io_uring: Fix current->fs handling in io_sq_wq_submit_work()
PATCH 1/1 io_uring: Fix current->fs handling in io_sq_wq_submit_work()
No upstream commit, this is a fix to a stable 5.4 specific backport.
The intention of backport commit cac68d12c531 ("io_uring: grab ->fs as part of async offload") as found in the stable 5.4 tree was to make io_sq_wq_submit_work() to switch the workqueue task's ->fs over to the submitting task's one during the IO operation.
However, due to a small logic error, this change turned out to not have any actual effect. From a high level, the relevant code in io_sq_wq_submit_work() looks like
old_fs_struct = current->fs;
do {
...
if (req->fs != current->fs && current->fs != old_fs_struct) {
task_lock(current);
if (req->fs)
current->fs = req->fs;
else
current->fs = old_fs_struct;
task_unlock(current);
}
...
} while (req);The if condition is supposed to cover the case that current->fs doesn't match what's needed for processing the request, but observe how it fails to ever evaluate to true due to the second clause:current->fs != old_fs_struct will be false in the first iteration as per the initialization of old_fs_struct and because this prevents current->fs from getting replaced, the same follows inductively for all subsequent iterations.
Fix said if condition such that
- if req->fs is set and doesn't match current->fs, the latter will be switched to the former
- or if req->fs is unset, the switch back to the initial old_fs_struct will be made, if necessary.
While at it, also correct the condition for the ->fs related cleanup right before the return of io_sq_wq_submit_work(): currently, old_fs_struct is restored only if it's non-NULL. It is always non-NULL though and thus, the if-condition is rendundant. Supposedly, the motivation had been to optimize and avoid switching current->fs back to the initial old_fs_struct in case it is found to have the desired value already. Make it so.
---
Tested on top of v5.4.90.
fs/io_uring.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)
fs/io_uring.c +3 -2
@@ -2226,7 +2226,8 @@ static void io_sq_wq_submit_work(struct work_struct *work)/* Ensure we clear previously set non-block flag */ req->rw.ki_flags &= ~IOCB_NOWAIT ; -if (req->fs != current->fs && current->fs != old_fs_struct) { +if ((req->fs && req->fs != current->fs) || + (!req->fs && current->fs != old_fs_struct)) {task_lock (current);if (req->fs) current->fs = req->fs;
@@ -2351,7 +2352,7 @@ static void io_sq_wq_submit_work(struct work_struct *work)mmput (cur_mm); }revert_creds (old_cred); -if (old_fs_struct) { +if (old_fs_struct != current->fs) {task_lock (current); current->fs = old_fs_struct;task_unlock (current);
Signature · 1 line
2.26.2