[PATCH] D38901: [CUDA] Require libdevice only if needed

Hahnfeld updated this revision to Diff 118969. Hahnfeld added a comment. Fix one more condition that checks for `nvvm/libdevice` and add a test. https://reviews.llvm.org/D38901 Files: lib/Driver/ToolChains/Cuda.cpp test/Driver/Inputs/CUDA-nolibdevice/usr/local/cuda/bin/.keep test/Driver/I

Re: r313096 - [ubsan] Function Sanitizer: Don't require writable text segments

Hi Vedant, So this actually broke -fsanitize=function on linux. Han is working up a testcase for it, but letting you know for now that we'll probably need some change here. -eric On Tue, Sep 12, 2017 at 5:05 PM Vedant Kumar via cfe-commits < cfe-commits@lists.llvm.org> wrote: > Author: vedantk

[PATCH] D38901: [CUDA] Require libdevice only if needed

tra accepted this revision. tra added a comment. This revision is now accepted and ready to land. Looks good. Thank you. https://reviews.llvm.org/D38901 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/li

Re: r313096 - [ubsan] Function Sanitizer: Don't require writable text segments

Hey Eric, I'm sorry for the breakage. I made sure to check the run-time tests in compiler-rt but we could have missing coverage there. The original version of this patch restricted the prologue data changes to Darwin only. We can switch back to that easily, just let me know. vedant > On Oct

[PATCH] D38816: Convert clang::LangAS to a strongly typed enum

yaxunl added inline comments. Comment at: lib/AST/TypePrinter.cpp:1323 OS << "address_space("; -OS << T->getEquivalentType().getAddressSpace(); +OS << T->getEquivalentType() + .getQualifiers() arichardson wrote: > arichardson wrote: > >

Re: r313096 - [ubsan] Function Sanitizer: Don't require writable text segments

> On Oct 13, 2017, at 1:39 PM, Vedant Kumar wrote: > > Hey Eric, > > I'm sorry for the breakage. I made sure to check the run-time tests in > compiler-rt but we could have missing coverage there. > > The original version of this patch restricted the prologue data changes to > Darwin only. We

Re: r313096 - [ubsan] Function Sanitizer: Don't require writable text segments

On Fri, Oct 13, 2017 at 1:42 PM Vedant Kumar wrote: > On Oct 13, 2017, at 1:39 PM, Vedant Kumar wrote: > > Hey Eric, > > I'm sorry for the breakage. I made sure to check the run-time tests in > compiler-rt but we could have missing coverage there. > > The original version of this patch restricte

[PATCH] D23963: [analyzer] pr28449 - Move literal rvalue construction away from RegionStore.

This revision was automatically updated to reflect the committed changes. Closed by commit rL315750: [analyzer] pr28449: Fix support for various array initializers. (authored by dergachev). Changed prior to commit: https://reviews.llvm.org/D23963?vs=118625&id=118974#toc Repository: rL LLVM

r315750 - [analyzer] pr28449: Fix support for various array initializers.

Author: dergachev Date: Fri Oct 13 13:54:56 2017 New Revision: 315750 URL: http://llvm.org/viewvc/llvm-project?rev=315750&view=rev Log: [analyzer] pr28449: Fix support for various array initializers. In some cases the analyzer didn't expect an array-type variable to be initialized with anything o

[PATCH] D38824: [X86] Synchronize the CPU predefined macros with gcc

craig.topper updated this revision to Diff 118973. craig.topper added a comment. Only define "corei7" on nehalem/westmere to match gcc. Don't define anything for the CPUs newer than that. Add comments to the CPUs where gcc has two sets of defines and we have only one. https://reviews.llvm.org/

[PATCH] D38900: libunwind: document tested FreeBSD configs and sort OS list

emaste added a subscriber: bsdjhb. emaste added a comment. In https://reviews.llvm.org/D38900#897304, @krytarowski wrote: > @joerg might have insight on ppc, sparc64, arm on NetBSD. And @bsdjhb for other architectures on FreeBSD. But I wanted to at least list x86_64 and ARM64 for FreeBSD, wher

r315755 - Fix -Woverloaded-virtual warning in clang-refactor

Author: arphaman Date: Fri Oct 13 14:15:25 2017 New Revision: 315755 URL: http://llvm.org/viewvc/llvm-project?rev=315755&view=rev Log: Fix -Woverloaded-virtual warning in clang-refactor Modified: cfe/trunk/tools/clang-refactor/ClangRefactor.cpp Modified: cfe/trunk/tools/clang-refactor/ClangR

[PATCH] D38900: libunwind: document tested FreeBSD configs and sort OS list

bsdjhb added a comment. FreeBSD/i386 also uses llvm libunwind by default. https://reviews.llvm.org/D38900 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D38903: [ubsan] Only use indirect RTTI in prologues on Darwin

vsk created this revision. Using a layer of indirection to point to RTTI through function prologues is not supported on some setups. One reported error message is: error: Cannot represent a difference across sections This is a regression. This patch limits the indirect RTTI behavior to Darwin,

Re: r313096 - [ubsan] Function Sanitizer: Don't require writable text segments

> On Oct 13, 2017, at 1:44 PM, Eric Christopher wrote: > > > > On Fri, Oct 13, 2017 at 1:42 PM Vedant Kumar > wrote: >> On Oct 13, 2017, at 1:39 PM, Vedant Kumar > > wrote: >> >> Hey Eric, >> >> I'm sorry for the breakage. I made sure to check t

Re: r313096 - [ubsan] Function Sanitizer: Don't require writable text segments

On Fri, Oct 13, 2017 at 2:50 PM Vedant Kumar wrote: > On Oct 13, 2017, at 1:44 PM, Eric Christopher wrote: > > > > On Fri, Oct 13, 2017 at 1:42 PM Vedant Kumar wrote: > >> On Oct 13, 2017, at 1:39 PM, Vedant Kumar wrote: >> >> Hey Eric, >> >> I'm sorry for the breakage. I made sure to check th

[PATCH] D38903: [ubsan] Only use indirect RTTI in prologues on Darwin

echristo added a comment. Given you were the last one in this code it seems reasonable to let you go for it :) That said, I didn't notice anything in particular that stuck out at me. https://reviews.llvm.org/D38903 ___ cfe-commits mailing list cfe

[PATCH] D38904: Allow building libFuzzer in two-stage compiler-rt build

george.karpenkov created this revision. Herald added subscribers: mgorny, dberris. When `LLVM_BUILD_EXTERNAL_COMPILER_RT` option is set to true, all of projects in compiler-rt are built with a freshly-built compiler using a recursive CMake invocation. (e.g. that's how compiler-rt is used in Swift)

[PATCH] D38903: [ubsan] Only use indirect RTTI in prologues on Darwin

vsk added a comment. Sounds good. This doesn't seem too controversial, since it just takes us back to the old behavior on all platforms except Darwin. I'll wait an hour or so before committing in case there are any more comments. https://reviews.llvm.org/D38903 _

r315771 - Allow building libFuzzer in two-stage compiler-rt build

Author: george.karpenkov Date: Fri Oct 13 15:03:09 2017 New Revision: 315771 URL: http://llvm.org/viewvc/llvm-project?rev=315771&view=rev Log: Allow building libFuzzer in two-stage compiler-rt build When LLVM_BUILD_EXTERNAL_COMPILER_RT option is set to true, all of projects in compiler-rt are bui

[PATCH] D38904: Allow building libFuzzer in two-stage compiler-rt build

This revision was automatically updated to reflect the committed changes. Closed by commit rL315771: Allow building libFuzzer in two-stage compiler-rt build (authored by george.karpenkov). Changed prior to commit: https://reviews.llvm.org/D38904?vs=118977&id=118979#toc Repository: rL LLVM h

Re: r313096 - [ubsan] Function Sanitizer: Don't require writable text segments

> On Oct 13, 2017, at 2:52 PM, Eric Christopher wrote: > > > > On Fri, Oct 13, 2017 at 2:50 PM Vedant Kumar > wrote: >> On Oct 13, 2017, at 1:44 PM, Eric Christopher > > wrote: >> >> >> >> On Fri, Oct 13, 2017 at 1:42 PM Vedant Kumar >

r315772 - [Sema] Avoid iterator invalidation when code completing.

Author: d0k Date: Fri Oct 13 15:14:34 2017 New Revision: 315772 URL: http://llvm.org/viewvc/llvm-project?rev=315772&view=rev Log: [Sema] Avoid iterator invalidation when code completing. It's possible for the code completion consumer to add new decls to the current scope while lookup happens on i

[PATCH] D38816: Convert clang::LangAS to a strongly typed enum

arichardson added inline comments. Comment at: lib/AST/TypePrinter.cpp:1323 OS << "address_space("; -OS << T->getEquivalentType().getAddressSpace(); +OS << T->getEquivalentType() + .getQualifiers() yaxunl wrote: > arichardson wrote: > >

r315774 - Revert r315738

Author: arphaman Date: Fri Oct 13 15:47:44 2017 New Revision: 315774 URL: http://llvm.org/viewvc/llvm-project?rev=315774&view=rev Log: Revert r315738 The ParsedSourceRange class does not work correctly on Windows with the ':' drive separators Removed: cfe/trunk/test/Refactor/tool-apply-repla

Re: r313096 - [ubsan] Function Sanitizer: Don't require writable text segments

Here's a small reproducer. struct A { virtual void f(...); }; struct B : virtual A { virtual void b(); virtual void f(...); }; void B::f(...) {} $ clang++ -fsanitize=function fsan.cpp -ffunction-sections -fdata-sections -c -o /dev/null fatal error: error in backend: Cannot represent a dif

Re: r313096 - [ubsan] Function Sanitizer: Don't require writable text segments

On Fri, Oct 13, 2017 at 4:06 PM, Peter Collingbourne wrote: > Here's a small reproducer. > > struct A { > virtual void f(...); > }; > > struct B : virtual A { > virtual void b(); > virtual void f(...); > }; > > void B::f(...) {} > > $ clang++ -fsanitize=function fsan.cpp -ffunction-sections

Re: r313096 - [ubsan] Function Sanitizer: Don't require writable text segments

> On Oct 13, 2017, at 4:08 PM, Peter Collingbourne wrote: > > On Fri, Oct 13, 2017 at 4:06 PM, Peter Collingbourne > wrote: > Here's a small reproducer. > > struct A { > virtual void f(...); > }; > > struct B : virtual A { > virtual void b(); > virtual void f(...

[PATCH] D38908: Do not link clang_rt.cfi on Android.

eugenis created this revision. Herald added a subscriber: srhines. The OS provides cross-dso CFI support starting with Android O. Trapping mode does not require any runtime at all, and diagnostic mode requires just ubsan-standalone. https://reviews.llvm.org/D38908 Files: clang/include/clang/D

Re: r313096 - [ubsan] Function Sanitizer: Don't require writable text segments

On Fri, Oct 13, 2017 at 4:19 PM, Vedant Kumar wrote: > > On Oct 13, 2017, at 4:08 PM, Peter Collingbourne wrote: > > On Fri, Oct 13, 2017 at 4:06 PM, Peter Collingbourne > wrote: > >> Here's a small reproducer. >> >> struct A { >> virtual void f(...); >> }; >> >> struct B : virtual A { >> v

[PATCH] D38773: [Sema] Add support for flexible array members in Obj-C.

vsapsai updated this revision to Diff 118990. vsapsai added a comment. - Address rjmccall review comment, move warn_variable_sized_ivar_visibility to DiagnoseVariableSizedIvars. https://reviews.llvm.org/D38773 Files: clang/include/clang/Basic/DiagnosticGroups.td clang/include/clang/Basic/D

r315776 - Allow building libFuzzer tests in two-stage compiler-rt build.

Author: george.karpenkov Date: Fri Oct 13 16:50:53 2017 New Revision: 315776 URL: http://llvm.org/viewvc/llvm-project?rev=315776&view=rev Log: Allow building libFuzzer tests in two-stage compiler-rt build. Modified: cfe/trunk/runtime/CMakeLists.txt Modified: cfe/trunk/runtime/CMakeLists.txt

[PATCH] D38757: [libc++] Fix PR34898 - vector iterator constructors and assign method perform push_back instead of emplace_back.

dlj requested changes to this revision. dlj added a comment. This revision now requires changes to proceed. Hmm, looking more at this change... while it does make the behaviour consistent for Forward and Input iterators, I think it's just making them both do the wrong thing. Specifically, based

[PATCH] D38678: [Sema] Warn about unused variables if we can constant evaluate the initializer.

rsmith accepted this revision. rsmith added a comment. This revision is now accepted and ready to land. Cunning, I like it :) https://reviews.llvm.org/D38678 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailm

[PATCH] D38908: Do not link clang_rt.cfi on Android.

pcc added inline comments. Comment at: clang/test/Driver/sanitizer-ld.c:605 -// CHECK-CFI-ANDROID: "{{(.*[^-.0-9A-Z_a-z])?}}ld{{(.exe)?}}" -// CHECK-CFI-ANDROID-NOT: libclang_rt.cfi -// CHECK-CFI-ANDROID-NOT: __cfi_check Why was this passing before? https://rev

[PATCH] D38913: [ubsan] Don't emit function signatures for virtual methods

vsk created this revision. The function sanitizer only checks indirect calls through function pointers. This excludes all non-static member functions (constructor calls, calls through thunks, etc all use a separate code path). Don't emit function signatures for functions that won't be checked. Ap

[PATCH] D38908: Do not link clang_rt.cfi on Android.

eugenis added inline comments. Comment at: clang/test/Driver/sanitizer-ld.c:605 -// CHECK-CFI-ANDROID: "{{(.*[^-.0-9A-Z_a-z])?}}ld{{(.exe)?}}" -// CHECK-CFI-ANDROID-NOT: libclang_rt.cfi -// CHECK-CFI-ANDROID-NOT: __cfi_check pcc wrote: > Why was this passing befo

r315784 - Fix backwards warning for use of C++17 attributes-on-namespaces-and-enumerators feature.

Author: rsmith Date: Fri Oct 13 17:56:24 2017 New Revision: 315784 URL: http://llvm.org/viewvc/llvm-project?rev=315784&view=rev Log: Fix backwards warning for use of C++17 attributes-on-namespaces-and-enumerators feature. Modified: cfe/trunk/include/clang/Basic/DiagnosticParseKinds.td cf

[PATCH] D38908: Do not link clang_rt.cfi on Android.

pcc accepted this revision. pcc added a comment. This revision is now accepted and ready to land. LGTM https://reviews.llvm.org/D38908 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

Re: r313096 - [ubsan] Function Sanitizer: Don't require writable text segments

> On Oct 13, 2017, at 4:33 PM, Peter Collingbourne wrote: > > > > On Fri, Oct 13, 2017 at 4:19 PM, Vedant Kumar > wrote: > >> On Oct 13, 2017, at 4:08 PM, Peter Collingbourne > > wrote: >> >> On Fri, Oct 13, 2017 at 4:06 PM, Peter Collingbourne

[PATCH] D38903: [ubsan] Only use indirect RTTI in prologues on Darwin

vsk added a comment. @pcc made an alternate suggestion which led to https://reviews.llvm.org/D38913. We're still discussing whether the new patch is a sufficient fix. https://reviews.llvm.org/D38903 ___ cfe-commits mailing list cfe-commits@lists.ll

Re: r313096 - [ubsan] Function Sanitizer: Don't require writable text segments

On Fri, Oct 13, 2017 at 5:59 PM, Vedant Kumar wrote: > > On Oct 13, 2017, at 4:33 PM, Peter Collingbourne wrote: > > > > On Fri, Oct 13, 2017 at 4:19 PM, Vedant Kumar wrote: > >> >> On Oct 13, 2017, at 4:08 PM, Peter Collingbourne wrote: >> >> On Fri, Oct 13, 2017 at 4:06 PM, Peter Collingbour

[PATCH] D38913: [ubsan] Don't emit function signatures for virtual methods

pcc accepted this revision. pcc added a comment. This revision is now accepted and ready to land. LGTM Small nit on the first line of the commit message: it should probably mention non-static member functions, rather than virtual methods. https://reviews.llvm.org/D38913

Re: r313096 - [ubsan] Function Sanitizer: Don't require writable text segments

> On Oct 13, 2017, at 6:05 PM, Peter Collingbourne wrote: > > > > On Fri, Oct 13, 2017 at 5:59 PM, Vedant Kumar > wrote: > >> On Oct 13, 2017, at 4:33 PM, Peter Collingbourne > > wrote: >> >> >> >> On Fri, Oct 13, 2017 at 4:19 PM, Vedant Kuma

[PATCH] D37436: Initial implementation of C attributes (WG14 N2137)

rsmith accepted this revision. rsmith added a comment. This revision is now accepted and ready to land. Thanks, LGTM :) Comment at: ../llvm/tools/clang/include/clang/Driver/Options.td:609-616 +def fdouble_square_bracket_attributes +: Flag<[ "-" ], "fdouble-square-bracket-at

r315785 - [Lex] Avoid out-of-bounds dereference in SkipLineComment

Author: arphaman Date: Fri Oct 13 18:18:30 2017 New Revision: 315785 URL: http://llvm.org/viewvc/llvm-project?rev=315785&view=rev Log: [Lex] Avoid out-of-bounds dereference in SkipLineComment Credit to OSS-Fuzz for discovery: https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=3145 rdar://3452

[PATCH] D38913: [ubsan] Don't emit function signatures for virtual methods

This revision was automatically updated to reflect the committed changes. Closed by commit rL315786: [ubsan] Don't emit function signatures for non-static member functions (authored by vedantk). Changed prior to commit: https://reviews.llvm.org/D38913?vs=118999&id=119004#toc Repository: rL L

r315786 - [ubsan] Don't emit function signatures for non-static member functions

Author: vedantk Date: Fri Oct 13 18:23:30 2017 New Revision: 315786 URL: http://llvm.org/viewvc/llvm-project?rev=315786&view=rev Log: [ubsan] Don't emit function signatures for non-static member functions The function sanitizer only checks indirect calls through function pointers. This excludes a

r315787 - [Sema] Warn about unused variables if we can constant evaluate the initializer.

Author: d0k Date: Fri Oct 13 18:30:49 2017 New Revision: 315787 URL: http://llvm.org/viewvc/llvm-project?rev=315787&view=rev Log: [Sema] Warn about unused variables if we can constant evaluate the initializer. If the variable construction can be constant evaluated it doesn't have side effects, so

[PATCH] D38678: [Sema] Warn about unused variables if we can constant evaluate the initializer.

This revision was automatically updated to reflect the committed changes. Closed by commit rL315787: [Sema] Warn about unused variables if we can constant evaluate the initializer. (authored by d0k). Repository: rL LLVM https://reviews.llvm.org/D38678 Files: cfe/trunk/lib/Sema/SemaDecl.cpp

Re: r313096 - [ubsan] Function Sanitizer: Don't require writable text segments

On Fri, Oct 13, 2017 at 6:12 PM, Vedant Kumar wrote: > > On Oct 13, 2017, at 6:05 PM, Peter Collingbourne wrote: > > > > On Fri, Oct 13, 2017 at 5:59 PM, Vedant Kumar wrote: > >> >> On Oct 13, 2017, at 4:33 PM, Peter Collingbourne wrote: >> >> >> >> On Fri, Oct 13, 2017 at 4:19 PM, Vedant Kuma

[PATCH] D38816: Convert clang::LangAS to a strongly typed enum

yaxunl added inline comments. Comment at: lib/AST/TypePrinter.cpp:1323 OS << "address_space("; -OS << T->getEquivalentType().getAddressSpace(); +OS << T->getEquivalentType() + .getQualifiers() arichardson wrote: > yaxunl wrote: > > arich

Re: r313096 - [ubsan] Function Sanitizer: Don't require writable text segments

> On Oct 13, 2017, at 6:33 PM, Peter Collingbourne wrote: > > > > On Fri, Oct 13, 2017 at 6:12 PM, Vedant Kumar > wrote: > >> On Oct 13, 2017, at 6:05 PM, Peter Collingbourne > > wrote: >> >> >> >> On Fri, Oct 13, 2017 at 5:59 PM, Vedant Kuma

[PATCH] D38903: [ubsan] Only use indirect RTTI in prologues on Darwin

vsk abandoned this revision. vsk added a comment. https://reviews.llvm.org/D38913 should make this unnecessary. https://reviews.llvm.org/D38903 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cf

[PATCH] D37437: [analyzer] Fix some checker's output plist not containing the checker name

zaks.anna added a comment. Looks like the need to have the checker name in BugType along with the checker names not being initialized early enough, leads to worse checker-writer experience. Is there a way to ensure that the checker names are set at construction? Comment at:

<    1   2