Some small overall improvements to stdarch-gen-arm#2099
Some small overall improvements to stdarch-gen-arm#2099folkertdev merged 3 commits intorust-lang:mainfrom
Conversation
|
r? @folkertdev rustbot has assigned @folkertdev. Use Why was this reviewer chosen?The reviewer was selected based on:
|
sayantn
force-pushed
the
gen-arm-improvements
branch
from
ea24450 to
2613dd9
Compare
sayantn
force-pushed
the
gen-arm-improvements
branch
2 times, most recently
from
c1eb587 to
873463e
Compare
This comment has been minimized.
This comment has been minimized.
There was a problem hiding this comment.
I authored these changes so unsurprisingly they look good to me. The final commit's changes might only make sense in the context of the changes in #2063.
| not-big-endian: ¬-big-endian | ||
| FnCall: [not, ['target_endian = "big"']] |
There was a problem hiding this comment.
is this better than target_endian = "small"?
There was a problem hiding this comment.
Probably not - just reflects how I was thinking at the time - changed to target_endian = "small", which required some changes to the yaml anchors just because I don't think I can make an anchor just for a string.
davidtwco
force-pushed
the
gen-arm-improvements
branch
from
873463e to
ac1c267
Compare
This comment has been minimized.
This comment has been minimized.
davidtwco
force-pushed
the
gen-arm-improvements
branch
2 times, most recently
from
940d659 to
8e474e9
Compare
This comment has been minimized.
This comment has been minimized.
Earlier patches added `big_endian_inverse` and used `transmute` instead of using `vreinterpret` due to an LLVM bug that has since been resolved
Instead of duplicating the stability attribute, re-use the `neon-stable` anchor through
Some intrinsics optimise to different instructions on big endian which leads to `assert_instr` failing
sayantn
force-pushed
the
gen-arm-improvements
branch
from
8e474e9 to
8075df7
Compare
|
This PR was rebased onto a different main commit. Here's a range-diff highlighting what actually changed. Rebasing is a normal part of keeping PRs up to date, so no action is needed—this note is just to help reviewers. |
4 participants
All the edits are by @davidtwco, I have just cherry-picked it from his branch and putting the PR up.