1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
|
//===--- Cuda.cpp - Cuda Tool and ToolChain Implementations -----*- C++ -*-===//
//
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
// See https://llvm.org/LICENSE.txt for license information.
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
//
//===----------------------------------------------------------------------===//
#include "Cuda.h"
#include "CommonArgs.h"
#include "clang/Basic/Cuda.h"
#include "clang/Config/config.h"
#include "clang/Driver/Compilation.h"
#include "clang/Driver/Distro.h"
#include "clang/Driver/Driver.h"
#include "clang/Driver/DriverDiagnostic.h"
#include "clang/Driver/InputInfo.h"
#include "clang/Driver/Options.h"
#include "llvm/ADT/StringExtras.h"
#include "llvm/Option/ArgList.h"
#include "llvm/Support/FileSystem.h"
#include "llvm/Support/FormatAdapters.h"
#include "llvm/Support/FormatVariadic.h"
#include "llvm/Support/Path.h"
#include "llvm/Support/Process.h"
#include "llvm/Support/Program.h"
#include "llvm/Support/VirtualFileSystem.h"
#include "llvm/TargetParser/Host.h"
#include "llvm/TargetParser/TargetParser.h"
#include <system_error>
using namespace clang::driver;
using namespace clang::driver::toolchains;
using namespace clang::driver::tools;
using namespace clang;
using namespace llvm::opt;
namespace {
CudaVersion getCudaVersion(uint32_t raw_version) {
if (raw_version < 7050)
return CudaVersion::CUDA_70;
if (raw_version < 8000)
return CudaVersion::CUDA_75;
if (raw_version < 9000)
return CudaVersion::CUDA_80;
if (raw_version < 9010)
return CudaVersion::CUDA_90;
if (raw_version < 9020)
return CudaVersion::CUDA_91;
if (raw_version < 10000)
return CudaVersion::CUDA_92;
if (raw_version < 10010)
return CudaVersion::CUDA_100;
if (raw_version < 10020)
return CudaVersion::CUDA_101;
if (raw_version < 11000)
return CudaVersion::CUDA_102;
if (raw_version < 11010)
return CudaVersion::CUDA_110;
if (raw_version < 11020)
return CudaVersion::CUDA_111;
if (raw_version < 11030)
return CudaVersion::CUDA_112;
if (raw_version < 11040)
return CudaVersion::CUDA_113;
if (raw_version < 11050)
return CudaVersion::CUDA_114;
if (raw_version < 11060)
return CudaVersion::CUDA_115;
if (raw_version < 11070)
return CudaVersion::CUDA_116;
if (raw_version < 11080)
return CudaVersion::CUDA_117;
if (raw_version < 11090)
return CudaVersion::CUDA_118;
if (raw_version < 12010)
return CudaVersion::CUDA_120;
if (raw_version < 12020)
return CudaVersion::CUDA_121;
if (raw_version < 12030)
return CudaVersion::CUDA_122;
if (raw_version < 12040)
return CudaVersion::CUDA_123;
return CudaVersion::NEW;
}
CudaVersion parseCudaHFile(llvm::StringRef Input) {
// Helper lambda which skips the words if the line starts with them or returns
// std::nullopt otherwise.
auto StartsWithWords =
[](llvm::StringRef Line,
const SmallVector<StringRef, 3> words) -> std::optional<StringRef> {
for (StringRef word : words) {
if (!Line.consume_front(word))
return {};
Line = Line.ltrim();
}
return Line;
};
Input = Input.ltrim();
while (!Input.empty()) {
if (auto Line =
StartsWithWords(Input.ltrim(), {"#", "define", "CUDA_VERSION"})) {
uint32_t RawVersion;
Line->consumeInteger(10, RawVersion);
return getCudaVersion(RawVersion);
}
// Find next non-empty line.
Input = Input.drop_front(Input.find_first_of("\n\r")).ltrim();
}
return CudaVersion::UNKNOWN;
}
} // namespace
void CudaInstallationDetector::WarnIfUnsupportedVersion() {
if (Version > CudaVersion::PARTIALLY_SUPPORTED) {
std::string VersionString = CudaVersionToString(Version);
if (!VersionString.empty())
VersionString.insert(0, " ");
D.Diag(diag::warn_drv_new_cuda_version)
<< VersionString
<< (CudaVersion::PARTIALLY_SUPPORTED != CudaVersion::FULLY_SUPPORTED)
<< CudaVersionToString(CudaVersion::PARTIALLY_SUPPORTED);
} else if (Version > CudaVersion::FULLY_SUPPORTED)
D.Diag(diag::warn_drv_partially_supported_cuda_version)
<< CudaVersionToString(Version);
}
CudaInstallationDetector::CudaInstallationDetector(
const Driver &D, const llvm::Triple &HostTriple,
const llvm::opt::ArgList &Args)
: D(D) {
struct Candidate {
std::string Path;
bool StrictChecking;
Candidate(std::string Path, bool StrictChecking = false)
: Path(Path), StrictChecking(StrictChecking) {}
};
SmallVector<Candidate, 4> Candidates;
// In decreasing order so we prefer newer versions to older versions.
std::initializer_list<const char *> Versions = {"8.0", "7.5", "7.0"};
auto &FS = D.getVFS();
if (Args.hasArg(clang::driver::options::OPT_cuda_path_EQ)) {
Candidates.emplace_back(
Args.getLastArgValue(clang::driver::options::OPT_cuda_path_EQ).str());
} else if (HostTriple.isOSWindows()) {
for (const char *Ver : Versions)
Candidates.emplace_back(
D.SysRoot + "/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v" +
Ver);
} else {
if (!Args.hasArg(clang::driver::options::OPT_cuda_path_ignore_env)) {
// Try to find ptxas binary. If the executable is located in a directory
// called 'bin/', its parent directory might be a good guess for a valid
// CUDA installation.
// However, some distributions might installs 'ptxas' to /usr/bin. In that
// case the candidate would be '/usr' which passes the following checks
// because '/usr/include' exists as well. To avoid this case, we always
// check for the directory potentially containing files for libdevice,
// even if the user passes -nocudalib.
if (llvm::ErrorOr<std::string> ptxas =
llvm::sys::findProgramByName("ptxas")) {
SmallString<256> ptxasAbsolutePath;
llvm::sys::fs::real_path(*ptxas, ptxasAbsolutePath);
StringRef ptxasDir = llvm::sys::path::parent_path(ptxasAbsolutePath);
if (llvm::sys::path::filename(ptxasDir) == "bin")
Candidates.emplace_back(
std::string(llvm::sys::path::parent_path(ptxasDir)),
/*StrictChecking=*/true);
}
}
Candidates.emplace_back(D.SysRoot + "/usr/local/cuda");
for (const char *Ver : Versions)
Candidates.emplace_back(D.SysRoot + "/usr/local/cuda-" + Ver);
Distro Dist(FS, llvm::Triple(llvm::sys::getProcessTriple()));
if (Dist.IsDebian() || Dist.IsUbuntu())
// Special case for Debian to have nvidia-cuda-toolkit work
// out of the box. More info on http://bugs.debian.org/882505
Candidates.emplace_back(D.SysRoot + "/usr/lib/cuda");
}
bool NoCudaLib = Args.hasArg(options::OPT_nogpulib);
for (const auto &Candidate : Candidates) {
InstallPath = Candidate.Path;
if (InstallPath.empty() || !FS.exists(InstallPath))
continue;
BinPath = InstallPath + "/bin";
IncludePath = InstallPath + "/include";
LibDevicePath = InstallPath + "/nvvm/libdevice";
if (!(FS.exists(IncludePath) && FS.exists(BinPath)))
continue;
bool CheckLibDevice = (!NoCudaLib || Candidate.StrictChecking);
if (CheckLibDevice && !FS.exists(LibDevicePath))
continue;
Version = CudaVersion::UNKNOWN;
if (auto CudaHFile = FS.getBufferForFile(InstallPath + "/include/cuda.h"))
Version = parseCudaHFile((*CudaHFile)->getBuffer());
// As the last resort, make an educated guess between CUDA-7.0, which had
// old-style libdevice bitcode, and an unknown recent CUDA version.
if (Version == CudaVersion::UNKNOWN) {
Version = FS.exists(LibDevicePath + "/libdevice.10.bc")
? CudaVersion::NEW
: CudaVersion::CUDA_70;
}
if (Version >= CudaVersion::CUDA_90) {
// CUDA-9+ uses single libdevice file for all GPU variants.
std::string FilePath = LibDevicePath + "/libdevice.10.bc";
if (FS.exists(FilePath)) {
for (int Arch = (int)CudaArch::SM_30, E = (int)CudaArch::LAST; Arch < E;
++Arch) {
CudaArch GpuArch = static_cast<CudaArch>(Arch);
if (!IsNVIDIAGpuArch(GpuArch))
continue;
std::string GpuArchName(CudaArchToString(GpuArch));
LibDeviceMap[GpuArchName] = FilePath;
}
}
} else {
std::error_code EC;
for (llvm::vfs::directory_iterator LI = FS.dir_begin(LibDevicePath, EC),
LE;
!EC && LI != LE; LI = LI.increment(EC)) {
StringRef FilePath = LI->path();
StringRef FileName = llvm::sys::path::filename(FilePath);
// Process all bitcode filenames that look like
// libdevice.compute_XX.YY.bc
const StringRef LibDeviceName = "libdevice.";
if (!(FileName.starts_with(LibDeviceName) && FileName.ends_with(".bc")))
continue;
StringRef GpuArch = FileName.slice(
LibDeviceName.size(), FileName.find('.', LibDeviceName.size()));
LibDeviceMap[GpuArch] = FilePath.str();
// Insert map entries for specific devices with this compute
// capability. NVCC's choice of the libdevice library version is
// rather peculiar and depends on the CUDA version.
if (GpuArch == "compute_20") {
LibDeviceMap["sm_20"] = std::string(FilePath);
LibDeviceMap["sm_21"] = std::string(FilePath);
LibDeviceMap["sm_32"] = std::string(FilePath);
} else if (GpuArch == "compute_30") {
LibDeviceMap["sm_30"] = std::string(FilePath);
if (Version < CudaVersion::CUDA_80) {
LibDeviceMap["sm_50"] = std::string(FilePath);
LibDeviceMap["sm_52"] = std::string(FilePath);
LibDeviceMap["sm_53"] = std::string(FilePath);
}
LibDeviceMap["sm_60"] = std::string(FilePath);
LibDeviceMap["sm_61"] = std::string(FilePath);
LibDeviceMap["sm_62"] = std::string(FilePath);
} else if (GpuArch == "compute_35") {
LibDeviceMap["sm_35"] = std::string(FilePath);
LibDeviceMap["sm_37"] = std::string(FilePath);
} else if (GpuArch == "compute_50") {
if (Version >= CudaVersion::CUDA_80) {
LibDeviceMap["sm_50"] = std::string(FilePath);
LibDeviceMap["sm_52"] = std::string(FilePath);
LibDeviceMap["sm_53"] = std::string(FilePath);
}
}
}
}
// Check that we have found at least one libdevice that we can link in if
// -nocudalib hasn't been specified.
if (LibDeviceMap.empty() && !NoCudaLib)
continue;
IsValid = true;
break;
}
}
void CudaInstallationDetector::AddCudaIncludeArgs(
const ArgList &DriverArgs, ArgStringList &CC1Args) const {
if (!DriverArgs.hasArg(options::OPT_nobuiltininc)) {
// Add cuda_wrappers/* to our system include path. This lets us wrap
// standard library headers.
SmallString<128> P(D.ResourceDir);
llvm::sys::path::append(P, "include");
llvm::sys::path::append(P, "cuda_wrappers");
CC1Args.push_back("-internal-isystem");
CC1Args.push_back(DriverArgs.MakeArgString(P));
}
if (DriverArgs.hasArg(options::OPT_nogpuinc))
return;
if (!isValid()) {
D.Diag(diag::err_drv_no_cuda_installation);
return;
}
CC1Args.push_back("-include");
CC1Args.push_back("__clang_cuda_runtime_wrapper.h");
}
void CudaInstallationDetector::CheckCudaVersionSupportsArch(
CudaArch Arch) const {
if (Arch == CudaArch::UNKNOWN || Version == CudaVersion::UNKNOWN ||
ArchsWithBadVersion[(int)Arch])
return;
auto MinVersion = MinVersionForCudaArch(Arch);
auto MaxVersion = MaxVersionForCudaArch(Arch);
if (Version < MinVersion || Version > MaxVersion) {
ArchsWithBadVersion[(int)Arch] = true;
D.Diag(diag::err_drv_cuda_version_unsupported)
<< CudaArchToString(Arch) << CudaVersionToString(MinVersion)
<< CudaVersionToString(MaxVersion) << InstallPath
<< CudaVersionToString(Version);
}
}
void CudaInstallationDetector::print(raw_ostream &OS) const {
if (isValid())
OS << "Found CUDA installation: " << InstallPath << ", version "
<< CudaVersionToString(Version) << "\n";
}
namespace {
/// Debug info level for the NVPTX devices. We may need to emit different debug
/// info level for the host and for the device itselfi. This type controls
/// emission of the debug info for the devices. It either prohibits disable info
/// emission completely, or emits debug directives only, or emits same debug
/// info as for the host.
enum DeviceDebugInfoLevel {
DisableDebugInfo, /// Do not emit debug info for the devices.
DebugDirectivesOnly, /// Emit only debug directives.
EmitSameDebugInfoAsHost, /// Use the same debug info level just like for the
/// host.
};
} // anonymous namespace
/// Define debug info level for the NVPTX devices. If the debug info for both
/// the host and device are disabled (-g0/-ggdb0 or no debug options at all). If
/// only debug directives are requested for the both host and device
/// (-gline-directvies-only), or the debug info only for the device is disabled
/// (optimization is on and --cuda-noopt-device-debug was not specified), the
/// debug directves only must be emitted for the device. Otherwise, use the same
/// debug info level just like for the host (with the limitations of only
/// supported DWARF2 standard).
static DeviceDebugInfoLevel mustEmitDebugInfo(const ArgList &Args) {
const Arg *A = Args.getLastArg(options::OPT_O_Group);
bool IsDebugEnabled = !A || A->getOption().matches(options::OPT_O0) ||
Args.hasFlag(options::OPT_cuda_noopt_device_debug,
options::OPT_no_cuda_noopt_device_debug,
/*Default=*/false);
if (const Arg *A = Args.getLastArg(options::OPT_g_Group)) {
const Option &Opt = A->getOption();
if (Opt.matches(options::OPT_gN_Group)) {
if (Opt.matches(options::OPT_g0) || Opt.matches(options::OPT_ggdb0))
return DisableDebugInfo;
if (Opt.matches(options::OPT_gline_directives_only))
return DebugDirectivesOnly;
}
return IsDebugEnabled ? EmitSameDebugInfoAsHost : DebugDirectivesOnly;
}
return willEmitRemarks(Args) ? DebugDirectivesOnly : DisableDebugInfo;
}
void NVPTX::Assembler::ConstructJob(Compilation &C, const JobAction &JA,
const InputInfo &Output,
const InputInfoList &Inputs,
const ArgList &Args,
const char *LinkingOutput) const {
const auto &TC =
static_cast<const toolchains::NVPTXToolChain &>(getToolChain());
assert(TC.getTriple().isNVPTX() && "Wrong platform");
StringRef GPUArchName;
// If this is a CUDA action we need to extract the device architecture
// from the Job's associated architecture, otherwise use the -march=arch
// option. This option may come from -Xopenmp-target flag or the default
// value.
if (JA.isDeviceOffloading(Action::OFK_Cuda)) {
GPUArchName = JA.getOffloadingArch();
} else {
GPUArchName = Args.getLastArgValue(options::OPT_march_EQ);
if (GPUArchName.empty()) {
C.getDriver().Diag(diag::err_drv_offload_missing_gpu_arch)
<< getToolChain().getArchName() << getShortName();
return;
}
}
// Obtain architecture from the action.
CudaArch gpu_arch = StringToCudaArch(GPUArchName);
assert(gpu_arch != CudaArch::UNKNOWN &&
"Device action expected to have an architecture.");
// Check that our installation's ptxas supports gpu_arch.
if (!Args.hasArg(options::OPT_no_cuda_version_check)) {
TC.CudaInstallation.CheckCudaVersionSupportsArch(gpu_arch);
}
ArgStringList CmdArgs;
CmdArgs.push_back(TC.getTriple().isArch64Bit() ? "-m64" : "-m32");
DeviceDebugInfoLevel DIKind = mustEmitDebugInfo(Args);
if (DIKind == EmitSameDebugInfoAsHost) {
// ptxas does not accept -g option if optimization is enabled, so
// we ignore the compiler's -O* options if we want debug info.
CmdArgs.push_back("-g");
CmdArgs.push_back("--dont-merge-basicblocks");
CmdArgs.push_back("--return-at-end");
} else if (Arg *A = Args.getLastArg(options::OPT_O_Group)) {
// Map the -O we received to -O{0,1,2,3}.
//
// TODO: Perhaps we should map host -O2 to ptxas -O3. -O3 is ptxas's
// default, so it may correspond more closely to the spirit of clang -O2.
// -O3 seems like the least-bad option when -Osomething is specified to
// clang but it isn't handled below.
StringRef OOpt = "3";
if (A->getOption().matches(options::OPT_O4) ||
A->getOption().matches(options::OPT_Ofast))
OOpt = "3";
else if (A->getOption().matches(options::OPT_O0))
OOpt = "0";
else if (A->getOption().matches(options::OPT_O)) {
// -Os, -Oz, and -O(anything else) map to -O2, for lack of better options.
OOpt = llvm::StringSwitch<const char *>(A->getValue())
.Case("1", "1")
.Case("2", "2")
.Case("3", "3")
.Case("s", "2")
.Case("z", "2")
.Default("2");
}
CmdArgs.push_back(Args.MakeArgString(llvm::Twine("-O") + OOpt));
} else {
// If no -O was passed, pass -O0 to ptxas -- no opt flag should correspond
// to no optimizations, but ptxas's default is -O3.
CmdArgs.push_back("-O0");
}
if (DIKind == DebugDirectivesOnly)
CmdArgs.push_back("-lineinfo");
// Pass -v to ptxas if it was passed to the driver.
if (Args.hasArg(options::OPT_v))
CmdArgs.push_back("-v");
CmdArgs.push_back("--gpu-name");
CmdArgs.push_back(Args.MakeArgString(CudaArchToString(gpu_arch)));
CmdArgs.push_back("--output-file");
std::string OutputFileName = TC.getInputFilename(Output);
// If we are invoking `nvlink` internally we need to output a `.cubin` file.
// FIXME: This should hopefully be removed if NVIDIA updates their tooling.
if (!C.getInputArgs().getLastArg(options::OPT_c)) {
SmallString<256> Filename(Output.getFilename());
llvm::sys::path::replace_extension(Filename, "cubin");
OutputFileName = Filename.str();
}
if (Output.isFilename() && OutputFileName != Output.getFilename())
C.addTempFile(Args.MakeArgString(OutputFileName));
CmdArgs.push_back(Args.MakeArgString(OutputFileName));
for (const auto &II : Inputs)
CmdArgs.push_back(Args.MakeArgString(II.getFilename()));
for (const auto &A : Args.getAllArgValues(options::OPT_Xcuda_ptxas))
CmdArgs.push_back(Args.MakeArgString(A));
bool Relocatable;
if (JA.isOffloading(Action::OFK_OpenMP))
// In OpenMP we need to generate relocatable code.
Relocatable = Args.hasFlag(options::OPT_fopenmp_relocatable_target,
options::OPT_fnoopenmp_relocatable_target,
/*Default=*/true);
else if (JA.isOffloading(Action::OFK_Cuda))
// In CUDA we generate relocatable code by default.
Relocatable = Args.hasFlag(options::OPT_fgpu_rdc, options::OPT_fno_gpu_rdc,
/*Default=*/false);
else
// Otherwise, we are compiling directly and should create linkable output.
Relocatable = true;
if (Relocatable)
CmdArgs.push_back("-c");
const char *Exec;
if (Arg *A = Args.getLastArg(options::OPT_ptxas_path_EQ))
Exec = A->getValue();
else
Exec = Args.MakeArgString(TC.GetProgramPath("ptxas"));
C.addCommand(std::make_unique<Command>(
JA, *this,
ResponseFileSupport{ResponseFileSupport::RF_Full, llvm::sys::WEM_UTF8,
"--options-file"},
Exec, CmdArgs, Inputs, Output));
}
static bool shouldIncludePTX(const ArgList &Args, const char *gpu_arch) {
bool includePTX = true;
for (Arg *A : Args) {
if (!(A->getOption().matches(options::OPT_cuda_include_ptx_EQ) ||
A->getOption().matches(options::OPT_no_cuda_include_ptx_EQ)))
continue;
A->claim();
const StringRef ArchStr = A->getValue();
if (ArchStr == "all" || ArchStr == gpu_arch) {
includePTX = A->getOption().matches(options::OPT_cuda_include_ptx_EQ);
continue;
}
}
return includePTX;
}
// All inputs to this linker must be from CudaDeviceActions, as we need to look
// at the Inputs' Actions in order to figure out which GPU architecture they
// correspond to.
void NVPTX::FatBinary::ConstructJob(Compilation &C, const JobAction &JA,
const InputInfo &Output,
const InputInfoList &Inputs,
const ArgList &Args,
const char *LinkingOutput) const {
const auto &TC =
static_cast<const toolchains::CudaToolChain &>(getToolChain());
assert(TC.getTriple().isNVPTX() && "Wrong platform");
ArgStringList CmdArgs;
if (TC.CudaInstallation.version() <= CudaVersion::CUDA_100)
CmdArgs.push_back("--cuda");
CmdArgs.push_back(TC.getTriple().isArch64Bit() ? "-64" : "-32");
CmdArgs.push_back(Args.MakeArgString("--create"));
CmdArgs.push_back(Args.MakeArgString(Output.getFilename()));
if (mustEmitDebugInfo(Args) == EmitSameDebugInfoAsHost)
CmdArgs.push_back("-g");
for (const auto &II : Inputs) {
auto *A = II.getAction();
assert(A->getInputs().size() == 1 &&
"Device offload action is expected to have a single input");
const char *gpu_arch_str = A->getOffloadingArch();
assert(gpu_arch_str &&
"Device action expected to have associated a GPU architecture!");
CudaArch gpu_arch = StringToCudaArch(gpu_arch_str);
if (II.getType() == types::TY_PP_Asm &&
!shouldIncludePTX(Args, gpu_arch_str))
continue;
// We need to pass an Arch of the form "sm_XX" for cubin files and
// "compute_XX" for ptx.
const char *Arch = (II.getType() == types::TY_PP_Asm)
? CudaArchToVirtualArchString(gpu_arch)
: gpu_arch_str;
CmdArgs.push_back(
Args.MakeArgString(llvm::Twine("--image=profile=") + Arch +
",file=" + getToolChain().getInputFilename(II)));
}
for (const auto &A : Args.getAllArgValues(options::OPT_Xcuda_fatbinary))
CmdArgs.push_back(Args.MakeArgString(A));
const char *Exec = Args.MakeArgString(TC.GetProgramPath("fatbinary"));
C.addCommand(std::make_unique<Command>(
JA, *this,
ResponseFileSupport{ResponseFileSupport::RF_Full, llvm::sys::WEM_UTF8,
"--options-file"},
Exec, CmdArgs, Inputs, Output));
}
void NVPTX::Linker::ConstructJob(Compilation &C, const JobAction &JA,
const InputInfo &Output,
const InputInfoList &Inputs,
const ArgList &Args,
const char *LinkingOutput) const {
const auto &TC =
static_cast<const toolchains::NVPTXToolChain &>(getToolChain());
ArgStringList CmdArgs;
assert(TC.getTriple().isNVPTX() && "Wrong platform");
assert((Output.isFilename() || Output.isNothing()) && "Invalid output.");
if (Output.isFilename()) {
CmdArgs.push_back("-o");
CmdArgs.push_back(Output.getFilename());
}
if (mustEmitDebugInfo(Args) == EmitSameDebugInfoAsHost)
CmdArgs.push_back("-g");
if (Args.hasArg(options::OPT_v))
CmdArgs.push_back("-v");
StringRef GPUArch = Args.getLastArgValue(options::OPT_march_EQ);
if (GPUArch.empty()) {
C.getDriver().Diag(diag::err_drv_offload_missing_gpu_arch)
<< getToolChain().getArchName() << getShortName();
return;
}
CmdArgs.push_back("-arch");
CmdArgs.push_back(Args.MakeArgString(GPUArch));
// Add paths specified in LIBRARY_PATH environment variable as -L options.
addDirectoryList(Args, CmdArgs, "-L", "LIBRARY_PATH");
// Add paths for the default clang library path.
SmallString<256> DefaultLibPath =
llvm::sys::path::parent_path(TC.getDriver().Dir);
llvm::sys::path::append(DefaultLibPath, CLANG_INSTALL_LIBDIR_BASENAME);
CmdArgs.push_back(Args.MakeArgString(Twine("-L") + DefaultLibPath));
for (const auto &II : Inputs) {
if (II.getType() == types::TY_LLVM_IR || II.getType() == types::TY_LTO_IR ||
II.getType() == types::TY_LTO_BC || II.getType() == types::TY_LLVM_BC) {
C.getDriver().Diag(diag::err_drv_no_linker_llvm_support)
<< getToolChain().getTripleString();
continue;
}
// The 'nvlink' application performs RDC-mode linking when given a '.o'
// file and device linking when given a '.cubin' file. We always want to
// perform device linking, so just rename any '.o' files.
// FIXME: This should hopefully be removed if NVIDIA updates their tooling.
if (II.isFilename()) {
auto InputFile = getToolChain().getInputFilename(II);
if (llvm::sys::path::extension(InputFile) != ".cubin") {
// If there are no actions above this one then this is direct input and
// we can copy it. Otherwise the input is internal so a `.cubin` file
// should exist.
if (II.getAction() && II.getAction()->getInputs().size() == 0) {
const char *CubinF =
Args.MakeArgString(getToolChain().getDriver().GetTemporaryPath(
llvm::sys::path::stem(InputFile), "cubin"));
if (llvm::sys::fs::copy_file(InputFile, C.addTempFile(CubinF)))
continue;
CmdArgs.push_back(CubinF);
} else {
SmallString<256> Filename(InputFile);
llvm::sys::path::replace_extension(Filename, "cubin");
CmdArgs.push_back(Args.MakeArgString(Filename));
}
} else {
CmdArgs.push_back(Args.MakeArgString(InputFile));
}
} else if (!II.isNothing()) {
II.getInputArg().renderAsInput(Args, CmdArgs);
}
}
C.addCommand(std::make_unique<Command>(
JA, *this,
ResponseFileSupport{ResponseFileSupport::RF_Full, llvm::sys::WEM_UTF8,
"--options-file"},
Args.MakeArgString(getToolChain().GetProgramPath("nvlink")), CmdArgs,
Inputs, Output));
}
void NVPTX::getNVPTXTargetFeatures(const Driver &D, const llvm::Triple &Triple,
const llvm::opt::ArgList &Args,
std::vector<StringRef> &Features) {
if (Args.hasArg(options::OPT_cuda_feature_EQ)) {
StringRef PtxFeature =
Args.getLastArgValue(options::OPT_cuda_feature_EQ, "+ptx42");
Features.push_back(Args.MakeArgString(PtxFeature));
return;
}
CudaInstallationDetector CudaInstallation(D, Triple, Args);
// New CUDA versions often introduce new instructions that are only supported
// by new PTX version, so we need to raise PTX level to enable them in NVPTX
// back-end.
const char *PtxFeature = nullptr;
switch (CudaInstallation.version()) {
#define CASE_CUDA_VERSION(CUDA_VER, PTX_VER) \
case CudaVersion::CUDA_##CUDA_VER: \
PtxFeature = "+ptx" #PTX_VER; \
break;
CASE_CUDA_VERSION(123, 83);
CASE_CUDA_VERSION(122, 82);
CASE_CUDA_VERSION(121, 81);
CASE_CUDA_VERSION(120, 80);
CASE_CUDA_VERSION(118, 78);
CASE_CUDA_VERSION(117, 77);
CASE_CUDA_VERSION(116, 76);
CASE_CUDA_VERSION(115, 75);
CASE_CUDA_VERSION(114, 74);
CASE_CUDA_VERSION(113, 73);
CASE_CUDA_VERSION(112, 72);
CASE_CUDA_VERSION(111, 71);
CASE_CUDA_VERSION(110, 70);
CASE_CUDA_VERSION(102, 65);
CASE_CUDA_VERSION(101, 64);
CASE_CUDA_VERSION(100, 63);
CASE_CUDA_VERSION(92, 61);
CASE_CUDA_VERSION(91, 61);
CASE_CUDA_VERSION(90, 60);
#undef CASE_CUDA_VERSION
default:
PtxFeature = "+ptx42";
}
Features.push_back(PtxFeature);
}
/// NVPTX toolchain. Our assembler is ptxas, and our linker is nvlink. This
/// operates as a stand-alone version of the NVPTX tools without the host
/// toolchain.
NVPTXToolChain::NVPTXToolChain(const Driver &D, const llvm::Triple &Triple,
const llvm::Triple &HostTriple,
const ArgList &Args, bool Freestanding = false)
: ToolChain(D, Triple, Args), CudaInstallation(D, HostTriple, Args),
Freestanding(Freestanding) {
if (CudaInstallation.isValid())
getProgramPaths().push_back(std::string(CudaInstallation.getBinPath()));
// Lookup binaries into the driver directory, this is used to
// discover the 'nvptx-arch' executable.
getProgramPaths().push_back(getDriver().Dir);
}
/// We only need the host triple to locate the CUDA binary utilities, use the
/// system's default triple if not provided.
NVPTXToolChain::NVPTXToolChain(const Driver &D, const llvm::Triple &Triple,
const ArgList &Args)
: NVPTXToolChain(D, Triple, llvm::Triple(LLVM_HOST_TRIPLE), Args,
/*Freestanding=*/true) {}
llvm::opt::DerivedArgList *
NVPTXToolChain::TranslateArgs(const llvm::opt::DerivedArgList &Args,
StringRef BoundArch,
Action::OffloadKind OffloadKind) const {
DerivedArgList *DAL = ToolChain::TranslateArgs(Args, BoundArch, OffloadKind);
if (!DAL)
DAL = new DerivedArgList(Args.getBaseArgs());
const OptTable &Opts = getDriver().getOpts();
for (Arg *A : Args)
if (!llvm::is_contained(*DAL, A))
DAL->append(A);
// TODO: We should accept 'generic' as a valid architecture.
if (!DAL->hasArg(options::OPT_march_EQ) && OffloadKind != Action::OFK_None) {
DAL->AddJoinedArg(nullptr, Opts.getOption(options::OPT_march_EQ),
CudaArchToString(CudaArch::CudaDefault));
} else if (DAL->getLastArgValue(options::OPT_march_EQ) == "native") {
auto GPUsOrErr = getSystemGPUArchs(Args);
if (!GPUsOrErr) {
getDriver().Diag(diag::err_drv_undetermined_gpu_arch)
<< getArchName() << llvm::toString(GPUsOrErr.takeError()) << "-march";
} else {
if (GPUsOrErr->size() > 1)
getDriver().Diag(diag::warn_drv_multi_gpu_arch)
<< getArchName() << llvm::join(*GPUsOrErr, ", ") << "-march";
DAL->AddJoinedArg(nullptr, Opts.getOption(options::OPT_march_EQ),
Args.MakeArgString(GPUsOrErr->front()));
}
}
return DAL;
}
void NVPTXToolChain::addClangTargetOptions(
const llvm::opt::ArgList &DriverArgs, llvm::opt::ArgStringList &CC1Args,
Action::OffloadKind DeviceOffloadingKind) const {
// If we are compiling with a standalone NVPTX toolchain we want to try to
// mimic a standard environment as much as possible. So we enable lowering
// ctor / dtor functions to global symbols that can be registered.
if (Freestanding)
CC1Args.append({"-mllvm", "--nvptx-lower-global-ctor-dtor"});
}
bool NVPTXToolChain::supportsDebugInfoOption(const llvm::opt::Arg *A) const {
const Option &O = A->getOption();
return (O.matches(options::OPT_gN_Group) &&
!O.matches(options::OPT_gmodules)) ||
O.matches(options::OPT_g_Flag) ||
O.matches(options::OPT_ggdbN_Group) || O.matches(options::OPT_ggdb) ||
O.matches(options::OPT_gdwarf) || O.matches(options::OPT_gdwarf_2) ||
O.matches(options::OPT_gdwarf_3) || O.matches(options::OPT_gdwarf_4) ||
O.matches(options::OPT_gdwarf_5) ||
O.matches(options::OPT_gcolumn_info);
}
void NVPTXToolChain::adjustDebugInfoKind(
llvm::codegenoptions::DebugInfoKind &DebugInfoKind,
const ArgList &Args) const {
switch (mustEmitDebugInfo(Args)) {
case DisableDebugInfo:
DebugInfoKind = llvm::codegenoptions::NoDebugInfo;
break;
case DebugDirectivesOnly:
DebugInfoKind = llvm::codegenoptions::DebugDirectivesOnly;
break;
case EmitSameDebugInfoAsHost:
// Use same debug info level as the host.
break;
}
}
Expected<SmallVector<std::string>>
NVPTXToolChain::getSystemGPUArchs(const ArgList &Args) const {
// Detect NVIDIA GPUs availible on the system.
std::string Program;
if (Arg *A = Args.getLastArg(options::OPT_nvptx_arch_tool_EQ))
Program = A->getValue();
else
Program = GetProgramPath("nvptx-arch");
auto StdoutOrErr = executeToolChainProgram(Program);
if (!StdoutOrErr)
return StdoutOrErr.takeError();
SmallVector<std::string, 1> GPUArchs;
for (StringRef Arch : llvm::split((*StdoutOrErr)->getBuffer(), "\n"))
if (!Arch.empty())
GPUArchs.push_back(Arch.str());
if (GPUArchs.empty())
return llvm::createStringError(std::error_code(),
"No NVIDIA GPU detected in the system");
return std::move(GPUArchs);
}
/// CUDA toolchain. Our assembler is ptxas, and our "linker" is fatbinary,
/// which isn't properly a linker but nonetheless performs the step of stitching
/// together object files from the assembler into a single blob.
CudaToolChain::CudaToolChain(const Driver &D, const llvm::Triple &Triple,
const ToolChain &HostTC, const ArgList &Args)
: NVPTXToolChain(D, Triple, HostTC.getTriple(), Args), HostTC(HostTC) {}
void CudaToolChain::addClangTargetOptions(
const llvm::opt::ArgList &DriverArgs, llvm::opt::ArgStringList &CC1Args,
Action::OffloadKind DeviceOffloadingKind) const {
HostTC.addClangTargetOptions(DriverArgs, CC1Args, DeviceOffloadingKind);
StringRef GpuArch = DriverArgs.getLastArgValue(options::OPT_march_EQ);
assert(!GpuArch.empty() && "Must have an explicit GPU arch.");
assert((DeviceOffloadingKind == Action::OFK_OpenMP ||
DeviceOffloadingKind == Action::OFK_Cuda) &&
"Only OpenMP or CUDA offloading kinds are supported for NVIDIA GPUs.");
if (DeviceOffloadingKind == Action::OFK_Cuda) {
CC1Args.append(
{"-fcuda-is-device", "-mllvm", "-enable-memcpyopt-without-libcalls"});
// Unsized function arguments used for variadics were introduced in CUDA-9.0
// We still do not support generating code that actually uses variadic
// arguments yet, but we do need to allow parsing them as recent CUDA
// headers rely on that. https://github.com/llvm/llvm-project/issues/58410
if (CudaInstallation.version() >= CudaVersion::CUDA_90)
CC1Args.push_back("-fcuda-allow-variadic-functions");
}
if (DriverArgs.hasArg(options::OPT_nogpulib))
return;
if (DeviceOffloadingKind == Action::OFK_OpenMP &&
DriverArgs.hasArg(options::OPT_S))
return;
std::string LibDeviceFile = CudaInstallation.getLibDeviceFile(GpuArch);
if (LibDeviceFile.empty()) {
getDriver().Diag(diag::err_drv_no_cuda_libdevice) << GpuArch;
return;
}
CC1Args.push_back("-mlink-builtin-bitcode");
CC1Args.push_back(DriverArgs.MakeArgString(LibDeviceFile));
clang::CudaVersion CudaInstallationVersion = CudaInstallation.version();
if (DriverArgs.hasFlag(options::OPT_fcuda_short_ptr,
options::OPT_fno_cuda_short_ptr, false))
CC1Args.append({"-mllvm", "--nvptx-short-ptr"});
if (CudaInstallationVersion >= CudaVersion::UNKNOWN)
CC1Args.push_back(
DriverArgs.MakeArgString(Twine("-target-sdk-version=") +
CudaVersionToString(CudaInstallationVersion)));
if (DeviceOffloadingKind == Action::OFK_OpenMP) {
if (CudaInstallationVersion < CudaVersion::CUDA_92) {
getDriver().Diag(
diag::err_drv_omp_offload_target_cuda_version_not_support)
<< CudaVersionToString(CudaInstallationVersion);
return;
}
// Link the bitcode library late if we're using device LTO.
if (getDriver().isUsingLTO(/* IsOffload */ true))
return;
addOpenMPDeviceRTL(getDriver(), DriverArgs, CC1Args, GpuArch.str(),
getTriple());
}
}
llvm::DenormalMode CudaToolChain::getDefaultDenormalModeForType(
const llvm::opt::ArgList &DriverArgs, const JobAction &JA,
const llvm::fltSemantics *FPType) const {
if (JA.getOffloadingDeviceKind() == Action::OFK_Cuda) {
if (FPType && FPType == &llvm::APFloat::IEEEsingle() &&
DriverArgs.hasFlag(options::OPT_fgpu_flush_denormals_to_zero,
options::OPT_fno_gpu_flush_denormals_to_zero, false))
return llvm::DenormalMode::getPreserveSign();
}
assert(JA.getOffloadingDeviceKind() != Action::OFK_Host);
return llvm::DenormalMode::getIEEE();
}
void CudaToolChain::AddCudaIncludeArgs(const ArgList &DriverArgs,
ArgStringList &CC1Args) const {
// Check our CUDA version if we're going to include the CUDA headers.
if (!DriverArgs.hasArg(options::OPT_nogpuinc) &&
!DriverArgs.hasArg(options::OPT_no_cuda_version_check)) {
StringRef Arch = DriverArgs.getLastArgValue(options::OPT_march_EQ);
assert(!Arch.empty() && "Must have an explicit GPU arch.");
CudaInstallation.CheckCudaVersionSupportsArch(StringToCudaArch(Arch));
}
CudaInstallation.AddCudaIncludeArgs(DriverArgs, CC1Args);
}
std::string CudaToolChain::getInputFilename(const InputInfo &Input) const {
// Only object files are changed, for example assembly files keep their .s
// extensions. If the user requested device-only compilation don't change it.
if (Input.getType() != types::TY_Object || getDriver().offloadDeviceOnly())
return ToolChain::getInputFilename(Input);
// Replace extension for object files with cubin because nvlink relies on
// these particular file names.
SmallString<256> Filename(ToolChain::getInputFilename(Input));
llvm::sys::path::replace_extension(Filename, "cubin");
return std::string(Filename);
}
llvm::opt::DerivedArgList *
CudaToolChain::TranslateArgs(const llvm::opt::DerivedArgList &Args,
StringRef BoundArch,
Action::OffloadKind DeviceOffloadKind) const {
DerivedArgList *DAL =
HostTC.TranslateArgs(Args, BoundArch, DeviceOffloadKind);
if (!DAL)
DAL = new DerivedArgList(Args.getBaseArgs());
const OptTable &Opts = getDriver().getOpts();
// For OpenMP device offloading, append derived arguments. Make sure
// flags are not duplicated.
// Also append the compute capability.
if (DeviceOffloadKind == Action::OFK_OpenMP) {
for (Arg *A : Args)
if (!llvm::is_contained(*DAL, A))
DAL->append(A);
if (!DAL->hasArg(options::OPT_march_EQ)) {
StringRef Arch = BoundArch;
if (Arch.empty()) {
auto ArchsOrErr = getSystemGPUArchs(Args);
if (!ArchsOrErr) {
std::string ErrMsg =
llvm::formatv("{0}", llvm::fmt_consume(ArchsOrErr.takeError()));
getDriver().Diag(diag::err_drv_undetermined_gpu_arch)
<< llvm::Triple::getArchTypeName(getArch()) << ErrMsg << "-march";
Arch = CudaArchToString(CudaArch::CudaDefault);
} else {
Arch = Args.MakeArgString(ArchsOrErr->front());
}
}
DAL->AddJoinedArg(nullptr, Opts.getOption(options::OPT_march_EQ), Arch);
}
return DAL;
}
for (Arg *A : Args) {
DAL->append(A);
}
if (!BoundArch.empty()) {
DAL->eraseArg(options::OPT_march_EQ);
DAL->AddJoinedArg(nullptr, Opts.getOption(options::OPT_march_EQ),
BoundArch);
}
return DAL;
}
Tool *NVPTXToolChain::buildAssembler() const {
return new tools::NVPTX::Assembler(*this);
}
Tool *NVPTXToolChain::buildLinker() const {
return new tools::NVPTX::Linker(*this);
}
Tool *CudaToolChain::buildAssembler() const {
return new tools::NVPTX::Assembler(*this);
}
Tool *CudaToolChain::buildLinker() const {
return new tools::NVPTX::FatBinary(*this);
}
void CudaToolChain::addClangWarningOptions(ArgStringList &CC1Args) const {
HostTC.addClangWarningOptions(CC1Args);
}
ToolChain::CXXStdlibType
CudaToolChain::GetCXXStdlibType(const ArgList &Args) const {
return HostTC.GetCXXStdlibType(Args);
}
void CudaToolChain::AddClangSystemIncludeArgs(const ArgList &DriverArgs,
ArgStringList &CC1Args) const {
HostTC.AddClangSystemIncludeArgs(DriverArgs, CC1Args);
if (!DriverArgs.hasArg(options::OPT_nogpuinc) && CudaInstallation.isValid())
CC1Args.append(
{"-internal-isystem",
DriverArgs.MakeArgString(CudaInstallation.getIncludePath())});
}
void CudaToolChain::AddClangCXXStdlibIncludeArgs(const ArgList &Args,
ArgStringList &CC1Args) const {
HostTC.AddClangCXXStdlibIncludeArgs(Args, CC1Args);
}
void CudaToolChain::AddIAMCUIncludeArgs(const ArgList &Args,
ArgStringList &CC1Args) const {
HostTC.AddIAMCUIncludeArgs(Args, CC1Args);
}
SanitizerMask CudaToolChain::getSupportedSanitizers() const {
// The CudaToolChain only supports sanitizers in the sense that it allows
// sanitizer arguments on the command line if they are supported by the host
// toolchain. The CudaToolChain will actually ignore any command line
// arguments for any of these "supported" sanitizers. That means that no
// sanitization of device code is actually supported at this time.
//
// This behavior is necessary because the host and device toolchains
// invocations often share the command line, so the device toolchain must
// tolerate flags meant only for the host toolchain.
return HostTC.getSupportedSanitizers();
}
VersionTuple CudaToolChain::computeMSVCVersion(const Driver *D,
const ArgList &Args) const {
return HostTC.computeMSVCVersion(D, Args);
}
|