Skip to content

Fix the MacOS remote unwinder for VS4Mac #63405

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

Merged
jeffschwMSFT merged 2 commits into from
Jan 13, 2022
Merged

Fix the MacOS remote unwinder for VS4Mac #63405

Changes from 1 commit
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
647ee19
Fix the MacOS remote unwinder for VS4Mac
Jan 4, 2022
064cbe0
Fix native frame unwind in syscall on arm64 for VS4Mac crash report
Jan 13, 2022
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
Next Next commit
Fix the MacOS remote unwinder for VS4Mac 
The wrong module was being passed to the remote unwinder because the load bias for shared modules
was being calculated incorrectly.

Issue: #63309
  • Loading branch information
Mike McLaughlin committed Jan 13, 2022
commit 647ee19c4cc861d2f779a52e6f15df4caace52ed
5 changes: 2 additions & 3 deletions src/coreclr/debug/dbgutil/machoreader.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -229,9 +229,8 @@ MachOModule::ReadLoadCommands()
m_segments.push_back(segment);

// Calculate the load bias for the module. This is the value to add to the vmaddr of a
// segment to get the actual address. For shared modules, this is 0 since those segments
// are absolute address.
if (segment->fileoff == 0 && segment->filesize > 0)
// segment to get the actual address.
if (strcmp(segment->segname, SEG_TEXT) == 0)
Copy link
Member

@hoyosjs hoyosjs Jan 13, 2022

Choose a reason for hiding this comment

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

Is there any guarantee that the first segment is the text segment, and that there's only one ever (the spec seems to imply that there is always at least one, but nothing about there being multiple or the ordering)? Otherwise we may get here more than once or have a wrong offset.

Also, what happens with the bias of data segments? Granted, this was already never considered if the first segment was always text.

Copy link
Contributor Author

@mikem8361 mikem8361 Jan 13, 2022

Choose a reason for hiding this comment

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

Yes, there is only __TEXT segment and it always seems to be the first one.

{
m_loadBias = m_baseAddress - segment->vmaddr;
}
Expand Down