Commit ba6b372c authored by Sylwester Nawrocki's avatar Sylwester Nawrocki Committed by Mauro Carvalho Chehab

[media] s5p-fimc: Don't use mutex_lock_interruptible() in device release()

Use uninterruptible mutex_lock in the release() file op to make
sure all resources are properly freed when a process is being
terminated. Returning -ERESTARTSYS has no effect for a terminating
process and this caused driver resources not to be released. Not
releasing the buffer queue also prevented other drivers to free
memory, e.g. in MMAP -> USERPTR scenario.
This patch is required for stable kernels v3.6+.
Reported-by: default avatarKamil Debski <k.debski@samsung.com>
Reported-by: default avatarMarek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: default avatarSylwester Nawrocki <s.nawrocki@samsung.com>
Signed-off-by: default avatarKyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: default avatarMauro Carvalho Chehab <mchehab@redhat.com>
parent 97d66c47
...@@ -556,8 +556,7 @@ static int fimc_capture_close(struct file *file) ...@@ -556,8 +556,7 @@ static int fimc_capture_close(struct file *file)
dbg("pid: %d, state: 0x%lx", task_pid_nr(current), fimc->state); dbg("pid: %d, state: 0x%lx", task_pid_nr(current), fimc->state);
if (mutex_lock_interruptible(&fimc->lock)) mutex_lock(&fimc->lock);
return -ERESTARTSYS;
if (--fimc->vid_cap.refcnt == 0) { if (--fimc->vid_cap.refcnt == 0) {
clear_bit(ST_CAPT_BUSY, &fimc->state); clear_bit(ST_CAPT_BUSY, &fimc->state);
......
...@@ -718,8 +718,7 @@ static int fimc_m2m_release(struct file *file) ...@@ -718,8 +718,7 @@ static int fimc_m2m_release(struct file *file)
dbg("pid: %d, state: 0x%lx, refcnt= %d", dbg("pid: %d, state: 0x%lx, refcnt= %d",
task_pid_nr(current), fimc->state, fimc->m2m.refcnt); task_pid_nr(current), fimc->state, fimc->m2m.refcnt);
if (mutex_lock_interruptible(&fimc->lock)) mutex_lock(&fimc->lock);
return -ERESTARTSYS;
v4l2_m2m_ctx_release(ctx->m2m_ctx); v4l2_m2m_ctx_release(ctx->m2m_ctx);
fimc_ctrls_delete(ctx); fimc_ctrls_delete(ctx);
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment