There are a few questionable things about this intrinsic and existing
DAG implementation. For some reason the intrinsic hardcodes the second
operand to be scalar-only i32, and SelectionDAG builder makes a
legalization decision based on whether the operand is constant.
These are treated identically to value aggregates placed in the kernel
argument list. A %struct.foo or %struct.foo addrspace(4)*
byref(sizeof(%struct.foo)) align(alignof(%struct.foo)) argument should
produce the same offsets and argument metadata.
This handles all 3 kernel ABI implementations, and the two HSA
metadata emission paths.
getTargetShuffleMask is used by the various "SimplifyDemanded" folds so we can't assume that the bypassed extract_subvector can be safely simplified - getFauxShuffleMask performs a more general decode that allows us to more safely catch many of these cases so the impact is minimal.
The AMDGPU handling of f16 vectors is terrible still since it gets
scalarized even when the vector operation is legal.
The code is is essentially duplicated between the non-strict and
strict case. Apparently no other expansions are currently trying to do
this. This is mostly because I found the behavior of
getStrictFPOperationAction to be confusing. In the ARM case, it would
expand strict_fsub even though it shouldn't due to the later check. At
that point, the logic required to check for legality was more complex
than just duplicating the 2 instruction expansion.
SUMMARY:
when we call memset, memcopy,memmove etc(this are llvm intrinsic function) in the c source code. the llvm will generate IR
like call call void @llvm.memset.p0i8.i32(i8* align 4 bitcast (%struct.S* @s to i8*), i8 %1, i32 %2, i1 false)
for c source code
bash> cat test_memset.call
struct S{
int a;
int b;
};
extern struct S s;
void bar() {
memset(&s, s.b, s.b);
}
like
%struct.S = type { i32, i32 }
@s = external global %struct.S, align 4
; Function Attrs: noinline nounwind optnone
define void @bar() #0 {
entry:
%0 = load i32, i32* getelementptr inbounds (%struct.S, %struct.S* @s, i32 0, i32 1), align 4
%1 = trunc i32 %0 to i8
%2 = load i32, i32* getelementptr inbounds (%struct.S, %struct.S* @s, i32 0, i32 1), align 4
call void @llvm.memset.p0i8.i32(i8* align 4 bitcast (%struct.S* @s to i8*), i8 %1, i32 %2, i1 false)
ret void
}
declare void @llvm.memset.p0i8.i32(i8* nocapture writeonly, i8, i32, i1 immarg) #1
If we want to let the aix as assembly compile pass without -u
it need to has following assembly code.
.extern .memset
(we do not output extern linkage for llvm instrinsic function.
even if we output the extern linkage for llvm intrinsic function, we should not out .extern llvm.memset.p0i8.i32,
instead of we should emit .extern memset)
for other llvm buildin function floatdidf . even if we do not call these function floatdidf in the c source code(the generated IR also do not the call __floatdidf . the function call
was generated in the LLVM optimized.
the function is not in the functions list of Module, but we still need to emit extern .__floatdidf
The solution for it as :
We record all the lllvm intrinsic extern symbol when transformCallee(), and emit all these symbol in the AsmPrinter::doFinalization(Module &M)
Reviewers: jasonliu, Sean Fertile, hubert.reinterpretcast,
Differential Revision: https://reviews.llvm.org/D78929
Current tail duplication in machine block placement pass uses block frequency
information in cost model. But frequency number has only relative meaning
compared to other basic blocks in the same function. A large frequency number
doesn't mean it is hot and a small frequency number doesn't mean it is cold.
To overcome this problem, this patch uses profile count in cost model if it's
available. So we can tail duplicate real hot basic blocks.
Differential Revision: https://reviews.llvm.org/D83265
The default calling convention needs to save/restore the SVE callee
saves according to the SVE PCS when the function takes or returns
scalable types, even when the `aarch64_sve_vector_pcs` CC is not
specified for the function.
Reviewers: efriedma, paulwalker-arm, david-arm, rengolin
Reviewed By: paulwalker-arm
Differential Revision: https://reviews.llvm.org/D84041
Summary:
Teach LLVM to recognize the above pattern, where the operands are
either signed or unsigned types.
Subscribers: kristof.beyls, hiraditya, danielkiss, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D83777
These extra vcvt instructions were missed from 74ca67c109 because they
live in a different Domain, but should be treated in the same way.
Differential Revision: https://reviews.llvm.org/D83204
This isn't a natively supported operation, so convert it to a
mask+compare.
In addition to the operation itself, fix up some surrounding stuff to
make the testcase work: we need concat_vectors on i1 vectors, we need
legalization of i1 vector truncates, and we need to fix up all the
relevant uses of getVectorNumElements().
Differential Revision: https://reviews.llvm.org/D83811
Its effect could be achieved by
`-stop-after`,`-print-after`,`-print-after-all`. But a few tests need to
print MIR after ISel which could not be done with
`-print-after`/`-stop-after` since isel pass does not have commandline name.
That's the reason `--print-machineinstrs` is downgraded to
`--print-after-isel` in this patch. `--print-after-isel` could be
removed after we switch to new pass manager since isel pass would have a
commandline text name to use `print-after` or equivalent switches.
The motivation of this patch is to reduce tests dependency on
would-be-deprecated feature.
Reviewed By: arsenm, dsanders
Differential Revision: https://reviews.llvm.org/D83275
This was failing to add the size of LDS globals that weren't directly
used by an instruction. They could be used by constant expressions
which are transitively used by the function. This requires a better
search, but just abort on this for now for correctness.
This allows tracking the in-memory type of a pointer argument to a
function for ABI purposes. This is essentially a stripped down version
of byval to remove some of the stack-copy implications in its
definition.
This includes the base IR changes, and some tests for places where it
should be treated similarly to byval. Codegen support will be in a
future patch.
My original attempt at solving some of these problems was to repurpose
byval with a different address space from the stack. However, it is
technically permitted for the callee to introduce a write to the
argument, although nothing does this in reality. There is also talk of
removing and replacing the byval attribute, so a new attribute would
need to take its place anyway.
This is intended avoid some optimization issues with the current
handling of aggregate arguments, as well as fixes inflexibilty in how
frontends can specify the kernel ABI. The most honest representation
of the amdgpu_kernel convention is to expose all kernel arguments as
loads from constant memory. Today, these are raw, SSA Argument values
and codegen is responsible for turning these into loads.
Background:
There currently isn't a satisfactory way to represent how arguments
for the amdgpu_kernel calling convention are passed. In reality,
arguments are passed in a single, flat, constant memory buffer
implicitly passed to the function. It is also illegal to call this
function in the IR, and this is only ever invoked by a driver of some
kind.
It does not make sense to have a stack passed parameter in this
context as is implied by byval. It is never valid to write to the
kernel arguments, as this would corrupt the inputs seen by other
dispatches of the kernel. These argumets are also not in the same
address space as the stack, so a copy is needed to an alloca. From a
source C-like language, the kernel parameters are invisible.
Semantically, a copy is always required from the constant argument
memory to a mutable variable.
The current clang calling convention lowering emits raw values,
including aggregates into the function argument list, since using
byval would not make sense. This has some unfortunate consequences for
the optimizer. In the aggregate case, we end up with an aggregate
store to alloca, which both SROA and instcombine turn into a store of
each aggregate field. The optimizer never pieces this back together to
see that this is really just a copy from constant memory, so we end up
stuck with expensive stack usage.
This also means the backend dictates the alignment of arguments, and
arbitrarily picks the LLVM IR ABI type alignment. By allowing an
explicit alignment, frontends can make better decisions. For example,
there's real no advantage to an aligment higher than 4, so a frontend
could choose to compact the argument layout. Similarly, there is a
high penalty to using an alignment lower than 4, so a frontend could
opt into more padding for small arguments.
Another design consideration is when it is appropriate to expose the
fact that these arguments are all really passed in adjacent
memory. Currently we have a late IR optimization pass in codegen to
rewrite the kernel argument values into explicit loads to enable
vectorization. In most programs, unrelated argument loads can be
merged together. However, exposing this property directly from the
frontend has some disadvantages. We still need a way to track the
original argument sizes and alignments to report to the driver. I find
using some side-channel, metadata mechanism to track this
unappealing. If the kernel arguments were exposed as a single buffer
to begin with, alias analysis would be unaware that the padding bits
betewen arguments are meaningless. Another family of problems is there
are still some gaps in replacing all of the available parameter
attributes with metadata equivalents once lowered to loads.
The immediate plan is to start using this new attribute to handle all
aggregate argumets for kernels. Long term, it makes sense to migrate
all kernel arguments, including scalars, to be passed indirectly in
the same manner.
Additional context is in D79744.
It's useful for a debugger to be able to distinguish an @llvm.debugtrap
from a (noreturn) @llvm.trap, so this extends the existing Windows
behaviour to other platforms.
Add narrowScalarFor action.
Add narrow scalar for typeIndex == 0 for G_FPTOSI/G_FPTOUI.
Legalize using narrowScalarFor as s16->s32 G_FPTOSI/G_FPTOUI
followed by s32->s64 G_SEXT/G_ZEXT.
Differential Revision: https://reviews.llvm.org/D84010
fma reassoc A, B, C --> fadd (fmul A, B), C (when target has no FMA hardware)
C/C++ code may use explicit fma() calls (which become LLVM fma
intrinsics in IR) but then gets compiled with -ffast-math or similar.
For targets that do not have FMA hardware, we don't want to go out to
the math library for a precise but slow FMA result.
I tried this as a generic DAGCombine, but it caused infinite looping
on more than 1 other target, so there's likely some over-reaching fma
formation happening.
There's also a potential intersection of strict FP with fast-math here.
Deferring to current behavior for that case (assuming that strict-ness
overrides fast-ness).
Differential Revision: https://reviews.llvm.org/D83981
As far as I can tell, it should not be necessary for VCTP to be
unpredictable in tail predicated loops. Either it has a a valid loop
counter as a operand which will naturally keep it in the right loop, or
it doesn't and it won't be converted to a tail predicated loop. Not
marking it as having side effects allows it to be scheduled more cleanly
for cases where it is not expected to become a tail predicate loop.
Differential Revision: https://reviews.llvm.org/D83907
Summary:
In the `ppc-early-ret` pass, we have use `BuildMI` and `copyImplicitOps` when the branch instructions can do the early return. But the two functions will add implicit operands twice, this is not correct.
This patch is to remove the redundant implicit operands in `ppc-early-ret pass`.
Reviewed By: jsji
Differential Revision: https://reviews.llvm.org/D76042
Currently, BTF datasec type for .rodata is generated only if there are
user-defined readonly global variables which have debuginfo generated.
Certain readonly global variables may be generated from initialized
local variables. For example,
void foo(const void *);
int test() {
const struct {
unsigned a[4];
char b;
} val = { .a = {2, 3, 4, 5}, .b = 6 };
foo(&val);
return 0;
}
The clang will create a private linkage const global to store
the initialized value:
@__const.test.val = private unnamed_addr constant %struct.anon
{ [4 x i32] [i32 2, i32 3, i32 4, i32 5], i8 6 }, align 4
This global variable eventually is put in .rodata ELF section.
If there is .rodata ELF section, libbpf expects a BTF .rodata
datasec as well even though it may be empty meaning there are no
global readonly variables with proper debuginfo. Martin reported
a bug where without this empty BTF .rodata datasec, the bpftool
gen will exit with an error.
This patch fixed the issue by generating .rodata BTF datasec
if there exists local var intial data which will result in
.rodata ELF section.
Differential Revision: https://reviews.llvm.org/D84002
There were cases where a do-while loop would be converted to a while
loop before finding out that it would be unsafe to expand the SCEV in
this situation and then bailing out of hardware loop conversion.
This patch checks if it would be unsafe to expand the SCEV and if so stops converting the do-while into a while, allowing conversion to a hardware loop.
Differential Revision: https://reviews.llvm.org/D83953
As explained in the comment:
// For a FLAT instruction the hardware decides whether to access
// global/scratch/shared memory based on the high bits of vaddr,
// ignoring the offset field, so we have to ensure that when we add
// remainder to vaddr it still points into the same underlying object.
// The easiest way to do that is to make sure that we split the offset
// into two pieces that are both >= 0 or both <= 0.
In particular FLAT (as opposed to SCRATCH and GLOBAL) instructions have
an unsigned immediate offset field, so we can't use it to help split a
negative offset.
Differential Revision: https://reviews.llvm.org/D83394
tryLatency compares two sched candidates. For the top zone it prefers
the one with lesser depth, but only if that depth is greater than the
total latency of the instructions we've already scheduled -- otherwise
its latency would be hidden and there would be no stall.
Unfortunately it only tests the depth of one of the candidates. This can
lead to situations where the TopDepthReduce heuristic does not kick in,
but a lower priority heuristic chooses the other candidate, whose depth
*is* greater than the already scheduled latency, which causes a stall.
The fix is to apply the heuristic if the depth of *either* candidate is
greater than the already scheduled latency.
All this also applies to the BotHeightReduce heuristic in the bottom
zone.
Differential Revision: https://reviews.llvm.org/D72392
MBBs are not allowed to have non-terminator instructions after the first
terminator. Currently in some cases (see the modified test),
EmitSchedule can add DBG_VALUEs after the last terminator, for example
when referring a debug value that gets folded into a TCRETURN
instruction on ARM.
This patch updates EmitSchedule to move inserted DBG_VALUEs just before
the first terminator. I am not sure if there are terminators produce
values that can in turn be used by a DBG_VALUE. In that case, moving the
DBG_VALUE might result in referencing an undefined register. But in any
case, it seems like currently there is no way to insert a proper DBG_VALUEs
for such registers anyways.
Alternatively it might make sense to just remove those extra DBG_VALUES.
I am not too familiar with the details of debug info in the backend and
would appreciate any suggestions on how to address the issue in the best
possible way.
Reviewers: vsk, aprantl, jpaquette, efriedma, paquette
Reviewed By: aprantl
Differential Revision: https://reviews.llvm.org/D83561
I meant to do this in D83913, but missed it while updating the
feature list.
Interestingly I think this is disabling the postRA scheduler. But
it does match our default 64-bit behavior.
Reviewed By: echristo
Differential Revision: https://reviews.llvm.org/D83996
When SCC is dead, but VCC is required then replace s_and / s_andn2
with s_mov into VCC when mask value is 0 or -1.
Reviewed By: rampitec
Differential Revision: https://reviews.llvm.org/D83850
Accounting for the fact that Wasm function indices are 32-bit, but in wasm64 we want uniform 64-bit pointers.
Includes reloc types for 64-bit table indices.
Differential Revision: https://reviews.llvm.org/D83729
There was a lot of duplicate code here for checking the VT and
subtarget. Moving it into a helper avoids that.
It also fixes a bug that combineAdd reused Op0/Op1 after a call
to isHorizontalBinOp may have changed it. The new helper function
has its own local version of Op0/Op1 that aren't shared by other
code.
Fixes PR46455.
Reviewed By: spatel, bkramer
Differential Revision: https://reviews.llvm.org/D83971
