scalar_to_vector is difficult to make appear and test,
but I found one case where this makes an observable difference.
It fires more often than this in the test suite, but most of them
have no net result in the final code. This helps reduce regressions
in a future commit.
The lowering code was using the wrong chain value, which meant that
the 'smstart' after the call from streaming agnostic-ZA functions ->
non-streaming private-ZA functions was incorrectly removed from the DAG.
This extension adds eleven instructions to accelerate interrupt
servicing.
The current spec can be found at:
https://github.com/quic/riscv-unified-db/releases/latest
This patch adds assembler only support.
---------
Co-authored-by: Harsh Chandel <hchandel@qti.qualcomm.com>
When passing an instruction with a register mask, the machine copy
propagation pass was dropping the information about some copy
instructions which define a register which is preserved by the mask,
because that register overlaps a register which is partially clobbered
by it. This resulted in a miscompilation for AArch64, because this
caused a live copy to be considered dead.
The fix is to clobber register masks by finding the set of reg units
which is preserved by the mask, and clobbering all units not in that
set.
For targets with free domain moves, or AVX512 support, allow the use of VPMOVSX/ZX extension loads to reduce the load sizes.
I've limited this to extension to i32/i64 types as we're mostly interested in shuffle mask loading here, but we could include i16 types as well just as easily.
Inspired by a regression on #122485
This change implements import call optimization for AArch64 Windows
(equivalent to the undocumented MSVC `/d2ImportCallOptimization` flag).
Import call optimization adds additional data to the binary which can be
used by the Windows kernel loader to rewrite indirect calls to imported
functions as direct calls. It uses the same [Dynamic Value Relocation
Table mechanism that was leveraged on x64 to implement
`/d2GuardRetpoline`](https://techcommunity.microsoft.com/blog/windowsosplatform/mitigating-spectre-variant-2-with-retpoline-on-windows/295618).
The change to the obj file is to add a new `.impcall` section with the
following layout:
```cpp
// Per section that contains calls to imported functions:
// uint32_t SectionSize: Size in bytes for information in this section.
// uint32_t Section Number
// Per call to imported function in section:
// uint32_t Kind: the kind of imported function.
// uint32_t BranchOffset: the offset of the branch instruction in its
// parent section.
// uint32_t TargetSymbolId: the symbol id of the called function.
```
NOTE: If the import call optimization feature is enabled, then the
`.impcall` section must be emitted, even if there are no calls to
imported functions.
The implementation is split across a few parts of LLVM:
* During AArch64 instruction selection, the `GlobalValue` for each call
to a global is recorded into the Extra Information for that node.
* During lowering to machine instructions, the called global value for
each call is noted in its containing `MachineFunction`.
* During AArch64 asm printing, if the import call optimization feature
is enabled:
- A (new) `.impcall` directive is emitted for each call to an imported
function.
- The `.impcall` section is emitted with its magic header (but is not
filled in).
* During COFF object writing, the `.impcall` section is filled in based
on each `.impcall` directive that were encountered.
The `.impcall` section can only be filled in when we are writing the
COFF object as it requires the actual section numbers, which are only
assigned at that point (i.e., they don't exist during asm printing).
I had tried to avoid using the Extra Information during instruction
selection and instead implement this either purely during asm printing
or in a `MachineFunctionPass` (as suggested in [on the
forums](https://discourse.llvm.org/t/design-gathering-locations-of-instructions-to-emit-into-a-section/83729/3))
but this was not possible due to how loading and calling an imported
function works on AArch64. Specifically, they are emitted as `ADRP` +
`LDR` (to load the symbol) then a `BR` (to do the call), so at the point
when we have machine instructions, we would have to work backwards
through the instructions to discover what is being called. An initial
prototype did work by inspecting instructions; however, it didn't
correctly handle the case where the same function was called twice in a
row, which caused LLVM to elide the `ADRP` + `LDR` and reuse the
previously loaded address. Worse than that, sometimes for the
double-call case LLVM decided to spill the loaded address to the stack
and then reload it before making the second call. So, instead of trying
to implement logic to discover where the value in a register came from,
I instead recorded the symbol being called at the last place where it
was easy to do: instruction selection.
We want special handing for IGLP instructions in the scheduler but they
should still be treated like they have side effects by other passes. Add
a target hook to the ScheduleDAGInstrs DAG builder so that we have more
control over this.
With range and undef metadata on a call we can have vector AssertZExt
generated on a target with no vector operations. The AssertZExt needs to
scalarize to a normal `AssertZext tin, ValueType`. I have added
AssertSext too, although I do not have a test case.
Fixes#110374
If we have a CSEL instruction that depends on the flags set by a
(SUBS x c) instruction and the true and/or false expression is
(add (add x y) -c), we can reassociate the latter expression to
(add (SUBS x c) y) and save one instruction.
Proof for the basic transformation: https://alive2.llvm.org/ce/z/-337Pb
We can extend this transformation for slightly different constants. For
example, if we have (add (add x y) -(c-1)) and a the comparison x <u c,
we can transform the comparison to x <=u c-1 to eliminate the comparison
instruction, too. Similarly, we can transform (x == 0) to (x <u 1).
Proofs for the transformations that alter the constants:
https://alive2.llvm.org/ce/z/3nVqgRFixes#119606.
Under certain circumstances, lowering of other instructions can result in computeKnownBits being able to detect a constant that it couldn't previously.
Fixes#122580
Add a prologue to the kernel entry to handle cases where code designed
for kernarg preloading is executed on hardware equipped with
incompatible firmware. If hardware has compatible firmware the 256 bytes
at the start of the kernel entry will be skipped. This skipping is done
automatically by hardware that supports the feature.
A pass is added which is intended to be run at the very end of the
pipeline to avoid any optimizations that would assume the prologue is a
real predecessor block to the actual code start. In reality we have two
possible entry points for the function. 1. The optimized path that
supports kernarg preloading which begins at an offset of 256 bytes. 2.
The backwards compatible entry point which starts at offset 0.
## Changes
- Delete DirectX length intrinsic
- Delete HLSL length lang builtin
- Implement length algorithm entirely in the header.
## History
- In the past if an HLSL intrinsic lowered to either a spirv op code or
a DXIL opcode we represented it with intrinsics
## Why we are moving away?
- To make HLSL apis more portable the team decided that it makes sense
for some intrinsics to be defined only in the header.
- Since there tends to be more SPIRV opcodes than DXIL opcodes the plan
is to support SPIRV opcodes either with target specific builtins or via
pattern matching.
Use the probe loop structure to allocate vector code in the stack as
well. We add the pseudo instruction RISCV::PROBED_STACKALLOC_RVV to
differentiate from the normal loop.
PR #96083 added intrinsics for async copy of 'tensor' data
using TMA. Following a similar design, this PR adds intrinsics
for async copy of bulk data (non-tensor variants) through TMA.
* These intrinsics optionally support multicast and cache_hints,
as indicated by the boolean arguments at the end of the intrinsics.
* The backend looks through these flag arguments and lowers to the
appropriate PTX instructions.
* Lit tests are added for all combinations of these intrinsics in
cp-async-bulk.ll.
* The generated PTX is verified with a 12.3 ptxas executable.
* Added docs for these intrinsics in NVPTXUsage.rst file.
PTX Spec reference:
https://docs.nvidia.com/cuda/parallel-thread-execution/#data-movement-and-conversion-instructions-cp-async-bulk
Signed-off-by: Durgadoss R <durgadossr@nvidia.com>
HexagonISD::PFALSE and PTRUE patterns do not form independently in
general as they are treated like operands of all 0s or all 1s. Eg: i32 =
transfer HEXAGONISD::PFALSE.
In this case, v8i1 = HEXAGONISD::PFALSE is formed independently without
accompanying opcode.
This patch adds a pattern to transfer all 0s or all 1s to a scalar
register and then use that register and this PFALSE/PTRUE opcode to
transfer to a predicate register like v8i1.
This takes inspiration from AArch64 which does the same thing to assist
with zip/trn/etc.. Doing this recursion unconditionally when the mask
allows is slightly questionable, but seems to work out okay in practice.
As a bit of context, it's helpful to realize that we have existing logic
in both DAGCombine and InstCombine which mutates the element width of in
an analogous manner. However, that code has two restriction which
prevent it from handling the motivating cases here. First, it only
triggers if there is a bitcast involving a different element type.
Second, the matcher used considers a partially undef wide element to be
a non-match. I considered trying to relax those assumptions, but the
information loss for undef in mid-level opt seemed more likely to open a
can of worms than I wanted.
If we sink the and in and(load), CGP can hoist is back again to the
load, getting into an infinite loop. This prevents sinking the and in
this case.
Fixes#122074
Replace `uaddlv` with `addv` for popcount operation as it is simpler
operation.
On certain platforms like Cortex-A510, `addv` has a latency of 3 cycles
whereas `uaddlv` has a latency of 4 cycles
GCC generates `addv` as well:
https://godbolt.org/z/MnYG9jcEo
This replaces the existing `-wasm-enable-exnref` with
`-wasm-use-legacy-eh` option, in an effort to make the new standardized
exnref proposal the 'default' state and the legacy proposal needs to be
separately enabled an option. But given that most users haven't switched
to the new proposal and major web browsers haven't turned it on by
default, this `-wasm-use-legacy-eh` is turned on by default, so nothing
will change for now for the functionality perspective.
This also removes the restriction that `-wasm-enable-exnref` be only
used with `-wasm-enable-eh` because this option is enabled by default.
This option does not have any effect when `-wasm-enable-eh` is not used.
The trampoline size is 36 bytes on AArch64. The runtime function
__trampoline_setup aborts as it expects the trampoline size of at least 36
bytes, and the size passed is 20 bytes. Fix the inconsistency in
AArch64TargetLowering::LowerINIT_TRAMPOLINE.
If the pointer to be checked is statically known to be zero, the tag
check will always pass since:
1) the tag is zero
2) shadow memory for address 0 is initialized to 0 and never updated.
We can therefore elide the tag check.
We perform the elision in two places:
1) the HWASan pass
2) when lowering the CHECK_MEMACCESS intrinsic. Conceivably, the HWASan
pass may encounter a "cannot currently statically prove to be null"
pointer (and is therefore unable to omit the intrinsic) that later
optimization passes convert into a statically known-null pointer. As a
last line of defense, we perform elision here too.
This also updates the tests from
https://github.com/llvm/llvm-project/pull/122186
