Fix directfs restore for deleted regular file when read handle is not available #11919
+71
−20
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Fix directfs restore for deleted regular file when read handle is not available
In such a situation, we can not rely on traditional methods (like openHandle())
to open a readable handle. This is because, for directfs to open a handle, it
needs to re-walk the file via the parent using openat(parentFD, name). This
does not work for deleted files, it will fail with ENOENT. runsc gofer works
around this by using /proc/self/fd/ to re-open the control FD in the desired
mode. However, the sentry does not have access to any procfs instance (for
security). See fcbc289 ("runsc: umount /proc in the sandbox namespace").
This change makes directfs just use the control FD to fetch file data. The
control FD should not be used for IO. We make an exception here for S/R.
Fixes #11903