[PATCH] D152433: [ARM,AArch64] Add a full set of -mtp= options.

This revision was automatically updated to reflect the committed changes. Closed by commit rG10e42281144e: [ARM,AArch64] Add a full set of -mtp= options. (authored by simon_tatham). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D152433/new/ https://r

[PATCH] D152433: [ARM,AArch64] Add a full set of -mtp= options.

nickdesaulniers accepted this revision. nickdesaulniers added a comment. This revision is now accepted and ready to land. thanks for the patch! Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D152433/new/ https://reviews.llvm.org/D152433

[PATCH] D152433: [ARM,AArch64] Add a full set of -mtp= options.

simon_tatham added a comment. Right, this seems to be passing tests now, so I think @nickdesaulniers's issue is fixed, and I've also split up the tests as @MaskRay suggested. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D152433/new/ https://review

[PATCH] D152433: [ARM,AArch64] Add a full set of -mtp= options.

simon_tatham updated this revision to Diff 530830. simon_tatham added a comment. Rebased past rG34d7acd444b8 (which conflicted with it, though trivially) and attempted to fix the clang-format complaint in pre-merge checks.

[PATCH] D152433: [ARM,AArch64] Add a full set of -mtp= options.

simon_tatham updated this revision to Diff 530544. simon_tatham added a comment. Fixed the isReadTPSoft predicate (sorry – I always get confused by when Tablegen autogenerates those predicates and when it doesn't), and reorganised the tests. I couldn't find any existing test file, so I made a n

[PATCH] D152433: [ARM,AArch64] Add a full set of -mtp= options.

MaskRay added inline comments. Comment at: clang/test/Driver/clang-translation.c:132 +// ARMv7_THREAD_POINTER-HARD: "-target-feature" "+read-tp-tpidruro" + +// RUN: %clang -target armv7-linux -mtp=tpidruro -### -S %s 2>&1 | \ `clang-translation.c`'s job is shifti

[PATCH] D152433: [ARM,AArch64] Add a full set of -mtp= options.

nickdesaulniers added inline comments. Comment at: llvm/lib/Target/ARM/ARMBaseInstrInfo.cpp:4941-4943 if (LoadImmOpc == ARM::MRC || LoadImmOpc == ARM::t2MRC) { -assert(Subtarget.isReadTPHard() && +assert(!Subtarget.isReadTPSoft() && "TLS stack protector req

[PATCH] D152433: [ARM,AArch64] Add a full set of -mtp= options.

rengolin added a subscriber: olista01. rengolin added inline comments. Comment at: clang/include/clang/Driver/Options.td:3525 + "For AArch32: 'soft' uses a function call, or 'tpidrurw', 'tpidruro' or 'tpidrprw' use the three CP15 registers. 'cp15' is an alias for 'tpi

[PATCH] D152433: [ARM,AArch64] Add a full set of -mtp= options.

simon_tatham added a comment. In D152433#4405428 , @rengolin wrote: > The only minor visible difference is the removal of `read-tp-hard` option > from the LLVM side, which could be used by other downstream implementations. Yes. I wasn't sure how much th

[PATCH] D152433: [ARM,AArch64] Add a full set of -mtp= options.

rengolin added a comment. The only minor visible difference is the removal of `read-tp-hard` option from the LLVM side, which could be used by other downstream implementations. I personally don't think this is a big deal. First, we don't promise stability on that layer, and second, it would be

[PATCH] D152433: [ARM,AArch64] Add a full set of -mtp= options.

simon_tatham created this revision. simon_tatham added reviewers: nickdesaulniers, peter.smith, kristof.beyls, t.p.northover, rengolin. Herald added a subscriber: hiraditya. Herald added a project: All. simon_tatham requested review of this revision. Herald added subscribers: llvm-commits, cfe-com