Use the SelectionDAG::getValidShiftAmountConstant helper to get const/constsplat shift amounts, which allows us to drop the out of range shift amount early-out.
First step towards better non-uniform shift amount support in SimplifyDemandedBits.
I believe this was carried over from getELFKindForNamedSection since
the wasm backend originally used ELF object writing as a template.
Differential Revision: https://reviews.llvm.org/D74565
This includes both GEPs where the indexed type is a scalable vector, and
GEPs where the result type is a scalable vector.
Differential Revision: https://reviews.llvm.org/D73602
When analyzing PHIs, we gather the known bits for every operand and
merge them together to get the known bits of the result of the PHI.
It is not unusual that merging the information leads to know nothing
on the result (e.g., phi a: i8 3, b: i8 unknown, ..., after looking at the
second argument we know we will know nothing on the result), thus, as
soon as we reach that state, stop analyzing the following operand (i.e.,
on the previous example, we won't process anything after looking at `b`).
This improves compile time in particular with PHIs with a large number
of operands.
NFC.
Similar to what we already do with SimplifyDemandedVectorElts, call SimplifyDemandedBits across all the extracted elements of the source vector, treating it as single use.
There's a minor regression in store-weird-sizes.ll which will be addressed in an upcoming SimplifyDemandedBits patch.
Summary:
If the programmer adds static profile data to a branch---i.e. uses
"__builtin_expect()" or similar---then we should honor it. Otherwise,
"__builtin_expect()" is ignored in crucial situations. So we trust that
the programmer knows what they're doing until proven wrong.
Subscribers: hiraditya, JDevlieghere, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D74809
Instcombine folds (a + b <u a) to (a ^ -1 <u b) and that does not match
the expected pattern in CodeGenPerpare via UAddWithOverflow.
This causes a regression over Clang 7 on both X86 and AArch64:
https://gcc.godbolt.org/z/juhXYV
This patch extends UAddWithOverflow to also catch the XOR case, if the
XOR is only used in the ICMP. This covers just a single case, but I'd
like to make sure I am not missing anything before tackling the other
cases.
Reviewers: nikic, RKSimon, lebedev.ri, spatel
Reviewed By: nikic, lebedev.ri
Differential Revision: https://reviews.llvm.org/D74228
On some targets, like SPARC, forming overflow ops is only profitable if
the math result is used: https://godbolt.org/z/DxSmdB
This patch adds a new MathUsed parameter to allow the targets
to make the decision and defaults to only allowing it
if the math result is used. That is the conservative choice.
This patch also updates AArch64ISelLowering, X86ISelLowering,
ARMISelLowering.h, SystemZISelLowering.h to allow forming overflow
ops if the math result is not used. On those targets using the
overflow intrinsic for the overflow check only generates better code.
Reviewers: nikic, RKSimon, lebedev.ri, spatel
Reviewed By: lebedev.ri
Differential Revision: https://reviews.llvm.org/D74722
https://reviews.llvm.org/D67133
While investigating some non determinism (CSE doesn't produce wrong
code, it just doesn't CSE some times) in GISel CSE on an out of tree
target, I realized that the core issue was that there were lots of code
that mutates (setReg, setRegClass etc), but doesn't notify observers
(CSE in this case but this could be any other observer). In order to
make the Observer be available in various parts of code and to avoid
having to thread it through various API, the MachineFunction now has the
observer as field. This allows it to be easily used in helper functions
such as constrainOperandRegClass.
Also added some invariant verification method in CSEInfo which can
catch these issues (when CSE is enabled).
Summary:
Unlike normal calls, call_indirects have immediate arguments that
caused a MachineVerifier failure without a small tweak to loosen the
verifier's requirements for variadicOpsAreDefs instructions.
One nice thing about the new call_indirects is that they do not need
to participate in the PCALL_INDIRECT mechanism because their post-isel
hook handles moving the function pointer argument and adding the flags
and typeindex arguments itself.
Reviewers: aheejin
Subscribers: dschuff, sbc100, jgravelle-google, hiraditya, sunfish, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D74191
This reverts commit 649aba93a27170cb03a4b17c98a19b9237a880b8, now that
the approach started there has been shown to be workable in the patch
series culminating in https://reviews.llvm.org/D74192.
Summary:
Making `Scale` a `TypeSize` in AArch64InstrInfo::getMemOpInfo,
has the effect that all places where this information is used
(notably, TargetInstrInfo::getMemOperandWithOffset) will need
to consider Scale - and derived, Offset - possibly being scalable.
This patch adds a new operand `bool &OffsetIsScalable` to
TargetInstrInfo::getMemOperandWithOffset and fixes up all
the places where this function is used, to consider the
offset possibly being scalable.
In most cases, this means bailing out because the algorithm does not
(or cannot) support scalable offsets in places where it does some
form of alias checking for example.
Reviewers: rovka, efriedma, kristof.beyls
Reviewed By: efriedma
Subscribers: wuzish, kerbowa, MatzeB, arsenm, nemanjai, jvesely, nhaehnle, hiraditya, kbarton, javed.absar, asb, rbar, johnrusso, simoncook, sabuasal, niosHD, jrtc27, MaskRay, zzheng, edward-jones, rogfer01, MartinMosbeck, brucehoult, the_o, PkmX, jocewei, jsji, Jim, lenary, s.egerton, pzheng, sameer.abuasal, apazos, luismarques, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D72758
This patch enables the debug entry values feature.
- Remove the (CC1) experimental -femit-debug-entry-values option
- Enable it for x86, arm and aarch64 targets
- Resolve the test failures
- Leave the llc experimental option for targets that do not
support the CallSiteInfo yet
Differential Revision: https://reviews.llvm.org/D73534
Summary:
Backends should fold the subtraction into the comparison, but not all
seem to. Moreover, on targets where pointers are not integers, such as
CHERI, an integer subtraction is not appropriate. Instead we should just
compare the two pointers directly, as this should work everywhere and
potentially generate more efficient code.
Reviewers: bogner, lebedev.ri, efriedma, t.p.northover, uweigand, sunfish
Reviewed By: lebedev.ri
Subscribers: dschuff, sbc100, arichardson, jgravelle-google, hiraditya, aheejin, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D74454
For a file in WebKit, this brings the time spent in LiveDebugValues down
from 16 minutes to 2 minutes. The reduction comes from iterating the set
of open variable locations just once in transferRegisterDef. Post-patch,
the most expensive item inside of transferRegisterDef is a call to
VarLoc::isDescribedByReg, which we have to do.
Testing: I built LNT using the Os-g cmake cache with & without this
patch, then diffed the object files to verify there was no binary diff.
rdar://59446577
Differential Revision: https://reviews.llvm.org/D74633
These are going to be useful in TargetLowering::SimplifyDemandedBits, so expose these helpers outside of SelectionDAG.cpp
Also add an getValidShiftAmountConstant early-out to getValidMinimumShiftAmountConstant/getValidMaximumShiftAmountConstant so we can use them for scalar cases as well.
Produce an unmerge to a narrower type and introduce a narrower shift
if needed. I wasn't sure if there was a better way to parameterize the
target's preferred shift type for the GICombineRule, so manually call
the combine helper.
Summary:
This patch implements the part of the calling convention
where SVE Vectors are passed by reference. This means the
caller must allocate stack space for these objects and
pass the address to the callee.
Reviewers: efriedma, rovka, cameron.mcinally, c-rhodes, rengolin
Reviewed By: efriedma
Subscribers: tschuett, kristof.beyls, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D71216
This adds another pattern to the combiner for a case that we were not handling
to generate the REV16 instruction for ARM/Thumb2 and a bswap+ror on X86.
Differential Revision: https://reviews.llvm.org/D74032
Summary:
When using strict fp, it is required to update the
chain when performing integer type promotion of a
operand to a integer to floating point conversion.
Reviewers: craig.topper, john.brawn
Reviewed By: craig.topper
Subscribers: kristof.beyls, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D74597
Follow-up for D74006.
When the integrated assembler is used, we use SHF_LINK_ORDER. The
linked-to symbol is part of ELFSectionKey, thus we can omit the unique
ID.
In https://reviews.llvm.org/rG8b737688c21a9755cae14cb9343930e0882164ab I
switched the condition gating the creation of the descriptor symbol from
checking the MCAsmInfo if we need to support descriptors, to if the OS
was AIX. Technically the 2 should be interchangeable: if we are
targeting AIX then we need to emit XCOFF object files, and the MCAsmInfo
must return true for needing function descriptors.
This doesn't account for lit test with runsteps that only set the arch.
Eg: test/CodeGen/XCore/section-name.ll
which when run natively on AIX we end up with a target xcore-ibm-aix and
needFunctionDescriptors is false.
This patch reverts to using the MCAsmInfo and adds an assert that the
target OS must be AIX since that is the only target using the descriptor
hook.
Differential Revision: https://reviews.llvm.org/D74622
This is more or less directly ported from the AMDGPU custom lowering
for FP_TO_FP16. I made a few minor fixups (using G_UNMERGE_VALUES
instead of creating shift/trunc to extract the two halves, and zexting
an inverted compare instead of select_cc).
This also does not include the fast math expansion the DAG which
converts to f32 and then to f16. I think that belongs in a
pre-legalize combine instead.
Like COPY instructions explained in D70616, we don't check the constraints
when combining G_UNMERGE_VALUES. Use the same logic used in D70616 to check
if registers can be replaced, or a COPY instruction needs to be built.
https://reviews.llvm.org/D70564
The goal of this patch is to maximize CPU utilization on multi-socket or high core count systems, so that parallel computations such as LLD/ThinLTO can use all hardware threads in the system. Before this patch, on Windows, a maximum of 64 hardware threads could be used at most, in some cases dispatched only on one CPU socket.
== Background ==
Windows doesn't have a flat cpu_set_t like Linux. Instead, it projects hardware CPUs (or NUMA nodes) to applications through a concept of "processor groups". A "processor" is the smallest unit of execution on a CPU, that is, an hyper-thread if SMT is active; a core otherwise. There's a limit of 32-bit processors on older 32-bit versions of Windows, which later was raised to 64-processors with 64-bit versions of Windows. This limit comes from the affinity mask, which historically is represented by the sizeof(void*). Consequently, the concept of "processor groups" was introduced for dealing with systems with more than 64 hyper-threads.
By default, the Windows OS assigns only one "processor group" to each starting application, in a round-robin manner. If the application wants to use more processors, it needs to programmatically enable it, by assigning threads to other "processor groups". This also means that affinity cannot cross "processor group" boundaries; one can only specify a "preferred" group on start-up, but the application is free to allocate more groups if it wants to.
This creates a peculiar situation, where newer CPUs like the AMD EPYC 7702P (64-cores, 128-hyperthreads) are projected by the OS as two (2) "processor groups". This means that by default, an application can only use half of the cores. This situation could only get worse in the years to come, as dies with more cores will appear on the market.
== The problem ==
The heavyweight_hardware_concurrency() API was introduced so that only *one hardware thread per core* was used. Once that API returns, that original intention is lost, only the number of threads is retained. Consider a situation, on Windows, where the system has 2 CPU sockets, 18 cores each, each core having 2 hyper-threads, for a total of 72 hyper-threads. Both heavyweight_hardware_concurrency() and hardware_concurrency() currently return 36, because on Windows they are simply wrappers over std:🧵:hardware_concurrency() -- which can only return processors from the current "processor group".
== The changes in this patch ==
To solve this situation, we capture (and retain) the initial intention until the point of usage, through a new ThreadPoolStrategy class. The number of threads to use is deferred as late as possible, until the moment where the std::threads are created (ThreadPool in the case of ThinLTO).
When using hardware_concurrency(), setting ThreadCount to 0 now means to use all the possible hardware CPU (SMT) threads. Providing a ThreadCount above to the maximum number of threads will have no effect, the maximum will be used instead.
The heavyweight_hardware_concurrency() is similar to hardware_concurrency(), except that only one thread per hardware *core* will be used.
When LLVM_ENABLE_THREADS is OFF, the threading APIs will always return 1, to ensure any caller loops will be exercised at least once.
Differential Revision: https://reviews.llvm.org/D71775
