[Xtensa] Implement volatile load/store. #110292
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
arsenm
reviewed
Sep 29, 2024
llvm/test/CodeGen/Xtensa/volatile.ll
Outdated
Comment on lines
42
to
48
| %0 = load volatile i8, ptr @x_i8, align 4 | ||
| store volatile i8 %0, ptr @y_i8, align 4 | ||
| %1 = load volatile i16, ptr @x_i16, align 4 | ||
| store volatile i16 %1, ptr @y_i16, align 4 | ||
| %2 = load volatile i32, ptr @x_i32, align 4 | ||
| store volatile i32 %2, ptr @y_i32, align 4 |
There was a problem hiding this comment.
Split each of these into its own test function?
Also test other types? i64, float, double, vectors, exotic types?
There was a problem hiding this comment.
Thank you for comment. I added tests with other types.
llvm/test/CodeGen/Xtensa/volatile.ll
Outdated
| ; CHECK-NEXT: ret | ||
|
|
||
| entry: | ||
| %0 = load volatile i8, ptr @x_i8, align 4 |
| case Xtensa::L16UI: | ||
| case Xtensa::L32I: { | ||
| const MachineMemOperand &MMO = **MI.memoperands_begin(); | ||
| if (MMO.isVolatile()) { |
There was a problem hiding this comment.
Comment why this is needed? It's also possible memory operands get lost during selection, should defend against missing MMO and assume volatile if it's missing
There was a problem hiding this comment.
I added handling of the case when MMO is missing.
| @@ -1104,10 +1104,24 @@ XtensaTargetLowering::emitSelectCC(MachineInstr &MI, | |||
| MachineBasicBlock *XtensaTargetLowering::EmitInstrWithCustomInserter( | |||
| MachineInstr &MI, MachineBasicBlock *MBB) const { | |||
| DebugLoc DL = MI.getDebugLoc(); | |||
| const TargetInstrInfo &TII = *Subtarget.getInstrInfo(); | |||
There was a problem hiding this comment.
Use XTensaInstrInfo. Also this might be a member already?
There was a problem hiding this comment.
I added using XtensaInstrInfo. It is not a member yet, I'm not sure, whether we need to add it?
arsenm
approved these changes
Oct 2, 2024
Add a memory wait "MEMW" instruction before volatile load/store operations, as implemented in GCC.
andreisfr
force-pushed
the
xtensa_volatile
branch
from
6bc11f4 to
1db96f6
Compare
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/73/builds/6535 Here is the relevant piece of the build log for the reference |
xgupta
pushed a commit
to xgupta/llvm-project
that referenced
this pull request
Oct 4, 2024
Add a memory wait "MEMW" instruction before volatile load/store operations, as implemented in GCC.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Add a memory wait "MEMW" instruction before volatile load/store operations, as implemented in GCC.