I'm still curious about the source of the vptr diff, but that's a minor question, otherwise. LGTM

I don't think this is the right place for this at all. Look at X86Subtarget::initSubtargetFeatures please.

Can you change raw_fd_ostream::has_colors to memoize the result of FileDescriptorHasColors(FD) instead? That should give most of the performance gain without breaking the abstractions as much.

This issue is on my long term TODO list. If there are no 64bit atomics, it is a really stupid idea to use them for timekeeping. I think for the use case of Timer, just splitting it into second and subsecond with explicit overflow handling is perfectly reasonable and does not affect the correctness of the normal use as the aggregates are only accessed on completion?

Please don't change the style of the code, it messes up review and the full source. Please split the introduction and testing of the new relocation into one patch and using a different set of relocations to implement large code model. The latter part needs at least a test as well?

I don't agree with the justification at all, but it also seems that noone else cares about the build option creep here.

I see that more as a short-coming in the existing DEFAULT_SYSROOT behavior and less an argument for making more cases like it. So the general idea is that for turnkey toolchains,
we want to allow customizing the "default" target of the toolchain to hard-code options like --sysroot, --target, -Wl,-rpath etc. Those are all related, so when using a different target, they no longer make sense. One way to deal with all those options in a consistent manner is hook into the logic for determining the current target, if none is explicitly specified on the command line or implicit from the executable name, then prepend some additional flags on the command line based on some cmake variable or so. This flags shouldn't trigger the unused argument warnings, so you can always pass -Wl,-rpath, --sysroot etc, independent of whether the compiler is in preprocessing / compile / assemble / link mode. That seems to be a more general and systematic approach than adding many additional build-time options.

I do agree with the feature request, but I'm not sure about the implementation. It doesn't seem to work well with the cross-compiling support in the driver as clearly shown by the amount of tests that need patching. Is cross-compiling a concern for you at all? Otherwise I would suggest going a slightly different route and use default values iff no "-target" or "target-command" mechanism is active. Might need a way to temporarily disable unused flag warnings, but that way would be pretty much toolchain independent?

LGTM

ARM's data detection only works for unstripped binaries. Other architectures often don't even have such markers at all.

The more compact output is a lot harder to read when you have to find embedded data. I wouldn't mind dropping half the spaces to have columns of 4 hex digits, but anything more is really hard to read.

I'm not sure about this change. It might make sense on x86, but many RISC architectures love to put data into the instruction stream and being able to pick them out is quite important for understanding the disassembly.

LGTM

@ldionne I was updating libc++ from d42baff45d9700a199982ba0ac04dbc6c6d911bb and LLVM itself from 38aebe5c04ab4cb3695dc1bcc60b9a7b55215aff to 3d1424bc7a0e9a6f9887727e2bc93a10f50e1709 when it started failing. It likely has been present for a while.

This fixes the module build of clang for me.

If no (inline) assembler is involved, you can likely get away with IAS. Just to give two trivial examples that are not handled by IAS right now:

Louis, did I answer your questions?

We tend to test push .section and .pushsection in this file, so it would be nice to continue doing that.

Can you add a variation with .pushsection for @progbits and @nobits? Otherwise LGTM.

Ping?

This summarizes the IRC discussion.

The original code has a functional dependency between sz and bytes and whether they can be constant evaluated. But the code doesn't express that. I don't think we can enforce that in any sensible way. There are valid use cases after all where partial inlining would result in entirely sensible decisions, just think about the more typical case of __builtin_constant_p selecting between inline asm taking immediate operands and one taking register/memory operands. That's why I am saying that I consider it a lot more useful to provide reliable building blocks to express the dependency and make sure they work.

...in the outer condition, I meant.

Yes, builtin_expect is just noise here. The point I wanted to make is that both sz and bytes should be used in builtin_constant_p in the other condition.

IMO the root of the problem here is that the branches are mixing predicated variables (bytes) with non-predicated variables (sz). If users want deterministic behavior here, that shouldn't be done.

I was discussing this with Bill off-list to get a better idea of the original test case. Basically, with the new constant intrinsic pass, we can provide a stronger semantic: the default LLVM pass chain always constant folds expressions involving is.constant and performs DCE trivially dead branches resulting from the former folding. This means that if the original expression is adjusted from:

if (__builtin_expect(!!(sz >= 0 && sz < bytes), 0)) {
    if (!__builtin_constant_p(bytes))
    ...
}

to the functionally equivalent:

if (!(__builtin_constant_p(sz) && __builtin_constant_p(bytes) && sz >= 0 && sz >= bytes) && __builtin_expect(!!(sz >= 0 && sz < bytes), 0)) {
    if (!__builtin_constant_p(bytes))
    ...
}

then we can actually ensure proper DCE even with -O0 in the default pass chain. That depends over all on two properties:

• If __builtin_constant_p is constant folded by clang, it must DCE the appropiate places.
• the constant intrinsic pass is run

With -O1 or higher, this should work in general.

Committed as 0b999f76575f0196d3cd01c0781b2513b0a1af15 without link.

Require __STDCPP_NEW_ALIGNMENT__ in C++03 mode. Prefer it over max_align_t in a number of tests when allocation alignment is desired. Adjust some tests to do minimal sanity checking of the alignment for C++03 only. Add an explicit check that __STDCPP_NEW_ALIGNMENT__ is more general than max_align_t if both are present.

This looks much better than the in-tree state, thanks.

In D73245#1918287, @ldionne wrote:

In D73245#1918148, @EricWF wrote:

I've already stated my disapproval of this patch. Libc++ has never and will never provide nor value C++03 conformance.
Moving backwards to C++03 is inconsistent with the libraries general direction.

@EricWF makes a point here, we want to move away from C++03.

Ping

It still seems to be a perfectly reasonable optimisation under the semantics of is.constant. The code here seems to be an abuse of it though and makes assumptions that are not sensible.

I'm confused by that example. If jumpthreading makes it possible to answer true, it is no differerent from inlining and other use cases. The reverse is the problematic case, turning constant cases into non-constant cases?

libc++ has no idea what a correct max_align_t is. The internal definition works due to historic requirements that all three fundamental types are supported for new/delete, but we don't have any such guarantees for every other context. A correctly implemented stddef.h does not provide max_align_t in C++03 mode, since that would pollute the global namespace. This means that libc++ currently has two failure modes: on NetBSD, it outright tries to use a non-existing symbol. On other platforms it silently defines max_align_t in a way that can be subtle wrong.

Do not depend on max_align_t in C++03 mode in the test cases.

It is used both in <new> and <memory> and the use in the latter is currently unconditional AFAICT. I don't have a problem splitting the conditional to avoid the typedef. That would address the ODR concern?

Ping2?

Ping?

Let me clarify the situation for a moment:
(1) libc++ does try to work in C++03 mode. See the separate implementation of <functional> for pre-C++11. It is also desirable to support. This is completely beside the question of TR1 support.
(2) The only reason why max_align_t is currently necessary is because it is used as default alignment in <new>.
(3) Most compilers we care about already have a preprocessor symbol specifically for that purpose.

libc++ generally tries to play nice with C++03 code. It doesn't go out of its way to break it and keeping it usable helps dealing with a lot of rusty old code. That's what the patch is all about, not breaking things for no good reason.

Can this be restricted to Linux?

What's the fixup behavior of the linker here, e.g. does it make sense to just use absolute encoding here and let the linker fix it up?

This is normally done by using -Bstatic/-Bdynamic around the library. See tools::addOpenMPRuntime.

Just for the sake of correctness, this is provided by GNU ld as part of the map file (-Wl,-Map=foo.map). It lists all archive members and which object file and symbol triggered the inclusion.

Can we please get this or the other fix in RSN? Mesa hits this.

I'm a bit puzzled by the need for this change. The bump allocator already has logic to do power-of-two scaling for the allocation, so I wonder why it doesn't work properly here.

Adjust based on comments.

Why go back to the large tables for crc32? Just because JamCRC had that bug doesn't mean it should persist.

I wonder if we should actually enumerate evil here, i.e. give the situations in which inlining actually fails. As mentioned on IRC, I wonder if we shouldn't aim for the stronger semantics and at least warn by default of any situation that prevents always_inline from doing its job.

I'd make the argument that it should be strlcpy if present. That covers a lot of existing systems too. I'm not sure this optimisation should be done for freestanding mode though.

2nd ping. Chandler, care to check this please?

Ping.

Ping

LGTM

Can you declare a volatile variable and set that from f instead?

Switch the pass to use two rounds. The first round will collect all relevant intrinsics in RPO, the second one will translate them accordingly.