[PATCH] D54473: [sanitizers] Initial implementation for -fsanitize=init-locals

glider created this revision. glider added reviewers: kcc, rjmccall, rsmith. Herald added a subscriber: cfe-commits. This patch adds a new feature, -fsanitize=init-locals, which generates zero initializers for uninitialized local variables. There's been discussions in the security community

[PATCH] D54473: [sanitizers] Initial implementation for -fsanitize=init-locals

glider added a comment. In https://reviews.llvm.org/D54473#1297460, @kcc wrote: > This new flag inhibits the warnings from -Wuninitialized, right? > While this is fine for experimenting (and I want to have this in ToT to > enable wide experimentation) > we should clearly state (in the

[PATCH] D54473: [sanitizers] Initial implementation for -fsanitize=init-locals

glider added inline comments. Comment at: include/clang/Basic/Sanitizers.def:163 +// Initialize local variables. +SANITIZER("init-locals", InitLocals) + lebedev.ri wrote: > Unless i'm mistaken, I suspect you may see some surprising behavior here, >

[PATCH] D57898: CodeGen: Fix PR40605 by splitting constant struct initializers

glider added inline comments. Comment at: clang/test/CodeGenCXX/auto-var-init.cpp:130 +// PATTERN-NOT-O1: @__const.test_bool4_custom.custom +// ZERO-NOT-O1: @__const.test_bool4_custom.custom + jfb wrote: > `-NOT` is in the wrong place above. Hm, I wonder if lit

[PATCH] D57898: CodeGen: Fix PR40605 by splitting constant struct initializers

glider updated this revision to Diff 188694. glider marked 3 inline comments as done. glider retitled this revision from "CodeGen: Fix PR40605" to "CodeGen: Fix PR40605 by splitting constant struct initializers". CHANGES SINCE LAST ACTION https://reviews.llvm.org/D57898/new/

[PATCH] D57898: CodeGen: Fix PR40605

glider updated this revision to Diff 188548. glider marked an inline comment as done. glider retitled this revision from "CodeGen: Fix PR40605: split constant structures generated by -ftrivial-auto-var-init when emitting initializators" to "CodeGen: Fix PR40605". glider edited the summary of

[PATCH] D57898: CodeGen: Fix PR40605

glider added a comment. Rebased the code, updated the tests, added FIXME. CHANGES SINCE LAST ACTION https://reviews.llvm.org/D57898/new/ https://reviews.llvm.org/D57898 ___ cfe-commits mailing list cfe-commits@lists.llvm.org

[PATCH] D58885: Variable auto-init: split out small arrays

glider added a comment. The change itself looks good. It doesn't seem to regress kernel performance on ARM64. I haven't got to testing on x86 yet, but don't anticipate any problems either. Comment at: lib/CodeGen/CGDecl.cpp:1206 + bool canDoSingleStore =

[PATCH] D58885: Variable auto-init: split out small arrays

glider accepted this revision. glider added inline comments. This revision is now accepted and ready to land. Comment at: lib/CodeGen/CGDecl.cpp:1206 + bool canDoSingleStore = Ty->isIntOrIntVectorTy() || + Ty->isPtrOrPtrVectorTy() ||

[PATCH] D58885: Variable auto-init: split out small arrays

glider added inline comments. Comment at: test/CodeGenCXX/auto-var-init.cpp:1025 // PATTERN-LABEL: @test_intptr4_uninit() -// PATTERN: call void @llvm.memcpy{{.*}} @__const.test_intptr4_uninit.uninit -// ZERO-LABEL: @test_intptr4_uninit() -// ZERO: call void @llvm.memset{{.*}},

[PATCH] D57898: CodeGen: Fix PR40605 by splitting constant struct initializers

glider closed this revision. glider added a comment. Landed r355181, thank you! CHANGES SINCE LAST ACTION https://reviews.llvm.org/D57898/new/ https://reviews.llvm.org/D57898 ___ cfe-commits mailing list cfe-commits@lists.llvm.org

[PATCH] D57898: [RFC] Split constant structures generated by -ftrivial-auto-var-init when emitting initializators

glider marked 2 inline comments as not done. glider added a comment. Sorry, didn't mean to mark these done. CHANGES SINCE LAST ACTION https://reviews.llvm.org/D57898/new/ https://reviews.llvm.org/D57898 ___ cfe-commits mailing list

[PATCH] D57898: [RFC] Split constant structures generated by -ftrivial-auto-var-init when emitting initializators

glider marked 2 inline comments as done. glider added inline comments. Comment at: tools/clang/lib/CodeGen/CGDecl.cpp:1143 +const llvm::StructLayout *Layout = +CGM.getDataLayout().getStructLayout(cast(Ty)); +for (unsigned i = 0; i != constant->getNumOperands();

[PATCH] D57898: CodeGen: Fix PR40605: split constant structures generated by -ftrivial-auto-var-init when emitting initializators

glider added a comment. Re: case when STy != Loc.getElementType() - this is already covered by other Clang tests. I'm still unsure about the heuristic here. I believe that for auto-initialization we want to be quite aggressive with these splits (unlike for regular constant stores). Perhaps we

[PATCH] D57898: CodeGen: Fix PR40605: split constant structures generated by -ftrivial-auto-var-init when emitting initializators

glider updated this revision to Diff 185990. glider retitled this revision from "[RFC] Split constant structures generated by -ftrivial-auto-var-init when emitting initializators" to "CodeGen: Fix PR40605: split constant structures generated by -ftrivial-auto-var-init when emitting

[PATCH] D57898: CodeGen: Fix PR40605: split constant structures generated by -ftrivial-auto-var-init when emitting initializators

glider added a comment. > Can you add a link to bug 40605 in the commit message? It's in the title now, doesn't that count? :) CHANGES SINCE LAST ACTION https://reviews.llvm.org/D57898/new/ https://reviews.llvm.org/D57898 ___ cfe-commits

[PATCH] D57898: CodeGen: Fix PR40605: split constant structures generated by -ftrivial-auto-var-init when emitting initializators

glider updated this revision to Diff 186469. glider marked an inline comment as done. glider added a subscriber: pcc. glider added a comment. Added a helper function to decide whether we want to break the structure or not. Right now it only checks for optimization level and structure size. The

[PATCH] D57898: CodeGen: Fix PR40605: split constant structures generated by -ftrivial-auto-var-init when emitting initializators

glider marked 4 inline comments as done. glider added inline comments. Herald added a subscriber: jdoerfert. Comment at: tools/clang/test/CodeGenCXX/auto-var-init.cpp:496 // ZERO-LABEL: @test_smallpartinit_uninit() // ZERO: call void @llvm.memset{{.*}}, i8 0,

[PATCH] D57898: CodeGen: Fix PR40605: split constant structures generated by -ftrivial-auto-var-init when emitting initializators

glider added a comment. This patch shaved off 2.5% of performance slowdown on https://github.com/kamalmarhubi/linux-ipc-benchmarks/blob/master/af_inet_loopback.c on a kernel built in pattern-initialization mode. The .rodata size went down by only 12K (0.5%, since we now split only structures

[PATCH] D57898: CodeGen: Fix PR40605: split constant structures generated by -ftrivial-auto-var-init when emitting initializators

glider added inline comments. Comment at: tools/clang/lib/CodeGen/CGDecl.cpp:979 + if (CGM.getCodeGenOpts().OptimizationLevel == 0) +return false; + if (GlobalSize <= SizeLimit) jfb wrote: > The general 64-byte heuristic is fine with me. It's just a bit

[PATCH] D57898: CodeGen: Fix PR40605: split constant structures generated by -ftrivial-auto-var-init when emitting initializators

glider added a comment. Well, now we somewhat break padding initialization. E.g. for the following struct: struct s { int a; char b; long int c; }; we generate the following constant initializer with `-O0`: .L__const.foo.local: .long 2863311530 #

[PATCH] D57898: CodeGen: Fix PR40605: split constant structures generated by -ftrivial-auto-var-init when emitting initializators

glider updated this revision to Diff 186605. glider marked 4 inline comments as done. glider added a comment. Addressed the comments. CHANGES SINCE LAST ACTION https://reviews.llvm.org/D57898/new/ https://reviews.llvm.org/D57898 Files: tools/clang/lib/CodeGen/CGDecl.cpp

[PATCH] D57898: CodeGen: Fix PR40605: split constant structures generated by -ftrivial-auto-var-init when emitting initializators

glider added a comment. ... which happily skips the padding. It occurs to me that we need to properly handle the padding in `patternFor` before we'll be able to split the structures. CHANGES SINCE LAST ACTION https://reviews.llvm.org/D57898/new/ https://reviews.llvm.org/D57898

[PATCH] D57898: CodeGen: Fix PR40605: split constant structures generated by -ftrivial-auto-var-init when emitting initializators

glider added a comment. FWIW I think I've almost finished handling the padding: https://reviews.llvm.org/D58188 I haven't checked whether it works correctly with padding in custom initializers (like `struct s local = {1, 2}`), but TEST_UNINIT and TEST_BRACES are covered already. CHANGES

[PATCH] D54604: Automatic variable initialization

glider added a comment. Not sure if a similar problem was mentioned already or not, but the following program: void alloc_sock(int *err); int try_send(); int send(int len) { int err; if (len) { err = -1; alloc_sock(); err = try_send(); } return -1;

[PATCH] D54604: Automatic variable initialization

glider added a comment. (Overall, since this is an "unsupported" feature, is it ok to file Bugzilla bugs for it?) Repository: rL LLVM CHANGES SINCE LAST ACTION https://reviews.llvm.org/D54604/new/ https://reviews.llvm.org/D54604 ___

[PATCH] D54604: Automatic variable initialization

glider added a comment. Herald added a project: LLVM. I think doing separate stores of 0xAA into struct members instead of copying part of .rodata over it will let us do a much better job in optimizing away redundant stores. I've opened https://bugs.llvm.org/show_bug.cgi?id=40605 to track that.

[PATCH] D61664: [NewPM] Setup Passes for KASan and KMSan

glider added a comment. Rubberstamp LGTM Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D61664/new/ https://reviews.llvm.org/D61664 ___ cfe-commits mailing list cfe-commits@lists.llvm.org

[PATCH] D61879: WIP: Prototype of DSE optimizations for -ftrivial-auto-var-init

glider added inline comments. Comment at: llvm/lib/Transforms/Scalar/DeadStoreEliminationExpGlobal.cpp:20 + +static const GlobalValue::GUID GUIDS_LOOKUP[] = { +// grep -P -o "(?<=LOOKUP_FN: ).*" | sort -u > error: zero-size array ‘llvm::GUIDS_LOOKUP’ Can you

[PATCH] D61879: WIP: Prototype of DSE optimizations for -ftrivial-auto-var-init

glider added a comment. Vitaly, can you please rebase the patch? As far as I can see, you've submitted parts of it already. (not that I can't resolve the conflicts locally, but keeping it up-to-date may save others some time) Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION

[PATCH] D61879: WIP: Prototype of DSE optimizations for -ftrivial-auto-var-init

glider added inline comments. Comment at: llvm/lib/Transforms/Scalar/DeadStoreEliminationExp.cpp:709 + size_t argN) { + const Function *F = cast(Callee); + if (argN >= F->arg_size()) This cast fails in the following case: ``` %21 = call i64

[PATCH] D61879: WIP: Prototype of DSE optimizations for -ftrivial-auto-var-init

glider added inline comments. Comment at: llvm/lib/Transforms/Scalar/DeadStoreEliminationExp.cpp:556 + LocationSize LS = MemoryLocation::getForDest(MSI).Size; + ConstantRange Range(PointerSizeInBits, false); + if (!LS.isPrecise() || An assertion

[PATCH] D64742: Allow using -ftrivial-auto-var-init=zero in C mode without extra flags

glider created this revision. Herald added a project: clang. Herald added a subscriber: cfe-commits. Initially concerns have been raised that -ftrivial-auto-var-init=zero potentially defines a new dialect of C++, therefore this option was guarded with

[PATCH] D61879: WIP: Prototype of DSE optimizations for -ftrivial-auto-var-init

glider added inline comments. Comment at: llvm/lib/Transforms/Scalar/DeadStoreEliminationExpGlobal.cpp:27 +static const char GlobalArgumentMemAccess[] = { +// grep -P -o "(?<=FUNCTION_INFO: ).*" | sort -u > +//

[PATCH] D66948: [CodeGen]: fix error message for "=r" asm constraint

glider updated this revision to Diff 218037. glider added a comment. Minor comment fix. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D66948/new/ https://reviews.llvm.org/D66948 Files: clang/lib/CodeGen/CGStmt.cpp

[PATCH] D66948: [CodeGen]: fix error message for "=r" asm constraint

This revision was automatically updated to reflect the committed changes. Closed by commit rL370444: [CodeGen]: fix error message for =r asm constraint (authored by glider, committed by ). Herald added a project: LLVM. Herald added a subscriber: llvm-commits. Changed prior to commit:

[PATCH] D65234: [CodeGen]: don't treat structures returned in registers as memory inputs

This revision was automatically updated to reflect the committed changes. Closed by commit rL370335: [CodeGen]: dont treat structures returned in registers as memory inputs (authored by glider, committed by ). Herald added a project: LLVM. Herald added a subscriber: llvm-commits. Changed prior

[PATCH] D65234: [CodeGen]: don't treat structures returned in registers as memory inputs

glider added a comment. Landed r370335, thank you! Repository: rL LLVM CHANGES SINCE LAST ACTION https://reviews.llvm.org/D65234/new/ https://reviews.llvm.org/D65234 ___ cfe-commits mailing list cfe-commits@lists.llvm.org

[PATCH] D65234: [CodeGen]: don't treat structures returned in registers as memory inputs

glider updated this revision to Diff 217831. glider added a comment. Rebased the patch Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D65234/new/ https://reviews.llvm.org/D65234 Files: clang/lib/CodeGen/CGStmt.cpp clang/test/CodeGen/asm-attrs.c

[PATCH] D66948: [CodeGen]: fix error message for "=r" asm constraint

glider created this revision. glider added reviewers: eli.friedman, thakis. Herald added a project: clang. Herald added a subscriber: cfe-commits. Nico Weber reported that the following code: char buf[9]; asm("" : "=r" (buf)); yields the "impossible constraint in asm: can't store struct

[PATCH] D66948: [CodeGen]: fix error message for "=r" asm constraint

glider updated this revision to Diff 217887. glider added a comment. clang-format Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D66948/new/ https://reviews.llvm.org/D66948 Files: clang/lib/CodeGen/CGStmt.cpp clang/test/CodeGen/x86_64-PR42672.c

[PATCH] D61879: WIP: Prototype of DSE optimizations for -ftrivial-auto-var-init

glider added a comment. In D61879#1621423 , @vitalybuka wrote: > Fix https://godbolt.org/z/-PinQP Please consider adding a test. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D61879/new/

[PATCH] D65234: [CodeGen]: don't treat structures returned in registers as memory inputs

glider added a comment. +srhines Heads up: this patch reduces the size of the following Android kernel functions: _raw_spin_lock and _raw_spin_lock_bh: 132 bytes->68 bytes _raw_spin_lock_irq: 136 bytes->72 bytes _raw_spin_lock_irqsave: 144 bytes->80 bytes _raw_spin_trylock: 156

[PATCH] D65234: [CodeGen]: don't treat structures returned in registers as memory inputs

glider updated this revision to Diff 212293. glider added a comment. Test for =X somehow sneaked in - drop it. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D65234/new/ https://reviews.llvm.org/D65234 Files: clang/lib/CodeGen/CGStmt.cpp

[PATCH] D61879: WIP: Prototype of DSE optimizations for -ftrivial-auto-var-init

glider added a comment. Vitaly, what's the current status of these patches? What's your plan on submitting them? Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D61879/new/ https://reviews.llvm.org/D61879

[PATCH] D64742: Allow using -ftrivial-auto-var-init=zero in C mode without extra flags

glider added a comment. As a data point, Linus Torvalds suggested that we need a similar feature for GCC so that the "kernel C standard" mandates zero-initialization for locals: https://lkml.org/lkml/2019/7/28/206 Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION

[PATCH] D65234: [CodeGen]: don't treat structures returned in registers as memory inputs

glider updated this revision to Diff 213920. glider added a comment. Added test cases for =x, replaced grep with not Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D65234/new/ https://reviews.llvm.org/D65234 Files: clang/lib/CodeGen/CGStmt.cpp

[PATCH] D65234: [CodeGen]: don't treat structures returned in registers as memory inputs

glider marked an inline comment as done. glider added a comment. In D65234#1615649 , @efriedma wrote: > For the "=x" thing I was talking about, try the following testcase: > > void a() { struct S { unsigned x[4]; } z; asm volatile("%0":"=x"(z)); } >

[PATCH] D65234: [CodeGen]: don't treat structures returned in registers as memory inputs

glider marked 3 inline comments as done. glider added inline comments. Comment at: clang/test/CodeGen/PR42672.c:50 + struct { +long long int v1, v2, v3, v4; + } str; efriedma wrote: > glider wrote: > > The acceptable size actually depends on the target

[PATCH] D65234: [CodeGen]: don't treat structures returned in registers as memory inputs

glider updated this revision to Diff 212164. glider added a comment. Addressed Eli's comments, added a test for a packed struct Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D65234/new/ https://reviews.llvm.org/D65234 Files:

[PATCH] D65234: [CodeGen]: don't treat structures returned in registers as memory inputs

glider added a comment. Eli, any other comments? Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D65234/new/ https://reviews.llvm.org/D65234 ___ cfe-commits mailing list cfe-commits@lists.llvm.org

[PATCH] D65234: [CodeGen]: don't treat structures returned in registers as memory inputs

glider created this revision. Herald added a project: clang. Herald added a subscriber: cfe-commits. glider added reviewers: eli.friedman, jyknight. glider added a subscriber: vitalybuka. The "=r" output constraint for a structure variable passed to inline asm shouldn't be converted to "=*r", as

[PATCH] D65234: [CodeGen]: don't treat structures returned in registers as memory inputs

glider marked 2 inline comments as done. glider added inline comments. Comment at: clang/lib/CodeGen/CGStmt.cpp:1989 std::vector Args; + std::vector ResultTypeRequiresCast; nickdesaulniers wrote: > Are we able to use something other than `std::vector` here

[PATCH] D65234: [CodeGen]: don't treat structures returned in registers as memory inputs

glider updated this revision to Diff 211919. glider marked 2 inline comments as done. glider added a comment. Fixed comments from Eli and Nick, added tests for unusual struct sizes Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D65234/new/

[PATCH] D65234: [CodeGen]: don't treat structures returned in registers as memory inputs

glider updated this revision to Diff 211920. glider added a comment. Make big_struct() test triple-specific Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D65234/new/ https://reviews.llvm.org/D65234 Files: clang/lib/CodeGen/CGStmt.cpp

[PATCH] D64742: Allow using -ftrivial-auto-var-init=zero in C mode without extra flags

glider added subscribers: rsmith, dvyukov, dblaikie, arthur.j.odwyer, fwyzard, kristina, mehdi_amini, troyj. glider added a comment. Adding more people from the original discussion. Folks, we're now stuck in a situation where there's a potential buy-in from the Linux kernel community for stack

[PATCH] D64742: Allow using -ftrivial-auto-var-init=zero in C mode without extra flags

glider added a comment. In D64742#1717054 , @mehdi_amini wrote: > In D64742#1606244 , @glider wrote: > > > As a data point, Linus Torvalds suggested that we need a similar feature > > for GCC so that the "kernel C

[PATCH] D80668: [Clang][Sanitizers] Expect test failure on {arm,thumb}v7

glider added inline comments. Comment at: clang/test/CodeGen/sanitize-coverage.c:8 +// +// XFAIL: armv7, thumbv7 Is there a Bugzilla entry for this? Please add a link to the code and to the patch description. Repository: rG LLVM Github Monorepo CHANGES

[PATCH] D81390: [KernelAddressSanitizer] Make globals constructors compatible with kernel [v2]

glider accepted this revision. glider added a comment. This revision is now accepted and ready to land. LGTM assuming allyesconfig builds. Comment at: llvm/lib/Transforms/Instrumentation/AddressSanitizer.cpp:759 // do globals-gc. -UseCtorComdat(UseGlobalsGC &&

[PATCH] D80805: [KernelAddressSanitizer] Make globals constructors compatible with kernel

glider accepted this revision. glider added a comment. This revision is now accepted and ready to land. LGTM Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D80805/new/ https://reviews.llvm.org/D80805

[PATCH] D81390: [KernelAddressSanitizer] Make globals constructors compatible with kernel [v2]

glider added inline comments. Comment at: clang/test/CodeGen/asan-globals.cpp:16 +int aliased_global; +extern int __attribute__((alias("aliased_global"))) __global_alias; +int __special_global; Need a comment explaining this case a bit.

[PATCH] D108029: [clang][Codegen] Introduce the disable_sanitizer_instrumentation attribute

This revision was landed with ongoing or failed builds. This revision was automatically updated to reflect the committed changes. Closed by commit rGb0391dfc737e: [clang][Codegen] Introduce the disable_sanitizer_instrumentation attribute (authored by glider). Repository: rG LLVM Github

[PATCH] D108029: [clang][Codegen] Introduce the disable_sanitizer_instrumentation attribute

glider updated this revision to Diff 367759. glider added a comment. Updated LangRef.rst Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D108029/new/ https://reviews.llvm.org/D108029 Files: clang/include/clang/Basic/Attr.td

[PATCH] D108029: [clang][Codegen] Introduce the disable_sanitizer_instrumentation attribute

glider added a comment. In D108029#2954604 , @glider wrote: > In D108029#2954566 , @melver wrote: > >> llvm/docs/LangRef.rst also needs a corresponding change. > > Right, will do. I thought LangRef was missing

[PATCH] D108199: [msan] Add support for disable_sanitizer_instrumentation attribute

glider updated this revision to Diff 367767. glider added a comment. Updated after landing the disable_sanitizer_instrumentation to LLVM Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D108199/new/ https://reviews.llvm.org/D108199 Files:

[PATCH] D108199: [msan] Add support for disable_sanitizer_instrumentation attribute

glider updated this revision to Diff 367774. glider marked 2 inline comments as done. glider added a comment. Addressed Marco's comments Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D108199/new/ https://reviews.llvm.org/D108199 Files:

[PATCH] D108199: [msan] Add support for disable_sanitizer_instrumentation attribute

This revision was landed with ongoing or failed builds. This revision was automatically updated to reflect the committed changes. Closed by commit rG8dc7dcdca1e0: [msan] Add support for disable_sanitizer_instrumentation attribute (authored by glider). Repository: rG LLVM Github Monorepo

[PATCH] D108465: [msan] Hotfix clang/test/CodeGen/sanitize-memory-disable.c

This revision was landed with ongoing or failed builds. This revision was automatically updated to reflect the committed changes. Closed by commit rG417a49e78e73: [msan] Hotfix clang/test/CodeGen/sanitize-memory-disable.c (authored by glider). Repository: rG LLVM Github Monorepo CHANGES SINCE

[PATCH] D108199: [msan] Add support for disable_sanitizer_instrumentation attribute

glider added inline comments. Comment at: clang/docs/MemorySanitizer.rst:91 + +The ``disable_sanitizer_instrumentation`` attribute can be applied to a certain +function to prevent all kinds of instrumentation. This attribute overrides melver wrote: > Could apply

[PATCH] D108465: [msan] Hotfix clang/test/CodeGen/sanitize-memory-disable.c

glider created this revision. glider added reviewers: eugenis, melver. Herald added a subscriber: pengfei. glider requested review of this revision. Herald added a project: clang. Herald added a subscriber: cfe-commits. Because KMSAN is not supported on many architectures, explicitly build the

[PATCH] D108465: [msan] Hotfix clang/test/CodeGen/sanitize-memory-disable.c

glider added a comment. TBR Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D108465/new/ https://reviews.llvm.org/D108465 ___ cfe-commits mailing list cfe-commits@lists.llvm.org

[PATCH] D108465: [msan] Hotfix clang/test/CodeGen/sanitize-memory-disable.c

glider added a comment. In D108465#2957355 , @melver wrote: > Who reported the issue? Might be worth mentioning in commit message, > otherwise it appears to come out of nowhere (although it's semi-obvious given > x86-64 is only supported). Right.

[PATCH] D108465: [msan] Hotfix clang/test/CodeGen/sanitize-memory-disable.c

glider updated this revision to Diff 367783. glider added a comment. Updated patch description with links to bot failures. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D108465/new/ https://reviews.llvm.org/D108465 Files:

[PATCH] D108202: [tsan] Add support for disable_sanitizer_instrumentation attribute

glider updated this revision to Diff 368056. glider marked 5 inline comments as done. glider added a comment. Addressed Marco's comments Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D108202/new/ https://reviews.llvm.org/D108202 Files:

[PATCH] D108202: [tsan] Add support for disable_sanitizer_instrumentation attribute

This revision was landed with ongoing or failed builds. This revision was automatically updated to reflect the committed changes. Closed by commit rG8300d52e8cbf: [tsan] Add support for disable_sanitizer_instrumentation attribute (authored by glider). Repository: rG LLVM Github Monorepo

[PATCH] D108029: [clang][Codegen] Introduce the disable_sanitizer_instrumentation attribute

glider updated this revision to Diff 366845. glider added a comment. Address Aaron's comments Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D108029/new/ https://reviews.llvm.org/D108029 Files: clang/include/clang/Basic/Attr.td

[PATCH] D108029: [clang][Codegen] Introduce the disable_sanitizer_instrumentation attribute

glider added inline comments. Comment at: clang/include/clang/Basic/Attr.td:2945 + let Spellings = [Clang<"disable_sanitizer_instrumentation">]; + let Subjects = SubjectList<[Function]>; + let Documentation = [DisableSanitizerInstrumentationDocs];

[PATCH] D108029: [clang][Codegen] Introduce the disable_sanitizer_instrumentation attribute

glider updated this revision to Diff 366887. glider added a comment. Fix the docs Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D108029/new/ https://reviews.llvm.org/D108029 Files: clang/include/clang/Basic/Attr.td

[PATCH] D108199: [msan] Add support for disable_sanitizer_instrumentation attribute

glider created this revision. glider added reviewers: eugenis, melver. Herald added a subscriber: hiraditya. glider requested review of this revision. Herald added projects: clang, LLVM. Herald added subscribers: llvm-commits, cfe-commits. Unlike __attribute__((no_sanitize("memory"))), this one

[PATCH] D108029: [clang][Codegen] Introduce the disable_sanitizer_instrumentation attribute

glider added inline comments. Comment at: llvm/include/llvm/IR/Attributes.td:90 +/// Do not instrument function with sanitizers. +def DisableSanitizerInstrumentation: EnumAttr<"disable_sanitizer_instrumentation", [FnAttr]>; + melver wrote: > There's this

[PATCH] D108202: [tsan] Add support for disable_sanitizer_instrumentation attribute

glider created this revision. glider added a reviewer: melver. Herald added subscribers: jfb, hiraditya. glider requested review of this revision. Herald added projects: clang, LLVM. Herald added subscribers: llvm-commits, cfe-commits. Unlike __attribute__((no_sanitize("thread"))), this one will

[PATCH] D108029: [clang][Codegen] Introduce the no_sanitizer_instrumentation attribute

glider created this revision. glider added reviewers: eugenis, melver, browneee, dvyukov. Herald added subscribers: ormris, dexonsmith, jdoerfert, steven_wu, hiraditya. Herald added a reviewer: aaron.ballman. glider requested review of this revision. Herald added projects: clang, LLVM. Herald

[PATCH] D108029: [clang][Codegen] Introduce the no_sanitizer_instrumentation attribute

glider added inline comments. Comment at: clang/include/clang/Basic/AttrDocs.td:2601-2602 + +This is not the same as ``__attribute__((no_sanitize(...)))``, which depending +on the tool may still insert instrumentation to prevent false positive reports. + }];

[PATCH] D108029: [clang][Codegen] Introduce the no_sanitizer_instrumentation attribute

glider updated this revision to Diff 366256. glider added a comment. Added a comment to clang/lib/CodeGen/CodeGenFunction.h Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D108029/new/ https://reviews.llvm.org/D108029 Files:

[PATCH] D108555: [tsan] Make sanitize-thread-disable.c an X86-only test

glider added a comment. Not really sure what's the best solution here, but I think restricting the test to x86 should help. So far only ARM and PPC bots reported failures. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D108555/new/

[PATCH] D108555: [tsan] Make sanitize-thread-disable.c an X86-only test

glider added inline comments. Comment at: clang/test/CodeGen/X86/sanitize-thread-disable.c:22 int instrumented1(int *a, _Atomic int *b) { return *a + atomic_load(b); } melver wrote: > melver wrote: > > I think you do not need to use atomic_load. > > > >

[PATCH] D108555: [tsan] Make sanitize-thread-disable.c an X86-only test

glider updated this revision to Diff 368096. glider added a comment. Removed the header Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D108555/new/ https://reviews.llvm.org/D108555 Files: clang/test/CodeGen/sanitize-thread-disable.c Index:

[PATCH] D108555: [tsan] Make sanitize-thread-disable.c an X86-only test

glider added a comment. In D108555#2960034 , @melver wrote: > LGTM, thanks! > > Patch title ("...an X86-only test..") also needs adjustment. It's strange that Phab doesn't automatically update the title when I update the commit message. Repository:

[PATCH] D108555: [tsan] Make sanitize-thread-disable.c an X86-only test

glider created this revision. glider added a reviewer: melver. Herald added subscribers: pengfei, jfb, kristof.beyls. glider requested review of this revision. Herald added a project: clang. Herald added a subscriber: cfe-commits. Looks like non-x86 bots are unhappy with inclusion of e.g.:

[PATCH] D108555: [tsan] Do not include from sanitize-thread-disable.c

This revision was landed with ongoing or failed builds. This revision was automatically updated to reflect the committed changes. Closed by commit rGcdb391698bb2: [tsan] Do not include stdatomic.h from sanitize-thread-disable.c (authored by glider). Repository: rG LLVM Github Monorepo CHANGES

[PATCH] D108029: [clang][Codegen] Introduce the no_sanitizer_instrumentation attribute

glider marked 2 inline comments as done. glider added inline comments. Comment at: clang/include/clang/Basic/AttrDocs.td:2601-2602 + +This is not the same as ``__attribute__((no_sanitize(...)))``, which depending +on the tool may still insert instrumentation to prevent false

[PATCH] D108029: [clang][Codegen] Introduce the no_sanitizer_instrumentation attribute

glider updated this revision to Diff 366591. glider added a comment. Updated patch description Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D108029/new/ https://reviews.llvm.org/D108029 Files: clang/include/clang/Basic/Attr.td

[PATCH] D108029: [clang][Codegen] Introduce the no_sanitizer_instrumentation attribute

glider added a subscriber: krytarowski. glider marked 2 inline comments as done. glider added a comment. +krytarowski FYI Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D108029/new/ https://reviews.llvm.org/D108029

[PATCH] D108029: [clang][Codegen] Introduce the no_sanitizer_instrumentation attribute

glider updated this revision to Diff 366570. glider added a comment. Renamed no_sanitizer_instrumentation to disable_sanitizer_instrumentation Added bitcode attribute declaration and missing handling of the new attribute in switches. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST

[PATCH] D108029: [clang][Codegen] Introduce the disable_sanitizer_instrumentation attribute

glider added inline comments. Comment at: llvm/include/llvm/Bitcode/LLVMBitCodes.h:674 ATTR_KIND_ELEMENTTYPE = 77, + ATTR_DISABLE_SANITIZER_INSTRUMENTATION = 78, }; Missing "_KIND_" Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION

[PATCH] D108029: [clang][Codegen] Introduce the disable_sanitizer_instrumentation attribute

glider updated this revision to Diff 367480. glider added a comment. Updated BitcodeReader.cpp and several tests per Marco's suggestion Renamed ATTR_DISABLE_SANITIZER_INSTRUMENTATION to ATTR_KIND_DISABLE_SANITIZER_INSTRUMENTATION Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION

[PATCH] D108029: [clang][Codegen] Introduce the disable_sanitizer_instrumentation attribute

glider added a comment. In D108029#2954566 , @melver wrote: > llvm/docs/LangRef.rst also needs a corresponding change. Right, will do. I thought LangRef was missing the sanitizer bits as well, and was planning to add them together, but apparently I was

[PATCH] D114421: [asan] Add support for disable_sanitizer_instrumentation attribute

glider created this revision. glider added reviewers: melver, vitalybuka. glider requested review of this revision. Herald added a project: clang. Herald added a subscriber: cfe-commits. For ASan this will effectively serve as a synonym for __attribute__((no_sanitize("address"))) Repository:

[PATCH] D114421: [asan] Add support for disable_sanitizer_instrumentation attribute

glider added a comment. Vitaly, Evgenii, can one of you please take a look? Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D114421/new/ https://reviews.llvm.org/D114421 ___ cfe-commits mailing list

[PATCH] D114421: [asan] Add support for disable_sanitizer_instrumentation attribute

glider updated this revision to Diff 393168. glider added a comment. Rebase. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D114421/new/ https://reviews.llvm.org/D114421 Files: clang/docs/AddressSanitizer.rst

  1   2   >