-
-
Notifications
You must be signed in to change notification settings - Fork 5.5k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Bump Julia to LLVM 16 #51720
Bump Julia to LLVM 16 #51720
Conversation
8f117e0
to
213c89b
Compare
4c9834c
to
07faba7
Compare
cc: @maleadt |
LLVM deleted this functionality, claimed NFC for the change
|
You now need to check |
Co-authored-by: Gabriel Baraldi <baraldigabriel@gmail.com> Co-authored-by: Prem Chintalapudi <prem.chintalapudi@gmail.com> Co-authored-by: Jameson Nash <vtjnash@gmail.com>
07faba7
to
a21c7ed
Compare
Checking the command line isn't enough, because we want to respect wheter GPUCompiler added a value or not. |
766881a
to
9ba6de9
Compare
90962a0
to
63f4afb
Compare
@nanosoldier |
@nanosoldier |
f9a68cc
to
f76032e
Compare
Co-authored-by: Tim Besard <tim.besard@gmail.com>
f76032e
to
471dc6c
Compare
@nanosoldier |
The package evaluation job you requested has completed - possible new issues were detected. |
fa6e077
to
3687772
Compare
After lots of annoyances and setbacks, I think this is ready for review. PkgEval looks good. the AllocCheck segfault is weird but I'm taking a look at it. |
@@ -1051,6 +1051,20 @@ static AOTOutputs add_output_impl(Module &M, TargetMachine &SourceTM, ShardTimer | |||
// functions. We do so after optimization to avoid cloning these functions. | |||
|
|||
// Float16 conversion routines | |||
#if defined(_CPU_X86_64_) && defined(_OS_DARWIN_) && JL_LLVM_VERSION >= 160000 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Can we make this dependent on the target triple? Just to preserve the fiction that codegen is cross-compile capable.
src/cgutils.cpp
Outdated
@@ -911,7 +911,7 @@ static bool is_tupletype_homogeneous(jl_svec_t *t, bool allow_va = false) | |||
} | |||
|
|||
static bool for_each_uniontype_small( | |||
std::function<void(unsigned, jl_datatype_t*)> f, | |||
const std::function<void(unsigned, jl_datatype_t*)> &f, |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Could use an llvm::function_ref here.
void SetOpaquePointer(LLVMContext &ctx) { | ||
if (jl_opaque_ptrs_set) | ||
return; | ||
#ifndef JL_LLVM_OPAQUE_POINTERS |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Is anyone actually using this build flag, given we have a runtime option for setting it as well? I'm in favor of just deleting.
Congratulations on getting this over the line! |
Current Status:
MissingRoots.c
error: <unknown>:0: Cannot represent a difference across sections