Conversation
ktf
force-pushed
the
pr15818
branch
2 times, most recently
from
September 19, 2026 18:03
15fab9a to
94f37bc
Compare
Collaborator
|
Error while checking build/O2/fullCI_slc9 for 15fab9a at 2026-09-19 20:04: Full log here. |
Collaborator
|
Error while checking build/O2/fullCI_slc9 for 94f37bc at 2026-09-19 20:06: Full log here. |
Member
Author
|
@davidrohr first bunch. I need to disentangle the rest from the migration to I have also opened FB24852288 with the Apple support. I think I still have some bug report open from 2008, though, so I wouldn't hold my breath. I would call it a victory they added generic address space "only" a year later. |
ktf
changed the title
Collaborator
|
Error while checking build/O2/fullCI_slc9 for e59b7ff at 2026-09-20 01:26: Full log here. |
Collaborator
If there are genuine formatting issues, sure, please fix them. But I think there is also some bogus clang-format behavior in some cases with the GPU macros. For double: I'd prefer to use a typedef for double instead of messing with all the code and replacing double with GPUdouble. Generally, I'd like to leave the code standard c++ where possible. |
Same treatment the TPC constants already had. MSL requires every variable at program scope to name an address space, and diagnoses it at the declaration, so a header full of plain constexpr breaks any device translation unit that merely includes it -- whether or not the constants are used. GPUglobalconstexpr() expands to constexpr everywhere except Metal, where it adds constant. Class-scope static members need it too: MSL counts them as program scope. constexpr functions do not, and are left alone. The six vDrift and ExB calibration defaults are double, which does not exist in MSL at all. They are host-only -- nothing under GPU/ refers to them -- so they are now compiled out of device code rather than converted, which would have changed their precision on CUDA and HIP. Preprocessed output is unchanged for host, CUDA, HIP and cling; device code sees the same constants minus those six doubles. Together this takes the TRD headers from 144 errors to 0 in a Metal translation unit.
MSL requires every variable at program scope to name an address space and diagnoses it at the declaration, so these headers broke any device translation unit that included them. GPUglobalconstexpr() expands to constexpr everywhere except Metal, where it adds constant. Preprocessed declarations are unchanged for host, CUDA, HIP and cling.
MSL requires every variable at program scope, class-scope statics included, to name an address space. GPUglobalconstexpr() expands to constexpr everywhere except Metal, so this is a no-op for host, CUDA, HIP, OpenCL and cling.
MSL requires every variable at program scope, class-scope statics included, to name an address space. GPUglobalconstexpr() expands to constexpr everywhere except Metal, so this is a no-op for host, CUDA, HIP, OpenCL and cling.
…ress space MSL requires every variable at program scope, class-scope statics included, to name an address space. GPUglobalconstexpr() expands to constexpr everywhere except Metal, so this is a no-op for host, CUDA, HIP, OpenCL and cling.
MSL requires every variable at program scope, class-scope statics included, to name an address space. GPUglobalconstexpr() expands to constexpr everywhere except Metal, so this is a no-op for host, CUDA, HIP, OpenCL and cling.
…ress space MSL requires every variable at program scope, class-scope statics included, to name an address space. GPUglobalconstexpr() expands to constexpr everywhere except Metal, so this is a no-op for host, CUDA, HIP, OpenCL and cling.
MSL requires every variable at program scope, class-scope statics included, to name an address space. GPUglobalconstexpr() expands to constexpr everywhere except Metal, so this is a no-op for host, CUDA, HIP, OpenCL and cling.
… space MSL requires every variable at program scope, class-scope statics included, to name an address space. GPUglobalconstexpr() expands to constexpr everywhere except Metal, so this is a no-op for host, CUDA, HIP, OpenCL and cling.
MSL requires every variable at program scope, class-scope statics included, to name an address space. GPUglobalconstexpr() expands to constexpr everywhere except Metal, so this is a no-op for host, CUDA, HIP, OpenCL and cling.
MSL requires every variable at program scope, class-scope statics included, to name an address space and diagnoses it at the declaration, so these headers broke any device translation unit that included them regardless of whether the constants were used. GPUglobalconstexpr() expands to constexpr everywhere except Metal, where it adds constant. Last of the series: with this the Metal translation unit has no program-scope diagnostics left, down from 208.
…address space MSL requires every variable at program scope, class-scope statics included, to name an address space. GPUglobalconstexpr() expands to constexpr everywhere except Metal, so this is a no-op for host, CUDA, HIP, OpenCL and cling. These were missed by the earlier pass: the header they sit in only started being compiled once the errors ahead of it were gone.
…s space MSL requires every variable at program scope, class-scope statics included, to name an address space. GPUglobalconstexpr() expands to constexpr everywhere except Metal, so this is a no-op for host, CUDA, HIP, OpenCL and cling. These were missed by the earlier pass: the header they sit in only started being compiled once the errors ahead of it were gone.
… in the constant address space MSL requires every variable at program scope, class-scope statics included, to name an address space. GPUglobalconstexpr() expands to constexpr everywhere except Metal, so this is a no-op for host, CUDA, HIP, OpenCL and cling. These were missed by the earlier pass: the header they sit in only started being compiled once the errors ahead of it were gone.
MSL requires every variable at program scope, class-scope statics included, to name an address space. GPUglobalconstexpr() expands to constexpr everywhere except Metal, so this is a no-op for host, CUDA, HIP, OpenCL and cling. These were missed by the earlier pass: the header they sit in only started being compiled once the errors ahead of it were gone.
MSL requires every variable at program scope, class-scope statics included, to name an address space. GPUglobalconstexpr() expands to constexpr everywhere except Metal, so this is a no-op for host, CUDA, HIP, OpenCL and cling. These were missed by the earlier pass: the header they sit in only started being compiled once the errors ahead of it were gone.
…tant address space MSL requires every variable at program scope, class-scope statics included, to name an address space. GPUglobalconstexpr() expands to constexpr everywhere except Metal, so this is a no-op for host, CUDA, HIP, OpenCL and cling. These were missed by the earlier pass: the header they sit in only started being compiled once the errors ahead of it were gone.
Collaborator
Member
Author
|
Ok, I think I fixed all the reported clang format issues. That said, regarding double, the code actually needs three different types, all of which are aliased to doubles in backends which support double as fast as float (or thereabouts). For hardware lacking double support (i.e. M5) or with extremely slow double support (some gaming card, as I understand it) the code I have separates the different semantics of the double in three:
Of the three, only GPUdoubleValue could be dropped if we aliased float to double on Metal. The first one is needed to get the proper layout when converting host / device and the third one allows for an implementation which is numerically comparable to the double one. That said, I would still keep all three, just to avoid confusion and catch mismatches at compile time. |
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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.
MSL requires every variable at program scope, class-scope statics included, to
name an address space. GPUglobalconstexpr() expands to constexpr everywhere
except Metal, so this is a no-op for host, CUDA, HIP, OpenCL and cling.
These were missed by the earlier pass: the header they sit in only started
being compiled once the errors ahead of it were gone.