On Sun, Apr 14, 2019 at 11:44:17AM +0200, Arend Van Spriel wrote: > + Herbert > > On 3/16/2019 7:14 PM, Toke Høiland-Jørgensen wrote: > > Felix Fietkau <[email protected]> writes: > > > > > When using iTXQ, tx sequence number allocation and statistics are run at > > > dequeue time. Because of that, it is safe to enable NETIF_F_LLTX, which > > > allows tx handlers to run on multiple CPUs in parallel. > > > > Cool, didn't know about that flag. > > It is water under the bridge as this patch got applied already, but I > stumbled upon it just recently and didn't know about that flag either. So I > looked for more information about it and found the definition [1], but the > comment seemed important enough to send this reply. > > NETIF_F_LLTX_BIT, /* LockLess TX - deprecated. Please */ > /* do not use LLTX in new drivers */
The most obvious problem with LLTX is that it can cause AF_PACKET to see packets twice. But the more subtle issue is that with a proper design this lock should have no contention anyway. So please explain why you want to use this in your driver and where the contention is coming from? Cheers, -- Email: Herbert Xu <[email protected]> Home Page: http://gondor.apana.org.au/~herbert/ PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
