Skip to content

Support for Cuda Separate Compilation #508

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.

Sign up for GitHub

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

Closed
wants to merge 3 commits into from
Closed

Support for Cuda Separate Compilation #508

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
ee927f8
for cuda separate compilation
YoshikawaMasashi Jun 1, 2020
fdd5761
move to compile objects
YoshikawaMasashi Jun 2, 2020
5d51a1d
rename to __cc_internal_link.o
YoshikawaMasashi Jun 9, 2020
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
30 changes: 29 additions & 1 deletion src/lib.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1139,6 +1139,21 @@ impl Build {
for obj in objs {
self.compile_object(obj)?;
}

if self.cuda {
let compiler = self.try_get_compiler()?;
let mut cmd = compiler.to_command();
cmd.arg("-dlink");
for obj in objs {
cmd.arg(obj.dst.clone());
}
cmd.arg("-o");
let out_dir = self.get_out_dir()?;
let out_dir = out_dir.join("__cc_internal_link.o");
cmd.arg(out_dir);
run(&mut cmd, "nvcc")?;
}

Ok(())
}

Expand All @@ -1165,6 +1180,11 @@ impl Build {
.into_owned(),
)
};

if self.cuda {
cmd.arg("-dc");
}

let is_arm = target.contains("aarch64") || target.contains("arm");
command_add_output_file(&mut cmd, &obj.dst, self.cuda, msvc, clang, is_asm, is_arm);
// armasm and armasm64 don't requrie -c option
Expand Down Expand Up @@ -1713,7 +1733,15 @@ impl Build {
// appends to it, which we don't want.
let _ = fs::remove_file(&dst);

let objects: Vec<_> = objs.iter().map(|obj| obj.dst.clone()).collect();
let mut objects: Vec<_> = objs.iter().map(|obj| obj.dst.clone()).collect();

if self.cuda {
let out_dir = self.get_out_dir()?;
let out_dir = out_dir.join("__cc_internal_link.o");
objects.push(out_dir);
}

let objects = objects;
let target = self.get_target()?;
if target.contains("msvc") {
let (mut cmd, program) = self.get_ar()?;
Expand Down