Add Apple M4 host detection #117530
Add Apple M4 host detection #117530
Conversation
|
✅ With the latest revision this PR passed the C/C++ code formatter. |
madsmtm
force-pushed
the
apple-m4-host-detection
branch
from
7944594 to
82bd606
Compare
madsmtm
force-pushed
the
apple-m4-host-detection
branch
2 times, most recently
from
6074a26 to
7b5d723
Compare
madsmtm
force-pushed
the
apple-m4-host-detection
branch
2 times, most recently
from
78fd5d4 to
a9966d9
Compare
llvm/lib/TargetParser/Host.cpp
Outdated
| case CPUFAMILY_ARM_EVEREST_SAWTOOTH: // A16 | ||
| case CPUFAMILY_ARM_IBIZA: // M3 | ||
| case CPUFAMILY_ARM_PALMA: // M3 Max | ||
| case CPUFAMILY_ARM_COLL: // A17 |
There was a problem hiding this comment.
apple-a17 is it's own thing, this shouldn't get merged with apple-m3. See the ProcessorAlias<> entries in AArch64Processors.td
There was a problem hiding this comment.
Hmm, why is apple-m3 aliased to apple-a16 and not apple-a17 in the first place?
There was a problem hiding this comment.
apple-m3 shouldn't be (and isn't) aliased with apple-a17
There was a problem hiding this comment.
But why not? The Wikipedia article (which of course isn't a reliable resource) lists Apple M3 as a variant of Apple A17 Pro?
In any case, apple-a16 and apple-a17 don't seem to have any functional differences in the source code (yet), so the point is perhaps a bit moot anyways?
There was a problem hiding this comment.
Wikipedia isn't a reliable source for this at all. Those pages often quote llvm's tablegen files as a source of truth on things, which sometimes aren't precisely correct for various reasons, both upstream-visible, and sometimes not-upstream-visible ones that I'm not going to explain here. We have them as separate processor definitions in AArch64Processors.td for reasons, and as lame of a justification that is, this code should respect that.... please.
There was a problem hiding this comment.
Fair, apologies if it came out wrong, my intention wasn't to contest your knowledge, I was mostly just curious ;)
There was a problem hiding this comment.
Yea, sorry, these things sometimes tread close to topics we can't really discuss externally.
There was a problem hiding this comment.
No worries!
I've updated the PR, please verify that you think it's correct?
madsmtm
force-pushed
the
apple-m4-host-detection
branch
from
a9966d9 to
6aa2846
Compare
Also add support for older ARM families (this is likely never going to get used, since only macOS is officially supported as host OS, but nice to have for completeness sake). Error handling (checking CPUFAMILY_UNKNOWN) is also included here. Finally, add links to extra documentation to make it easier for others to update this in the future. These values should be up to date with Xcode 16.2 beta 3.
madsmtm
force-pushed
the
apple-m4-host-detection
branch
from
6aa2846 to
b4efe4b
Compare
|
No problem! (I can't merge it myself, btw, if you're expecting that ;) ). |
Add Apple M4 host detection, which fixes rust-lang/rust#133414.
Also add support for older ARM families (this is likely never going to get used, since only macOS is officially supported as host OS, but nice to have for completeness sake). Error handling (checking
CPUFAMILY_UNKNOWN) is also included here.Finally, add links to extra documentation to make it easier for others to update this in the future.
NOTE: These values are taken from
mach/machine.hthe Xcode 16.2 SDK, and has been confirmed on an M4 Max in rust-lang/rust#133414 (comment).