Dear Frederic, dear Francois,
Thank you for the patch and review.
Am 12.02.25 um 21:58 schrieb Frederic Weisbecker:
Le Wed, Feb 12, 2025 at 09:49:29PM +0100, Francois Romieu a écrit :
Frederic Weisbecker <frederic@xxxxxxxxxx> :
[...]
r8152 may call napi_schedule() on device resume time from a bare task
context without disabling softirqs as the following trace shows:
[...]
diff --git a/drivers/net/usb/r8152.c b/drivers/net/usb/r8152.c
index 468c73974046..1325460ae457 100644
--- a/drivers/net/usb/r8152.c
+++ b/drivers/net/usb/r8152.c
@@ -8537,8 +8537,11 @@ static int rtl8152_runtime_resume(struct r8152 *tp)
clear_bit(SELECTIVE_SUSPEND, &tp->flags);
smp_mb__after_atomic();
- if (!list_empty(&tp->rx_done))
+ if (!list_empty(&tp->rx_done)) {
+ local_bh_disable();
napi_schedule(&tp->napi);
+ local_bh_enable();
+ }
AFAIU drivers/net/usb/r8152.c::rtl_work_func_t exhibits the same
problem.
It's a workqueue function and softirqs don't seem to be disabled.
Looks like a goot catch!
Tested-by: Paul Menzel <pmenzel@xxxxxxxxxxxxx>
Are you going to send a v2, so it might get into Linux 6.14, or is it
too late anyway?
Kind regards,
Paul
