On 30/05/14 12:25, Jim Baxter wrote: > On 29/05/14 20:04, Eric Dumazet wrote: >> On Thu, 2014-05-29 at 18:12 +0100, Jim Baxter wrote: >> >> Note the patch is OK, but changelog a bit misleading ;) >> >> Kernel was not running out of memory, because truesize was correct. >> >> The problem here is that a frame was consuming more kernel memory than >> really needed, so chances of hitting socket sk_rcvbuf limit was high. >> >> BTW : >> #define NTB_OUT_SIZE 16384 >> >> alloc_skb(size) -> >> kmalloc(16384 + sizeof(struct skb_shared_info)) -> >> roundup() => 32768 >> >> So truesize of the skb was infact ~32KB, which is really insane indeed. >> After your patch, its back to ~2KB >> >> Acked-by: Eric Dumazet <edumazet@xxxxxxxxxx> >> >> >> > > Thank you, I will clarify the description. > > Jim > Hi Eric, Is this commit log more correct? This fixes a problem with dropped packets over 16k CDC-NCM when the connection is being heavily used. The issue was that the extracted frames cloned from the received frame were consuming more memory than necessary due to the truesize being ~32KB instead of ~2KB, this meant there was a high chance of reaching the sk_rcvbuf limit. Acked-by: Eric Dumazet <edumazet@xxxxxxxxxx> Signed-off-by: Jim Baxter <jim_baxter@xxxxxxxxxx> Thank you, Jim -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
