0.0164 +0.7969 1627422 1600798 665736 1196236 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<100>, NumWaitingThreads<15>, NumHighPrioTasks<4>>/32 -0.0095 +0.8362 3231500 3200840 1290017 2368789 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<100>, NumWaitingThreads<15>, NumHighPrioTasks<4>>/64 -0.0050 +0.7319 6433401 6401180 2747936 4759115 BM_1_atomic_multi_waiter_1_notifier<KeepNotifying, NumWaitingThreads<3>, NumHighPrioTasks<7>>/16 +0.0155 +0.0092 1177 1195 1171 1181 BM_1_atomic_multi_waiter_1_notifier<KeepNotifying, NumWaitingThreads<3>, NumHighPrioTasks<7>>/32 -0.0135 -0.0145 2103 2074 2095 2064 BM_1_atomic_multi_waiter_1_notifier<KeepNotifying, NumWaitingThreads<3>, NumHighPrioTasks<7>>/64 +0.0022 +0.0009 3832 3841 3820 3823 BM_1_atomic_multi_waiter_1_notifier<KeepNotifying, NumWaitingThreads<7>, NumHighPrioTasks<7>>/8 +13.9131 +9.5298 2074 30931 2041 21495 BM_1_atomic_multi_waiter_1_notifier<KeepNotifying, NumWaitingThreads<7>, NumHighPrioTasks<7>>/16 +5.9980 +3.9816 3168 22172 3124 15563 BM_1_atomic_multi_waiter_1_notifier<KeepNotifying, NumWaitingThreads<7>, NumHighPrioTasks<7>>/32 +3.8681 +2.3515 5412 26348 5321 17833 BM_1_atomic_multi_waiter_1_notifier<KeepNotifying, NumWaitingThreads<15>, NumHighPrioTasks<7>>/4 +0.1312 +0.4845 31938 36127 12666 18803 BM_1_atomic_multi_waiter_1_notifier<KeepNotifying, NumWaitingThreads<15>, NumHighPrioTasks<7>>/8 -0.0475 +0.0775 39196 37336 18078 19479 BM_1_atomic_multi_waiter_1_notifier<KeepNotifying, NumWaitingThreads<15>, NumHighPrioTasks<7>>/16 -0.3146 -0.3853 57548 39441 31743 19513 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<50>, NumWaitingThreads<3>, NumHighPrioTasks<7>>/8 -0.0012 -0.0916 400610 400149 399248 362679 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<50>, NumWaitingThreads<3>, NumHighPrioTasks<7>>/16 -0.0032 -0.0904 802940 800342 798964 726744 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<50>, NumWaitingThreads<3>, NumHighPrioTasks<7>>/32 -0.0030 -0.0911 1604860 1600044 1598235 1452647 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<50>, NumWaitingThreads<7>, NumHighPrioTasks<7>>/4 +0.0348 -0.3515 202073 209107 199452 129352 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<50>, NumWaitingThreads<7>, NumHighPrioTasks<7>>/8 -0.0004 -0.3628 406727 406545 400942 255464 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<50>, NumWaitingThreads<7>, NumHighPrioTasks<7>>/16 -0.0176 -0.3705 821725 807256 803722 505959 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<50>, NumWaitingThreads<15>, NumHighPrioTasks<7>>/2 +0.0575 +0.0699 138530 146498 79463 85020 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<50>, NumWaitingThreads<15>, NumHighPrioTasks<7>>/4 -0.2307 -0.4182 327417 251885 222502 129448 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<50>, NumWaitingThreads<15>, NumHighPrioTasks<7>>/8 -0.4166 -0.5733 765495 446598 535265 228384 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<100>, NumWaitingThreads<3>, NumHighPrioTasks<7>>/8 +0.0001 +0.0022 800108 800227 759501 761200 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<100>, NumWaitingThreads<3>, NumHighPrioTasks<7>>/16 +0.0002 +0.0052 1599998 1600327 1515336 1523162 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<100>, NumWaitingThreads<3>, NumHighPrioTasks<7>>/32 -0.0004 +0.0029 3201730 3200529 3037191 3045996 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<100>, NumWaitingThreads<7>, NumHighPrioTasks<7>>/4 -0.0063 +0.3625 402752 400231 231304 315156 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<100>, NumWaitingThreads<7>, NumHighPrioTasks<7>>/8 -0.0029 +0.5760 802313 799998 401474 632716 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<100>, NumWaitingThreads<7>, NumHighPrioTasks<7>>/16 -0.0014 +0.4607 1602184 1600012 877859 1282310 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<100>, NumWaitingThreads<15>, NumHighPrioTasks<7>>/2 -0.0492 +0.3586 212875 202398 100437 136457 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<100>, NumWaitingThreads<15>, NumHighPrioTasks<7>>/4 -0.0927 +0.4432 444857 403606 181089 261350 BM_1_atomic_multi_waiter_1_notifier<NotifyEveryNus<100>, NumWaitingThreads<15>, NumHighPrioTasks<7>>/8 -0.0704 +0.8210 861808 801099 318774 580489 BM_N_atomics_N_waiter_N_notifier<KeepNotifying, NumberOfAtomics<2>, NumHighPrioTasks<0>>/4096 -0.0730 -0.0762 333804 309427 333180 307803 BM_N_atomics_N_waiter_N_notifier<KeepNotifying, NumberOfAtomics<2>, NumHighPrioTasks<0>>/8192 -0.0775 -0.0795 701228 646853 700065 644381 BM_N_atomics_N_waiter_N_notifier<KeepNotifying, NumberOfAtomics<2>, NumHighPrioTasks<0>>/16384 +0.0245 +0.0229 1328777 1361291 1326360 1356745 BM_N_atomics_N_waiter_N_notifier<KeepNotifying, NumberOfAtomics<3>, NumHighPrioTasks<0>>/1024 -0.0541 -0.0562 201559 190662 201259 189940 BM_N_atomics_N_waiter_N_notifier<KeepNotifying, NumberOfAtomics<3>, NumHighPrioTasks<0>>/2048 -0.1959 -0.1986 416092 334584 415412 332927 BM_N_atomics_N_waiter_N_notifier<KeepNotifying, NumberOfAtomics<3>, NumHighPrioTasks<0>>/4096 -0.1699 -0.1710 811966 674040 810157 671584 BM_N_atomics_N_waiter_N_notifier<KeepNotifying, NumberOfAtomics<5>, NumHighPrioTasks<0>>/1024 +0.1383 +0.1301 379893 432426 377756 426885 BM_N_atomics_N_waiter_N_notifier<KeepNotifying, NumberOfAtomics<5>, NumHighPrioTasks<0>>/2048 +0.0396 +0.0339 822384 854937 818110 845866 BM_N_atomics_N_waiter_N_notifier<KeepNotifying, NumberOfAtomics<5>, NumHighPrioTasks<0>>/4096 +0.2499 +0.2451 1350161 1687588 1345121 1674845 BM_N_atomics_N_waiter_N_notifier<KeepNotifying, NumberOfAtomics<7>, NumHighPrioTasks<0>>/256 +0.0042 +0.0101 213598 214487 199282 201303 BM_N_atomics_N_waiter_N_notifier<KeepNotifying, NumberOfAtomics<7>, NumHighPrioTasks<0>>/512 -0.1034 -0.1065 428033 383755 409546 365945 BM_N_atomics_N_waiter_N_notifier<KeepNotifying, NumberOfAtomics<7>, NumHighPrioTasks<0>>/1024 -0.0972 -0.1064 833189 752165 810146 723952 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<50>, NumberOfAtomics<2>, NumHighPrioTasks<0>>/1024 +0.0001 -0.1103 51201684 51204581 51124714 45485867 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<50>, NumberOfAtomics<2>, NumHighPrioTasks<0>>/2048 -0.0000 -0.1202 102409167 102405120 102243857 89953750 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<50>, NumberOfAtomics<2>, NumHighPrioTasks<0>>/4096 +0.0000 -0.1166 204807125 204813833 204453333 180618500 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<50>, NumberOfAtomics<3>, NumHighPrioTasks<0>>/256 +0.0002 -0.1623 12803624 12806161 12778727 10704806 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<50>, NumberOfAtomics<3>, NumHighPrioTasks<0>>/512 -0.0002 -0.1414 25607327 25603223 25551852 21939152 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<50>, NumberOfAtomics<3>, NumHighPrioTasks<0>>/1024 -0.0002 -0.1653 51212196 51202776 51126643 42673625 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<50>, NumberOfAtomics<5>, NumHighPrioTasks<0>>/256 -0.0002 -0.0709 12805016 12802157 12784636 11878785 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<50>, NumberOfAtomics<5>, NumHighPrioTasks<0>>/512 -0.0002 -0.1182 25611565 25606346 25560815 22540033 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<50>, NumberOfAtomics<5>, NumHighPrioTasks<0>>/1024 -0.0002 -0.0813 51220762 51208122 51121071 46963571 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<50>, NumberOfAtomics<7>, NumHighPrioTasks<0>>/64 +0.0012 -0.2125 3219858 3223858 3194027 2515373 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<50>, NumberOfAtomics<7>, NumHighPrioTasks<0>>/128 -0.0370 -0.2643 6668396 6421601 6563402 4828970 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<50>, NumberOfAtomics<7>, NumHighPrioTasks<0>>/256 -0.0288 -0.2220 13220067 12839487 13073964 10172062 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<100>, NumberOfAtomics<2>, NumHighPrioTasks<0>>/256 -0.0000 -0.0105 25602159 25600917 24178138 23923759 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<100>, NumberOfAtomics<2>, NumHighPrioTasks<0>>/512 +0.0000 -0.0175 51201819 51203125 48569867 47718143 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<100>, NumberOfAtomics<2>, NumHighPrioTasks<0>>/1024 +0.0001 -0.0118 102404155 102414482 96908714 95760857 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<100>, NumberOfAtomics<3>, NumHighPrioTasks<0>>/256 +0.0000 -0.0574 25599943 25600621 25326679 23871733 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<100>, NumberOfAtomics<3>, NumHighPrioTasks<0>>/512 +0.0001 -0.0813 51200525 51206978 50459500 46355867 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<100>, NumberOfAtomics<3>, NumHighPrioTasks<0>>/1024 +0.0001 -0.0774 102400405 102409875 101483571 93631000 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<100>, NumberOfAtomics<5>, NumHighPrioTasks<0>>/128 -0.0002 +0.0456 12802792 12800864 11731131 12265881 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<100>, NumberOfAtomics<5>, NumHighPrioTasks<0>>/256 -0.0000 +0.0649 25601667 25601070 22686065 24157862 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<100>, NumberOfAtomics<5>, NumHighPrioTasks<0>>/512 -0.0005 +0.0513 51224453 51200650 45549867 47885533 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<100>, NumberOfAtomics<7>, NumHighPrioTasks<0>>/64 -0.0014 +0.2205 6408711 6400039 4698868 5734934 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<100>, NumberOfAtomics<7>, NumHighPrioTasks<0>>/128 +0.0155 +0.2459 12810413 13009276 9163080 11416117 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<100>, NumberOfAtomics<7>, NumHighPrioTasks<0>>/256 +0.0081 +0.2304 25603646 25811111 1877978 23106867 BM_N_atomics_N_waiter_N_notifier<KeepNotifying, NumberOfAtomics<2>, NumHighPrioTasks<4>>/128 -0.1103 -0.1108 24307 21625 24256 21568 BM_N_atomics_N_waiter_N_notifier<KeepNotifying, NumberOfAtomics<2>, NumHighPrioTasks<4>>/256 +0.0637 +0.0574 45588 48491 45498 48112 BM_N_atomics_N_waiter_N_notifier<KeepNotifying, NumberOfAtomics<2>, NumHighPrioTasks<4>>/512 -0.0519 -0.0539 90764 86054 90527 85648 BM_N_atomics_N_waiter_N_notifier<KeepNotifying, NumberOfAtomics<3>, NumHighPrioTasks<4>>/128 +0.1161 +0.1083 28810 32155 28722 31832 BM_N_atomics_N_waiter_N_notifier<KeepNotifying, NumberOfAtomics<3>, NumHighPrioTasks<4>>/256 +0.1152 +0.1094 64670 72123 64461 71512 BM_N_atomics_N_waiter_N_notifier<KeepNotifying, NumberOfAtomics<3>, NumHighPrioTasks<4>>/512 -0.0804 -0.0993 125916 115796 125476 113010 BM_N_atomics_N_waiter_N_notifier<KeepNotifying, NumberOfAtomics<5>, NumHighPrioTasks<4>>/64 +0.2682 -0.2446 53787 68210 51896 39203 BM_N_atomics_N_waiter_N_notifier<KeepNotifying, NumberOfAtomics<5>, NumHighPrioTasks<4>>/128 +0.5732 -0.4832 103915 163474 100825 52105 BM_N_atomics_N_waiter_N_notifier<KeepNotifying, NumberOfAtomics<5>, NumHighPrioTasks<4>>/256 +0.1283 -0.4606 211518 238645 203852 109957 BM_N_atomics_N_waiter_N_notifier<KeepNotifying, NumberOfAtomics<7>, NumHighPrioTasks<4>>/16 -0.1526 +0.1523 59673 50567 23275 26819 BM_N_atomics_N_waiter_N_notifier<KeepNotifying, NumberOfAtomics<7>, NumHighPrioTasks<4>>/32 -0.0492 +0.7075 82796 78719 24187 41298 BM_N_atomics_N_waiter_N_notifier<KeepNotifying, NumberOfAtomics<7>, NumHighPrioTasks<4>>/64 -0.0712 +0.0764 150268 139570 55304 59527 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<50>, NumberOfAtomics<2>, NumHighPrioTasks<4>>/128 -0.0004 -0.0828 6402859 6400308 6380145 5851557 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<50>, NumberOfAtomics<2>, NumHighPrioTasks<4>>/256 -0.0002 -0.0370 12802978 12801020 12769107 12296293 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<50>, NumberOfAtomics<2>, NumHighPrioTasks<4>>/512 -0.0028 -0.0799 25674170 25601862 25612667 23566586 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<50>, NumberOfAtomics<3>, NumHighPrioTasks<4>>/128 -0.0004 -0.0672 6402990 6400344 6382100 5953248 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<50>, NumberOfAtomics<3>, NumHighPrioTasks<4>>/256 -0.0004 -0.0841 12806197 12801334 12765891 11691661 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<50>, NumberOfAtomics<3>, NumHighPrioTasks<4>>/512 -0.0006 -0.0574 25615708 25601085 25533250 24067828 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<50>, NumberOfAtomics<5>, NumHighPrioTasks<4>>/32 -0.0163 -0.2801 1645647 1618805 1614735 1162471 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<50>, NumberOfAtomics<5>, NumHighPrioTasks<4>>/64 -0.0211 -0.2501 3285234 3216045 3217295 2412509 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<50>, NumberOfAtomics<5>, NumHighPrioTasks<4>>/128 -0.0502 -0.2956 6755976 6416549 6653264 4686407 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<50>, NumberOfAtomics<7>, NumHighPrioTasks<4>>/8 -0.0815 -0.2227 534476 490942 337482 262341 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<50>, NumberOfAtomics<7>, NumHighPrioTasks<4>>/16 +0.0973 -0.0629 1071127 1175390 664897 623053 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<50>, NumberOfAtomics<7>, NumHighPrioTasks<4>>/32 -0.2263 -0.3717 2297477 1777444 1488023 934861 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<100>, NumberOfAtomics<2>, NumHighPrioTasks<4>>/64 -0.0000 +0.0183 6400348 6400261 6145171 6257342 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<100>, NumberOfAtomics<2>, NumHighPrioTasks<4>>/128 +0.0000 +0.0194 12800545 12800759 12279474 12517804 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<100>, NumberOfAtomics<2>, NumHighPrioTasks<4>>/256 +0.0001 +0.0111 25601568 25602976 24636179 24909821 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<100>, NumberOfAtomics<3>, NumHighPrioTasks<4>>/64 -0.0000 +0.0545 6400600 6400444 5795288 6111077 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<100>, NumberOfAtomics<3>, NumHighPrioTasks<4>>/128 +0.0001 +0.0474 12800507 12801355 11566729 12114860 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<100>, NumberOfAtomics<3>, NumHighPrioTasks<4>>/256 +0.0000 +0.0423 25601503 25601760 23281967 24267276 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<100>, NumberOfAtomics<5>, NumHighPrioTasks<4>>/32 +0.0005 +0.2842 3201968 3203421 2175379 2793615 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<100>, NumberOfAtomics<5>, NumHighPrioTasks<4>>/64 -0.0003 +0.3807 6402555 6400496 4052465 5595309 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<100>, NumberOfAtomics<5>, NumHighPrioTasks<4>>/128 -0.0003 +0.3827 12804155 12800925 8114370 11219400 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<100>, NumberOfAtomics<7>, NumHighPrioTasks<4>>/8 +0.0262 +0.1272 821954 843475 503297 567320 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<100>, NumberOfAtomics<7>, NumHighPrioTasks<4>>/16 +0.0298 +0.3134 1634476 1683172 901978 1184619 BM_N_atomics_N_waiter_N_notifier<NotifyEveryNus<100>, NumberOfAtomics<7>, NumHighPrioTasks<4>>/32 +0.0147 +0.2925 3244262 3291994 1721000 2224350 OVERALL_GEOMEAN +0.0185 -0.1876 0 0 0 0 ``` --------- Co-authored-by: Hui Xie <[email protected]> Co-authored-by: Hui Xie <[email protected]>
…the other way around. NFC (llvm#123452) It's better to cast away constness on the reference being returned than to cast away constness on the this pointer.
…lvm#123432) This generates basically the same code with the operands commuted, but gets there with less legalization steps.
…return a range. NFC (llvm#123453)
`m_c_XorLike` matches either: `(xor L, R)`, `(xor R, L)`, or `(sub nuw R, L)` iff `R.isMask()`. This is in preperation for dropping the fold from: `(sub C_Mask, X)` -> `(xor X, C_Mask)`
Note that PointerUnion::dyn_cast has been soft deprecated in PointerUnion.h: // FIXME: Replace the uses of is(), get() and dyn_cast() with // isa<T>, cast<T> and the llvm::dyn_cast<T> Literal migration would result in dyn_cast_if_present (see the definition of PointerUnion::dyn_cast), but this patch uses dyn_cast because we expect ValueOrInherited to be nonnull. Note that isSet checks to see if ValueOrInherited is nonnull.
Note that PointerUnion::dyn_cast has been soft deprecated in PointerUnion.h: // FIXME: Replace the uses of is(), get() and dyn_cast() with // isa<T>, cast<T> and the llvm::dyn_cast<T> Literal migration would result in dyn_cast_if_present (see the definition of PointerUnion::dyn_cast), but this patch uses dyn_cast because we expect D to be nonnull.
…larations with explicit `this`" (llvm#123455) This reverts commit c3a935e. The only change to the reverted commit is that this also updates the OCaml bindings according to the C debug-info API changes. The build failure originally introduced was: ``` FAILED: bindings/ocaml/debuginfo/debuginfo_ocaml.o /b/1/llvm-clang-x86_64-expensive-checks-debian/build/bindings/ocaml/debuginfo/debuginfo_ocaml.o cd /b/1/llvm-clang-x86_64-expensive-checks-debian/build/bindings/ocaml/debuginfo && /usr/bin/ocamlfind ocamlc -c /b/1/llvm-clang-x86_64-expensive-checks-debian/build/bindings/ocaml/debuginfo/debuginfo_ocaml.c -ccopt "-I/b/1/llvm-clang-x86_64-expensive-checks-debian/llvm-project/llvm/bindings/ocaml/debuginfo/../llvm -D_GNU_SOURCE -D_DEBUG -D_GLIBCXX_ASSERTIONS -DEXPENSIVE_CHECKS -D_GLIBCXX_DEBUG -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -D__STDC_LIMIT_MACROS -I/b/1/llvm-clang-x86_64-expensive-checks-debian/build/include -I/b/1/llvm-clang-x86_64-expensive-checks-debian/llvm-project/llvm/include -DNDEBUG " /b/1/llvm-clang-x86_64-expensive-checks-debian/build/bindings/ocaml/debuginfo/debuginfo_ocaml.c: In function ‘llvm_dibuild_create_object_pointer_type’: /b/1/llvm-clang-x86_64-expensive-checks-debian/build/bindings/ocaml/debuginfo/debuginfo_ocaml.c:620:30: error: too few arguments to function ‘LLVMDIBuilderCreateObjectPointerType’ 620 | LLVMMetadataRef Metadata = LLVMDIBuilderCreateObjectPointerType( | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ In file included from /b/1/llvm-clang-x86_64-expensive-checks-debian/build/bindings/ocaml/debuginfo/debuginfo_ocaml.c:23: /b/1/llvm-clang-x86_64-expensive-checks-debian/llvm-project/llvm/include/llvm-c/DebugInfo.h:880:17: note: declared here 880 | LLVMMetadataRef LLVMDIBuilderCreateObjectPointerType(LLVMDIBuilderRef Builder, | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ```
…he test suite (llvm#123464) In preparation of making `-Wreturn-type` default to an error (as there is virtually no situation where you’d *want* to fall off the end of a function that is supposed to return a value), this patch fixes tests that have relied on this being only a warning, of which there seem to be 3 kinds: 1. Tests which for no apparent reason have a function that triggers the warning. I suspect that a lot of these were on accident (or from before the warning was introduced), since a lot of people will open issues w/ their problematic code in the `main` function (which is the one case where you don’t need to return from a non-void function, after all...), which someone will then copy, possibly into a namespace, possibly renaming it, the end result of that being that you end up w/ something that definitely is not `main` anymore, but which still is declared as returning `int`, and which still has no return statement (another reason why I think this might apply to a lot of these is because usually the actual return type of such problematic functions is quite literally `int`). A lot of these are really old tests that don’t use `-verify`, which is why no-one noticed or had to care about the extra warning that was already being emitted by them until now. 2. Tests which test either `-Wreturn-type`, `[[noreturn]]`, or what codegen and sanitisers do whenever you do fall off the end of a function. 3. Tests where I struggle to figure out what is even being tested (usually because they’re Objective-C tests, and I don’t know Objective-C), whether falling off the end of a function matters in the first place, and tests where actually spelling out an expression to return would be rather cumbersome (e.g. matrix types currently don’t support list initialisation, so I can’t write e.g. `return {}`). For tests that fall into categories 2 and 3, I just added `-Wno-error=return-type` to the `RUN` lines and called it a day. This was especially necessary for the former since `-Wreturn-type` is an analysis-based warning, meaning that it is currently impossible to test for more than one occurrence of it in the same compilation if it defaults to an error since the analysis pass is skipped for subsequent functions as soon as an error is emitted. I’ve also added `-Werror=return-type` to a few tests that I had already updated as this patch was previously already making the warning an error by default, but we’ve decided to split that into two patches instead.
Co-authored-by: Nikita Popov <[email protected]>
ConstantRange's constructor asserts that the range not empty, except if lower/upper are min or max values. Check if the length is strictly positive instead of just non-negative so std::nullopt is returned when the size is 0. If that's the case, the access doesn't initialize anything. This should fix a crash when building on macOS with ASan & UBsan after llvm#97373 / llvm#117104 landed: https://green.lab.llvm.org/job/llvm.org/job/clang-stage2-cmake-RgSan/664/console
This allows optimization of more signed floor implementations when the divisor is a known power of two to an arithmetic shift. Proof for the implemented optimizations: https://alive2.llvm.org/ce/z/j6C-Nz Proof for the test cases: https://alive2.llvm.org/ce/z/M_PBjw --------- Co-authored-by: Jacob Young <[email protected]>
This PR extends the existing poison support in https://mlir.llvm.org/docs/Dialects/UBOps/ by representing poison mask values in `vector.shuffle`. Similar to LLVM (see https://github.com/llvm/llvm-project/blob/main/llvm/include/llvm/IR/Instructions.h#L1884) this requires defining an integer value (`-1`) to represent poison in the `vector.shuffle` mask.
…dentation of `export { ... }` (llvm#110381) `export { ... }` blocks can get a bit long, so I thought it would make sense to have an option that makes it so their contents are not indented (basically the same argument as for namespaces). This is based on the `NamespaceIndentation` option, except that there is no option to control the behaviour of `export` blocks when nested because nesting them doesn’t really make sense. Additionally, brace wrapping of short `export { ... }` blocks is now controlled by the `AllowShortBlocksOnASingleLine` option. There is no separate option just for `export` blocks because you can just write e.g. `export int x;` instead of `export { int x; }`. This closes llvm#121723.
…x targets. (llvm#123449) Mark the `offset_range` test as UNSUPPORTED for the `armv7-unknown-linux-gnueabihf` target (32-bit). Ref PR llvm#122798
In the unlikely case where the stack size is greater than 4GB, we may run into the situation where the local stack size and the callee saved registers stack size get combined incorrectly when restoring the callee saved registers. This happens because the stack size in shouldCombineCSRLocalStackBumpInEpilogue is represented as an 'unsigned', but is passed in as an 'int64_t'. We end up with something like $fp, $lr = frame-destroy LDPXi $sp, 536870912 This change just makes 'shouldCombineCSRLocalStackBumpInEpilogue' match 'shouldCombineCSRLocalStackBump' where 'StackBumpBytes' is an 'uint64_t'
Tests for crash caused by initial version of llvm#112147.
This reverts the revert commit 58326f1. The build failure in sanitizer stage2 builds has been fixed with 0d39fe6. Original commit message: Model updating IV users directly in VPlan, replace fixupIVUsers. Now simple extracts are created for all phis in the exit block during initial VPlan construction. A later VPlan transform (optimizeInductionExitUsers) replaces extracts of inductions with their pre-computed values if possible. This completes the transition towards modeling all live-outs directly in VPlan. There are a few follow-ups: * emit extracts initially also for resume phis, and optimize them tougher with IV exit users * support for VPlans with multiple exits in optimizeInductionExitUsers. Depends on llvm#110004, llvm#109975 and llvm#112145.
Handle the one unittest executable that I've missed in llvm#120966.
In order to meaningfully generate getters and setters from IRDL, it makes sense to embed the names of operands, results, etc. in the IR definition. This PR introduces this feature. Names are constrained similarly to TableGen names.
…y using(). NFC Many of these are indexing BitVectors or something where we can't using MCRegister and need the register number.
This commit promotes the SPIR-V backend from experimental to official status. As a result, SPIR-V will be built by default, simplifying integration and increasing accessibility for downstream projects. Discussion and RFC on Discourse: https://discourse.llvm.org/t/rfc-promoting-spir-v-to-an-official-target/83614
This is needed as of 6301394 for MachO::getCPUType and MachO::getCPUSubType.
This commit changes the branch emission logic so that instead of throwing the "branch target out of range" error, we emit a relocation instead.
… relocs (llvm#121330) If linker relaxation enabled, relaxable code sequence expanded from pseudos should avoid being separated by instruction scheduling. This commit tags scheduling boundary for them to avoid being scheduled. (Except for `tls_le/tls_ie` and `call36/tail36`. Because `tls_le/tls_ie` can be scheduled and have no influence to relax, `call36/tail36` are expanded later in `LoongArchExpandPseudo` pass.) A new mask target-flag is added to attach relax relocs to the relaxable code sequence. (No need to add it for `tls_le` and `call36/tail36` because we can simply add relax relocs for them according to their relocs. But for other code sequence, such as `PCALA_{HI20/LO12}`, we must use the mask flag, mainly because relax should not be added when code model is large.) Because of the new mask target-flag, get "direct" flags is necessary when using their target-flags. In addition, code sequence after being optimized by `MergeBaseOffset` pass may not relaxable any more, so the relax "bitmask" flag should be removed.
When we did the initial AMDGCNSPIRV commits we left the initialisation of the feature map in a relatively disorderly state. This change corrects that oversight: - We make sure that AMDGCNSPIRV actually advertises the union of all AMDGCN features, as some were not included; - We keep feature initialisation in sorted order to make it easy to pick an insertion point when features are added in the future.
This functionality is needed for inliner pass and also for correct dyno stats. Needed for [PR](llvm#120187)
FreeBSD going forward will not have 32-bit arch support. Also missed a spot with removing riscv32 support.
`void InlinerPass::runOnOperation()` doesn't need to have `return;` at the end of the procedure.
) Add the required IMAGE_REL_MIPS_PAIR relocation after IMAGE_REL_MIPS_REFHI/IMAGE_REL_MIPS_SECRELHI Microsoft PE/COFF specification says that the IMAGE_REL_MIPS_REFHI relocation contains "the high 16 bits of the target's 32-bit virtual address. [...] This relocation must be immediately followed by a PAIR relocation whose SymbolTableIndex contains a 16-bit displacement which is added to the upper 16 bits taken from the location being relocated." Microsoft PE/COFF specification says that the IMAGE_REL_MIPS_SECRELHI relocation contains "the high 16 bits of the 32-bit offset of the target from the beginning of its section. A PAIR relocation must immediately follow this on. The SymbolTableIndex of the PAIR relocation contains a 16-bit displacement, which is added to the upper 16 bits taken from the location being relocated." Behavior has been checked against Microsoft C compiler for MIPS.
Also add support for new relocation types required by debug information. Constants have been taken from CodeView Symbolic Debug Information Specification.
Now that we have it, use it.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.