Skip to content

AICOMRCCL-708 fix rccl unit test failures on mi300a #3943

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
prasanna-amd wants to merge 1 commit into
base: develop
Choose a base branch
from
+16 −0
Open

AICOMRCCL-708 fix rccl unit test failures on mi300a #3943

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
14 changes: 14 additions & 0 deletions projects/rccl/src/misc/recorder.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -609,8 +609,22 @@ void parseJsonEntry(const char* entry, std::vector<rcclApiCall>& calls)
// TODO: parse comma too
rcclApiCall call;
std::string str(entry);

// Handle empty string case
if (str.empty()) {
printf("[ERROR] parseJsonEntry: empty entry string\n");
return;
}

size_t begin = str.find_first_not_of(' ');
size_t end = str.find(" : ");

// Validate that we found the delimiter and positions are valid
if (begin == std::string::npos || end == std::string::npos || end <= begin) {
printf("[ERROR] parseJsonEntry: invalid entry format: '%s'\n", entry);
return;
}

rcclCall_t type = getFuncType(str.substr(begin, end-begin));
call.type = type;
switch(type) {
Expand Down
2 changes: 2 additions & 0 deletions projects/rccl/test/ArgCheckTests.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -479,11 +479,13 @@ TEST(ArgCheckTest, ArgsCheck_InvalidCommunicatorPointers)
{
ASSERT_EQ(hipFree((void*)env.info->sendbuff), hipSuccess);
env.info->sendbuff = nullptr; // Invalid send buffer
env.sendDevicePtr = nullptr; // Prevent double-free in cleanup
}
if(env.info->recvbuff)
{
ASSERT_EQ(hipFree((void*)env.info->recvbuff), hipSuccess);
env.info->recvbuff = nullptr; // Invalid receive buffer
env.recvDevicePtr = nullptr; // Prevent double-free in cleanup
}

ncclResult_t result = ArgsCheck(env.info);
Expand Down
Loading