1..
2  -------------------------------------------------------------------
3  NOTE: This file is automatically generated by running clang-tblgen
4  -gen-opt-docs. Do not edit this file by hand!!
5  -------------------------------------------------------------------
6
7=====================================
8Clang command line argument reference
9=====================================
10.. contents::
11   :local:
12
13Introduction
14============
15
16This page lists the command line arguments currently supported by the
17GCC-compatible ``clang`` and ``clang++`` drivers.
18
19
20.. program:: clang
21.. option:: -B<prefix>, --prefix <arg>, --prefix=<arg>
22
23Search $prefix/$triple-$file and $prefix$file for executables, libraries, includes, and data files used by the compiler. $prefix may or may not be a directory
24
25.. option:: -F<arg>
26
27Add directory to framework include search path
28
29.. option:: -ObjC
30
31Treat source input files as Objective-C inputs
32
33.. program:: clang1
34.. option:: -ObjC++
35.. program:: clang
36
37Treat source input files as Objective-C++ inputs
38
39.. option:: -Qn, -fno-ident
40
41Do not emit metadata containing compiler name and version
42
43.. option:: -Qunused-arguments
44
45Don't emit warning for unused driver arguments
46
47.. option:: -Qy, -fident
48
49Emit metadata containing compiler name and version
50
51.. option:: -Wa,<arg>,<arg2>...
52
53Pass the comma separated arguments in <arg> to the assembler
54
55.. option:: -Wlarge-by-value-copy=<arg>
56
57.. option:: -Xarch\_<arg1> <arg2>
58
59.. program:: clang1
60.. option:: -Xarch\_device <arg>
61.. program:: clang
62
63Pass <arg> to the CUDA/HIP device compilation
64
65.. program:: clang2
66.. option:: -Xarch\_host <arg>
67.. program:: clang
68
69Pass <arg> to the CUDA/HIP host compilation
70
71.. option:: -Xcuda-fatbinary <arg>
72
73Pass <arg> to fatbinary invocation
74
75.. option:: -Xcuda-ptxas <arg>
76
77Pass <arg> to the ptxas assembler
78
79.. option:: -Xopenmp-target <arg>
80
81Pass <arg> to the target offloading toolchain.
82
83.. program:: clang1
84.. option:: -Xopenmp-target=<triple> <arg>
85.. program:: clang
86
87Pass <arg> to the target offloading toolchain identified by <triple>.
88
89.. option:: -Z<arg>
90
91.. option:: -a<arg>, --profile-blocks
92
93.. option:: -all\_load
94
95.. option:: -allowable\_client <arg>
96
97.. option:: --analyze
98
99Run the static analyzer
100
101.. option:: --analyzer-no-default-checks
102
103.. option:: --analyzer-output<arg>
104
105Static analyzer report output format (html\|plist\|plist-multi-file\|plist-html\|sarif\|sarif-html\|text).
106
107.. option:: -arch <arg>
108
109.. program:: clang1
110.. option:: -arch\_errors\_fatal
111.. program:: clang
112
113.. program:: clang2
114.. option:: -arch\_only <arg>
115.. program:: clang
116
117.. option:: -arcmt-migrate-emit-errors
118
119Emit ARC errors even if the migrator can fix them
120
121.. option:: -arcmt-migrate-report-output <arg>
122
123Output path for the plist report
124
125.. option:: --autocomplete=<arg>
126
127.. option:: -bind\_at\_load
128
129.. option:: -bundle
130
131.. program:: clang1
132.. option:: -bundle\_loader <arg>
133.. program:: clang
134
135.. option:: -client\_name<arg>
136
137.. option:: -compatibility\_version<arg>
138
139.. option:: --config <arg>
140
141Specifies configuration file
142
143.. option:: --constant-cfstrings
144
145.. option:: --cuda-compile-host-device
146
147Compile CUDA code for both host and device (default).  Has no effect on non-CUDA compilations.
148
149.. option:: --cuda-device-only
150
151Compile CUDA code for device only
152
153.. option:: --cuda-host-only
154
155Compile CUDA code for host only.  Has no effect on non-CUDA compilations.
156
157.. option:: --cuda-include-ptx=<arg>, --no-cuda-include-ptx=<arg>
158
159Include PTX for the following GPU architecture (e.g. sm\_35) or 'all'. May be specified more than once.
160
161.. option:: --cuda-noopt-device-debug, --no-cuda-noopt-device-debug
162
163Enable device-side debug info generation. Disables ptxas optimizations.
164
165.. option:: -cuid=<arg>
166
167An ID for compilation unit, which should be the same for the same compilation unit but different for different compilation units. It is used to externalize device-side static variables for single source offloading languages CUDA and HIP so that they can be accessed by the host code of the same compilation unit.
168
169.. option:: -current\_version<arg>
170
171.. option:: -dead\_strip
172
173.. option:: -dependency-dot <arg>
174
175Filename to write DOT-formatted header dependencies to
176
177.. option:: -dependency-file <arg>
178
179Filename (or -) to write dependency output to
180
181.. option:: -dsym-dir<dir>
182
183Directory to output dSYM's (if any) to
184
185.. option:: -dumpmachine
186
187.. option:: -dumpversion
188
189.. option:: --dyld-prefix=<arg>, --dyld-prefix <arg>
190
191.. option:: -dylib\_file <arg>
192
193.. option:: -dylinker
194
195.. program:: clang1
196.. option:: -dylinker\_install\_name<arg>
197.. program:: clang
198
199.. option:: -dynamic
200
201.. option:: -dynamiclib
202
203.. option:: -emit-ast
204
205Emit Clang AST files for source inputs
206
207.. option:: --emit-static-lib
208
209Enable linker job to emit a static library.
210
211.. option:: -enable-trivial-auto-var-init-zero-knowing-it-will-be-removed-from-clang
212
213Trivial automatic variable initialization to zero is only here for benchmarks, it'll eventually be removed, and I'm OK with that because I'm only using it to benchmark
214
215.. option:: -exported\_symbols\_list <arg>
216
217.. option:: -faligned-new=<arg>
218
219.. option:: -ffixed-r19
220
221Reserve register r19 (Hexagon only)
222
223.. option:: -fgpu-flush-denormals-to-zero, -fcuda-flush-denormals-to-zero, -fno-gpu-flush-denormals-to-zero
224
225Flush denormal floating point values to zero in CUDA/HIP device mode.
226
227.. option:: -fheinous-gnu-extensions
228
229.. option:: -flat\_namespace
230
231.. option:: -fopenmp-targets=<arg1>,<arg2>...
232
233Specify comma-separated list of triples OpenMP offloading targets to be supported
234
235.. option:: -force\_cpusubtype\_ALL
236
237.. program:: clang1
238.. option:: -force\_flat\_namespace
239.. program:: clang
240
241.. program:: clang2
242.. option:: -force\_load <arg>
243.. program:: clang
244
245.. option:: -framework <arg>
246
247.. option:: -frtlib-add-rpath, -fno-rtlib-add-rpath
248
249Add -rpath with architecture-specific resource directory to the linker flags
250
251.. option:: -fsanitize-system-ignorelist=<arg>, -fsanitize-system-blacklist=<arg>
252
253Path to system ignorelist file for sanitizers
254
255.. option:: -fshow-skipped-includes
256
257#include files may be "skipped" due to include guard optimization
258             or #pragma once. This flag makes -H show also such includes.
259
260.. option:: -fsystem-module
261
262Build this module as a system module. Only used with -emit-module
263
264.. option:: -fuse-cuid=<arg>
265
266Method to generate ID's for compilation units for single source offloading languages CUDA and HIP: 'hash' (ID's generated by hashing file path and command line options) \| 'random' (ID's generated as random numbers) \| 'none' (disabled). Default is 'hash'. This option will be overridden by option '-cuid=\[ID\]' if it is specified.
267
268.. option:: --gcc-toolchain=<arg>, -gcc-toolchain <arg>
269
270Search for GCC installation in the specified directory on targets which commonly use GCC. The directory usually contains 'lib{,32,64}/gcc{,-cross}/$triple' and 'include'. If specified, sysroot is skipped for GCC detection. Note: executables (e.g. ld) used by the compiler are not overridden by the selected GCC installation
271
272.. option:: -gcodeview
273
274Generate CodeView debug information
275
276.. option:: -gcodeview-ghash, -gno-codeview-ghash
277
278Emit type record hashes in a .debug$H section
279
280.. option:: --gpu-instrument-lib=<arg>
281
282Instrument device library for HIP, which is a LLVM bitcode containing \_\_cyg\_profile\_func\_enter and \_\_cyg\_profile\_func\_exit
283
284.. option:: --gpu-max-threads-per-block=<arg>
285
286Default max threads per block for kernel launch bounds for HIP
287
288.. option:: -headerpad\_max\_install\_names<arg>
289
290.. option:: -help, --help
291
292Display available options
293
294.. option:: --help-hidden
295
296Display help for hidden options
297
298.. option:: --hip-link
299
300Link clang-offload-bundler bundles for HIP
301
302.. option:: --hip-version=<arg>
303
304HIP version in the format of major.minor.patch
305
306.. option:: -ibuiltininc
307
308Enable builtin #include directories even when -nostdinc is used before or after -ibuiltininc. Using -nobuiltininc after the option disables it
309
310.. option:: -image\_base <arg>
311
312.. option:: -index-header-map
313
314Make the next included directory (-I or -F) an indexer header map
315
316.. option:: -init <arg>
317
318.. option:: -install\_name <arg>
319
320.. option:: -interface-stub-version=<arg>
321
322.. option:: -keep\_private\_externs
323
324.. option:: -lazy\_framework <arg>
325
326.. program:: clang1
327.. option:: -lazy\_library <arg>
328.. program:: clang
329
330.. option:: -mbig-endian, -EB
331
332.. option:: -mbranch-protection=<arg>
333
334Enforce targets of indirect branches and function returns
335
336.. option:: -menable-unsafe-fp-math
337
338Allow unsafe floating-point math optimizations which may decrease precision
339
340.. option:: -mharden-sls=<arg>
341
342Select straight-line speculation hardening scope
343
344.. option:: --migrate
345
346Run the migrator
347
348.. option:: -mios-simulator-version-min=<arg>, -miphonesimulator-version-min=<arg>
349
350.. option:: -mlinker-version=<arg>
351
352.. option:: -mlittle-endian, -EL
353
354.. option:: -mllvm <arg>
355
356Additional arguments to forward to LLVM's option processing
357
358.. option:: -module-dependency-dir <arg>
359
360Directory to dump module dependencies to
361
362.. option:: -mtvos-simulator-version-min=<arg>, -mappletvsimulator-version-min=<arg>
363
364.. option:: -multi\_module
365
366.. option:: -multiply\_defined <arg>
367
368.. program:: clang1
369.. option:: -multiply\_defined\_unused <arg>
370.. program:: clang
371
372.. option:: -mwatchos-simulator-version-min=<arg>, -mwatchsimulator-version-min=<arg>
373
374.. option:: --no-cuda-version-check
375
376Don't error out if the detected version of the CUDA install is too low for the requested CUDA gpu architecture.
377
378.. option:: -no-integrated-cpp, --no-integrated-cpp
379
380.. option:: -no\_dead\_strip\_inits\_and\_terms
381
382.. option:: -nobuiltininc
383
384Disable builtin #include directories
385
386.. option:: -nodefaultlibs
387
388.. option:: -nofixprebinding
389
390.. option:: -nogpuinc, -nocudainc
391
392Do not add include paths for CUDA/HIP and do not include the default CUDA/HIP wrapper headers
393
394.. option:: -nogpulib, -nocudalib
395
396Do not link device library for CUDA/HIP device compilation
397
398.. option:: -nolibc
399
400.. option:: -nomultidefs
401
402.. option:: -nopie, -no-pie
403
404.. option:: -noprebind
405
406.. option:: -noprofilelib
407
408.. option:: -noseglinkedit
409
410.. option:: -nostdinc, --no-standard-includes
411
412.. program:: clang1
413.. option:: -nostdinc++
414.. program:: clang
415
416Disable standard #include directories for the C++ standard library
417
418.. option:: -nostdlib++
419
420.. option:: -nostdlibinc
421
422.. option:: -o<file>, --output <arg>, --output=<arg>
423
424Write output to <file>
425
426.. option:: -objcmt-atomic-property
427
428Make migration to 'atomic' properties
429
430.. option:: -objcmt-migrate-all
431
432Enable migration to modern ObjC
433
434.. option:: -objcmt-migrate-annotation
435
436Enable migration to property and method annotations
437
438.. option:: -objcmt-migrate-designated-init
439
440Enable migration to infer NS\_DESIGNATED\_INITIALIZER for initializer methods
441
442.. option:: -objcmt-migrate-instancetype
443
444Enable migration to infer instancetype for method result type
445
446.. option:: -objcmt-migrate-literals
447
448Enable migration to modern ObjC literals
449
450.. option:: -objcmt-migrate-ns-macros
451
452Enable migration to NS\_ENUM/NS\_OPTIONS macros
453
454.. option:: -objcmt-migrate-property
455
456Enable migration to modern ObjC property
457
458.. option:: -objcmt-migrate-property-dot-syntax
459
460Enable migration of setter/getter messages to property-dot syntax
461
462.. option:: -objcmt-migrate-protocol-conformance
463
464Enable migration to add protocol conformance on classes
465
466.. option:: -objcmt-migrate-readonly-property
467
468Enable migration to modern ObjC readonly property
469
470.. option:: -objcmt-migrate-readwrite-property
471
472Enable migration to modern ObjC readwrite property
473
474.. option:: -objcmt-migrate-subscripting
475
476Enable migration to modern ObjC subscripting
477
478.. option:: -objcmt-ns-nonatomic-iosonly
479
480Enable migration to use NS\_NONATOMIC\_IOSONLY macro for setting property's 'atomic' attribute
481
482.. option:: -objcmt-returns-innerpointer-property
483
484Enable migration to annotate property with NS\_RETURNS\_INNER\_POINTER
485
486.. option:: -objcmt-whitelist-dir-path=<arg>, -objcmt-white-list-dir-path=<arg>
487
488Only modify files with a filename contained in the provided directory path
489
490.. option:: -object
491
492.. option:: --offload-arch=<arg>, --cuda-gpu-arch=<arg>, --no-offload-arch=<arg>
493
494CUDA offloading device architecture (e.g. sm\_35), or HIP offloading target ID in the form of a device architecture followed by target ID features delimited by a colon. Each target ID feature is a pre-defined string followed by a plus or minus sign (e.g. gfx908:xnack+:sramecc-).  May be specified more than once.
495
496.. option:: -p, --profile
497
498.. option:: -pagezero\_size<arg>
499
500.. option:: -pg
501
502Enable mcount instrumentation
503
504.. option:: -pipe, --pipe
505
506Use pipes between commands, when possible
507
508.. option:: -prebind
509
510.. program:: clang1
511.. option:: -prebind\_all\_twolevel\_modules
512.. program:: clang
513
514.. option:: -preload
515
516.. option:: --print-diagnostic-categories
517
518.. option:: -print-effective-triple, --print-effective-triple
519
520Print the effective target triple
521
522.. option:: -print-file-name=<file>, --print-file-name=<file>, --print-file-name <arg>
523
524Print the full library path of <file>
525
526.. option:: -print-ivar-layout
527
528Enable Objective-C Ivar layout bitmap print trace
529
530.. option:: -print-libgcc-file-name, --print-libgcc-file-name
531
532Print the library path for the currently used compiler runtime library ("libgcc.a" or "libclang\_rt.builtins.\*.a")
533
534.. option:: -print-multi-directory, --print-multi-directory
535
536.. option:: -print-multi-lib, --print-multi-lib
537
538.. option:: -print-multiarch, --print-multiarch
539
540Print the multiarch target triple
541
542.. option:: -print-prog-name=<name>, --print-prog-name=<name>, --print-prog-name <arg>
543
544Print the full program path of <name>
545
546.. option:: -print-resource-dir, --print-resource-dir
547
548Print the resource directory pathname
549
550.. option:: -print-rocm-search-dirs, --print-rocm-search-dirs
551
552Print the paths used for finding ROCm installation
553
554.. option:: -print-runtime-dir, --print-runtime-dir
555
556Print the directory pathname containing clangs runtime libraries
557
558.. option:: -print-search-dirs, --print-search-dirs
559
560Print the paths used for finding libraries and programs
561
562.. option:: -print-target-triple, --print-target-triple
563
564Print the normalized target triple
565
566.. option:: -print-targets, --print-targets
567
568Print the registered targets
569
570.. option:: -private\_bundle
571
572.. option:: -pthread, -no-pthread
573
574Support POSIX threads in generated code
575
576.. option:: -pthreads
577
578.. option:: -read\_only\_relocs <arg>
579
580.. option:: -relocatable-pch, --relocatable-pch
581
582Whether to build a relocatable precompiled header
583
584.. option:: -remap
585
586.. option:: -rewrite-legacy-objc
587
588Rewrite Legacy Objective-C source to C++
589
590.. option:: -rtlib=<arg>, --rtlib=<arg>, --rtlib <arg>
591
592Compiler runtime library to use
593
594.. option:: -save-stats=<arg>, --save-stats=<arg>, -save-stats (equivalent to -save-stats=cwd), --save-stats (equivalent to -save-stats=cwd)
595
596Save llvm statistics.
597
598.. option:: -save-temps=<arg>, --save-temps=<arg>, -save-temps (equivalent to -save-temps=cwd), --save-temps (equivalent to -save-temps=cwd)
599
600Save intermediate compilation results.
601
602.. option:: -sectalign <arg1> <arg2> <arg3>
603
604.. option:: -sectcreate <arg1> <arg2> <arg3>
605
606.. option:: -sectobjectsymbols <arg1> <arg2>
607
608.. option:: -sectorder <arg1> <arg2> <arg3>
609
610.. option:: -seg1addr<arg>
611
612.. option:: -seg\_addr\_table <arg>
613
614.. program:: clang1
615.. option:: -seg\_addr\_table\_filename <arg>
616.. program:: clang
617
618.. option:: -segaddr <arg1> <arg2>
619
620.. option:: -segcreate <arg1> <arg2> <arg3>
621
622.. option:: -seglinkedit
623
624.. option:: -segprot <arg1> <arg2> <arg3>
625
626.. option:: -segs\_read\_<arg>
627
628.. program:: clang1
629.. option:: -segs\_read\_only\_addr <arg>
630.. program:: clang
631
632.. program:: clang2
633.. option:: -segs\_read\_write\_addr <arg>
634.. program:: clang
635
636.. option:: -serialize-diagnostics <arg>, --serialize-diagnostics <arg>
637
638Serialize compiler diagnostics to a file
639
640.. option:: -shared-libgcc
641
642.. option:: -shared-libsan, -shared-libasan
643
644Dynamically link the sanitizer runtime
645
646.. option:: -single\_module
647
648.. option:: -static-libgcc
649
650.. option:: -static-libsan
651
652Statically link the sanitizer runtime
653
654.. option:: -static-libstdc++
655
656.. option:: -static-openmp
657
658Use the static host OpenMP runtime while linking.
659
660.. option:: -std-default=<arg>
661
662.. option:: -stdlib=<arg>, --stdlib=<arg>, --stdlib <arg>
663
664C++ standard library to use
665
666.. option:: -sub\_library<arg>
667
668.. program:: clang1
669.. option:: -sub\_umbrella<arg>
670.. program:: clang
671
672.. option:: --sysroot=<arg>, --sysroot <arg>
673
674.. option:: --target-help
675
676.. option:: --target=<arg>, -target <arg>
677
678Generate code for the given target
679
680.. option:: -time
681
682Time individual commands
683
684.. option:: -traditional, --traditional
685
686.. option:: -traditional-cpp, --traditional-cpp
687
688Enable some traditional CPP emulation
689
690.. option:: -twolevel\_namespace
691
692.. program:: clang1
693.. option:: -twolevel\_namespace\_hints
694.. program:: clang
695
696.. option:: -umbrella <arg>
697
698.. option:: -unexported\_symbols\_list <arg>
699
700.. option:: -unwindlib=<arg>, --unwindlib=<arg>
701
702Unwind library to use
703
704.. option:: -v, --verbose
705
706Show commands to run and use verbose output
707
708.. option:: --verify-debug-info
709
710Verify the binary representation of debug output
711
712.. option:: --version
713
714Print version information
715
716.. option:: -w, --no-warnings
717
718Suppress all warnings
719
720.. option:: -weak-l<arg>
721
722.. option:: -weak\_framework <arg>
723
724.. program:: clang1
725.. option:: -weak\_library <arg>
726.. program:: clang
727
728.. program:: clang2
729.. option:: -weak\_reference\_mismatches <arg>
730.. program:: clang
731
732.. option:: -whatsloaded
733
734.. option:: -why\_load, -whyload
735
736.. option:: -working-directory<arg>, -working-directory=<arg>
737
738Resolve file paths relative to the specified directory
739
740.. option:: -x<language>, --language <arg>, --language=<arg>
741
742Treat subsequent input files as having type <language>
743
744.. option:: -y<arg>
745
746Actions
747=======
748The action to perform on the input.
749
750.. option:: -E, --preprocess
751
752Only run the preprocessor
753
754.. option:: -S, --assemble
755
756Only run preprocess and compilation steps
757
758.. option:: -c, --compile
759
760Only run preprocess, compile, and assemble steps
761
762.. option:: -emit-interface-stubs
763
764Generate Interface Stub Files.
765
766.. option:: -emit-llvm
767
768Use the LLVM representation for assembler and object files
769
770.. option:: -emit-merged-ifs
771
772Generate Interface Stub Files, emit merged text not binary.
773
774.. option:: -fsyntax-only
775
776.. option:: -module-file-info
777
778Provide information about a particular module file
779
780.. option:: --precompile
781
782Only precompile the input
783
784.. option:: -rewrite-objc
785
786Rewrite Objective-C source to C++
787
788.. option:: -verify-pch
789
790Load and verify that a pre-compiled header file is not stale
791
792Compilation flags
793=================
794
795Flags controlling the behavior of Clang during compilation. These flags have
796no effect during actions that do not perform compilation.
797
798.. option:: -Xassembler <arg>
799
800Pass <arg> to the assembler
801
802.. option:: -Xclang <arg>
803
804Pass <arg> to the clang compiler
805
806.. option:: -ansi, --ansi
807
808.. option:: -fc++-abi=<arg>
809
810C++ ABI to use. This will override the target C++ ABI.
811
812.. option:: -fclang-abi-compat=<version>
813
814Attempt to match the ABI of Clang <version>
815
816.. option:: -fcomment-block-commands=<arg>,<arg2>...
817
818Treat each comma separated argument in <arg> as a documentation comment block command
819
820.. option:: -fcomplete-member-pointers, -fno-complete-member-pointers
821
822Require member pointer base types to be complete if they would be significant under the Microsoft ABI
823
824.. option:: -fcrash-diagnostics-dir=<dir>
825
826Put crash-report files in <dir>
827
828.. option:: -fdeclspec, -fno-declspec
829
830Allow \_\_declspec as a keyword
831
832.. option:: -fdepfile-entry=<arg>
833
834.. option:: -fdiagnostics-fixit-info, -fno-diagnostics-fixit-info
835
836.. option:: -fdiagnostics-format=<arg>
837
838.. option:: -fdiagnostics-parseable-fixits
839
840Print fix-its in machine parseable form
841
842.. option:: -fdiagnostics-print-source-range-info
843
844Print source range spans in numeric form
845
846.. option:: -fdiagnostics-show-category=<arg>
847
848.. option:: -fdiscard-value-names, -fno-discard-value-names
849
850Discard value names in LLVM IR
851
852.. option:: -fexperimental-relative-c++-abi-vtables, -fno-experimental-relative-c++-abi-vtables
853
854Use the experimental C++ class ABI for classes with virtual tables
855
856.. option:: -fexperimental-strict-floating-point
857
858Enables experimental strict floating point in LLVM.
859
860.. option:: -ffine-grained-bitfield-accesses, -fno-fine-grained-bitfield-accesses
861
862Use separate accesses for consecutive bitfield runs with legal widths and alignments.
863
864.. option:: -fglobal-isel, -fexperimental-isel, -fno-global-isel
865
866Enables the global instruction selector
867
868.. option:: -finline-functions, -fno-inline-functions
869
870Inline suitable functions
871
872.. option:: -finline-hint-functions
873
874Inline functions which are (explicitly or implicitly) marked inline
875
876.. option:: -flegacy-pass-manager, -fno-experimental-new-pass-manager, -fno-legacy-pass-manager
877
878Use the legacy pass manager in LLVM
879
880.. option:: -fno-crash-diagnostics
881
882Disable auto-generation of preprocessed source files and a script for reproduction during a clang crash
883
884.. option:: -fno-sanitize-ignorelist, -fno-sanitize-blacklist
885
886Don't use ignorelist file for sanitizers
887
888.. option:: -fparse-all-comments
889
890.. option:: -frecord-command-line, -fno-record-command-line, -frecord-gcc-switches
891
892.. option:: -fsanitize-address-destructor=<arg>
893
894Set destructor type used in ASan instrumentation
895
896.. option:: -fsanitize-address-field-padding=<arg>
897
898Level of field padding for AddressSanitizer
899
900.. option:: -fsanitize-address-globals-dead-stripping
901
902Enable linker dead stripping of globals in AddressSanitizer
903
904.. option:: -fsanitize-address-poison-custom-array-cookie, -fno-sanitize-address-poison-custom-array-cookie
905
906Enable poisoning array cookies when using custom operator new\[\] in AddressSanitizer
907
908.. option:: -fsanitize-address-use-after-return=<mode>
909
910Select the mode of detecting stack use-after-return in AddressSanitizer
911
912.. option:: -fsanitize-address-use-after-scope, -fno-sanitize-address-use-after-scope
913
914Enable use-after-scope detection in AddressSanitizer
915
916.. option:: -fsanitize-address-use-odr-indicator, -fno-sanitize-address-use-odr-indicator
917
918Enable ODR indicator globals to avoid false ODR violation reports in partially sanitized programs at the cost of an increase in binary size
919
920.. option:: -fsanitize-cfi-canonical-jump-tables, -fno-sanitize-cfi-canonical-jump-tables
921
922Make the jump table addresses canonical in the symbol table
923
924.. option:: -fsanitize-cfi-cross-dso, -fno-sanitize-cfi-cross-dso
925
926Enable control flow integrity (CFI) checks for cross-DSO calls.
927
928.. option:: -fsanitize-cfi-icall-generalize-pointers
929
930Generalize pointers in CFI indirect call type signature checks
931
932.. option:: -fsanitize-coverage-allowlist=<arg>, -fsanitize-coverage-whitelist=<arg>
933
934Restrict sanitizer coverage instrumentation exclusively to modules and functions that match the provided special case list, except the blocked ones
935
936.. option:: -fsanitize-coverage-ignorelist=<arg>, -fsanitize-coverage-blacklist=<arg>
937
938Disable sanitizer coverage instrumentation for modules and functions that match the provided special case list, even the allowed ones
939
940.. option:: -fsanitize-coverage=<arg1>,<arg2>..., -fno-sanitize-coverage=<arg1>,<arg2>...
941
942Specify the type of coverage instrumentation for Sanitizers
943
944.. option:: -fsanitize-hwaddress-abi=<arg>
945
946Select the HWAddressSanitizer ABI to target (interceptor or platform, default interceptor). This option is currently unused.
947
948.. option:: -fsanitize-hwaddress-experimental-aliasing, -fno-sanitize-hwaddress-experimental-aliasing
949
950Enable aliasing mode in HWAddressSanitizer
951
952.. option:: -fsanitize-ignorelist=<arg>, -fsanitize-blacklist=<arg>
953
954Path to ignorelist file for sanitizers
955
956.. option:: -fsanitize-link-c++-runtime, -fno-sanitize-link-c++-runtime
957
958.. option:: -fsanitize-link-runtime, -fno-sanitize-link-runtime
959
960.. option:: -fsanitize-memory-track-origins, -fno-sanitize-memory-track-origins
961
962Enable origins tracking in MemorySanitizer
963
964.. program:: clang1
965.. option:: -fsanitize-memory-track-origins=<arg>
966.. program:: clang
967
968Enable origins tracking in MemorySanitizer
969
970.. option:: -fsanitize-memory-use-after-dtor, -fno-sanitize-memory-use-after-dtor
971
972Enable use-after-destroy detection in MemorySanitizer
973
974.. option:: -fsanitize-minimal-runtime, -fno-sanitize-minimal-runtime
975
976.. option:: -fsanitize-recover=<arg1>,<arg2>..., -fno-sanitize-recover=<arg1>,<arg2>..., -fsanitize-recover (equivalent to -fsanitize-recover=all)
977
978Enable recovery for specified sanitizers
979
980.. option:: -fsanitize-stats, -fno-sanitize-stats
981
982Enable sanitizer statistics gathering.
983
984.. option:: -fsanitize-thread-atomics, -fno-sanitize-thread-atomics
985
986Enable atomic operations instrumentation in ThreadSanitizer (default)
987
988.. option:: -fsanitize-thread-func-entry-exit, -fno-sanitize-thread-func-entry-exit
989
990Enable function entry/exit instrumentation in ThreadSanitizer (default)
991
992.. option:: -fsanitize-thread-memory-access, -fno-sanitize-thread-memory-access
993
994Enable memory access instrumentation in ThreadSanitizer (default)
995
996.. option:: -fsanitize-trap=<arg1>,<arg2>..., -fno-sanitize-trap=<arg1>,<arg2>..., -fsanitize-trap (equivalent to -fsanitize-trap=all), -fsanitize-undefined-trap-on-error (equivalent to -fsanitize-trap=undefined)
997
998Enable trapping for specified sanitizers
999
1000.. option:: -fsanitize-undefined-strip-path-components=<number>
1001
1002Strip (or keep only, if negative) a given number of path components when emitting check metadata.
1003
1004.. option:: -fsanitize=<check>,<arg2>..., -fno-sanitize=<arg1>,<arg2>...
1005
1006Turn on runtime checks for various forms of undefined or suspicious behavior. See user manual for available checks
1007
1008.. option:: -moutline, -mno-outline
1009
1010Enable function outlining (AArch64 only)
1011
1012.. option:: -moutline-atomics, -mno-outline-atomics
1013
1014Generate local calls to out-of-line atomic operations
1015
1016.. option:: --param <arg>, --param=<arg>
1017
1018.. option:: -print-supported-cpus, --print-supported-cpus, -mcpu=?, -mtune=?
1019
1020Print supported cpu models for the given target (if target is not specified, it will print the supported cpus for the default target)
1021
1022.. option:: -std=<arg>, --std=<arg>, --std <arg>
1023
1024Language standard to compile for
1025
1026Preprocessor flags
1027~~~~~~~~~~~~~~~~~~
1028
1029Flags controlling the behavior of the Clang preprocessor.
1030
1031.. option:: -C, --comments
1032
1033Include comments in preprocessed output
1034
1035.. option:: -CC, --comments-in-macros
1036
1037Include comments from within macros in preprocessed output
1038
1039.. option:: -D<macro>=<value>, --define-macro <arg>, --define-macro=<arg>
1040
1041Define <macro> to <value> (or 1 if <value> omitted)
1042
1043.. option:: -H, --trace-includes
1044
1045Show header includes and nesting depth
1046
1047.. option:: -P, --no-line-commands
1048
1049Disable linemarker output in -E mode
1050
1051.. option:: -U<macro>, --undefine-macro <arg>, --undefine-macro=<arg>
1052
1053Undefine macro <macro>
1054
1055.. option:: -Wp,<arg>,<arg2>...
1056
1057Pass the comma separated arguments in <arg> to the preprocessor
1058
1059.. option:: -Xpreprocessor <arg>
1060
1061Pass <arg> to the preprocessor
1062
1063.. option:: -fmacro-prefix-map=<arg>
1064
1065remap file source paths in predefined preprocessor macros
1066
1067Include path management
1068-----------------------
1069
1070Flags controlling how ``#include``\s are resolved to files.
1071
1072.. option:: -I<dir>, --include-directory <arg>, --include-directory=<arg>
1073
1074Add directory to include search path. For C++ inputs, if
1075there are multiple -I options, these directories are searched
1076in the order they are given before the standard system directories
1077are searched. If the same directory is in the SYSTEM include search
1078paths, for example if also specified with -isystem, the -I option
1079will be ignored
1080
1081.. option:: -I-, --include-barrier
1082
1083Restrict all prior -I flags to double-quoted inclusion and remove current directory from include path
1084
1085.. option:: --amdgpu-arch-tool=<arg>
1086
1087Tool used for detecting AMD GPU arch in the system.
1088
1089.. option:: --cuda-path-ignore-env
1090
1091Ignore environment variables to detect CUDA installation
1092
1093.. option:: --cuda-path=<arg>
1094
1095CUDA installation path
1096
1097.. option:: -cxx-isystem<directory>
1098
1099Add directory to the C++ SYSTEM include search path
1100
1101.. option:: -fbuild-session-file=<file>
1102
1103Use the last modification time of <file> as the build session timestamp
1104
1105.. option:: -fbuild-session-timestamp=<time since Epoch in seconds>
1106
1107Time when the current build session started
1108
1109.. option:: -fmodule-file=\[<name>=\]<file>
1110
1111Specify the mapping of module name to precompiled module file, or load a module file if name is omitted.
1112
1113.. option:: -fmodules-cache-path=<directory>
1114
1115Specify the module cache path
1116
1117.. option:: -fmodules-disable-diagnostic-validation
1118
1119Disable validation of the diagnostic options when loading the module
1120
1121.. option:: -fmodules-prune-after=<seconds>
1122
1123Specify the interval (in seconds) after which a module file will be considered unused
1124
1125.. option:: -fmodules-prune-interval=<seconds>
1126
1127Specify the interval (in seconds) between attempts to prune the module cache
1128
1129.. option:: -fmodules-user-build-path <directory>
1130
1131Specify the module user build path
1132
1133.. option:: -fmodules-validate-once-per-build-session
1134
1135Don't verify input files for the modules if the module has been successfully validated or loaded during this build session
1136
1137.. option:: -fmodules-validate-system-headers, -fno-modules-validate-system-headers
1138
1139Validate the system headers that a module depends on when loading the module
1140
1141.. option:: -fprebuilt-module-path=<directory>
1142
1143Specify the prebuilt module path
1144
1145.. option:: --hip-path=<arg>
1146
1147HIP runtime installation path, used for finding HIP version and adding HIP include path.
1148
1149.. option:: -idirafter<arg>, --include-directory-after <arg>, --include-directory-after=<arg>
1150
1151Add directory to AFTER include search path
1152
1153.. option:: -iframework<arg>
1154
1155Add directory to SYSTEM framework search path
1156
1157.. option:: -iframeworkwithsysroot<directory>
1158
1159Add directory to SYSTEM framework search path, absolute paths are relative to -isysroot
1160
1161.. option:: -imacros<file>, --imacros<file>, --imacros=<arg>
1162
1163Include macros from file before parsing
1164
1165.. option:: -include<file>, --include<file>, --include=<arg>
1166
1167Include file before parsing
1168
1169.. option:: -include-pch <file>
1170
1171Include precompiled header file
1172
1173.. option:: -iprefix<dir>, --include-prefix <arg>, --include-prefix=<arg>
1174
1175Set the -iwithprefix/-iwithprefixbefore prefix
1176
1177.. option:: -iquote<directory>
1178
1179Add directory to QUOTE include search path
1180
1181.. option:: -isysroot<dir>
1182
1183Set the system root directory (usually /)
1184
1185.. option:: -isystem<directory>
1186
1187Add directory to SYSTEM include search path
1188
1189.. option:: -isystem-after<directory>
1190
1191Add directory to end of the SYSTEM include search path
1192
1193.. option:: -ivfsoverlay<arg>
1194
1195Overlay the virtual filesystem described by file over the real file system
1196
1197.. option:: -iwithprefix<dir>, --include-with-prefix <arg>, --include-with-prefix-after <arg>, --include-with-prefix-after=<arg>, --include-with-prefix=<arg>
1198
1199Set directory to SYSTEM include search path with prefix
1200
1201.. option:: -iwithprefixbefore<dir>, --include-with-prefix-before <arg>, --include-with-prefix-before=<arg>
1202
1203Set directory to include search path with prefix
1204
1205.. option:: -iwithsysroot<directory>
1206
1207Add directory to SYSTEM include search path, absolute paths are relative to -isysroot
1208
1209.. option:: --libomptarget-amdgcn-bc-path=<arg>
1210
1211Path to libomptarget-amdgcn bitcode library
1212
1213.. option:: --libomptarget-nvptx-bc-path=<arg>
1214
1215Path to libomptarget-nvptx bitcode library
1216
1217.. option:: --ptxas-path=<arg>
1218
1219Path to ptxas (used for compiling CUDA code)
1220
1221.. option:: --rocm-path=<arg>
1222
1223ROCm installation path, used for finding and automatically linking required bitcode libraries.
1224
1225.. program:: clang1
1226.. option:: -stdlib++-isystem<directory>
1227.. program:: clang
1228
1229Use directory as the C++ standard library include path
1230
1231.. option:: --system-header-prefix=<prefix>, --no-system-header-prefix=<prefix>, --system-header-prefix <arg>
1232
1233Treat all #include paths starting with <prefix> as including a system header.
1234
1235Dependency file generation
1236--------------------------
1237
1238Flags controlling generation of a dependency file for ``make``-like build
1239systems.
1240
1241.. option:: -M, --dependencies
1242
1243Like -MD, but also implies -E and writes to stdout by default
1244
1245.. option:: -MD, --write-dependencies
1246
1247Write a depfile containing user and system headers
1248
1249.. option:: -MF<file>
1250
1251Write depfile output from -MMD, -MD, -MM, or -M to <file>
1252
1253.. option:: -MG, --print-missing-file-dependencies
1254
1255Add missing headers to depfile
1256
1257.. option:: -MJ<arg>
1258
1259Write a compilation database entry per input
1260
1261.. option:: -MM, --user-dependencies
1262
1263Like -MMD, but also implies -E and writes to stdout by default
1264
1265.. option:: -MMD, --write-user-dependencies
1266
1267Write a depfile containing user headers
1268
1269.. option:: -MP
1270
1271Create phony target for each dependency (other than main file)
1272
1273.. option:: -MQ<arg>
1274
1275Specify name of main file output to quote in depfile
1276
1277.. option:: -MT<arg>
1278
1279Specify name of main file output in depfile
1280
1281.. option:: -MV
1282
1283Use NMake/Jom format for the depfile
1284
1285Dumping preprocessor state
1286--------------------------
1287
1288Flags allowing the state of the preprocessor to be dumped in various ways.
1289
1290.. option:: -d
1291
1292.. program:: clang1
1293.. option:: -d<arg>
1294.. program:: clang
1295
1296.. option:: -dD
1297
1298Print macro definitions in -E mode in addition to normal output
1299
1300.. option:: -dI
1301
1302Print include directives in -E mode in addition to normal output
1303
1304.. option:: -dM
1305
1306Print macro definitions in -E mode instead of normal output
1307
1308Diagnostic flags
1309~~~~~~~~~~~~~~~~
1310
1311Flags controlling which warnings, errors, and remarks Clang will generate.
1312See the :doc:`full list of warning and remark flags <DiagnosticsReference>`.
1313
1314.. option:: -R<remark>
1315
1316Enable the specified remark
1317
1318.. option:: -Rpass-analysis=<arg>
1319
1320Report transformation analysis from optimization passes whose name matches the given POSIX regular expression
1321
1322.. option:: -Rpass-missed=<arg>
1323
1324Report missed transformations by optimization passes whose name matches the given POSIX regular expression
1325
1326.. option:: -Rpass=<arg>
1327
1328Report transformations performed by optimization passes whose name matches the given POSIX regular expression
1329
1330.. option:: -W<warning>, --extra-warnings, --warn-<arg>, --warn-=<arg>
1331
1332Enable the specified warning
1333
1334.. option:: -Wdeprecated, -Wno-deprecated
1335
1336Enable warnings for deprecated constructs and define \_\_DEPRECATED
1337
1338.. option:: -Wnonportable-cfstrings<arg>, -Wno-nonportable-cfstrings<arg>
1339
1340Target-independent compilation options
1341~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1342.. option:: -Wframe-larger-than=<arg>
1343
1344.. option:: -fPIC, -fno-PIC
1345
1346.. option:: -fPIE, -fno-PIE
1347
1348.. option:: -faccess-control, -fno-access-control
1349
1350.. option:: -faddrsig, -fno-addrsig
1351
1352Emit an address-significance table
1353
1354.. option:: -falign-functions, -fno-align-functions
1355
1356.. program:: clang1
1357.. option:: -falign-functions=<arg>
1358.. program:: clang
1359
1360.. program:: clang1
1361.. option:: -faligned-allocation, -faligned-new, -fno-aligned-allocation
1362.. program:: clang
1363
1364Enable C++17 aligned allocation functions
1365
1366.. option:: -fallow-editor-placeholders, -fno-allow-editor-placeholders
1367
1368Treat editor placeholders as valid source code
1369
1370.. option:: -fallow-unsupported
1371
1372.. option:: -faltivec, -fno-altivec
1373
1374.. option:: -fansi-escape-codes
1375
1376Use ANSI escape codes for diagnostics
1377
1378.. option:: -fapple-kext, -findirect-virtual-calls, -fterminated-vtables
1379
1380Use Apple's kernel extensions ABI
1381
1382.. option:: -fapple-link-rtlib
1383
1384Force linking the clang builtins runtime library
1385
1386.. option:: -fapple-pragma-pack, -fno-apple-pragma-pack
1387
1388Enable Apple gcc-compatible #pragma pack handling
1389
1390.. option:: -fapplication-extension, -fno-application-extension
1391
1392Restrict code to those available for App Extensions
1393
1394.. option:: -fasm, -fno-asm
1395
1396.. option:: -fasm-blocks, -fno-asm-blocks
1397
1398.. option:: -fassociative-math, -fno-associative-math
1399
1400.. option:: -fassume-sane-operator-new, -fno-assume-sane-operator-new
1401
1402.. option:: -fast
1403
1404.. option:: -fastcp
1405
1406.. option:: -fastf
1407
1408.. option:: -fasync-exceptions, -fno-async-exceptions
1409
1410Enable EH Asynchronous exceptions
1411
1412.. option:: -fasynchronous-unwind-tables, -fno-asynchronous-unwind-tables
1413
1414.. option:: -fautolink, -fno-autolink
1415
1416.. option:: -fbasic-block-sections=<arg>
1417
1418Generate labels for each basic block or place each basic block or a subset of basic blocks in its own section.
1419
1420.. option:: -fbinutils-version=<major.minor>
1421
1422Produced object files can use all ELF features supported by this binutils version and newer. If -fno-integrated-as is specified, the generated assembly will consider GNU as support. 'none' means that all ELF features can be used, regardless of binutils support. Defaults to 2.26.
1423
1424.. option:: -fblocks, -fno-blocks
1425
1426Enable the 'blocks' language feature
1427
1428.. option:: -fbootclasspath=<arg>, --bootclasspath <arg>, --bootclasspath=<arg>
1429
1430.. option:: -fborland-extensions, -fno-borland-extensions
1431
1432Accept non-standard constructs supported by the Borland compiler
1433
1434.. option:: -fbracket-depth=<arg>
1435
1436.. option:: -fbuiltin, -fno-builtin
1437
1438.. option:: -fbuiltin-module-map
1439
1440Load the clang builtins module map file.
1441
1442.. program:: clang1
1443.. option:: -fc++-static-destructors, -fno-c++-static-destructors
1444.. program:: clang
1445
1446.. option:: -fcaret-diagnostics, -fno-caret-diagnostics
1447
1448.. option:: -fcf-protection=<arg>, -fcf-protection (equivalent to -fcf-protection=full)
1449
1450Instrument control-flow architecture protection. Options: return, branch, full, none.
1451
1452.. option:: -fcf-runtime-abi=<arg>
1453
1454.. option:: -fchar8\_t, -fno-char8\_t
1455
1456Enable C++ builtin type char8\_t
1457
1458.. option:: -fclasspath=<arg>, --CLASSPATH <arg>, --CLASSPATH=<arg>, --classpath <arg>, --classpath=<arg>
1459
1460.. option:: -fcolor-diagnostics, -fno-color-diagnostics
1461
1462Enable colors in diagnostics
1463
1464.. option:: -fcommon, -fno-common
1465
1466Place uninitialized global variables in a common block
1467
1468.. option:: -fcompile-resource=<arg>, --resource <arg>, --resource=<arg>
1469
1470.. option:: -fconstant-cfstrings, -fno-constant-cfstrings
1471
1472.. option:: -fconstant-string-class=<arg>
1473
1474.. option:: -fconstexpr-backtrace-limit=<arg>
1475
1476.. option:: -fconstexpr-depth=<arg>
1477
1478.. option:: -fconstexpr-steps=<arg>
1479
1480.. option:: -fconvergent-functions
1481
1482Assume functions may be convergent
1483
1484.. option:: -fcoroutines-ts, -fno-coroutines-ts
1485
1486Enable support for the C++ Coroutines TS
1487
1488.. option:: -fcoverage-compilation-dir=<arg>
1489
1490The compilation directory to embed in the coverage mapping.
1491
1492.. option:: -fcoverage-mapping, -fno-coverage-mapping
1493
1494Generate coverage mapping to enable code coverage analysis
1495
1496.. option:: -fcoverage-prefix-map=<arg>
1497
1498remap file source paths in coverage mapping
1499
1500.. option:: -fcreate-profile
1501
1502.. option:: -fcs-profile-generate
1503
1504Generate instrumented code to collect context sensitive execution counts into default.profraw (overridden by LLVM\_PROFILE\_FILE env var)
1505
1506.. program:: clang1
1507.. option:: -fcs-profile-generate=<directory>
1508.. program:: clang
1509
1510Generate instrumented code to collect context sensitive execution counts into <directory>/default.profraw (overridden by LLVM\_PROFILE\_FILE env var)
1511
1512.. option:: -fcuda-approx-transcendentals, -fno-cuda-approx-transcendentals
1513
1514Use approximate transcendental functions
1515
1516.. option:: -fcuda-short-ptr, -fno-cuda-short-ptr
1517
1518Use 32-bit pointers for accessing const/local/shared address spaces
1519
1520.. option:: -fcxx-exceptions, -fno-cxx-exceptions
1521
1522Enable C++ exceptions
1523
1524.. option:: -fcxx-modules, -fno-cxx-modules
1525
1526.. option:: -fdata-sections, -fno-data-sections
1527
1528Place each data in its own section
1529
1530.. option:: -fdebug-compilation-dir=<arg>, -fdebug-compilation-dir <arg>
1531
1532The compilation directory to embed in the debug info
1533
1534.. option:: -fdebug-default-version=<arg>
1535
1536Default DWARF version to use, if a -g option caused DWARF debug info to be produced
1537
1538.. option:: -fdebug-info-for-profiling, -fno-debug-info-for-profiling
1539
1540Emit extra debug info to make sample profile more accurate
1541
1542.. option:: -fdebug-macro, -fno-debug-macro
1543
1544Emit macro debug information
1545
1546.. option:: -fdebug-pass-arguments
1547
1548.. option:: -fdebug-pass-structure
1549
1550.. option:: -fdebug-prefix-map=<arg>
1551
1552remap file source paths in debug info
1553
1554.. option:: -fdebug-ranges-base-address, -fno-debug-ranges-base-address
1555
1556Use DWARF base address selection entries in .debug\_ranges
1557
1558.. option:: -fdebug-types-section, -fno-debug-types-section
1559
1560Place debug types in their own section (ELF Only)
1561
1562.. option:: -fdelayed-template-parsing, -fno-delayed-template-parsing
1563
1564Parse templated function definitions at the end of the translation unit
1565
1566.. option:: -fdelete-null-pointer-checks, -fno-delete-null-pointer-checks
1567
1568Treat usage of null pointers as undefined behavior (default)
1569
1570.. option:: -fdenormal-fp-math=<arg>
1571
1572.. option:: -fdiagnostics-absolute-paths
1573
1574Print absolute paths in diagnostics
1575
1576.. option:: -fdiagnostics-color, -fno-diagnostics-color
1577
1578.. program:: clang1
1579.. option:: -fdiagnostics-color=<arg>
1580.. program:: clang
1581
1582.. option:: -fdiagnostics-hotness-threshold=<value>
1583
1584Prevent optimization remarks from being output if they do not have at least this profile count. Use 'auto' to apply the threshold from profile summary
1585
1586.. option:: -fdiagnostics-show-hotness, -fno-diagnostics-show-hotness
1587
1588Enable profile hotness information in diagnostic line
1589
1590.. option:: -fdiagnostics-show-note-include-stack, -fno-diagnostics-show-note-include-stack
1591
1592Display include stacks for diagnostic notes
1593
1594.. option:: -fdiagnostics-show-option, -fno-diagnostics-show-option
1595
1596Print option name with mappable diagnostics
1597
1598.. option:: -fdiagnostics-show-template-tree
1599
1600Print a template comparison tree for differing templates
1601
1602.. option:: -fdigraphs, -fno-digraphs
1603
1604Enable alternative token representations '<:', ':>', '<%', '%>', '%:', '%:%:' (default)
1605
1606.. option:: -fdirect-access-external-data, -fno-direct-access-external-data
1607
1608Don't use GOT indirection to reference external data symbols
1609
1610.. option:: -fdollars-in-identifiers, -fno-dollars-in-identifiers
1611
1612Allow '$' in identifiers
1613
1614.. option:: -fdouble-square-bracket-attributes, -fno-double-square-bracket-attributes
1615
1616Enable '\[\[\]\]' attributes in all C and C++ language modes
1617
1618.. option:: -fdwarf-directory-asm, -fno-dwarf-directory-asm
1619
1620.. option:: -fdwarf-exceptions
1621
1622Use DWARF style exceptions
1623
1624.. option:: -felide-constructors, -fno-elide-constructors
1625
1626.. option:: -feliminate-unused-debug-symbols, -fno-eliminate-unused-debug-symbols
1627
1628.. option:: -feliminate-unused-debug-types, -fno-eliminate-unused-debug-types
1629
1630Do not emit  debug info for defined but unused types
1631
1632.. option:: -fembed-bitcode=<option>, -fembed-bitcode (equivalent to -fembed-bitcode=all), -fembed-bitcode-marker (equivalent to -fembed-bitcode=marker)
1633
1634Embed LLVM bitcode (option: off, all, bitcode, marker)
1635
1636.. option:: -femit-all-decls
1637
1638Emit all declarations, even if unused
1639
1640.. option:: -femulated-tls, -fno-emulated-tls
1641
1642Use emutls functions to access thread\_local variables
1643
1644.. option:: -fenable-matrix
1645
1646Enable matrix data type and related builtin functions
1647
1648.. option:: -fencoding=<arg>, --encoding <arg>, --encoding=<arg>
1649
1650.. option:: -ferror-limit=<arg>
1651
1652.. option:: -fescaping-block-tail-calls, -fno-escaping-block-tail-calls
1653
1654.. option:: -fexceptions, -fno-exceptions
1655
1656Enable support for exception handling
1657
1658.. option:: -fexec-charset=<arg>
1659
1660.. option:: -fexperimental-new-constant-interpreter
1661
1662Enable the experimental new constant interpreter
1663
1664.. option:: -fextdirs=<arg>, --extdirs <arg>, --extdirs=<arg>
1665
1666.. option:: -fextend-arguments=<arg>
1667
1668Controls how scalar integer arguments are extended in calls to unprototyped and varargs functions
1669
1670.. option:: -ffast-math, -fno-fast-math
1671
1672Allow aggressive, lossy floating-point optimizations
1673
1674.. option:: -ffile-compilation-dir=<arg>
1675
1676The compilation directory to embed in the debug info and coverage mapping.
1677
1678.. option:: -ffile-prefix-map=<arg>
1679
1680remap file source paths in debug info and predefined preprocessor macros
1681
1682.. option:: -ffinite-loops, -fno-finite-loops
1683
1684Assume all loops are finite.
1685
1686.. option:: -ffinite-math-only, -fno-finite-math-only
1687
1688.. option:: -ffixed-point, -fno-fixed-point
1689
1690Enable fixed point types
1691
1692.. option:: -ffor-scope, -fno-for-scope
1693
1694.. option:: -fforce-dwarf-frame, -fno-force-dwarf-frame
1695
1696Always emit a debug frame section
1697
1698.. option:: -fforce-emit-vtables, -fno-force-emit-vtables
1699
1700Emits more virtual tables to improve devirtualization
1701
1702.. option:: -fforce-enable-int128, -fno-force-enable-int128
1703
1704Enable support for int128\_t type
1705
1706.. option:: -ffp-contract=<arg>
1707
1708Form fused FP ops (e.g. FMAs): fast (fuses across statements disregarding pragmas) \| on (only fuses in the same statement unless dictated by pragmas) \| off (never fuses) \| fast-honor-pragmas (fuses across statements unless dictated by pragmas). Default is 'fast' for CUDA, 'fast-honor-pragmas' for HIP, and 'on' otherwise.
1709
1710.. option:: -ffp-exception-behavior=<arg>
1711
1712Specifies the exception behavior of floating-point operations.
1713
1714.. option:: -ffp-model=<arg>
1715
1716Controls the semantics of floating-point calculations.
1717
1718.. option:: -ffreestanding
1719
1720Assert that the compilation takes place in a freestanding environment
1721
1722.. option:: -ffunction-sections, -fno-function-sections
1723
1724Place each function in its own section
1725
1726.. option:: -fgnu-inline-asm, -fno-gnu-inline-asm
1727
1728.. option:: -fgnu-keywords, -fno-gnu-keywords
1729
1730Allow GNU-extension keywords regardless of language standard
1731
1732.. option:: -fgnu-runtime
1733
1734Generate output compatible with the standard GNU Objective-C runtime
1735
1736.. option:: -fgnu89-inline, -fno-gnu89-inline
1737
1738Use the gnu89 inline semantics
1739
1740.. option:: -fgnuc-version=<arg>
1741
1742Sets various macros to claim compatibility with the given GCC version (default is 4.2.1)
1743
1744.. option:: -fgpu-allow-device-init, -fno-gpu-allow-device-init
1745
1746Allow device side init function in HIP (experimental)
1747
1748.. option:: -fgpu-defer-diag, -fno-gpu-defer-diag
1749
1750Defer host/device related diagnostic messages for CUDA/HIP
1751
1752.. option:: -fgpu-rdc, -fcuda-rdc, -fno-gpu-rdc
1753
1754Generate relocatable device code, also known as separate compilation mode
1755
1756.. option:: -fgpu-sanitize, -fno-gpu-sanitize
1757
1758Enable sanitizer for AMDGPU target
1759
1760.. option:: -fhip-fp32-correctly-rounded-divide-sqrt, -fno-hip-fp32-correctly-rounded-divide-sqrt
1761
1762Specify that single precision floating-point divide and sqrt used in the program source are correctly rounded (HIP device compilation only)
1763
1764.. option:: -fhip-new-launch-api, -fno-hip-new-launch-api
1765
1766Use new kernel launching API for HIP
1767
1768.. option:: -fhonor-infinities, -fhonor-infinites, -fno-honor-infinities
1769
1770.. option:: -fhonor-nans, -fno-honor-nans
1771
1772.. option:: -fhosted
1773
1774.. option:: -fignore-exceptions
1775
1776Enable support for ignoring exception handling constructs
1777
1778.. option:: -fimplicit-module-maps, -fmodule-maps, -fno-implicit-module-maps
1779
1780Implicitly search the file system for module map files.
1781
1782.. option:: -fimplicit-modules, -fno-implicit-modules
1783
1784.. option:: -finput-charset=<arg>
1785
1786Specify the default character set for source files
1787
1788.. option:: -finstrument-function-entry-bare
1789
1790Instrument function entry only, after inlining, without arguments to the instrumentation call
1791
1792.. option:: -finstrument-functions
1793
1794Generate calls to instrument function entry and exit
1795
1796.. option:: -finstrument-functions-after-inlining
1797
1798Like -finstrument-functions, but insert the calls after inlining
1799
1800.. option:: -fintegrated-as, -fno-integrated-as, -integrated-as
1801
1802Enable the integrated assembler
1803
1804.. option:: -fintegrated-cc1, -fno-integrated-cc1
1805
1806Run cc1 in-process
1807
1808.. option:: -fjump-tables, -fno-jump-tables
1809
1810Use jump tables for lowering switches
1811
1812.. option:: -fkeep-static-consts, -fno-keep-static-consts
1813
1814Keep static const variables if unused
1815
1816.. option:: -flax-vector-conversions=<arg>, -flax-vector-conversions (equivalent to -flax-vector-conversions=integer), -fno-lax-vector-conversions (equivalent to -flax-vector-conversions=none)
1817
1818Enable implicit vector bit-casts
1819
1820.. option:: -flimited-precision=<arg>
1821
1822.. option:: -flto, -fno-lto
1823
1824Enable LTO in 'full' mode
1825
1826.. option:: -flto-jobs=<arg>
1827
1828Controls the backend parallelism of -flto=thin (default of 0 means the number of threads will be derived from the number of CPUs detected)
1829
1830.. program:: clang1
1831.. option:: -flto=<arg>
1832.. program:: clang
1833
1834Set LTO mode to either 'full' or 'thin'
1835
1836.. program:: clang2
1837.. option:: -flto=auto
1838.. program:: clang
1839
1840.. program:: clang3
1841.. option:: -flto=jobserver
1842.. program:: clang
1843
1844.. option:: -fmacro-backtrace-limit=<arg>
1845
1846.. option:: -fmath-errno, -fno-math-errno
1847
1848Require math functions to indicate errors by setting errno
1849
1850.. option:: -fmax-tokens=<arg>
1851
1852Max total number of preprocessed tokens for -Wmax-tokens.
1853
1854.. option:: -fmax-type-align=<arg>
1855
1856Specify the maximum alignment to enforce on pointers lacking an explicit alignment
1857
1858.. option:: -fmemory-profile, -fno-memory-profile
1859
1860Enable heap memory profiling
1861
1862.. program:: clang1
1863.. option:: -fmemory-profile=<directory>
1864.. program:: clang
1865
1866Enable heap memory profiling and dump results into <directory>
1867
1868.. option:: -fmerge-all-constants, -fno-merge-all-constants
1869
1870Allow merging of constants
1871
1872.. option:: -fmessage-length=<arg>
1873
1874Format message diagnostics so that they fit within N columns
1875
1876.. option:: -fmodule-file-deps, -fno-module-file-deps
1877
1878.. option:: -fmodule-map-file=<file>
1879
1880Load this module map file
1881
1882.. option:: -fmodule-name=<name>, -fmodule-implementation-of <arg>
1883
1884Specify the name of the module to build
1885
1886.. option:: -fmodules, -fno-modules
1887
1888Enable the 'modules' language feature
1889
1890.. option:: -fmodules-decluse, -fno-modules-decluse
1891
1892Require declaration of modules used within a module
1893
1894.. option:: -fmodules-ignore-macro=<arg>
1895
1896Ignore the definition of the given macro when building and loading modules
1897
1898.. option:: -fmodules-search-all, -fno-modules-search-all
1899
1900Search even non-imported modules to resolve references
1901
1902.. option:: -fmodules-strict-decluse
1903
1904Like -fmodules-decluse but requires all headers to be in modules
1905
1906.. option:: -fmodules-ts
1907
1908Enable support for the C++ Modules TS
1909
1910.. option:: -fmodules-validate-input-files-content
1911
1912Validate PCM input files based on content if mtime differs
1913
1914.. option:: -fms-compatibility, -fno-ms-compatibility
1915
1916Enable full Microsoft Visual C++ compatibility
1917
1918.. option:: -fms-compatibility-version=<arg>
1919
1920Dot-separated value representing the Microsoft compiler version number to report in \_MSC\_VER (0 = don't define it (default))
1921
1922.. option:: -fms-extensions, -fno-ms-extensions
1923
1924Accept some non-standard constructs supported by the Microsoft compiler
1925
1926.. option:: -fms-memptr-rep=<arg>
1927
1928.. option:: -fms-volatile
1929
1930.. option:: -fmsc-version=<arg>
1931
1932Microsoft compiler version number to report in \_MSC\_VER (0 = don't define it (default))
1933
1934.. option:: -fmudflap
1935
1936.. option:: -fmudflapth
1937
1938.. option:: -fnested-functions
1939
1940.. option:: -fnew-alignment=<align>, -fnew-alignment <arg>
1941
1942Specifies the largest alignment guaranteed by '::operator new(size\_t)'
1943
1944.. option:: -fnew-infallible
1945
1946Treats throwing global C++ operator new as always returning valid memory (annotates with \_\_attribute\_\_((returns\_nonnull)) and throw()). This is detectable in source.
1947
1948.. option:: -fnext-runtime
1949
1950.. option:: -fno-builtin-<arg>
1951
1952Disable implicit builtin knowledge of a specific function
1953
1954.. option:: -fno-elide-type
1955
1956Do not elide types when printing diagnostics
1957
1958.. option:: -fno-max-type-align
1959
1960.. option:: -fno-strict-modules-decluse
1961
1962.. option:: -fno-temp-file
1963
1964Directly create compilation output files. This may lead to incorrect incremental builds if the compiler crashes
1965
1966.. option:: -fno-working-directory
1967
1968.. option:: -fno\_modules-validate-input-files-content
1969
1970.. program:: clang1
1971.. option:: -fno\_pch-validate-input-files-content
1972.. program:: clang
1973
1974.. option:: -fnoxray-link-deps
1975
1976.. option:: -fobjc-abi-version=<arg>
1977
1978.. option:: -fobjc-arc, -fno-objc-arc
1979
1980Synthesize retain and release calls for Objective-C pointers
1981
1982.. option:: -fobjc-arc-exceptions, -fno-objc-arc-exceptions
1983
1984Use EH-safe code when synthesizing retains and releases in -fobjc-arc
1985
1986.. option:: -fobjc-convert-messages-to-runtime-calls, -fno-objc-convert-messages-to-runtime-calls
1987
1988.. option:: -fobjc-disable-direct-methods-for-testing
1989
1990Ignore attribute objc\_direct so that direct methods can be tested
1991
1992.. option:: -fobjc-encode-cxx-class-template-spec, -fno-objc-encode-cxx-class-template-spec
1993
1994Fully encode c++ class template specialization
1995
1996.. option:: -fobjc-exceptions, -fno-objc-exceptions
1997
1998Enable Objective-C exceptions
1999
2000.. option:: -fobjc-infer-related-result-type, -fno-objc-infer-related-result-type
2001
2002.. option:: -fobjc-legacy-dispatch, -fno-objc-legacy-dispatch
2003
2004.. option:: -fobjc-link-runtime
2005
2006.. option:: -fobjc-nonfragile-abi, -fno-objc-nonfragile-abi
2007
2008.. option:: -fobjc-nonfragile-abi-version=<arg>
2009
2010.. option:: -fobjc-runtime=<arg>
2011
2012Specify the target Objective-C runtime kind and version
2013
2014.. option:: -fobjc-sender-dependent-dispatch
2015
2016.. option:: -fobjc-weak, -fno-objc-weak
2017
2018Enable ARC-style weak references in Objective-C
2019
2020.. option:: -foffload-lto, -fno-offload-lto
2021
2022Enable LTO in 'full' mode for offload compilation
2023
2024.. program:: clang1
2025.. option:: -foffload-lto=<arg>
2026.. program:: clang
2027
2028Set LTO mode to either 'full' or 'thin' for offload compilation
2029
2030.. option:: -fomit-frame-pointer, -fno-omit-frame-pointer
2031
2032.. option:: -fopenmp, -fno-openmp
2033
2034Parse OpenMP pragmas and generate parallel code.
2035
2036.. option:: -fopenmp-simd, -fno-openmp-simd
2037
2038Emit OpenMP code only for SIMD-based constructs.
2039
2040.. option:: -fopenmp-version=<arg>
2041
2042.. program:: clang1
2043.. option:: -fopenmp=<arg>
2044.. program:: clang
2045
2046.. option:: -foperator-arrow-depth=<arg>
2047
2048.. option:: -foperator-names, -fno-operator-names
2049
2050.. option:: -foptimization-record-file=<file>
2051
2052Specify the output name of the file containing the optimization remarks. Implies -fsave-optimization-record. On Darwin platforms, this cannot be used with multiple -arch <arch> options.
2053
2054.. option:: -foptimization-record-passes=<regex>
2055
2056Only include passes which match a specified regular expression in the generated optimization record (by default, include all passes)
2057
2058.. option:: -foptimize-sibling-calls, -fno-optimize-sibling-calls
2059
2060.. option:: -forder-file-instrumentation
2061
2062Generate instrumented code to collect order file into default.profraw file (overridden by '=' form of option or LLVM\_PROFILE\_FILE env var)
2063
2064.. option:: -foutput-class-dir=<arg>, --output-class-directory <arg>, --output-class-directory=<arg>
2065
2066.. option:: -fpack-struct, -fno-pack-struct
2067
2068.. program:: clang1
2069.. option:: -fpack-struct=<arg>
2070.. program:: clang
2071
2072Specify the default maximum struct packing alignment
2073
2074.. option:: -fpascal-strings, -fno-pascal-strings, -mpascal-strings
2075
2076Recognize and construct Pascal-style string literals
2077
2078.. option:: -fpass-plugin=<dsopath>
2079
2080Load pass plugin from a dynamic shared object file (only with new pass manager).
2081
2082.. option:: -fpatchable-function-entry=<N,M>
2083
2084Generate M NOPs before function entry and N-M NOPs after function entry
2085
2086.. option:: -fpcc-struct-return
2087
2088Override the default ABI to return all structs on the stack
2089
2090.. option:: -fpch-codegen, -fno-pch-codegen
2091
2092Generate code for uses of this PCH that assumes an explicit object file will be built for the PCH
2093
2094.. option:: -fpch-debuginfo, -fno-pch-debuginfo
2095
2096Generate debug info for types in an object file built from this PCH and do not generate them elsewhere
2097
2098.. option:: -fpch-instantiate-templates, -fno-pch-instantiate-templates
2099
2100Instantiate templates already while building a PCH
2101
2102.. option:: -fpch-preprocess
2103
2104.. option:: -fpch-validate-input-files-content
2105
2106Validate PCH input files based on content if mtime differs
2107
2108.. option:: -fpic, -fno-pic
2109
2110.. option:: -fpie, -fno-pie
2111
2112.. option:: -fplt, -fno-plt
2113
2114.. option:: -fplugin=<dsopath>
2115
2116Load the named plugin (dynamic shared object)
2117
2118.. option:: -fprebuilt-implicit-modules, -fno-prebuilt-implicit-modules
2119
2120Look up implicit modules in the prebuilt module path
2121
2122.. option:: -fpreserve-as-comments, -fno-preserve-as-comments
2123
2124.. option:: -fproc-stat-report<arg>
2125
2126Print subprocess statistics
2127
2128.. program:: clang1
2129.. option:: -fproc-stat-report=<arg>
2130.. program:: clang
2131
2132Save subprocess statistics to the given file
2133
2134.. option:: -fprofile-arcs, -fno-profile-arcs
2135
2136.. option:: -fprofile-dir=<arg>
2137
2138.. option:: -fprofile-exclude-files=<arg>
2139
2140Instrument only functions from files where names don't match all the regexes separated by a semi-colon
2141
2142.. option:: -fprofile-filter-files=<arg>
2143
2144Instrument only functions from files where names match any regex separated by a semi-colon
2145
2146.. option:: -fprofile-generate, -fno-profile-generate
2147
2148Generate instrumented code to collect execution counts into default.profraw (overridden by LLVM\_PROFILE\_FILE env var)
2149
2150.. program:: clang1
2151.. option:: -fprofile-generate=<directory>
2152.. program:: clang
2153
2154Generate instrumented code to collect execution counts into <directory>/default.profraw (overridden by LLVM\_PROFILE\_FILE env var)
2155
2156.. option:: -fprofile-instr-generate, -fno-profile-instr-generate
2157
2158Generate instrumented code to collect execution counts into default.profraw file (overridden by '=' form of option or LLVM\_PROFILE\_FILE env var)
2159
2160.. program:: clang1
2161.. option:: -fprofile-instr-generate=<file>
2162.. program:: clang
2163
2164Generate instrumented code to collect execution counts into <file> (overridden by LLVM\_PROFILE\_FILE env var)
2165
2166.. option:: -fprofile-instr-use, -fno-profile-instr-use, -fprofile-use
2167
2168.. program:: clang1
2169.. option:: -fprofile-instr-use=<arg>
2170.. program:: clang
2171
2172Use instrumentation data for profile-guided optimization
2173
2174.. option:: -fprofile-list=<arg>
2175
2176Filename defining the list of functions/files to instrument
2177
2178.. option:: -fprofile-remapping-file=<file>
2179
2180Use the remappings described in <file> to match the profile data against names in the program
2181
2182.. option:: -fprofile-sample-accurate, -fauto-profile-accurate, -fno-profile-sample-accurate
2183
2184Specifies that the sample profile is accurate. If the sample
2185               profile is accurate, callsites without profile samples are marked
2186               as cold. Otherwise, treat callsites without profile samples as if
2187               we have no profile
2188
2189.. option:: -fprofile-sample-use, -fauto-profile, -fno-profile-sample-use
2190
2191.. program:: clang1
2192.. option:: -fprofile-sample-use=<arg>, -fauto-profile=<arg>
2193.. program:: clang
2194
2195Enable sample-based profile guided optimizations
2196
2197.. option:: -fprofile-update=<method>
2198
2199Set update method of profile counters (atomic,prefer-atomic,single)
2200
2201.. program:: clang1
2202.. option:: -fprofile-use=<pathname>
2203.. program:: clang
2204
2205Use instrumentation data for profile-guided optimization. If pathname is a directory, it reads from <pathname>/default.profdata. Otherwise, it reads from file <pathname>.
2206
2207.. option:: -fpseudo-probe-for-profiling, -fno-pseudo-probe-for-profiling
2208
2209Emit pseudo probes for sample profiling
2210
2211.. option:: -freciprocal-math, -fno-reciprocal-math
2212
2213Allow division operations to be reassociated
2214
2215.. option:: -freg-struct-return
2216
2217Override the default ABI to return small structs in registers
2218
2219.. option:: -fregister-global-dtors-with-atexit, -fno-register-global-dtors-with-atexit
2220
2221Use atexit or \_\_cxa\_atexit to register global destructors
2222
2223.. option:: -frelaxed-template-template-args, -fno-relaxed-template-template-args
2224
2225Enable C++17 relaxed template template argument matching
2226
2227.. option:: -freroll-loops, -fno-reroll-loops
2228
2229Turn on loop reroller
2230
2231.. option:: -fretain-comments-from-system-headers
2232
2233.. option:: -frewrite-imports, -fno-rewrite-imports
2234
2235.. option:: -frewrite-includes, -fno-rewrite-includes
2236
2237.. option:: -frewrite-map-file=<arg>
2238
2239.. option:: -fropi, -fno-ropi
2240
2241Generate read-only position independent code (ARM only)
2242
2243.. option:: -frounding-math, -fno-rounding-math
2244
2245.. option:: -frtti, -fno-rtti
2246
2247.. option:: -frtti-data, -fno-rtti-data
2248
2249.. option:: -frwpi, -fno-rwpi
2250
2251Generate read-write position independent code (ARM only)
2252
2253.. option:: -fsave-optimization-record, -fno-save-optimization-record
2254
2255Generate a YAML optimization record file
2256
2257.. program:: clang1
2258.. option:: -fsave-optimization-record=<format>
2259.. program:: clang
2260
2261Generate an optimization record file in a specific format
2262
2263.. option:: -fseh-exceptions
2264
2265Use SEH style exceptions
2266
2267.. option:: -fsemantic-interposition, -fno-semantic-interposition
2268
2269.. option:: -fshort-enums, -fno-short-enums
2270
2271Allocate to an enum type only as many bytes as it needs for the declared range of possible values
2272
2273.. option:: -fshort-wchar, -fno-short-wchar
2274
2275Force wchar\_t to be a short unsigned int
2276
2277.. option:: -fshow-column, -fno-show-column
2278
2279.. option:: -fshow-overloads=<arg>
2280
2281Which overload candidates to show when overload resolution fails: best\|all; defaults to all
2282
2283.. option:: -fshow-source-location, -fno-show-source-location
2284
2285.. option:: -fsignaling-math, -fno-signaling-math
2286
2287.. option:: -fsigned-bitfields
2288
2289.. option:: -fsigned-char, -fno-signed-char, --signed-char
2290
2291char is signed
2292
2293.. option:: -fsigned-zeros, -fno-signed-zeros
2294
2295.. option:: -fsized-deallocation, -fno-sized-deallocation
2296
2297Enable C++14 sized global deallocation functions
2298
2299.. option:: -fsjlj-exceptions
2300
2301Use SjLj style exceptions
2302
2303.. option:: -fslp-vectorize, -fno-slp-vectorize, -ftree-slp-vectorize
2304
2305Enable the superword-level parallelism vectorization passes
2306
2307.. option:: -fspell-checking, -fno-spell-checking
2308
2309.. option:: -fspell-checking-limit=<arg>
2310
2311.. option:: -fsplit-dwarf-inlining, -fno-split-dwarf-inlining
2312
2313Provide minimal debug info in the object/executable to facilitate online symbolication/stack traces in the absence of .dwo/.dwp files when using Split DWARF
2314
2315.. option:: -fsplit-lto-unit, -fno-split-lto-unit
2316
2317Enables splitting of the LTO unit
2318
2319.. option:: -fsplit-machine-functions, -fno-split-machine-functions
2320
2321Enable late function splitting using profile information (x86 ELF)
2322
2323.. option:: -fsplit-stack, -fno-split-stack
2324
2325Use segmented stack
2326
2327.. option:: -fstack-clash-protection, -fno-stack-clash-protection
2328
2329Enable stack clash protection
2330
2331.. option:: -fstack-protector, -fno-stack-protector
2332
2333Enable stack protectors for some functions vulnerable to stack smashing. This uses a loose heuristic which considers functions vulnerable if they contain a char (or 8bit integer) array or constant sized calls to alloca , which are of greater size than ssp-buffer-size (default: 8 bytes). All variable sized calls to alloca are considered vulnerable. A function with a stack protector has a guard value added to the stack frame that is checked on function exit. The guard value must be positioned in the stack frame such that a buffer overflow from a vulnerable variable will overwrite the guard value before overwriting the function's return address. The reference stack guard value is stored in a global variable.
2334
2335.. option:: -fstack-protector-all
2336
2337Enable stack protectors for all functions
2338
2339.. option:: -fstack-protector-strong
2340
2341Enable stack protectors for some functions vulnerable to stack smashing. Compared to -fstack-protector, this uses a stronger heuristic that includes functions containing arrays of any size (and any type), as well as any calls to alloca or the taking of an address from a local variable
2342
2343.. option:: -fstack-size-section, -fno-stack-size-section
2344
2345Emit section containing metadata on function stack sizes
2346
2347.. option:: -fstack-usage
2348
2349Emit .su file containing information on function stack sizes
2350
2351.. option:: -fstandalone-debug, -fno-limit-debug-info, -fno-standalone-debug
2352
2353Emit full debug info for all types used by the program
2354
2355.. option:: -fstrict-aliasing, -fno-strict-aliasing
2356
2357.. option:: -fstrict-enums, -fno-strict-enums
2358
2359Enable optimizations based on the strict definition of an enum's value range
2360
2361.. option:: -fstrict-float-cast-overflow, -fno-strict-float-cast-overflow
2362
2363Assume that overflowing float-to-int casts are undefined (default)
2364
2365.. option:: -fstrict-overflow, -fno-strict-overflow
2366
2367.. option:: -fstrict-return, -fno-strict-return
2368
2369.. option:: -fstrict-vtable-pointers, -fno-strict-vtable-pointers
2370
2371Enable optimizations based on the strict rules for overwriting polymorphic C++ objects
2372
2373.. option:: -fstruct-path-tbaa, -fno-struct-path-tbaa
2374
2375.. option:: -fsymbol-partition=<arg>
2376
2377.. option:: -ftabstop=<arg>
2378
2379.. option:: -ftemplate-backtrace-limit=<arg>
2380
2381.. option:: -ftemplate-depth-<arg>
2382
2383.. option:: -ftemplate-depth=<arg>
2384
2385.. option:: -ftest-coverage, -fno-test-coverage
2386
2387.. option:: -fthin-link-bitcode=<arg>
2388
2389Write minimized bitcode to <file> for the ThinLTO thin link only
2390
2391.. option:: -fthinlto-index=<arg>
2392
2393Perform ThinLTO importing using provided function summary index
2394
2395.. option:: -fthreadsafe-statics, -fno-threadsafe-statics
2396
2397.. option:: -ftime-report
2398
2399.. program:: clang1
2400.. option:: -ftime-report=<arg>
2401.. program:: clang
2402
2403(For new pass manager) "per-pass": one report for each pass; "per-pass-run": one report for each pass invocation
2404
2405.. option:: -ftime-trace
2406
2407
2408Turn on time profiler. Generates JSON file based on output filename. Results
2409can be analyzed with chrome://tracing or `Speedscope App
2410<https://www.speedscope.app>`_ for flamegraph visualization.
2411
2412.. option:: -ftime-trace-granularity=<arg>
2413
2414Minimum time granularity (in microseconds) traced by time profiler
2415
2416.. option:: -ftls-model=<arg>
2417
2418.. option:: -ftrap-function=<arg>
2419
2420Issue call to specified function rather than a trap instruction
2421
2422.. option:: -ftrapping-math, -fno-trapping-math
2423
2424.. option:: -ftrapv
2425
2426Trap on integer overflow
2427
2428.. option:: -ftrapv-handler <arg>
2429
2430.. program:: clang1
2431.. option:: -ftrapv-handler=<function name>
2432.. program:: clang
2433
2434Specify the function to be called on overflow
2435
2436.. option:: -ftrigraphs, -fno-trigraphs, -trigraphs, --trigraphs
2437
2438Process trigraph sequences
2439
2440.. option:: -ftrivial-auto-var-init-stop-after=<arg>
2441
2442Stop initializing trivial automatic stack variables after the specified number of instances
2443
2444.. option:: -ftrivial-auto-var-init=<arg>
2445
2446Initialize trivial automatic stack variables: uninitialized (default) \| pattern
2447
2448.. option:: -funique-basic-block-section-names, -fno-unique-basic-block-section-names
2449
2450Use unique names for basic block sections (ELF Only)
2451
2452.. option:: -funique-internal-linkage-names, -fno-unique-internal-linkage-names
2453
2454Uniqueify Internal Linkage Symbol Names by appending the MD5 hash of the module path
2455
2456.. option:: -funique-section-names, -fno-unique-section-names
2457
2458.. option:: -funit-at-a-time, -fno-unit-at-a-time
2459
2460.. option:: -funroll-loops, -fno-unroll-loops
2461
2462Turn on loop unroller
2463
2464.. option:: -funsafe-math-optimizations, -fno-unsafe-math-optimizations
2465
2466.. option:: -funsigned-bitfields
2467
2468.. option:: -funsigned-char, -fno-unsigned-char, --unsigned-char
2469
2470.. option:: -funwind-tables, -fno-unwind-tables
2471
2472.. option:: -fuse-cxa-atexit, -fno-use-cxa-atexit
2473
2474.. option:: -fuse-init-array, -fno-use-init-array
2475
2476.. option:: -fuse-ld=<arg>
2477
2478.. option:: -fuse-line-directives, -fno-use-line-directives
2479
2480Use #line in preprocessed output
2481
2482.. option:: -fminimize-whitespace, -fno-minimize-whitespace
2483
2484Ignore the whitespace from the input file when emitting preprocessor
2485output. It will only contain whitespace when necessary, e.g. to keep two
2486minus signs from merging into to an increment operator. Useful with the
2487-P option to normalize whitespace such that two files with only formatting
2488changes are equal.
2489
2490Only valid with -E on C-like inputs and incompatible with -traditional-cpp.
2491
2492.. option:: -fvalidate-ast-input-files-content
2493
2494Compute and store the hash of input files used to build an AST. Files with mismatching mtime's are considered valid if both contents is identical
2495
2496.. option:: -fveclib=<arg>
2497
2498Use the given vector functions library
2499
2500.. option:: -fvectorize, -fno-vectorize, -ftree-vectorize
2501
2502Enable the loop vectorization passes
2503
2504.. option:: -fverbose-asm, -dA, -fno-verbose-asm
2505
2506Generate verbose assembly output
2507
2508.. option:: -fvirtual-function-elimination, -fno-virtual-function-elimination
2509
2510Enables dead virtual function elimination optimization. Requires -flto=full
2511
2512.. option:: -fvisibility-dllexport=<arg>
2513
2514The visibility for dllexport definitions \[-fvisibility-from-dllstorageclass\]
2515
2516.. option:: -fvisibility-externs-dllimport=<arg>
2517
2518The visibility for dllimport external declarations \[-fvisibility-from-dllstorageclass\]
2519
2520.. option:: -fvisibility-externs-nodllstorageclass=<arg>
2521
2522The visibility for external declarations without an explicit DLL dllstorageclass \[-fvisibility-from-dllstorageclass\]
2523
2524.. option:: -fvisibility-from-dllstorageclass, -fno-visibility-from-dllstorageclass
2525
2526Set the visibility of symbols in the generated code from their DLL storage class
2527
2528.. option:: -fvisibility-global-new-delete-hidden
2529
2530Give global C++ operator new and delete declarations hidden visibility
2531
2532.. option:: -fvisibility-inlines-hidden, -fno-visibility-inlines-hidden
2533
2534Give inline C++ member functions hidden visibility by default
2535
2536.. option:: -fvisibility-inlines-hidden-static-local-var, -fno-visibility-inlines-hidden-static-local-var
2537
2538When -fvisibility-inlines-hidden is enabled, static variables in inline C++ member functions will also be given hidden visibility by default
2539
2540.. option:: -fvisibility-ms-compat
2541
2542Give global types 'default' visibility and global functions and variables 'hidden' visibility by default
2543
2544.. option:: -fvisibility-nodllstorageclass=<arg>
2545
2546The visibility for definitions without an explicit DLL export class \[-fvisibility-from-dllstorageclass\]
2547
2548.. option:: -fvisibility=<arg>
2549
2550Set the default symbol visibility for all global declarations
2551
2552.. option:: -fwarn-stack-size=<arg>
2553
2554.. option:: -fwasm-exceptions
2555
2556Use WebAssembly style exceptions
2557
2558.. option:: -fwhole-program-vtables, -fno-whole-program-vtables
2559
2560Enables whole-program vtable optimization. Requires -flto
2561
2562.. option:: -fwrapv, -fno-wrapv
2563
2564Treat signed integer overflow as two's complement
2565
2566.. option:: -fwritable-strings
2567
2568Store string literals as writable data
2569
2570.. option:: -fxl-pragma-pack, -fno-xl-pragma-pack
2571
2572Enable IBM XL #pragma pack handling
2573
2574.. option:: -fxray-always-emit-customevents, -fno-xray-always-emit-customevents
2575
2576Always emit \_\_xray\_customevent(...) calls even if the containing function is not always instrumented
2577
2578.. option:: -fxray-always-emit-typedevents, -fno-xray-always-emit-typedevents
2579
2580Always emit \_\_xray\_typedevent(...) calls even if the containing function is not always instrumented
2581
2582.. option:: -fxray-always-instrument=<arg>
2583
2584DEPRECATED: Filename defining the whitelist for imbuing the 'always instrument' XRay attribute.
2585
2586.. option:: -fxray-attr-list=<arg>
2587
2588Filename defining the list of functions/types for imbuing XRay attributes.
2589
2590.. option:: -fxray-function-groups=<arg>
2591
2592Only instrument 1 of N groups
2593
2594.. option:: -fxray-function-index, -fno-xray-function-index
2595
2596.. option:: -fxray-ignore-loops, -fno-xray-ignore-loops
2597
2598Don't instrument functions with loops unless they also meet the minimum function size
2599
2600.. option:: -fxray-instruction-threshold<arg>
2601
2602.. program:: clang1
2603.. option:: -fxray-instruction-threshold=<arg>
2604.. program:: clang
2605
2606Sets the minimum function size to instrument with XRay
2607
2608.. option:: -fxray-instrument, -fno-xray-instrument
2609
2610Generate XRay instrumentation sleds on function entry and exit
2611
2612.. option:: -fxray-instrumentation-bundle=<arg>
2613
2614Select which XRay instrumentation points to emit. Options: all, none, function-entry, function-exit, function, custom. Default is 'all'.  'function' includes both 'function-entry' and 'function-exit'.
2615
2616.. option:: -fxray-link-deps
2617
2618Tells clang to add the link dependencies for XRay.
2619
2620.. option:: -fxray-modes=<arg>
2621
2622List of modes to link in by default into XRay instrumented binaries.
2623
2624.. option:: -fxray-never-instrument=<arg>
2625
2626DEPRECATED: Filename defining the whitelist for imbuing the 'never instrument' XRay attribute.
2627
2628.. option:: -fxray-selected-function-group=<arg>
2629
2630When using -fxray-function-groups, select which group of functions to instrument. Valid range is 0 to fxray-function-groups - 1
2631
2632.. option:: -fzero-initialized-in-bss, -fno-zero-initialized-in-bss
2633
2634.. option:: -fzvector, -fno-zvector, -mzvector
2635
2636Enable System z vector language extension
2637
2638.. option:: --gpu-bundle-output, --no-gpu-bundle-output
2639
2640Bundle output files of HIP device compilation
2641
2642.. option:: -pedantic, --pedantic, -no-pedantic, --no-pedantic
2643
2644Warn on language extensions
2645
2646.. option:: -pedantic-errors, --pedantic-errors
2647
2648OpenCL flags
2649------------
2650.. option:: -cl-denorms-are-zero
2651
2652OpenCL only. Allow denormals to be flushed to zero.
2653
2654.. option:: -cl-fast-relaxed-math
2655
2656OpenCL only. Sets -cl-finite-math-only and -cl-unsafe-math-optimizations, and defines \_\_FAST\_RELAXED\_MATH\_\_.
2657
2658.. option:: -cl-finite-math-only
2659
2660OpenCL only. Allow floating-point optimizations that assume arguments and results are not NaNs or +-Inf.
2661
2662.. option:: -cl-fp32-correctly-rounded-divide-sqrt
2663
2664OpenCL only. Specify that single precision floating-point divide and sqrt used in the program source are correctly rounded.
2665
2666.. option:: -cl-kernel-arg-info
2667
2668OpenCL only. Generate kernel argument metadata.
2669
2670.. option:: -cl-mad-enable
2671
2672OpenCL only. Allow use of less precise MAD computations in the generated binary.
2673
2674.. option:: -cl-no-signed-zeros
2675
2676OpenCL only. Allow use of less precise no signed zeros computations in the generated binary.
2677
2678.. option:: -cl-no-stdinc
2679
2680OpenCL only. Disables all standard includes containing non-native compiler types and functions.
2681
2682.. option:: -cl-opt-disable
2683
2684OpenCL only. This option disables all optimizations. By default optimizations are enabled.
2685
2686.. option:: -cl-single-precision-constant
2687
2688OpenCL only. Treat double precision floating-point constant as single precision constant.
2689
2690.. option:: -cl-std=<arg>
2691
2692OpenCL language standard to compile for.
2693
2694.. option:: -cl-strict-aliasing
2695
2696OpenCL only. This option is added for compatibility with OpenCL 1.0.
2697
2698.. option:: -cl-uniform-work-group-size
2699
2700OpenCL only. Defines that the global work-size be a multiple of the work-group size specified to clEnqueueNDRangeKernel
2701
2702.. option:: -cl-unsafe-math-optimizations
2703
2704OpenCL only. Allow unsafe floating-point optimizations.  Also implies -cl-no-signed-zeros and -cl-mad-enable.
2705
2706SYCL flags
2707----------
2708.. option:: -fsycl, -fno-sycl
2709
2710Enables SYCL kernels compilation for device
2711
2712.. option:: -sycl-std=<arg>
2713
2714SYCL language standard to compile for.
2715
2716Target-dependent compilation options
2717~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2718.. option:: -G<size>, -G=<arg>, -msmall-data-limit=<arg>, -msmall-data-threshold=<arg>
2719
2720Put objects of at most <size> bytes into small data section (MIPS / Hexagon)
2721
2722.. option:: -ffixed-x1
2723
2724Reserve the x1 register (AArch64/RISC-V only)
2725
2726.. option:: -ffixed-x10
2727
2728Reserve the x10 register (AArch64/RISC-V only)
2729
2730.. option:: -ffixed-x11
2731
2732Reserve the x11 register (AArch64/RISC-V only)
2733
2734.. option:: -ffixed-x12
2735
2736Reserve the x12 register (AArch64/RISC-V only)
2737
2738.. option:: -ffixed-x13
2739
2740Reserve the x13 register (AArch64/RISC-V only)
2741
2742.. option:: -ffixed-x14
2743
2744Reserve the x14 register (AArch64/RISC-V only)
2745
2746.. option:: -ffixed-x15
2747
2748Reserve the x15 register (AArch64/RISC-V only)
2749
2750.. option:: -ffixed-x16
2751
2752Reserve the x16 register (AArch64/RISC-V only)
2753
2754.. option:: -ffixed-x17
2755
2756Reserve the x17 register (AArch64/RISC-V only)
2757
2758.. option:: -ffixed-x18
2759
2760Reserve the x18 register (AArch64/RISC-V only)
2761
2762.. option:: -ffixed-x19
2763
2764Reserve the x19 register (AArch64/RISC-V only)
2765
2766.. option:: -ffixed-x2
2767
2768Reserve the x2 register (AArch64/RISC-V only)
2769
2770.. option:: -ffixed-x20
2771
2772Reserve the x20 register (AArch64/RISC-V only)
2773
2774.. option:: -ffixed-x21
2775
2776Reserve the x21 register (AArch64/RISC-V only)
2777
2778.. option:: -ffixed-x22
2779
2780Reserve the x22 register (AArch64/RISC-V only)
2781
2782.. option:: -ffixed-x23
2783
2784Reserve the x23 register (AArch64/RISC-V only)
2785
2786.. option:: -ffixed-x24
2787
2788Reserve the x24 register (AArch64/RISC-V only)
2789
2790.. option:: -ffixed-x25
2791
2792Reserve the x25 register (AArch64/RISC-V only)
2793
2794.. option:: -ffixed-x26
2795
2796Reserve the x26 register (AArch64/RISC-V only)
2797
2798.. option:: -ffixed-x27
2799
2800Reserve the x27 register (AArch64/RISC-V only)
2801
2802.. option:: -ffixed-x28
2803
2804Reserve the x28 register (AArch64/RISC-V only)
2805
2806.. option:: -ffixed-x29
2807
2808Reserve the x29 register (AArch64/RISC-V only)
2809
2810.. option:: -ffixed-x3
2811
2812Reserve the x3 register (AArch64/RISC-V only)
2813
2814.. option:: -ffixed-x30
2815
2816Reserve the x30 register (AArch64/RISC-V only)
2817
2818.. option:: -ffixed-x31
2819
2820Reserve the x31 register (AArch64/RISC-V only)
2821
2822.. option:: -ffixed-x4
2823
2824Reserve the x4 register (AArch64/RISC-V only)
2825
2826.. option:: -ffixed-x5
2827
2828Reserve the x5 register (AArch64/RISC-V only)
2829
2830.. option:: -ffixed-x6
2831
2832Reserve the x6 register (AArch64/RISC-V only)
2833
2834.. option:: -ffixed-x7
2835
2836Reserve the x7 register (AArch64/RISC-V only)
2837
2838.. option:: -ffixed-x8
2839
2840Reserve the x8 register (AArch64/RISC-V only)
2841
2842.. option:: -ffixed-x9
2843
2844Reserve the x9 register (AArch64/RISC-V only)
2845
2846.. option:: -m16
2847
2848.. option:: -m32
2849
2850.. option:: -m64
2851
2852.. option:: -mabi=<arg>
2853
2854.. program:: clang1
2855.. option:: -mabi=vec-default
2856.. program:: clang
2857
2858Enable the default Altivec ABI on AIX (AIX only). Uses only volatile vector registers.
2859
2860.. program:: clang2
2861.. option:: -mabi=vec-extabi
2862.. program:: clang
2863
2864Enable the extended Altivec ABI on AIX (AIX only). Uses volatile and nonvolatile vector registers
2865
2866.. option:: -maix-struct-return
2867
2868Return all structs in memory (PPC32 only)
2869
2870.. option:: -malign-branch-boundary=<arg>
2871
2872Specify the boundary's size to align branches
2873
2874.. option:: -malign-branch=<arg1>,<arg2>...
2875
2876Specify types of branches to align
2877
2878.. option:: -malign-double
2879
2880Align doubles to two words in structs (x86 only)
2881
2882.. option:: -mamdgpu-ieee, -mno-amdgpu-ieee
2883
2884Sets the IEEE bit in the expected default floating point  mode register. Floating point opcodes that support exception flag gathering quiet and propagate signaling NaN inputs per IEEE 754-2008. This option changes the ABI. (AMDGPU only)
2885
2886.. option:: -march=<arg>
2887
2888.. option:: -masm=<arg>
2889
2890.. option:: -mbackchain, -mno-backchain
2891
2892Link stack frames through backchain on System Z
2893
2894.. option:: -mbranches-within-32B-boundaries
2895
2896Align selected branches (fused, jcc, jmp) within 32-byte boundary
2897
2898.. option:: -mcmodel=<arg>, -mcmodel=medany (equivalent to -mcmodel=medium), -mcmodel=medlow (equivalent to -mcmodel=small)
2899
2900.. option:: -mcode-object-v3, -mno-code-object-v3
2901
2902Legacy option to specify code object ABI V3 (AMDGPU only)
2903
2904.. option:: -mcode-object-version=<version>
2905
2906Specify code object ABI version. Defaults to 3. (AMDGPU only)
2907
2908.. option:: -mconsole<arg>
2909
2910.. program:: clang1
2911.. option:: -mcpu=<arg>, -mv5 (equivalent to -mcpu=hexagonv5), -mv55 (equivalent to -mcpu=hexagonv55), -mv60 (equivalent to -mcpu=hexagonv60), -mv62 (equivalent to -mcpu=hexagonv62), -mv65 (equivalent to -mcpu=hexagonv65), -mv66 (equivalent to -mcpu=hexagonv66), -mv67 (equivalent to -mcpu=hexagonv67), -mv67t (equivalent to -mcpu=hexagonv67t), -mv68 (equivalent to -mcpu=hexagonv68)
2912.. program:: clang
2913
2914.. option:: -mcrc, -mno-crc
2915
2916Allow use of CRC instructions (ARM/Mips only)
2917
2918.. option:: -mdefault-build-attributes<arg>, -mno-default-build-attributes<arg>
2919
2920.. option:: -mdll<arg>
2921
2922.. option:: -mdouble=<arg>
2923
2924Force double to be 32 bits or 64 bits
2925
2926.. option:: -mdynamic-no-pic<arg>
2927
2928.. option:: -meabi <arg>
2929
2930Set EABI type, e.g. 4, 5 or gnu (default depends on triple)
2931
2932.. option:: -menable-experimental-extensions
2933
2934Enable use of experimental RISC-V extensions.
2935
2936.. option:: -mfentry
2937
2938Insert calls to fentry at function entry (x86/SystemZ only)
2939
2940.. option:: -mfloat-abi=<arg>
2941
2942.. option:: -mfpmath=<arg>
2943
2944.. option:: -mfpu=<arg>
2945
2946.. option:: -mglobal-merge, -mno-global-merge
2947
2948Enable merging of globals
2949
2950.. option:: -mhard-float
2951
2952.. option:: -mhwdiv=<arg>, --mhwdiv <arg>, --mhwdiv=<arg>
2953
2954.. option:: -mhwmult=<arg>
2955
2956.. option:: -miamcu, -mno-iamcu
2957
2958Use Intel MCU ABI
2959
2960.. option:: -mignore-xcoff-visibility
2961
2962Not emit the visibility attribute for asm in AIX OS or give all symbols 'unspecified' visibility in XCOFF object file
2963
2964.. option:: -mimplicit-float, -mno-implicit-float
2965
2966.. option:: -mimplicit-it=<arg>
2967
2968.. option:: -mincremental-linker-compatible, -mno-incremental-linker-compatible
2969
2970(integrated-as) Emit an object file which can be used with an incremental linker
2971
2972.. option:: -miphoneos-version-min=<arg>, -mios-version-min=<arg>
2973
2974.. option:: -mkernel
2975
2976.. option:: -mlong-calls, -mno-long-calls
2977
2978Generate branches with extended addressability, usually via indirect jumps.
2979
2980.. option:: -mlvi-cfi, -mno-lvi-cfi
2981
2982Enable only control-flow mitigations for Load Value Injection (LVI)
2983
2984.. option:: -mlvi-hardening, -mno-lvi-hardening
2985
2986Enable all mitigations for Load Value Injection (LVI)
2987
2988.. option:: -mmacosx-version-min=<arg>, -mmacos-version-min=<arg>
2989
2990Set Mac OS X deployment target
2991
2992.. option:: -mmcu=<arg>
2993
2994.. option:: -mms-bitfields, -mno-ms-bitfields
2995
2996Set the default structure layout to be compatible with the Microsoft compiler standard
2997
2998.. option:: -mnop-mcount
2999
3000Generate mcount/\_\_fentry\_\_ calls as nops. To activate they need to be patched in.
3001
3002.. option:: -momit-leaf-frame-pointer, -mno-omit-leaf-frame-pointer
3003
3004Omit frame pointer setup for leaf functions
3005
3006.. option:: -moslib=<arg>
3007
3008.. option:: -mpacked-stack, -mno-packed-stack
3009
3010Use packed stack layout (SystemZ only).
3011
3012.. option:: -mpad-max-prefix-size=<arg>
3013
3014Specify maximum number of prefixes to use for padding
3015
3016.. option:: -mprefer-vector-width=<arg>
3017
3018Specifies preferred vector width for auto-vectorization. Defaults to 'none' which allows target specific decisions.
3019
3020.. option:: -mqdsp6-compat
3021
3022Enable hexagon-qdsp6 backward compatibility
3023
3024.. option:: -mrecip
3025
3026.. program:: clang1
3027.. option:: -mrecip=<arg1>,<arg2>...
3028.. program:: clang
3029
3030.. option:: -mrecord-mcount
3031
3032Generate a \_\_mcount\_loc section entry for each \_\_fentry\_\_ call.
3033
3034.. option:: -mred-zone, -mno-red-zone
3035
3036.. option:: -mregparm=<arg>
3037
3038.. option:: -mrelax, -mno-relax
3039
3040Enable linker relaxation
3041
3042.. option:: -mrelax-all, -mno-relax-all
3043
3044(integrated-as) Relax all machine instructions
3045
3046.. option:: -mretpoline, -mno-retpoline
3047
3048.. option:: -mrtd, -mno-rtd
3049
3050Make StdCall calling convention the default
3051
3052.. option:: -mseses, -mno-seses
3053
3054Enable speculative execution side effect suppression (SESES). Includes LVI control flow integrity mitigations
3055
3056.. option:: -msign-return-address=<arg>
3057
3058Select return address signing scope
3059
3060.. option:: -msim
3061
3062.. option:: -msoft-float, -mno-soft-float
3063
3064Use software floating point
3065
3066.. option:: -mspeculative-load-hardening, -mno-speculative-load-hardening
3067
3068.. option:: -mstack-alignment=<arg>
3069
3070Set the stack alignment
3071
3072.. option:: -mstack-arg-probe, -mno-stack-arg-probe
3073
3074Enable stack probes
3075
3076.. option:: -mstack-probe-size=<arg>
3077
3078Set the stack probe size
3079
3080.. option:: -mstack-protector-guard-offset=<arg>
3081
3082Use the given offset for addressing the stack-protector guard
3083
3084.. option:: -mstack-protector-guard-reg=<arg>
3085
3086Use the given reg for addressing the stack-protector guard
3087
3088.. option:: -mstack-protector-guard=<arg>
3089
3090Use the given guard (global, tls) for addressing the stack-protector guard
3091
3092.. option:: -mstackrealign, -mno-stackrealign
3093
3094Force realign the stack at entry to every function
3095
3096.. option:: -msvr4-struct-return
3097
3098Return small structs in registers (PPC32 only)
3099
3100.. option:: -mthread-model <arg>
3101
3102The thread model to use, e.g. posix, single (posix by default)
3103
3104.. option:: -mthreads<arg>
3105
3106.. option:: -mthumb, -mno-thumb
3107
3108.. option:: -mtls-direct-seg-refs, -mno-tls-direct-seg-refs
3109
3110Enable direct TLS access through segment registers (default)
3111
3112.. option:: -mtls-size=<arg>
3113
3114Specify bit size of immediate TLS offsets (AArch64 ELF only): 12 (for 4KB) \| 24 (for 16MB, default) \| 32 (for 4GB) \| 48 (for 256TB, needs -mcmodel=large)
3115
3116.. program:: clang1
3117.. option:: -mtune=<arg>
3118.. program:: clang
3119
3120Only supported on X86 and RISC-V. Otherwise accepted for compatibility with GCC.
3121
3122.. option:: -mtvos-version-min=<arg>, -mappletvos-version-min=<arg>
3123
3124.. option:: -municode<arg>
3125
3126.. option:: -munsafe-fp-atomics, -mno-unsafe-fp-atomics
3127
3128Enable unsafe floating point atomic instructions (AMDGPU only)
3129
3130.. option:: -mvx, -mno-vx
3131
3132.. option:: -mwarn-nonportable-cfstrings, -mno-warn-nonportable-cfstrings
3133
3134.. option:: -mwatchos-version-min=<arg>
3135
3136.. option:: -mwavefrontsize64, -mno-wavefrontsize64
3137
3138Specify wavefront size 64 mode (AMDGPU only)
3139
3140.. option:: -mwindows<arg>
3141
3142.. option:: -mx32
3143
3144AARCH64
3145-------
3146.. option:: -fcall-saved-x10
3147
3148Make the x10 register call-saved (AArch64 only)
3149
3150.. option:: -fcall-saved-x11
3151
3152Make the x11 register call-saved (AArch64 only)
3153
3154.. option:: -fcall-saved-x12
3155
3156Make the x12 register call-saved (AArch64 only)
3157
3158.. option:: -fcall-saved-x13
3159
3160Make the x13 register call-saved (AArch64 only)
3161
3162.. option:: -fcall-saved-x14
3163
3164Make the x14 register call-saved (AArch64 only)
3165
3166.. option:: -fcall-saved-x15
3167
3168Make the x15 register call-saved (AArch64 only)
3169
3170.. option:: -fcall-saved-x18
3171
3172Make the x18 register call-saved (AArch64 only)
3173
3174.. option:: -fcall-saved-x8
3175
3176Make the x8 register call-saved (AArch64 only)
3177
3178.. option:: -fcall-saved-x9
3179
3180Make the x9 register call-saved (AArch64 only)
3181
3182.. option:: -mfix-cortex-a53-835769, -mno-fix-cortex-a53-835769
3183
3184Workaround Cortex-A53 erratum 835769 (AArch64 only)
3185
3186.. option:: -mgeneral-regs-only
3187
3188Generate code which only uses the general purpose registers (AArch64 only)
3189
3190.. option:: -mmark-bti-property
3191
3192Add .note.gnu.property with BTI to assembly files (AArch64 only)
3193
3194.. option:: -msve-vector-bits=<arg>
3195
3196Specify the size in bits of an SVE vector register. Defaults to the vector length agnostic value of "scalable". (AArch64 only)
3197
3198AMDGPU
3199------
3200.. option:: -mcumode, -mno-cumode
3201
3202Specify CU wavefront execution mode (AMDGPU only)
3203
3204.. option:: -mtgsplit, -mno-tgsplit
3205
3206Enable threadgroup split execution mode (AMDGPU only)
3207
3208ARM
3209---
3210.. option:: -faapcs-bitfield-load
3211
3212Follows the AAPCS standard that all volatile bit-field write generates at least one load. (ARM only).
3213
3214.. option:: -faapcs-bitfield-width, -fno-aapcs-bitfield-width
3215
3216Follow the AAPCS standard requirement stating that volatile bit-field width is dictated by the field container type. (ARM only).
3217
3218.. option:: -ffixed-r9
3219
3220Reserve the r9 register (ARM only)
3221
3222.. option:: -mcmse
3223
3224Allow use of CMSE (Armv8-M Security Extensions)
3225
3226.. option:: -mexecute-only, -mno-execute-only, -mpure-code
3227
3228Disallow generation of data access to code sections (ARM only)
3229
3230.. option:: -mno-movt
3231
3232Disallow use of movt/movw pairs (ARM only)
3233
3234.. option:: -mno-neg-immediates
3235
3236Disallow converting instructions with negative immediates to their negation or inversion.
3237
3238.. option:: -mnocrc
3239
3240Disallow use of CRC instructions (ARM only)
3241
3242.. option:: -mrestrict-it, -mno-restrict-it
3243
3244Disallow generation of deprecated IT blocks for ARMv8. It is on by default for ARMv8 Thumb mode.
3245
3246.. option:: -mtp=<arg>
3247
3248Thread pointer access method (AArch32/AArch64 only)
3249
3250.. option:: -munaligned-access, -mno-unaligned-access
3251
3252Allow memory accesses to be unaligned (AArch32/AArch64 only)
3253
3254Hexagon
3255-------
3256.. option:: -mieee-rnd-near
3257
3258.. option:: -mmemops, -mno-memops
3259
3260Enable generation of memop instructions
3261
3262.. option:: -mnvj, -mno-nvj
3263
3264Enable generation of new-value jumps
3265
3266.. option:: -mnvs, -mno-nvs
3267
3268Enable generation of new-value stores
3269
3270.. option:: -mpackets, -mno-packets
3271
3272Enable generation of instruction packets
3273
3274Hexagon
3275-------
3276.. option:: -mhvx, -mno-hvx
3277
3278Enable Hexagon Vector eXtensions
3279
3280.. option:: -mhvx-length=<arg>
3281
3282Set Hexagon Vector Length
3283
3284.. program:: clang1
3285.. option:: -mhvx=<arg>
3286.. program:: clang
3287
3288Enable Hexagon Vector eXtensions
3289
3290M68k
3291----
3292.. option:: -ffixed-a0
3293
3294Reserve the a0 register (M68k only)
3295
3296.. option:: -ffixed-a1
3297
3298Reserve the a1 register (M68k only)
3299
3300.. option:: -ffixed-a2
3301
3302Reserve the a2 register (M68k only)
3303
3304.. option:: -ffixed-a3
3305
3306Reserve the a3 register (M68k only)
3307
3308.. option:: -ffixed-a4
3309
3310Reserve the a4 register (M68k only)
3311
3312.. option:: -ffixed-a5
3313
3314Reserve the a5 register (M68k only)
3315
3316.. option:: -ffixed-a6
3317
3318Reserve the a6 register (M68k only)
3319
3320.. option:: -ffixed-d0
3321
3322Reserve the d0 register (M68k only)
3323
3324.. option:: -ffixed-d1
3325
3326Reserve the d1 register (M68k only)
3327
3328.. option:: -ffixed-d2
3329
3330Reserve the d2 register (M68k only)
3331
3332.. option:: -ffixed-d3
3333
3334Reserve the d3 register (M68k only)
3335
3336.. option:: -ffixed-d4
3337
3338Reserve the d4 register (M68k only)
3339
3340.. option:: -ffixed-d5
3341
3342Reserve the d5 register (M68k only)
3343
3344.. option:: -ffixed-d6
3345
3346Reserve the d6 register (M68k only)
3347
3348.. option:: -ffixed-d7
3349
3350Reserve the d7 register (M68k only)
3351
3352.. option:: -m68000
3353
3354.. option:: -m68010
3355
3356.. option:: -m68020
3357
3358.. option:: -m68030
3359
3360.. option:: -m68040
3361
3362.. option:: -m68060
3363
3364MIPS
3365----
3366.. option:: -mabicalls, -mno-abicalls
3367
3368Enable SVR4-style position-independent code (Mips only)
3369
3370.. option:: -mabs=<arg>
3371
3372.. option:: -mcheck-zero-division, -mno-check-zero-division
3373
3374.. option:: -mcompact-branches=<arg>
3375
3376.. option:: -mdouble-float
3377
3378.. option:: -mdsp, -mno-dsp
3379
3380.. option:: -mdspr2, -mno-dspr2
3381
3382.. option:: -membedded-data, -mno-embedded-data
3383
3384Place constants in the .rodata section instead of the .sdata section even if they meet the -G <size> threshold (MIPS)
3385
3386.. option:: -mextern-sdata, -mno-extern-sdata
3387
3388Assume that externally defined data is in the small data if it meets the -G <size> threshold (MIPS)
3389
3390.. option:: -mfp32
3391
3392Use 32-bit floating point registers (MIPS only)
3393
3394.. option:: -mfp64
3395
3396Use 64-bit floating point registers (MIPS only)
3397
3398.. option:: -mginv, -mno-ginv
3399
3400.. option:: -mgpopt, -mno-gpopt
3401
3402Use GP relative accesses for symbols known to be in a small data section (MIPS)
3403
3404.. option:: -mindirect-jump=<arg>
3405
3406Change indirect jump instructions to inhibit speculation
3407
3408.. option:: -mips16
3409
3410.. option:: -mldc1-sdc1, -mno-ldc1-sdc1
3411
3412.. option:: -mlocal-sdata, -mno-local-sdata
3413
3414Extend the -G behaviour to object local data (MIPS)
3415
3416.. option:: -mmadd4, -mno-madd4
3417
3418Enable the generation of 4-operand madd.s, madd.d and related instructions.
3419
3420.. option:: -mmicromips, -mno-micromips
3421
3422.. option:: -mmsa, -mno-msa
3423
3424Enable MSA ASE (MIPS only)
3425
3426.. option:: -mmt, -mno-mt
3427
3428Enable MT ASE (MIPS only)
3429
3430.. option:: -mnan=<arg>
3431
3432.. option:: -mno-mips16
3433
3434.. option:: -msingle-float
3435
3436.. option:: -mvirt, -mno-virt
3437
3438.. option:: -mxgot, -mno-xgot
3439
3440PowerPC
3441-------
3442.. option:: -maltivec, -mno-altivec
3443
3444.. option:: -mcmpb, -mno-cmpb
3445
3446.. option:: -mcrbits, -mno-crbits
3447
3448.. option:: -mcrypto, -mno-crypto
3449
3450.. option:: -mdirect-move, -mno-direct-move
3451
3452.. option:: -mefpu2
3453
3454.. option:: -mfloat128, -mno-float128
3455
3456.. option:: -mfprnd, -mno-fprnd
3457
3458.. option:: -mhtm, -mno-htm
3459
3460.. option:: -minvariant-function-descriptors, -mno-invariant-function-descriptors
3461
3462.. option:: -misel, -mno-isel
3463
3464.. option:: -mlongcall, -mno-longcall
3465
3466.. option:: -mmfocrf, -mmfcrf, -mno-mfocrf
3467
3468.. option:: -mmma, -mno-mma
3469
3470.. option:: -mpaired-vector-memops, -mno-paired-vector-memops
3471
3472.. option:: -mpcrel, -mno-pcrel
3473
3474.. option:: -mpopcntd, -mno-popcntd
3475
3476.. option:: -mpower10-vector, -mno-power10-vector
3477
3478.. option:: -mpower8-vector, -mno-power8-vector
3479
3480.. option:: -mpower9-vector, -mno-power9-vector
3481
3482.. option:: -mprefixed, -mno-prefixed
3483
3484.. option:: -mprivileged
3485
3486.. option:: -mrop-protect
3487
3488.. option:: -msecure-plt
3489
3490.. option:: -mspe, -mno-spe
3491
3492.. option:: -mvsx, -mno-vsx
3493
3494WebAssembly
3495-----------
3496.. option:: -matomics, -mno-atomics
3497
3498.. option:: -mbulk-memory, -mno-bulk-memory
3499
3500.. option:: -mexception-handling, -mno-exception-handling
3501
3502.. option:: -mmultivalue, -mno-multivalue
3503
3504.. option:: -mmutable-globals, -mno-mutable-globals
3505
3506.. option:: -mnontrapping-fptoint, -mno-nontrapping-fptoint
3507
3508.. option:: -mreference-types, -mno-reference-types
3509
3510.. option:: -msign-ext, -mno-sign-ext
3511
3512.. option:: -msimd128, -mno-simd128
3513
3514.. option:: -mtail-call, -mno-tail-call
3515
3516WebAssembly Driver
3517------------------
3518.. option:: -mexec-model=<arg>
3519
3520Execution model (WebAssembly only)
3521
3522X86
3523---
3524.. option:: -m3dnow, -mno-3dnow
3525
3526.. option:: -m3dnowa, -mno-3dnowa
3527
3528.. option:: -madx, -mno-adx
3529
3530.. option:: -maes, -mno-aes
3531
3532.. option:: -mamx-bf16, -mno-amx-bf16
3533
3534.. option:: -mamx-int8, -mno-amx-int8
3535
3536.. option:: -mamx-tile, -mno-amx-tile
3537
3538.. option:: -mavx, -mno-avx
3539
3540.. option:: -mavx2, -mno-avx2
3541
3542.. option:: -mavx512bf16, -mno-avx512bf16
3543
3544.. option:: -mavx512bitalg, -mno-avx512bitalg
3545
3546.. option:: -mavx512bw, -mno-avx512bw
3547
3548.. option:: -mavx512cd, -mno-avx512cd
3549
3550.. option:: -mavx512dq, -mno-avx512dq
3551
3552.. option:: -mavx512er, -mno-avx512er
3553
3554.. option:: -mavx512f, -mno-avx512f
3555
3556.. option:: -mavx512fp16, -mno-avx512fp16
3557
3558.. option:: -mavx512ifma, -mno-avx512ifma
3559
3560.. option:: -mavx512pf, -mno-avx512pf
3561
3562.. option:: -mavx512vbmi, -mno-avx512vbmi
3563
3564.. option:: -mavx512vbmi2, -mno-avx512vbmi2
3565
3566.. option:: -mavx512vl, -mno-avx512vl
3567
3568.. option:: -mavx512vnni, -mno-avx512vnni
3569
3570.. option:: -mavx512vp2intersect, -mno-avx512vp2intersect
3571
3572.. option:: -mavx512vpopcntdq, -mno-avx512vpopcntdq
3573
3574.. option:: -mavxvnni, -mno-avxvnni
3575
3576.. option:: -mbmi, -mno-bmi
3577
3578.. option:: -mbmi2, -mno-bmi2
3579
3580.. option:: -mcldemote, -mno-cldemote
3581
3582.. option:: -mclflushopt, -mno-clflushopt
3583
3584.. option:: -mclwb, -mno-clwb
3585
3586.. option:: -mclzero, -mno-clzero
3587
3588.. option:: -mcx16, -mno-cx16
3589
3590.. option:: -menqcmd, -mno-enqcmd
3591
3592.. option:: -mf16c, -mno-f16c
3593
3594.. option:: -mfma, -mno-fma
3595
3596.. option:: -mfma4, -mno-fma4
3597
3598.. option:: -mfsgsbase, -mno-fsgsbase
3599
3600.. option:: -mfxsr, -mno-fxsr
3601
3602.. option:: -mgfni, -mno-gfni
3603
3604.. option:: -mhreset, -mno-hreset
3605
3606.. option:: -minvpcid, -mno-invpcid
3607
3608.. option:: -mkl, -mno-kl
3609
3610.. option:: -mlwp, -mno-lwp
3611
3612.. option:: -mlzcnt, -mno-lzcnt
3613
3614.. option:: -mmmx, -mno-mmx
3615
3616.. option:: -mmovbe, -mno-movbe
3617
3618.. option:: -mmovdir64b, -mno-movdir64b
3619
3620.. option:: -mmovdiri, -mno-movdiri
3621
3622.. option:: -mmwaitx, -mno-mwaitx
3623
3624.. option:: -mpclmul, -mno-pclmul
3625
3626.. option:: -mpconfig, -mno-pconfig
3627
3628.. option:: -mpku, -mno-pku
3629
3630.. option:: -mpopcnt, -mno-popcnt
3631
3632.. option:: -mprefetchwt1, -mno-prefetchwt1
3633
3634.. option:: -mprfchw, -mno-prfchw
3635
3636.. option:: -mptwrite, -mno-ptwrite
3637
3638.. option:: -mrdpid, -mno-rdpid
3639
3640.. option:: -mrdrnd, -mno-rdrnd
3641
3642.. option:: -mrdseed, -mno-rdseed
3643
3644.. option:: -mretpoline-external-thunk, -mno-retpoline-external-thunk
3645
3646.. option:: -mrtm, -mno-rtm
3647
3648.. option:: -msahf, -mno-sahf
3649
3650.. option:: -mserialize, -mno-serialize
3651
3652.. option:: -msgx, -mno-sgx
3653
3654.. option:: -msha, -mno-sha
3655
3656.. option:: -mshstk, -mno-shstk
3657
3658.. option:: -msse, -mno-sse
3659
3660.. option:: -msse2, -mno-sse2
3661
3662.. option:: -msse3, -mno-sse3
3663
3664.. option:: -msse4.1, -mno-sse4.1
3665
3666.. program:: clang1
3667.. option:: -msse4.2, -mno-sse4.2, -msse4
3668.. program:: clang
3669
3670.. option:: -msse4a, -mno-sse4a
3671
3672.. option:: -mssse3, -mno-ssse3
3673
3674.. option:: -mtbm, -mno-tbm
3675
3676.. option:: -mtsxldtrk, -mno-tsxldtrk
3677
3678.. option:: -muintr, -mno-uintr
3679
3680.. option:: -mvaes, -mno-vaes
3681
3682.. option:: -mvpclmulqdq, -mno-vpclmulqdq
3683
3684.. option:: -mvzeroupper, -mno-vzeroupper
3685
3686.. option:: -mwaitpkg, -mno-waitpkg
3687
3688.. option:: -mwbnoinvd, -mno-wbnoinvd
3689
3690.. option:: -mwidekl, -mno-widekl
3691
3692.. option:: -mx87, -m80387, -mno-x87
3693
3694.. option:: -mxop, -mno-xop
3695
3696.. option:: -mxsave, -mno-xsave
3697
3698.. option:: -mxsavec, -mno-xsavec
3699
3700.. option:: -mxsaveopt, -mno-xsaveopt
3701
3702.. option:: -mxsaves, -mno-xsaves
3703
3704RISCV
3705-----
3706.. option:: -msave-restore, -mno-save-restore
3707
3708Enable using library calls for save and restore
3709
3710Long double flags
3711-----------------
3712Selects the long double implementation
3713
3714.. option:: -mlong-double-128
3715
3716Force long double to be 128 bits
3717
3718.. option:: -mlong-double-64
3719
3720Force long double to be 64 bits
3721
3722.. option:: -mlong-double-80
3723
3724Force long double to be 80 bits, padded to 128 bits for storage
3725
3726Optimization level
3727~~~~~~~~~~~~~~~~~~
3728
3729Flags controlling how much optimization should be performed.
3730
3731.. option:: -O<arg>, -O (equivalent to -O1), --optimize, --optimize=<arg>
3732
3733.. option:: -Ofast<arg>
3734
3735Debug information generation
3736~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3737
3738Flags controlling how much and what kind of debug information should be
3739generated.
3740
3741Kind and level of debug information
3742-----------------------------------
3743.. option:: -g, --debug, --debug=<arg>
3744
3745Generate source-level debug information
3746
3747.. option:: -gdwarf
3748
3749Generate source-level debug information with the default dwarf version
3750
3751.. option:: -gdwarf-2
3752
3753Generate source-level debug information with dwarf version 2
3754
3755.. option:: -gdwarf-3
3756
3757Generate source-level debug information with dwarf version 3
3758
3759.. option:: -gdwarf-4
3760
3761Generate source-level debug information with dwarf version 4
3762
3763.. option:: -gdwarf-5
3764
3765Generate source-level debug information with dwarf version 5
3766
3767.. option:: -gdwarf32
3768
3769Enables DWARF32 format for ELF binaries, if debug information emission is enabled.
3770
3771.. option:: -gdwarf64
3772
3773Enables DWARF64 format for ELF binaries, if debug information emission is enabled.
3774
3775.. option:: -gfull
3776
3777.. option:: -ginline-line-tables, -gno-inline-line-tables
3778
3779.. option:: -gused
3780
3781Debug level
3782___________
3783.. option:: -g0
3784
3785.. option:: -g2
3786
3787.. option:: -g3
3788
3789.. option:: -ggdb0
3790
3791.. option:: -ggdb1
3792
3793.. option:: -ggdb2
3794
3795.. option:: -ggdb3
3796
3797.. option:: -gline-directives-only
3798
3799Emit debug line info directives only
3800
3801.. option:: -gline-tables-only, -g1, -gmlt
3802
3803Emit debug line number tables only
3804
3805.. option:: -gmodules
3806
3807Generate debug info with external references to clang modules or precompiled headers
3808
3809Debugger to tune debug information for
3810______________________________________
3811.. option:: -gdbx
3812
3813.. option:: -ggdb
3814
3815.. option:: -glldb
3816
3817.. option:: -gsce
3818
3819Debug information flags
3820-----------------------
3821.. option:: -gcolumn-info, -gno-column-info
3822
3823.. option:: -gdwarf-aranges
3824
3825.. option:: -gembed-source, -gno-embed-source
3826
3827Embed source text in DWARF debug sections
3828
3829.. option:: -ggnu-pubnames, -gno-gnu-pubnames
3830
3831.. option:: -gpubnames, -gno-pubnames
3832
3833.. option:: -grecord-command-line, -gno-record-command-line, -grecord-gcc-switches
3834
3835.. option:: -gsplit-dwarf, -gno-split-dwarf
3836
3837.. program:: clang1
3838.. option:: -gsplit-dwarf=<arg>
3839.. program:: clang
3840
3841Set DWARF fission mode to either 'split' or 'single'
3842
3843.. option:: -gstrict-dwarf, -gno-strict-dwarf
3844
3845.. option:: -gz=<arg>, -gz (equivalent to -gz=zlib)
3846
3847DWARF debug sections compression type
3848
3849Static analyzer flags
3850=====================
3851
3852Flags controlling the behavior of the Clang Static Analyzer.
3853
3854.. option:: -Xanalyzer <arg>
3855
3856Pass <arg> to the static analyzer
3857
3858Fortran compilation flags
3859=========================
3860
3861Flags that will be passed onto the ``gfortran`` compiler when Clang is given
3862a Fortran input.
3863
3864.. option:: -A<arg>, --assert <arg>, --assert=<arg>
3865
3866.. option:: -A-<arg>
3867
3868.. option:: -faggressive-function-elimination, -fno-aggressive-function-elimination
3869
3870.. option:: -falign-commons, -fno-align-commons
3871
3872.. option:: -fall-intrinsics, -fno-all-intrinsics
3873
3874.. option:: -fautomatic, -fno-automatic
3875
3876.. option:: -fbacktrace, -fno-backtrace
3877
3878.. option:: -fblas-matmul-limit=<arg>
3879
3880.. option:: -fbounds-check, -fno-bounds-check
3881
3882.. option:: -fcheck-array-temporaries, -fno-check-array-temporaries
3883
3884.. option:: -fcheck=<arg>
3885
3886.. option:: -fcoarray=<arg>
3887
3888.. option:: -fconvert=<arg>
3889
3890.. option:: -fcray-pointer, -fno-cray-pointer
3891
3892.. option:: -fd-lines-as-code, -fno-d-lines-as-code
3893
3894.. option:: -fd-lines-as-comments, -fno-d-lines-as-comments
3895
3896.. option:: -fdollar-ok, -fno-dollar-ok
3897
3898.. option:: -fdump-fortran-optimized, -fno-dump-fortran-optimized
3899
3900.. option:: -fdump-fortran-original, -fno-dump-fortran-original
3901
3902.. option:: -fdump-parse-tree, -fno-dump-parse-tree
3903
3904.. option:: -fexternal-blas, -fno-external-blas
3905
3906.. option:: -ff2c, -fno-f2c
3907
3908.. option:: -ffpe-trap=<arg>
3909
3910.. option:: -ffree-line-length-<arg>
3911
3912.. option:: -ffrontend-optimize, -fno-frontend-optimize
3913
3914.. option:: -finit-character=<arg>
3915
3916.. option:: -finit-integer=<arg>
3917
3918.. option:: -finit-local-zero, -fno-init-local-zero
3919
3920.. option:: -finit-logical=<arg>
3921
3922.. option:: -finit-real=<arg>
3923
3924.. option:: -finteger-4-integer-8, -fno-integer-4-integer-8
3925
3926.. option:: -fmax-array-constructor=<arg>
3927
3928.. option:: -fmax-errors=<arg>
3929
3930.. option:: -fmax-identifier-length, -fno-max-identifier-length
3931
3932.. option:: -fmax-stack-var-size=<arg>
3933
3934.. option:: -fmax-subrecord-length=<arg>
3935
3936.. option:: -fmodule-private, -fno-module-private
3937
3938.. option:: -fpack-derived, -fno-pack-derived
3939
3940.. option:: -fprotect-parens, -fno-protect-parens
3941
3942.. option:: -frange-check, -fno-range-check
3943
3944.. option:: -freal-4-real-10, -fno-real-4-real-10
3945
3946.. option:: -freal-4-real-16, -fno-real-4-real-16
3947
3948.. option:: -freal-4-real-8, -fno-real-4-real-8
3949
3950.. option:: -freal-8-real-10, -fno-real-8-real-10
3951
3952.. option:: -freal-8-real-16, -fno-real-8-real-16
3953
3954.. option:: -freal-8-real-4, -fno-real-8-real-4
3955
3956.. option:: -frealloc-lhs, -fno-realloc-lhs
3957
3958.. option:: -frecord-marker=<arg>
3959
3960.. option:: -frecursive, -fno-recursive
3961
3962.. option:: -frepack-arrays, -fno-repack-arrays
3963
3964.. option:: -fsecond-underscore, -fno-second-underscore
3965
3966.. option:: -fsign-zero, -fno-sign-zero
3967
3968.. option:: -fstack-arrays, -fno-stack-arrays
3969
3970.. option:: -funderscoring, -fno-underscoring
3971
3972.. option:: -fwhole-file, -fno-whole-file
3973
3974.. option:: -imultilib <arg>
3975
3976.. option:: -static-libgfortran
3977
3978Linker flags
3979============
3980Flags that are passed on to the linker
3981
3982.. option:: -L<dir>, --library-directory <arg>, --library-directory=<arg>
3983
3984Add directory to library search path
3985
3986.. option:: -Mach
3987
3988.. option:: -T<script>
3989
3990Specify <script> as linker script
3991
3992.. option:: -Tbss<addr>
3993
3994Set starting address of BSS to <addr>
3995
3996.. option:: -Tdata<addr>
3997
3998Set starting address of DATA to <addr>
3999
4000.. option:: -Ttext<addr>
4001
4002Set starting address of TEXT to <addr>
4003
4004.. option:: -Wl,<arg>,<arg2>...
4005
4006Pass the comma separated arguments in <arg> to the linker
4007
4008.. option:: -X
4009
4010.. option:: -Xlinker <arg>, --for-linker <arg>, --for-linker=<arg>
4011
4012Pass <arg> to the linker
4013
4014.. program:: clang1
4015.. option:: -Z
4016.. program:: clang
4017
4018.. option:: -coverage, --coverage
4019
4020.. option:: -e<arg>, --entry
4021
4022.. option:: -filelist <arg>
4023
4024.. option:: --hip-device-lib=<arg>
4025
4026HIP device library
4027
4028.. option:: -l<arg>
4029
4030.. option:: --ld-path=<arg>
4031
4032.. option:: -nostartfiles
4033
4034.. program:: clang1
4035.. option:: -nostdlib, --no-standard-libraries
4036.. program:: clang
4037
4038.. option:: -pie
4039
4040.. option:: -r
4041
4042.. option:: -rdynamic
4043
4044.. option:: --rocm-device-lib-path=<arg>, --hip-device-lib-path=<arg>
4045
4046ROCm device library path. Alternative to rocm-path.
4047
4048.. option:: -rpath <arg>
4049
4050.. option:: -s
4051
4052.. option:: -shared, --shared
4053
4054.. option:: -specs=<arg>, --specs=<arg>
4055
4056.. option:: -static, --static
4057
4058.. option:: -static-pie
4059
4060.. option:: -t
4061
4062.. option:: -u<arg>, --force-link <arg>, --force-link=<arg>
4063
4064.. option:: -undef
4065
4066undef all system defines
4067
4068.. option:: -undefined<arg>, --no-undefined
4069
4070.. option:: -z <arg>
4071
4072Pass -z <arg> to the linker
4073
4074