`llvm.trap` will be convert as unreachable which is terminator.
Instruction after terminator will cause validation failed.
This PR introduces a pass to clean instruction after terminator.
Fixes: https://github.com/llvm/llvm-project/issues/68770
Reapply: #90207
`llvm.trap` will be convert as `unreachable` which is terminator.
Instruction after terminator will cause validation failed.
This PR introduces a pass to clean instruction after terminator.
Fixes: #68770.
Currently we check `Subtarget->hasReferenceTypes()` to decide whether to
run `RefTypeMem2Local` pass:
6133878227/llvm/lib/Target/WebAssembly/WebAssemblyTargetMachine.cpp (L491-L495)
This works fine when `-mattr=+reference-types` is given in the command
line (of `llc` or of `wasm-ld` in case of LTO). This also works fine if
the backend is called by Clang, because Clang's feature set will be
passed to the backend when creating a `TargetMachine`:
ac791888bb/clang/lib/CodeGen/BackendUtil.cpp (L549-L550)ac791888bb/clang/lib/CodeGen/BackendUtil.cpp (L561-L562)
But if the backend compilation is called by `llc`, a `TargetMachine` is
created here:
bf1ad1d267/llvm/tools/llc/llc.cpp (L554-L555)
And if the backend is called by `wasm-ld`'s LTO, a `TargetMachine` is
created here:
ac791888bb/llvm/lib/LTO/LTOBackend.cpp (L513)
At this point, in the both places, the created `TargetMachine` only has
access to target features given by the command line with `-mattr=` and
doesn't have access to bitcode functions' `target-features` attribute.
We later gather the target features used by functions and store that
info in the `TargetMachine` in `CoalesceFeaturesAndStripAtomics`,
ac791888bb/llvm/lib/Target/WebAssembly/WebAssemblyTargetMachine.cpp (L202-L206)
but this runs in the pass pipeline driven by the pass manager, so this
has not run by the time we check `Subtarget->hasReferenceTypes()` in
`WebAssemblyPassConfig::addISelPrepare`. So currently `RefTypeMem2Local`
would not run on those functions with
`"target-features"="+reference-types"` attributes if the backend is
called by `llc` or `wasm-ld`.
So this makes `RefTypeMem2Local` pass run unconditionally, and checks
`target-featurs` function attribute to decide whether to run the pass on
each function. This allows the pass to run with `wasm-ld` + LTO and
`llc`, even if `-mattr=+reference-types` is not explicitly given in the
command line again, as long as `+reference-types` is in the function's
`target-features` attribute.
This also covers the case we give the target features by the command
line like `llc -mattr=+reference-types` and not in the bitcode
function's attribute, because attributes given in the command line will
be stored in the function's attributes anyway:
bd28889732/llvm/lib/CodeGen/CommandFlags.cpp (L673-L674)bd28889732/llvm/lib/CodeGen/CommandFlags.cpp (L732-L733)
With this PR,
- `lto0.test_externref_emjs`
- `thinlto0.test_externref_emjs`,
- `lto0.test_externref_emjs_dynlink`,
- `thinlto0.test_externref_emjs_dynlnk`
pass. These currently fail but don't get checked in the CI. I think they
used to pass but started to fail after #83196, because we used to run
mem2reg even with `-O0` before that.
(`ltoN` (N > 0) tests are not affected because they run mem2reg anyway
so they don't need `RefTypeMem2Local`)
Multivalue feature of WebAssembly has been standardized for several
years now. I think it makes sense to be able to enable it in the feature
section by default for our clang/llvm-produced binaries so that the
multivalue feature can be used as necessary when necessary within our
toolchain and also when running other optimizers (e.g. wasm-opt) after
the LLVM code generation.
But some WebAssembly toolchains, such as Emscripten, do not provide both
mulvalue-returning and not-multivalue-returning versions of libraries.
Also allowing the uses of multivalue in the features section does not
necessarily mean we generate them whenever we can to the fullest, which
is a different code generation / optimization option.
So this makes the lowering of multivalue returns conditional on the use
of 'experimental-mv' target ABI. This ABI is turned off by default and
turned on by passing `-Xclang -target-abi -Xclang experimental-mv` to
`clang`, or `-target-abi experimental-mv` to `clang -cc1` or `llc`.
But the purpose of this PR is not tying the multivalue lowering to this
specific 'experimental-mv'. 'experimental-mv' is just one multivalue ABI
we currently have, and it is still experimental, meaning it is not very
well optimized or tuned for performance. (e.g. it does not have the
limitation of the max number of multivalue-lowered values, which can be
detrimental to performance.) We may change the name of this ABI, or
improve it, or add a new multivalue ABI in the future. Also I heard that
WASI is planning to add their multivalue ABI soon. So the plan is,
whenever any one of multivalue ABIs is enabled, we enable the lowering
of multivalue returns in the backend. We currently have only
'experimental-mv' in the repo so we only check for that in this PR.
Related past discussions:
#82714https://github.com/WebAssembly/tool-conventions/pull/223#issuecomment-2008298652
Remove `llvm.threadlocal.address` intrinsic usage when disabling TLS.
This fixes errors revealed by the stricter IR verification introduced in
PR #87841.
When reference-types feature is enabled, forcing mem2reg unconditionally
even in `-O0` has some problems described in #81575. This uses
RefTypeMem2Local pass added in #81965 instead. This also removes
`IsForced` parameter added in
890146b192
given that we don't need it anymore.
This may still hurt debug info related to reference type variables a
little during the backend transformation given that they are not stored
in memory anymore, but reference type variables are presumably rare and
it would be still a lot less damage than forcing mem2reg on the whole
program. Also this fixes the EH problem described in #81575.
Fixes#81575.
This reverts commit 6e6bf9f81756ba6655b4eea8dc45469a47f89b39.
It turned out the multivalue feature had active outside users and it
could cause some disruptions to them, so I'd like to investigate more
about the workarounds before doing this.
This adds `WebAssemblyRefTypeMem2Local` pass, which changes the address
spaces of reference type `alloca`s to `addrspace(1)`. This in turn
changes the address spaces of all `load` and `store` instructions that
use the `alloca`s.
`addrspace(1)` is `WASM_ADDRESS_SPACE_VAR`, and loads and stores to this
address space become `local.get`s and `local.set`s, thanks to the Wasm
local IR support added in
82f92e35c6.
In a follow-up PR, I am planning to replace the usage of mem2reg pass
with this to solve the reference type `alloca` problems described in
#81575.
We plan to enable multivalue in the features section soon (#80923) for
other reasons, such as the feature having been standardized for many
years and other features being developed (e.g. EH) depending on it. This
is separate from enabling Clang experimental multivalue ABI (`-Xclang
-target-abi -Xclang experimental-mv`), but it turned out we generate
some multivalue code in the backend as well if it is enabled in the
features section.
Given that our backend multivalue generation still has not been much
used nor tested, and enabling the feature in the features section can be
a separate decision from how much multialue (including none) we decide
to generate for now, I'd like to temporarily disable the actual
generation of multivalue in our backend. To do that, this adds an
internal flag `-wasm-emit-multivalue` that defaults to false. All our
existing multivalue tests can use this to test multivalue code. This
flag can be removed later when we are confident the multivalue
generation is well tested.
In `CoalesceFeaturesAndStripAtomics`, feature string is converted to FeatureBitset and back to feature string. It will lose information about explicit diasbled features.
Currently for any wasm target, llvm will make a pass that removes
irreducible control flow. (See
[here](https://llvm.org/doxygen/WebAssemblyFixIrreducibleControlFlow_8cpp.html)).
This can result in O(NumBlocks * NumNestedLoops * NumIrreducibleLoops +
NumLoops * NumLoops) build time, which has resulted in exceedingly long
build times when testing. This PR introduces a hidden flag to skip this
pass, which brings some of our build times down from 30 minutes to ~6
seconds.
In the WebAssembly back end, the TrapUnreachable option is currently
load-bearing for correctness, inserting wasm `unreachable` instructions
where needed to create valid wasm. There is another option,
NoTrapAfterNoreturn, that removes some of those traps and causes
incorrect wasm to be emitted.
This turns off `NoTrapAfterNoreturn` for the Wasm backend and adds new
tests.
This will make it easy for callers to see issues with and fix up calls
to createTargetMachine after a future change to the params of
TargetMachine.
This matches other nearby enums.
For downstream users, this should be a fairly straightforward
replacement,
e.g. s/CodeGenOpt::Aggressive/CodeGenOptLevel::Aggressive
or s/CGFT_/CodeGenFileType::
Move WebAssemblyUtilities from Utils to the CodeGen library. It
primarily deals in MIR layer types, so it really lives in the CodeGen
library.
Move a variety of other things around to try create better separation.
See issue #64166 for more info on layering.
Move llvm/include/CodeGen/WasmAddressSpaces.h back to
llvm/lib/Target/WebAssembly/Utils.
Differential Revision: https://reviews.llvm.org/D156472
This patch introduces a new type __externref_t that denotes a WebAssembly opaque
reference type. It also implements builtin __builtin_wasm_ref_null_extern(),
that returns a null value of __externref_t. This lays the ground work
for further builtins and reference types.
Reviewed By: aaron.ballman
Differential Revision: https://reviews.llvm.org/D122215
This patch introduces a new type __externref_t that denotes a WebAssembly opaque
reference type. It also implements builtin __builtin_wasm_ref_null_extern(),
that returns a null value of __externref_t. This lays the ground work
for further builtins and reference types.
Differential Revision: https://reviews.llvm.org/D122215
Currently default simd alignment is specified by Clang specific TargetInfo
class. This class cannot be reused for LLVM Flang. If we move the default
alignment field into TargetMachine class then we can create TargetMachine
objects and query them to find SIMD alignment.
Scope of changes:
1) Added information about maximal allowed SIMD alignment to TargetMachine
classes.
2) Removed getSimdDefaultAlign function from Clang TargetInfo class.
3) Refactored createTargetMachine function.
Reviewed By: jsjodin
Differential Revision: https://reviews.llvm.org/D138496
This adds the missing handling for defs for target index operands, as is
already done for registers.
There are two kinds of target indices: local indices and stack operands.
- Locals are something similar to registers in Wasm-land. For local
indices, we can check for local-defining instructions (`local.set` or
`local.tee`).
- Wasm is a stack machine, so we have values in certain Wasm value stack
location, which change when Wasm instructions produce or consume
values. So basically any value-producing instrucion, i.e., instruction
with defs, can change values in the Wasm stack. But I think we don't
need to worry about this here, because `WebAssemblyDebugFixup`, which
runs right before this analysis, makes sure to insert terminating
`DBG_VALUE $noreg` instructions whenever a stack value gets popped.
After `WebAssemblyDebugFixup`, there shouldn't be any `DBG_VALUE`s for
stack operands that don't have a terminating `DBG_VALUE $noreg` within
the same BB.
So this CL only works on `DBG_VALUE`s for locals. When we encounter a
`local.set` or `local.tee` instructions, we delete `DBG_VALUE`s for
those target index locations from the open range set, so they will not
be availble in `OutLocs`. For example,
```
bb.0:
successors: %bb.1
DBG_VALUE target-index(wasm-local) + 2, $noreg, "var", ...
...
local.set 2 ...
bb.1:
; predecessors: %bb.0
; We shouldn't add `DBG_VALUE target (wasm-local) + 2 here because
; it was killed by 'local.set' in bb.0
```
After disabling register coalescing at -O1, the average PC ranges
covered for Emscripten core benchmarks is currently 20.6% in the LLVM
tot. After applying D138943 and this CL, the coverage goes up to 57%.
This also enables LiveDebugValues analysis in the Wasm pipeline by
default.
Reviewed By: dschuff, jmorse
Differential Revision: https://reviews.llvm.org/D140373
Follow up to the series:
1. https://reviews.llvm.org/D140161
2. https://reviews.llvm.org/D140349
3. https://reviews.llvm.org/D140331
4. https://reviews.llvm.org/D140323
Completes the work from the previous two for remaining targets.
This creates the following named passes that can be run via
`llc -{start|stop}-{before|after}`:
- arc-isel
- arm-isel
- avr-isel
- bpf-isel
- csky-isel
- hexagon-isel
- lanai-isel
- loongarch-isel
- m68k-isel
- msp430-isel
- mips-isel
- nvptx-isel
- ppc-codegen
- riscv-isel
- sparc-isel
- systemz-isel
- ve-isel
- wasm-isel
- xcore-isel
A nice way to write tests for SelectionDAGISel might be to use a RUN:
line like:
llc -mtriple=<triple> -start-before=<arch>-isel -stop-after=finalize-isel -o -
Fixes: https://github.com/llvm/llvm-project/issues/59538
Reviewed By: asb, zixuan-wu
Differential Revision: https://reviews.llvm.org/D140364
This fixes what I consider to be an API flaw I've tripped over
multiple times. The point this is constructed isn't well defined, so
depending on where this is first called, you can conclude different
information based on the MachineFunction. For example, the AMDGPU
implementation inspected the MachineFrameInfo on construction for the
stack objects and if the frame has calls. This kind of worked in
SelectionDAG which visited all allocas up front, but broke in
GlobalISel which hasn't visited any of the IR when arguments are
lowered.
I've run into similar problems before with the MIR parser and trying
to make use of other MachineFunction fields, so I think it's best to
just categorically disallow dependency on the MachineFunction state in
the constructor and to always construct this at the same time as the
MachineFunction itself.
A missing feature I still could use is a way to access an custom
analysis pass on the IR here.
These passes were lying around but weren't initialized, so they weren't
showing up in -print-after-all.
Differential Revision: https://reviews.llvm.org/D139440
This reverts commit 122efef8ee9be57055d204d52c38700fe933c033.
- Patch fixed to not reuse definitions from predecessors in EH landing pads.
- Late review suggestions (by MaskRay) have been addressed.
- M68k/pipeline.ll test updated.
- Init captures added in processBlock() to avoid capturing structured bindings.
- RISCV has this disabled for now.
Original commit message:
A new pass MachineLateInstrsCleanup is added to be run after PEI.
This is a simple pass that removes redundant and identical instructions
whenever found by scanning the MF once while keeping track of register
definitions in a map. These instructions are typically immediate loads
resulting from rematerialization, and address loads emitted by target in
eliminateFrameInde().
This is enabled by default, but a target could easily disable it by means of
'disablePass(&MachineLateInstrsCleanupID);'.
This late cleanup is naturally not "optimal" in removing instructions as it
is done by looking at phys-regs, but still quite effective. It would be
desirable to improve other parts of CodeGen and avoid these redundant
instructions in the first place, but there are no ideas for this yet.
Differential Revision: https://reviews.llvm.org/D123394
Reviewed By: RKSimon, foad, craig.topper, arsenm, asb
Init captures added in processBlock() to avoid capturing structured bindings,
which caused the build problems (with clang).
RISCV has this disabled for now until problems relating to post RA pseudo
expansions are resolved.
A new pass MachineLateInstrsCleanup is added to be run after PEI.
This is a simple pass that removes redundant and identical instructions
whenever found by scanning the MF once while keeping track of register
definitions in a map. These instructions are typically immediate loads
resulting from rematerialization, and address loads emitted by target in
eliminateFrameInde().
This is enabled by default, but a target could easily disable it by means of
'disablePass(&MachineLateInstrsCleanupID);'.
This late cleanup is naturally not "optimal" in removing instructions as it
is done by looking at phys-regs, but still quite effective. It would be
desirable to improve other parts of CodeGen and avoid these redundant
instructions in the first place, but there are no ideas for this yet.
Differential Revision: https://reviews.llvm.org/D123394
Reviewed By: RKSimon, foad, craig.topper, arsenm, asb
This disables `RegisterCoalescer` pass at -O1, which currently runs for
all levels except for -O0, as a part of common optimization pipeline.
`RegisterCoalescer` pass degrades Wasm debug info quality by a
significant margin. When I use `LiveDebugValue` analysis, disabling this
increases the average PC ranges covered by 15% on Emscripten core
benchmarks (52% -> 66.8%). (Our code is currently not using
`LiveDebugValues` analysis at the moment, and the experiment was done on
a local setting that enabled it. I'm planning to upstream it soon.)
In Emscripten core benchmarks, disabling this at -O1 causes +4.5% in
code size and +1% in the number of locals. The number of globals stays
the same. I believe this tradeoff is acceptable given that -O1 is not
usually used in production builds and is often used for debugging when
the application size is very large.
The plan is to investigate and fix what's causing the degradation in
that pass, but for now disabling it seems like a low-hanging quick fix.
Reviewed By: dschuff
Differential Revision: https://reviews.llvm.org/D138455
The MachineFunctionInfo here is a bit awkward because
WasmEHInfo is in the MachineFunction but handled from
the target code. Either everything should move into WebAssembly
or into the MachineFunction for MIR serialization.
With f3b4f99007cdcb3306484c9a39d31addc20aaa69, the exclusive source of
truth for whether threads are supported is the -matomics flag.
Accordingly, strip TLS flags when -matomic is not specified, even if
bulk-memory is specified and it would theoretically be supportable.
This allows the backend to compile TLS variables when -mbulk-memory is
enabled but threads are not enabled.
Differential Revision: https://reviews.llvm.org/D125730
FixIrreducibleControlFlow pass adds dispatch blocks with a `br_table`
that has multiple predecessors and successors, because it serves as
something like a traffic hub for BBs. As a result of this, there can be
register uses that are not dominated by a def in every path from the
entry block. For example, suppose register %a is defined in BB1 and used
in BB2, and there is a single path from BB1 and BB2:
```
BB1 -> ... -> BB2
```
After FixIrreducibleControlFlow runs, there can be a dispatch block
between these two BBs:
```
BB1 -> ... -> Dispatch -> ... -> BB2
```
And this dispatch block has multiple predecessors, now
there is a path to BB2 that does not first visit BB1, and in that path
%a is not dominated by a def anymore.
To fix this problem, we have been adding `IMPLICIT_DEF`s to all
registers in PrepareForLiveInternals pass, and then remove unnecessary
ones in OptimizeLiveIntervals pass after computing `LiveIntervals`. But
FixIrreducibleControlFlow pass itself ends up violating register use-def
relationship, resulting in invalid code. This was OK so far because
MIR verifier apparently didn't check this in validation. But @arsenm
fixed this and it caught this bug in validation
(https://github.com/llvm/llvm-project/issues/55249).
This CL moves the `IMPLICIT_DEF` adding routine from
PrepareForLiveInternals to FixIrreducibleControlFlow. We only run it
when FixIrreducibleControlFlow changes the code. And then
PrepareForLiveInternals doesn't do anything other than setting
`TracksLiveness` property, which is a prerequisite for running
`LiveIntervals` analysis, which is required by the next pass
OptimizeLiveIntervals.
But in our backend we don't seem to do anything that invalidates this up
until OptimizeLiveIntervals, and I'm not sure why we are calling
`invalidateLiveness` in ReplacePhysRegs pass, because what that pass
does is to replace physical registers with virtual ones 1-to-1. I
deleted the `invalidateLiveness` call there and we don't need to set
that flag explicitly, which obviates all the need for
PrepareForLiveInternals.
(By the way, This 'Liveness' here is different from `LiveIntervals`
analysis. Setting this only means BBs' live-in info is correct, all uses
are dominated by defs, `kill` flag is conservatively correct, which
means if there is a `kill` flag set it should be the last use. See
2a0837aab1/llvm/include/llvm/CodeGen/MachineFunction.h (L125-L134)
for details.)
So this CL removes PrepareForLiveInternals pass altogether. Something
similar to this was attempted by D56091 long ago but that came short of
actually removing the pass, and I couldn't land it because
FixIrreducibleControlFlow violated use-def relationship, which this CL
fixes.
This doesn't change output in any meaningful way. All test changes
except `irreducible-cfg.mir` are register numbering.
Also this will likely to reduce compilation time, because we have been
adding `IMPLICIT_DEF` for all registers every time `-O2` is given, but
now we do that only when there is irreducible control flow, which is
rare.
Fixes https://github.com/llvm/llvm-project/issues/55249.
Reviewed By: dschuff, kripken
Differential Revision: https://reviews.llvm.org/D125515
static_cast is a little safer here since the compiler will
ensure we're casting to a class derived from
yaml::MachineFunctionInfo.
I believe this first appeared on AMDGPU and was copied to the
other two targets.
Spotted when it was being copied to RISCV in D123178.
Differential Revision: https://reviews.llvm.org/D123260
For MachO, lower `@llvm.global_dtors` into `@llvm_global_ctors` with
`__cxa_atexit` calls to avoid emitting the deprecated `__mod_term_func`.
Reuse the existing `WebAssemblyLowerGlobalDtors.cpp` to accomplish this.
Enable fallback to the old behavior via Clang driver flag
(`-fregister-global-dtors-with-atexit`) or llc / code generation flag
(`-lower-global-dtors-via-cxa-atexit`). This escape hatch will be
removed in the future.
Differential Revision: https://reviews.llvm.org/D121736
For MachO, lower `@llvm.global_dtors` into `@llvm_global_ctors` with
`__cxa_atexit` calls to avoid emitting the deprecated `__mod_term_func`.
Reuse the existing `WebAssemblyLowerGlobalDtors.cpp` to accomplish this.
Enable fallback to the old behavior via Clang driver flag
(`-fregister-global-dtors-with-atexit`) or llc / code generation flag
(`-lower-global-dtors-via-cxa-atexit`). This escape hatch will be
removed in the future.
Differential Revision: https://reviews.llvm.org/D121736
For MachO, lower `@llvm.global_dtors` into `@llvm_global_ctors` with
`__cxa_atexit` calls to avoid emitting the deprecated `__mod_term_func`.
Reuse the existing `WebAssemblyLowerGlobalDtors.cpp` to accomplish this.
Enable fallback to the old behavior via Clang driver flag
(`-fregister-global-dtors-with-atexit`) or llc / code generation flag
(`-lower-global-dtors-via-cxa-atexit`). This escape hatch will be
removed in the future.
Differential Revision: https://reviews.llvm.org/D121327
