Messages in this thread | | | From | Vinicius Costa Gomes <> | Subject | Re: [Intel-wired-lan] [PATCH net v2] igc: Prevent garbled TX queue with XDP ZEROCOPY | Date | Thu, 29 Jun 2023 09:25:43 -0700 |
| |
Florian Kauer <florian.kauer@linutronix.de> writes:
> Hi Vinicius, > > On 28.06.23 23:34, Vinicius Costa Gomes wrote: >> Florian Kauer <florian.kauer@linutronix.de> writes: >> >>> In normal operation, each populated queue item has >>> next_to_watch pointing to the last TX desc of the packet, >>> while each cleaned item has it set to 0. In particular, >>> next_to_use that points to the next (necessarily clean) >>> item to use has next_to_watch set to 0. >>> >>> When the TX queue is used both by an application using >>> AF_XDP with ZEROCOPY as well as a second non-XDP application >>> generating high traffic, the queue pointers can get in >>> an invalid state where next_to_use points to an item >>> where next_to_watch is NOT set to 0. >>> >>> However, the implementation assumes at several places >>> that this is never the case, so if it does hold, >>> bad things happen. In particular, within the loop inside >>> of igc_clean_tx_irq(), next_to_clean can overtake next_to_use. >>> Finally, this prevents any further transmission via >>> this queue and it never gets unblocked or signaled. >>> Secondly, if the queue is in this garbled state, >>> the inner loop of igc_clean_tx_ring() will never terminate, >>> completely hogging a CPU core. >>> >>> The reason is that igc_xdp_xmit_zc() reads next_to_use >>> before acquiring the lock, and writing it back >>> (potentially unmodified) later. If it got modified >>> before locking, the outdated next_to_use is written >>> pointing to an item that was already used elsewhere >>> (and thus next_to_watch got written). >>> >>> Fixes: 9acf59a752d4 ("igc: Enable TX via AF_XDP zero-copy") >>> Signed-off-by: Florian Kauer <florian.kauer@linutronix.de> >>> Reviewed-by: Kurt Kanzenbach <kurt@linutronix.de> >>> Tested-by: Kurt Kanzenbach <kurt@linutronix.de> >>> --- >> >> This patch doesn't directly apply because there's a small conflict with >> commit 95b681485563 ("igc: Avoid transmit queue timeout for XDP"), >> but really easy to solve. >> >> Anyway, good catch: >> >> Acked-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> > > I am sorry, that was bad timing. I prepared the initial patch on Friday and overlooked the merge. > Shall I send a v3 or will someone else take care of the conflict > resolution?
I think it's easier if you send a v3.
Cheers, -- Vinicius
| |