[PGO] Ensure vp data in indexed profile always sorted
Done in InstrProfWriter to eliminate the need for client
code to do the sorting. The operation is done once and reused
many times so it is more efficient. Update unit test to remove
sorting. Also update expected output of affected tests.
Kyle Butt [Fri, 8 Jan 2016 02:06:19 +0000 (02:06 +0000)]
Add call sequence start and end for __tls_get_addr
This is a fix for bug http://llvm.org/bugs/show_bug.cgi?id=25839.
For a PIC TLS variable access in a function, prologue (mflr followed by std and
stdu) gets scheduled after a tls_get_addr call. tls_get_addr messed up LR but
no one saves/restores it.
Also added a test for save/restore clobbered registers during calling __tls_get_addr.
Kyle Butt [Fri, 8 Jan 2016 01:55:13 +0000 (01:55 +0000)]
[Vectorization] Actually return from error case in isStridedPtr
The early return seems to be missed. This causes a radical and wrong loop
optimization on powerpc. It isn't reproducible on x86_64, because
"UseInterleaved" is false.
Sanjay Patel [Fri, 8 Jan 2016 01:39:16 +0000 (01:39 +0000)]
[InstCombine] insert a new shuffle in a safe place (PR25999)
Limit this transform to a basic block and guard against PHIs.
Hopefully, this fixes the remaining failures in PR25999:
https://llvm.org/bugs/show_bug.cgi?id=25999
JF Bastien [Thu, 7 Jan 2016 23:18:29 +0000 (23:18 +0000)]
WebAssembly: use .skip instead of .zero directive
.zero is confusing when used with two arguments. Documentation:
This directive emits SIZE 0-valued bytes. SIZE must be an absolute
expression. This directive is actually an alias for the '.skip'
directive so in can take an optional second argument of the value to
store in the bytes instead of zero. Using '.zero' in this way would be
confusing however.
Keno Fischer [Thu, 7 Jan 2016 22:18:37 +0000 (22:18 +0000)]
[Verifier] Check that debug values have proper size
Summary:
Teach the Verifier to make sure that the storage size given to llvm.dbg.declare
or the value size given to llvm.dbg.value agree with what is declared in
DebugInfo. This is implicitly assumed in a number of passes (e.g. in SROA).
Additionally this catches a number of common mistakes, such as passing a
pointer when a value was intended or vice versa.
One complication comes from stack coloring which modifies the original IR when
it merges allocas in order to make sure that if AA falls back to the IR it gets
the correct result. However, given this new invariant, indiscriminately
replacing one alloca by a different (differently sized one) is no longer valid.
Fix this by just undefing out any use of the alloca in a dbg.declare in this
case.
Additionally, I had to fix a number of test cases. Of particular note:
- I regenerated dbg-changes-codegen-branch-folding.ll from the given source as
it was affected by the bug fixed in r256077
- two-cus-from-same-file.ll was changed to avoid having a variable-typed debug
variable as that would depend on the target, even though this test is
supposed to be generic
- I had to manually declared size/align for reference type. See also the
discussion for D14275/r253186.
- fpstack-debuginstr-kill.ll required changing `double` to `long double`
- most others were just a question of adding OP_deref
Dimitry Andric [Thu, 7 Jan 2016 22:09:12 +0000 (22:09 +0000)]
Turn off lldb debug tuning by default for FreeBSD
Summary:
In rL242338, debugger tuning was introduced, and the tuning for FreeBSD
was set to lldb by default. However, for the foreseeable future we
still need to default to gdb tuning, since lldb is not ready for all of
FreeBSD's architectures, and some system tools (like objcopy, etc) have
not yet been adapted to cope with the lldb tuned format, which has
.apple sections.
Therefore, let FreeBSD use gdb by default for now.
David Majnemer [Thu, 7 Jan 2016 21:36:16 +0000 (21:36 +0000)]
[SCCP] Don't violate the lattice invariants
We marked values which are 'undef' as constant instead of undefined
which violates SCCP's invariants. If we can figure out that a
computation results in 'undef', leave it in the undefined state.
Coverage mapping data may reference names of functions
that are skipped by FE (e.g, unused inline functions). Since
those functions are skipped, normal instr-prof function lowering
pass won't put those names in the right section, so special
handling is needed to walk through coverage mapping structure
and recollect the references.
With this patch, only names that are skipped are processed. This
simplifies the lowering code and it no longer needs to make
assumptions coverage mapping data layout. It should also be
more efficient.
David Majnemer [Thu, 7 Jan 2016 19:25:39 +0000 (19:25 +0000)]
[SCCP] Can't go from overdefined to constant
The fix for PR23999 made us mark loads of null as producing the constant
undef which upsets the lattice. Instead, keep the load as "undefined".
This fixes PR26044.
Dan Gohman [Thu, 7 Jan 2016 18:49:53 +0000 (18:49 +0000)]
[WebAssembly] Use the default private label prefixes.
The MC assembler doesn't like using the empty string as a private label
prefix because then it treats all labels as private. This commit reverts
back to the default prefix, which is .L, which is common in ELF targets
and consistent with the LLVM name mangler.
Nicolai Haehnle [Thu, 7 Jan 2016 17:10:29 +0000 (17:10 +0000)]
AMDGPU/SI: Fold operands with sub-registers
Summary:
Multi-dword constant loads generated unnecessary moves from SGPRs into VGPRs,
increasing the code size and VGPR pressure. These moves are now folded away.
Note that this lack of operand folding was not a problem for VMEM loads,
because COPY nodes from VReg_Nnn to VGPR32 are eliminated by the register
coalescer.
Some tests are updated, note that the fsub.ll test explicitly checks that
the move is elided.
With the IR generated by current Mesa, the changes are obviously relatively
minor:
Nicolai Haehnle [Thu, 7 Jan 2016 17:10:20 +0000 (17:10 +0000)]
AMDGPU/SI: xnack_mask is always reserved on VI
Summary:
Somehow, I first interpreted the docs as saying space for xnack_mask is only
reserved when XNACK is enabled via SH_MEM_CONFIG. I felt uneasy about this and
went back to actually test what is happening, and it turns out that xnack_mask
is always reserved at least on Tonga and Carrizo, in the sense that flat_scr
is always fixed below the SGPRs that are used to implement xnack_mask, whether
or not they are actually used.
I confirmed this by writing a shader using inline assembly to tease out the
aliasing between flat_scratch and regular SGPRs. For example, on Tonga, where
we fix the number of SGPRs to 80, s[74:75] aliases flat_scratch (so
xnack_mask is s[76:77] and vcc is s[78:79]).
This patch changes both the calculation of the total number of SGPRs and the
various register reservations to account for this.
It ought to be possible to use the gap left by xnack_mask when the feature
isn't used, but this patch doesn't try to do that. (Note that the same applies
to vcc.)
Note that previously, even before my earlier change in r256794, the SGPRs that
alias to xnack_mask could end up being used as well when flat_scr was unused
and the total number of SGPRs happened to fall on the right alignment
(e.g. highest regular SGPR being used s29 and VCC used would lead to number
of SGPRs being 32, where s28 and s29 alias with xnack_mask). So if there
were some conflict due to such aliasing, we should have noticed that already.
Silviu Baranga [Thu, 7 Jan 2016 14:56:08 +0000 (14:56 +0000)]
[InstCombine] Look through PHIs, GEPs, IntToPtrs and PtrToInts to expose more constants when comparing GEPs
Summary:
When comparing two GEP instructions which have the same base pointer
and one of them has a constant index, it is possible to only compare
indices, transforming it to a compare with a constant. This removes
one use for the GEP instruction with the constant index, can reduce
register pressure and can sometimes lead to removing the comparisson
entirely.
InstCombine was already doing this when comparing two GEPs if the
base pointers were the same. However, in the case where we have
complex pointer arithmetic (GEPs applied to GEPs, PHIs of GEPs,
conversions to or from integers, etc) the value of the original
base pointer will be hidden to the optimizer and this transformation
will be disabled.
This change detects when the two sides of the comparison can be
expressed as GEPs with the same base pointer, even if they don't
appear as such in the IR. The transformation will convert all the
pointer arithmetic to arithmetic done on indices and all the
relevant uses of GEPs to GEPs with a common base pointer. The
GEP comparison will be converted to a comparison done on indices.
James Molloy [Thu, 7 Jan 2016 13:33:28 +0000 (13:33 +0000)]
[GlobalsAA] Partially back out r248576
See PR25822 for a more full summary, but we were conflating the concepts of "capture" and "escape". We were proving nocapture and using that proof to infer noescape, which is not true. Escaped-ness is a function-local property - as soon as a value is used in a call argument it escapes. Capturedness is a related but distinct property. It implies a *temporally limited* escape. Consider:
static int a;
int b;
int g(int * nocapture arg);
int f() {
a = 2; // Even though a escapes to g, it is not captured so can be treated as non-escaping here.
g(&a); // But here it must be treated as escaping.
g(&b); // Now that g(&a) has returned we know it was not captured so we can treat it as non-escaping again.
}
The original commit did not sufficiently understand this nuance and so caused PR25822 and PR26046.
r248576 included both a performance improvement (which has been backed out) and a related conformance fix (which has been kept along with its testcase).
Mark arm as the 32bit variant of aarch64 in Triple
Change Triple::get32BitArchVariant to return arm/armeb as the 32bit
variant of aarch64/aarch64_be and do the same change for the oppoiste
direction in Triple::get64BitArchVariant.
Simon Pilgrim [Thu, 7 Jan 2016 10:24:19 +0000 (10:24 +0000)]
[X86][SSE} Add INSERTPS as a target shuffle
Follow up to D15378, added INSERTPS to the list of decodable target shuffles and enabled XFormVExtractWithShuffleIntoLoad to handle target shuffles with SentinelZero and tested this with INSERTPS.
Tim Northover [Thu, 7 Jan 2016 09:03:03 +0000 (09:03 +0000)]
ARM: support TLS accesses on Darwin platforms
Darwin TLS accesses most closely resemble ELF's general-dynamic situation,
since they have to be able to handle all possible situations. The descriptors
and so on are obviously slightly different though.
Jonas Paulsson [Thu, 7 Jan 2016 07:20:55 +0000 (07:20 +0000)]
[SystemZ] Add hasSideEffects flag on Serialize instruction.
Serialize will perform a hardware serialization operation, and is
acting as a memory barrier. Therefore it must have the hasSideEffects
flag set so it will be treated as a global memory object.
Philip Reames [Thu, 7 Jan 2016 04:15:31 +0000 (04:15 +0000)]
[Statepoints] Add test cases around vectors and stablize test
Unlike my comment in 257022 said, it turns out we do handle constant vectors in the statepoint lowering, but only because SelectionDAG doesn't actually produce constants for them. Add a couple of tests which show this working.
Also, add a triple to the same test file to hopefully fix a failing bot.
Philip Reames [Thu, 7 Jan 2016 03:32:11 +0000 (03:32 +0000)]
[Statepoints] Initial support for relocating vectors of pointers
Currently, we try to split vectors of pointers back into their component pointer elements during rewrite-statepoints-for-gc. This is less than ideal since presumably the vectorizer chose to vectorize for a reason. :) It's also been a source of bugs - in particular, the relocation logic as currently implemented was recently discovered to be wrong.
The alternate approach is to allow gc.relocates of vector-of-pointer type and update the backend to handle them. That's what this patch tries to do. This won't actually enable vector-of-pointers in practice - there are some RS4GC changes needed - but the lowering is standalone and testable so it makes sense to separate.
Note that there are some known cases around vector constants which this patch does not handle. Once this is in, I'll send another patch with individual fixes and test cases.
Philip Reames [Thu, 7 Jan 2016 02:20:11 +0000 (02:20 +0000)]
[RS4GC] Add an option to suppress vector splitting
At the moment, this is essentially a diangostic option so that I can start collecting failing test cases, but we will eventually migrate to removing the vector splitting code entirely.
[ShrinkWrapping] Give up on irreducible CFGs.
We need to know whether or not a given basic block is in a loop for the analysis
to be correct.
Loop information may be incomplete on irreducible CFGs, therefore we may
generate incorrect code if we use it in those situations.
Andrew Wilkins [Thu, 7 Jan 2016 00:18:56 +0000 (00:18 +0000)]
tools/llvm-config: improve shared library support
Summary:
r252532 added support for reporting the monolithic library
when LLVM_BUILD_LLVM_DYLIB is used. This would only be done
if the individual components were not found, and the dynamic
library is found.
This diff extends this as follows:
- If LLVM_LINK_LLVM_DYLIB is set, then prefer the shared
library, even if all component libraries exist.
- Two flags, --link-shared and --link-static are introduced
to provide explicit guidance. If --link-shared is passed
and the shared library does not exist, an error results.
Additionally, changed the expected shared library names from
(e.g.) LLVM-3.8.0 to LLVM-3.8. The former exists only in an
installation (and then only in CMake builds I think?), and not
in the build tree; this breaks usage of llvm-config during
builds, e.g. by llvm-go.
Teresa Johnson [Thu, 7 Jan 2016 00:06:27 +0000 (00:06 +0000)]
Always treat DISubprogram reached by DIImportedEntity as needed.
It is illegal to have a null entity in a DIImportedEntity, so
we must link in a DISubprogram metadata node referenced by one,
even if the associated function is not linked in or inlined anywhere.
Simon Pilgrim [Wed, 6 Jan 2016 23:24:40 +0000 (23:24 +0000)]
[X86] Determine if target shuffle can contain zero elements
getTargetShuffleMask may return shuffle masks with SM_SentinelZero (-2) values (currently just for PSHUFB but VPERM2X128 as well with this patch). Although some calling functions can make use of this (mainly for shuffle combining), others can not and their inclusion makes shuffle mask comparisons more difficult.
This patch adds a flag to getTargetShuffleMask to indicate if the calling function can't handle SM_SentinelZero; getTargetShuffleMask will then return false if it occurs to make handling much easier.
I've tidied up some uses of getTargetShuffleMask to better indicate what is going on - more could be done but at present I don't have test cases to demonstrate it.
Some upcoming patches will make use of this to both support more uses where SM_SentinelZero is not permitted (e.g. combineShuffleToAddSub), and also will allow us to add INSERTPS support to getTargetShuffleMask as part of better zero handling discussed in D14261.
Justin Bogner [Wed, 6 Jan 2016 23:16:37 +0000 (23:16 +0000)]
Bitcode: Move these tests into compatibility.ll
I added a couple of tests in r256982, but vedantk suggested that they
fit better into compatibility.ll, since they could catch format breaks
later on there.
Justin Bogner [Wed, 6 Jan 2016 22:31:32 +0000 (22:31 +0000)]
Bitcode: Fix reading and writing of ConstantDataVectors of halfs
In r254991 I allowed ConstantDataVectors to contain elements of
HalfTy, but I missed updating the bitcode reader and writer to handle
this, so now we crash if we try to emit bitcode on programs that have
constant vectors of half.
This fixes the issue and adds test coverage for reading and writing
constant sequences in bitcode.
Nicolai Haehnle [Wed, 6 Jan 2016 22:01:04 +0000 (22:01 +0000)]
AMDGPU/SI: Fix crash when inline assembly is used in a graphics shader
Summary:
This is admittedly something that you could only run into by manually
playing around with shader assembly because the SITypeWriter pass is
skipped for compute.
Chen Li [Wed, 6 Jan 2016 20:32:05 +0000 (20:32 +0000)]
[SplitLandingPadPredecessors] Create a PHINode for the original landingpad only if it has some uses
Summary: This patch adds a check in SplitLandingPadPredecessors to see if the original landingpad instruction has any uses. If not, we don't need to create a PHINode for it in the joint block since it's gonna be a dead code anyway. The motivation for this patch is that we found a bug that SplitLandingPadPredecessors created a PHINode of token type landingpad, which failed the verifier since PHINode can not be token type. However, the created PHINode will never be used in our code pattern. This patch will workaround this bug, and we might add supports in SplitLandingPadPredecessors to handle token type landingpad with uses in the future.
Philip Reames [Wed, 6 Jan 2016 19:33:12 +0000 (19:33 +0000)]
Consolidate MemRefs handling from BranchFolding and correct latent bug
Move the logic from BranchFolding to use the shared infrastructure for merging MMOs introduced in 256909. This has the effect of making BranchFolding more capable.
In the process, fix a latent bug. The existing handling for merging didn't handle the case where one of the instructions being merged had overflowed and dropped MemRefs. This was a latent bug in the places the code was commoned from, but potentially reachable in BranchFolding.
Once this is in, we're left with a single place to consider implementing MMO unique-ing as proposed in http://reviews.llvm.org/D15230.
[X86] Correctly model TLS calls w.r.t. frame requirements.
TLS calls need the stack frame to be properly set up and this
implies that such calls need ADJUSTSTACK_xxx markers.
Nico Weber [Wed, 6 Jan 2016 19:05:19 +0000 (19:05 +0000)]
Make WinCOFFObjectWriter.cpp's timestamp writing not use ENABLE_TIMESTAMPS
LLVM_ENABLE_TIMESTAMPS controls if timestamps are embedded into llvm's
binaries. Turning it off is useful for deterministic builds.
r246905 made it so that the define suddenly also controls if the binaries that
the llvm binaries _create_ embed timestamps or not – but this shouldn't be a
configure-time option. r256203/r256204 added a driver option to toggle this on
and off, so this patch now passes this driver option in LLVM_ENABLE_TIMESTAMPS
builds so that if LLVM_ENABLE_TIMESTAMPS is set, the build of LLVM is
deterministic – but the built clang can still write timestamps into other
executables when requested.
This also allows removing some of the test machinery added in r292012 to work
around this problem.
See PR24740 for background.
http://reviews.llvm.org/D15783
[ShrinkWrap] Fix FindIDom to only have one kind of failure.
FindIDom() can fail in two different ways - it can either return nullptr or the
block itself, depending on the circumstances. Some users of FindIDom() check
one error condition, while others check the other.
Change it to always return nullptr on failure.
This fixes PR26004.
Dan Gohman [Wed, 6 Jan 2016 18:29:35 +0000 (18:29 +0000)]
[WebAssembly] Don't use range-based loop for a list that's being modified
The first instruction in a block is what the rend() iterator points to, so
if it moves, we need to re-evaluate rend() so that we continue to iterate
through the rest of the instructions.
Weiming Zhao [Wed, 6 Jan 2016 18:20:25 +0000 (18:20 +0000)]
Filtering IR printing for print-after-all/print-before-all
Summary:
This patch implements "-print-funcs" option to support function filtering for IR printing like -print-after-all, -print-before etc.
Examples:
-print-after-all -print-funcs=foo,bar
projects / llvm / log