arm64 osx: support byte sizes from lowering to codegen. #43024
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
sandreenko
added
the
area-CodeGen-coreclr
sandreenko
force-pushed
the
Arm64Apple-Jit-Part3
branch
from
3899ec9 to
14576d9
Compare
sandreenko
force-pushed
the
Arm64Apple-Jit-Part3
branch
from
14576d9 to
96cf3c5
Compare
sandreenko
force-pushed
the
Arm64Apple-Jit-Part3
branch
from
96cf3c5 to
4602f28
Compare
sandreenko
commented
Oct 30, 2020
| // Get the number of bytes that this argument is occupying on the stack. | ||
| // Get the number of bytes that this argument is occupying on the stack, | ||
| // including padding up to the target pointer size for platforms | ||
| // where a stack argument can't take less. |
There was a problem hiding this comment.
it could be confusing when we have arguments that require more than a pointer size, for example:
call(arguments to occupy registers, byte a, struct with 16 bytes alignment b)
in this example we will have such offsets on arm64 not-apple:
a - offset 0, GetStackByteSize() = 8,
b - offset 16, GetStackByteSize() = 16.
so the space between offset 1 and 8 is padding for a, the space between 8 and 16 is an alignment for b.
sandreenko
commented
Oct 30, 2020
| if (!layout->HasGCPtr() && (size & (XMM_REGSIZE_BYTES - 1)) != 0) | ||
| { | ||
| regMaskTP regMask = allRegs(TYP_INT); | ||
|
|
||
| #ifdef TARGET_X86 | ||
| if ((size % 2) != 0) |
There was a problem hiding this comment.
size was putArgStk->gtNumSlots * TARGET_POINTER_SIZE where TARGET_POINTER_SIZE % 2 == 0 so that condition was always false.
sandreenko
force-pushed
the
Arm64Apple-Jit-Part3
branch
from
4602f28 to
703636e
Compare
|
PTAL @dotnet/jit-contrib |
This was referenced Oct 30, 2020
sandreenko
force-pushed
the
Arm64Apple-Jit-Part3
branch
from
703636e to
a394784
Compare
|
ping @dotnet/jit-contrib |
CarolEidt
reviewed
Nov 11, 2020
|
/azp run runtime-coreclr outerloop |
|
Azure Pipelines successfully started running 1 pipeline(s). |
|
The PR was updated, the failures in stress modes are not relevant. |
CarolEidt
approved these changes
Nov 14, 2020
echesakov
approved these changes
Nov 14, 2020
sandreenko
force-pushed
the
Arm64Apple-Jit-Part3
branch
from
4e49956 to
5281a9d
Compare
ghost
locked as resolved and limited conversation to collaborators
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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.
Keep precise byte sizes and offsets for call arguments from lowering to codegen. It is a continuation of #42503.
Contributes to #41130.
No diffs.