Changes in asan-globals.cpp

70455193 - Add missing sanitizer metadata plumbing from CFE.

Wed, 13 Jul 2022 15:39:31 +0000

Add missing sanitizer metadata plumbing from CFE.clang misses attaching sanitizer metadata for external globals.Reviewed By: eugenisDifferential Revision: https://reviews.llvm.org/D129492 List of files: /llvm-project-15.0.7/clang/test/CodeGen/asan-globals.cpp

a2095d1a - Allow mangled names in sanitizer clang IR gen tests.

Tue, 28 Jun 2022 16:26:36 +0000

Allow mangled names in sanitizer clang IR gen tests.Looks like with https://reviews.llvm.org/D127911, Windows emits moreglobals with mangled names into the IR. Relax the tests in order toallow these mangled names. List of files: /llvm-project-15.0.7/clang/test/CodeGen/asan-globals.cpp

dacfa24f - Delete 'llvm.asan.globals' for global metadata.

Mon, 27 Jun 2022 21:40:03 +0000

Delete 'llvm.asan.globals' for global metadata.Now that we have the sanitizer metadata that is actually on the globalvariable, and now that we use debuginfo in order to do symbolization ofglobals, we can delete the 'llvm.asan.globals' IR synthesis.This patch deletes the 'location' part of the __asan_global that'sembedded in the binary as well, because it's unnecessary. This savesabout ~1.7% of the optimised non-debug with-asserts clang binary.Reviewed By: vitalybukaDifferential Revision: https://reviews.llvm.org/D127911 List of files: /llvm-project-15.0.7/clang/test/CodeGen/asan-globals.cpp

77475ffd - Reland "Add sanitizer metadata attributes to clang IR gen."

Mon, 13 Jun 2022 19:12:31 +0000

Reland "Add sanitizer metadata attributes to clang IR gen."RE-LAND (reverts a revert):This reverts commit 8e1f47b596b28fbc88cf32e8f46eb2fecb145fb2.This patch adds generation of sanitizer metadata attributes (which wereadded in D126100) to the clang frontend.We still currently generate the llvm.asan.globals that's consumed bythe IR pass, but the plan is to eventually migrate off of that ontopurely debuginfo and these IR attributes.Reviewed By: vitalybuka, kstoimenovDifferential Revision: https://reviews.llvm.org/D126929 List of files: /llvm-project-15.0.7/clang/test/CodeGen/asan-globals.cpp

8e1f47b5 - Revert "Add sanitizer metadata attributes to clang IR gen."

Mon, 13 Jun 2022 19:10:40 +0000

Revert "Add sanitizer metadata attributes to clang IR gen."This reverts commit e7766972a6790e25dbb4ce3481f57e9792b49269.Broke the Windows buildbots. List of files: /llvm-project-15.0.7/clang/test/CodeGen/asan-globals.cpp

e7766972 - Add sanitizer metadata attributes to clang IR gen.

Mon, 13 Jun 2022 18:18:55 +0000

Add sanitizer metadata attributes to clang IR gen.This patch adds generation of sanitizer metadata attributes (which wereadded in D126100) to the clang frontend.We still currently generate the `llvm.asan.globals` that's consumed bythe IR pass, but the plan is to eventually migrate off of that ontopurely debuginfo and these IR attributes.Reviewed By: vitalybuka, kstoimenovDifferential Revision: https://reviews.llvm.org/D126929 List of files: /llvm-project-15.0.7/clang/test/CodeGen/asan-globals.cpp

c85a2645 - [asan] Add support for disable_sanitizer_instrumentation attribute

Thu, 17 Feb 2022 14:48:35 +0000

[asan] Add support for disable_sanitizer_instrumentation attributeFor ASan this will effectively serve as a synonym for__attribute__((no_sanitize("address"))).Adding the disable_sanitizer_instrumentation to functions will drop thesanitize_XXX attributes on the IR level.This is the third reland of https://reviews.llvm.org/D114421.Now that TSan test is fixed (https://reviews.llvm.org/D120050) thereshould be no deadlocks.Differential Revision: https://reviews.llvm.org/D120055 List of files: /llvm-project-15.0.7/clang/test/CodeGen/asan-globals.cpp

05ee1f4a - Revert "[asan] Add support for disable_sanitizer_instrumentation attribute"

Tue, 15 Feb 2022 13:57:28 +0000

Revert "[asan] Add support for disable_sanitizer_instrumentation attribute"This reverts commit dd145f953db3dafbc019f1d3783bb4f09a28af92.https://reviews.llvm.org/D119726, like https://reviews.llvm.org/D114421,still causes TSan to fail, see https://lab.llvm.org/buildbot/#/builders/70/builds/18020Differential Revision: https://reviews.llvm.org/D119838 List of files: /llvm-project-15.0.7/clang/test/CodeGen/asan-globals.cpp

dd145f95 - [asan] Add support for disable_sanitizer_instrumentation attribute

Tue, 23 Nov 2021 09:32:35 +0000

[asan] Add support for disable_sanitizer_instrumentation attributeFor ASan this will effectively serve as a synonym for__attribute__((no_sanitize("address")))This is a reland of https://reviews.llvm.org/D114421Reviewed By: melver, eugenisDifferential Revision: https://reviews.llvm.org/D119726 List of files: /llvm-project-15.0.7/clang/test/CodeGen/asan-globals.cpp

6398903a - Extend the `uwtable` attribute with unwind table kind

Mon, 14 Feb 2022 13:41:34 +0000

Extend the `uwtable` attribute with unwind table kindWe have the `clang -cc1` command-line option `-funwind-tables=1|2` andthe codegen option `VALUE_CODEGENOPT(UnwindTables, 2, 0) ///< Unwindtables (1) or asynchronous unwind tables (2)`. However, this isencoded in LLVM IR by the presence or the absence of the `uwtable`attribute, i.e. we lose the information whether to generate want justsome unwind tables or asynchronous unwind tables.Asynchronous unwind tables take more space in the runtime image, I'destimate something like 80-90% more, as the difference is addingroughly the same number of CFI directives as for prologues, only a bitsimpler (e.g. `.cfi_offset reg, off` vs. `.cfi_restore reg`). Or evenmore, if you consider tail duplication of epilogue blocks.Asynchronous unwind tables could also restrict code generation tohaving only a finite number of frame pointer adjustments (an exampleof *not* having a finite number of `SP` adjustments is on AArch64 whenuntagging the stack (MTE) in some cases the compiler can modify `SP`in a loop).Having the CFI precise up to an instruction generally also means onecannot bundle together CFI instructions once the prologue is done,they need to be interspersed with ordinary instructions, which meansextra `DW_CFA_advance_loc` commands, further increasing the unwindtables size.That is to say, async unwind tables impose a non-negligible overhead,yet for the most common use cases (like C++ exceptions), they are noteven needed.This patch extends the `uwtable` attribute with an optionalvalue: - `uwtable` (default to `async`) - `uwtable(sync)`, synchronous unwind tables - `uwtable(async)`, asynchronous (instruction precise) unwind tablesReviewed By: MaskRayDifferential Revision: https://reviews.llvm.org/D114543 List of files: /llvm-project-15.0.7/clang/test/CodeGen/asan-globals.cpp

7c004c2b - Revert "[asan] Add support for disable_sanitizer_instrumentation attribute"

Fri, 10 Dec 2021 22:28:36 +0000

Revert "[asan] Add support for disable_sanitizer_instrumentation attribute"This reverts commit 2b554920f11c8b763cd9ed9003f4e19b919b8e1f.This change causes tsan test timeout on x86_64-linux-autoconf.The timeout can be reproduced by: git clone https://github.com/llvm/llvm-zorg.git BUILDBOT_CLOBBER= BUILDBOT_REVISION=eef8f3f85679c5b1ae725bade1c23ab7bb6b924f llvm-zorg/zorg/buildbot/builders/sanitizers/buildbot_standard.sh List of files: /llvm-project-15.0.7/clang/test/CodeGen/asan-globals.cpp

2b554920 - [asan] Add support for disable_sanitizer_instrumentation attribute

Tue, 23 Nov 2021 09:32:35 +0000

[asan] Add support for disable_sanitizer_instrumentation attributeFor ASan this will effectively serve as a synonym for__attribute__((no_sanitize("address")))Differential Revision: https://reviews.llvm.org/D114421 List of files: /llvm-project-15.0.7/clang/test/CodeGen/asan-globals.cpp

8924ba3b - [NFC][clang] Inclusive terms: replace uses of blacklist in clang/test/

Wed, 17 Nov 2021 12:40:55 +0000

[NFC][clang] Inclusive terms: replace uses of blacklist in clang/test/Replace filenames, variable names, check prefixes uses of blacklist with ignore list.Reviewed By: jkorousDifferential Revision: https://reviews.llvm.org/D113211 List of files: /llvm-project-15.0.7/clang/test/CodeGen/asan-globals.cpp

7647a841 - Fix -fno-unwind-tables -fasynchronous-unwind-tables to emit unwind tables

Thu, 23 Sep 2021 23:15:40 +0000

Fix -fno-unwind-tables -fasynchronous-unwind-tables to emit unwind tablesThis matches GCC.Change the CC1 option to encode the unwind table level (1: needed by exceptions,2: asynchronous) so that we can support two modes in the future. List of files: /llvm-project-15.0.7/clang/test/CodeGen/asan-globals.cpp

dbc641de - [sanitizer] Reduce redzone size for small size global objects

Thu, 20 May 2021 02:18:49 +0000

[sanitizer] Reduce redzone size for small size global objectsCurrently 1 byte global object has a ridiculous 63 bytes redzone.This patch reduces the redzone size to be less than 32 if the size of global object is less than or equal to half of 32 (the minimal size of redzone).A 12 bytes object has a 20 bytes redzone, a 20 bytes object has a 44 bytes redzone.Reviewed By: MaskRay, #sanitizers, vitalybukaDifferential Revision: https://reviews.llvm.org/D102469 List of files: /llvm-project-15.0.7/clang/test/CodeGen/asan-globals.cpp

775a9483 - [IR][sanitizer] Set nounwind on module ctor/dtor, additionally set uwtable if -fasynchronous-unwind-tables

Wed, 21 Apr 2021 22:58:20 +0000

[IR][sanitizer] Set nounwind on module ctor/dtor, additionally set uwtable if -fasynchronous-unwind-tablesOn ELF targets, if a function has uwtable or personality, or does not havenounwind (`needsUnwindTableEntry`), it marks that `.eh_frame` is needed in the module.Then, a function gets `.eh_frame` if `needsUnwindTableEntry` or `-g[123]` is specified.(i.e. If -g[123], every function gets `.eh_frame`.This behavior is strange but that is the status quo on GCC and Clang.)Let's take asan as an example. Other sanitizers are similar.`asan.module_[cd]tor` has no attribute. `needsUnwindTableEntry` returns true,so every function gets `.eh_frame` if `-g[123]` is specified.This is the root cause that`-fno-exceptions -fno-asynchronous-unwind-tables -g` produces .debug_framewhile`-fno-exceptions -fno-asynchronous-unwind-tables -g -fsanitize=address` produces .eh_frame.This patch* sets the nounwind attribute on sanitizer module ctor/dtor.* let Clang emit a module flag metadata "uwtable" for -fasynchronous-unwind-tables. If "uwtable" is set, sanitizer module ctor/dtor additionally get the uwtable attribute.The "uwtable" mechanism is generic: synthesized functions not cloned/specializedfrom existing ones should consider `Function::createWithDefaultAttr` instead of`Function::create` if they want to get some default attributes whichhave more of module semantics.Other candidates: "frame-pointer" (https://github.com/ClangBuiltLinux/linux/issues/955https://github.com/ClangBuiltLinux/linux/issues/1238), dso_local, etc.Differential Revision: https://reviews.llvm.org/D100251 List of files: /llvm-project-15.0.7/clang/test/CodeGen/asan-globals.cpp

52cae05e - [ASan][Test] Fix expected strings for globals test

Wed, 10 Jun 2020 18:20:23 +0000

[ASan][Test] Fix expected strings for globals testThe expected strings would previously not catch bugs when redzones wereadded when they were not actually expected. Fix by adding "global "before the type. List of files: /llvm-project-15.0.7/clang/test/CodeGen/asan-globals.cpp

0f04f104 - [ASan][Test] Split out global alias test

Wed, 10 Jun 2020 17:55:51 +0000

[ASan][Test] Split out global alias testSome platforms do not support aliases. Split the test, and pass explicittriple to avoid test failure.Reviewed By: thakisTags: #clangDifferential Revision: https://reviews.llvm.org/D81591 List of files: /llvm-project-15.0.7/clang/test/CodeGen/asan-globals.cpp

d3f89314 - [KernelAddressSanitizer] Make globals constructors compatible with kernel [v2]

Wed, 10 Jun 2020 13:01:40 +0000

[KernelAddressSanitizer] Make globals constructors compatible with kernel [v2][ v1 was reverted by c6ec352a6bde1995794c523adc2ebab802ccdf0a due to modpost failing; v2 fixes this. More info: https://github.com/ClangBuiltLinux/linux/issues/1045#issuecomment-640381783 ]This makes -fsanitize=kernel-address emit the correct globalsconstructors for the kernel. We had to do the following:* Disable generation of constructors that rely on linker features such as dead-global elimination.* Only instrument globals *not* in explicit sections. The kernel uses sections for special globals, which we should not touch.* Do not instrument globals that are prefixed with "__" nor that are aliased by a symbol that is prefixed with "__". For example, modpost relies on specially named aliases to find globals and checks their contents. Unfortunately modpost relies on size stored as ELF debug info and any padding of globals currently causes the debug info to cause size reported to be *with* redzone which throws modpost off.Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=203493Tested:* With 'clang/test/CodeGen/asan-globals.cpp'.* With test_kasan.ko, we can see: BUG: KASAN: global-out-of-bounds in kasan_global_oob+0xb3/0xba [test_kasan]* allyesconfig, allmodconfig (x86_64)Reviewed By: gliderTags: #clang, #llvmDifferential Revision: https://reviews.llvm.org/D81390 List of files: /llvm-project-15.0.7/clang/test/CodeGen/asan-globals.cpp

c6ec352a - Revert "[KernelAddressSanitizer] Make globals constructors compatible with kernel"

Mon, 08 Jun 2020 08:30:50 +0000

Revert "[KernelAddressSanitizer] Make globals constructors compatible with kernel"This reverts commit 866ee2353f7d0224644799d0d1faed53c7f3a06d.Building the kernel results in modpost failures due to modpost relyingon debug info and inspecting kernel modules' globals:https://github.com/ClangBuiltLinux/linux/issues/1045#issuecomment-640381783 List of files: /llvm-project-15.0.7/clang/test/CodeGen/asan-globals.cpp