Skip to content

[Driver][NFC] Check pointer Arg before dereferencing. #21287

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

Open
srividya-sundaram wants to merge 3 commits into
base: sycl
Choose a base branch
from
Open

[Driver][NFC] Check pointer Arg before dereferencing. #21287

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
ab7817a
[Driver][NFC] Check pointer Arg before dereferencing.
srividya-sundaram Feb 13, 2026
9941ebd
Keep intel/llvm specific changes only in this patch.
srividya-sundaram Feb 16, 2026
4a42f4e
Fix clang format.
srividya-sundaram Feb 16, 2026
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
9 changes: 5 additions & 4 deletions clang/lib/Driver/ToolChains/Clang.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5993,6 +5993,8 @@ void Clang::ConstructJob(Compilation &C, const JobAction &JA,
options::OPT_no_offload_new_driver, false);
Arg *SYCLSplitMode =
Args.getLastArg(options::OPT_fsycl_device_code_split_EQ);
Arg *OffloadLTO = Args.getLastArg(options::OPT_foffload_lto,
options::OPT_foffload_lto_EQ);
Copy link
Contributor

@mdtoguchi mdtoguchi Feb 13, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This looks like something that would benefit the community as it isn't specific to SYCL. Would this be better to just upstream instead of doing intel/llvm?

Copy link
Contributor Author

@srividya-sundaram srividya-sundaram Feb 14, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

There are to SYCL specific checks here before some of these diagnostics are emitted (like SYCLSplitMode). Unless I am mis-understanding what you are referring to.

Copy link
Contributor

@mdtoguchi mdtoguchi Feb 14, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yes, there are SYCL specific things here, but the fix itself can be applied to upstream and be pulled into intel/llvm as it also impacts OpenMP upstream behaviors.

Copy link
Contributor Author

@srividya-sundaram srividya-sundaram Feb 16, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

but the fix itself can be applied to upstream and be pulled into intel/llvm as it also impacts OpenMP upstream behaviors.

err_drv_sycl_thinlto_split_off is specific to intel/llvm only and is therefore handled here, while the other two diagnostics are addressed in the upstream patch.

bool IsDeviceCodeSplitDisabled =
SYCLSplitMode && StringRef(SYCLSplitMode->getValue()) == "off";
bool IsSYCLLTOSupported =
Expand All @@ -6014,12 +6016,11 @@ void Clang::ConstructJob(Compilation &C, const JobAction &JA,
->getAsString(Args)
<< "-fno-gpu-rdc";
} else if (JA.isDeviceOffloading(Action::OFK_SYCL) &&
IsDeviceCodeSplitDisabled && LTOMode == LTOK_Thin) {
IsDeviceCodeSplitDisabled && LTOMode == LTOK_Thin &&
OffloadLTO) {
D.Diag(diag::err_drv_sycl_thinlto_split_off)
<< SYCLSplitMode->getAsString(Args)
<< Args.getLastArg(options::OPT_foffload_lto,
options::OPT_foffload_lto_EQ)
->getAsString(Args);
<< OffloadLTO->getAsString(Args);
} else {
assert(LTOMode == LTOK_Full || LTOMode == LTOK_Thin);
CmdArgs.push_back(Args.MakeArgString(
Expand Down
Loading