[bpf-next PATCH 3/7] sockmap: convert refcnt to an atomic refcnt

sk_callback_lock() here (on every send) seems less than ideal, also it may sleep in cases where we hit memory pressure. Instead of dealing with these issues in some clever way simply make the reference counting a refcnt_t type and do proper atomic ops. Signed-off-by: John Fastabend --- kernel/bpf

[bpf-next PATCH 1/7] net: add a UID to use for ULP socket assignment

ULP is known and done on the kernel side. In this case the named lookup is not needed. Remove pr_notice, user gets an error code back and should check that rather than rely on logs. Signed-off-by: John Fastabend --- include/net/tcp.h |5 + net/ipv4/tcp_ulp.c | 51

[bpf-next PATCH 0/7] Add BPF_PROG_TYPE_SK_MSG and attach pt

e automated side. We can push this as an independent patch set. --- John Fastabend (7): net: add a UID to use for ULP socket assignment sock: make static tls function alloc_sg generic sock helper sockmap: convert refcnt to an atomic refcnt net: do_tcp_sendpages flag to avo

[bpf-next PATCH v3 4/7] bpf: sockmap sample, report bytes/sec

Report bytes/sec sent as well as total bytes. Useful to get rough idea how different configurations and usage patterns perform with sockmap. Signed-off-by: John Fastabend --- samples/sockmap/sockmap_user.c | 37 - 1 file changed, 32 insertions(+), 5

[bpf-next PATCH v3 6/7] bpf: sockmap put client sockets in blocking mode

-by: John Fastabend --- samples/sockmap/sockmap_user.c |2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/samples/sockmap/sockmap_user.c b/samples/sockmap/sockmap_user.c index c3295a7..818766b 100644 --- a/samples/sockmap/sockmap_user.c +++ b/samples/sockmap/sockmap_user.c

[bpf-next PATCH v3 7/7] bpf: sockmap set rlimit

Avoid extra step of setting limit from cmdline and do it directly in the program. Signed-off-by: John Fastabend --- samples/sockmap/sockmap_user.c |7 +++ 1 file changed, 7 insertions(+) diff --git a/samples/sockmap/sockmap_user.c b/samples/sockmap/sockmap_user.c index 818766b..a6dab97

[bpf-next PATCH v3 5/7] bpf: sockmap sample add base test without any BPF for comparison

Add a base test that does not use BPF hooks to test baseline case. Signed-off-by: John Fastabend --- samples/sockmap/sockmap_user.c | 26 +- 1 file changed, 21 insertions(+), 5 deletions(-) diff --git a/samples/sockmap/sockmap_user.c b/samples/sockmap/sockmap_user.c

[bpf-next PATCH v3 2/7] bpf: add sendmsg option for testing BPF programs

supported, but more can be added as needed. The new help argument gives the following, Usage: ./sockmap --cgroup options: --help -h --cgroup -c --rate -r --verbose -v --iov_count-i --length -l --test -t Signed-off-by: John Fastabend

[bpf-next PATCH v3 3/7] bpf: sockmap sample, use fork() for send and recv

get many GBps of data which helps exercise the sockmap code. Signed-off-by: John Fastabend --- samples/sockmap/sockmap_user.c | 58 +--- 1 file changed, 42 insertions(+), 16 deletions(-) diff --git a/samples/sockmap/sockmap_user.c b/samples/sockmap

[bpf-next PATCH v3 1/7] bpf: refactor sockmap sample program update for arg parsing

future. Signed-off-by: John Fastabend --- samples/sockmap/sockmap_user.c | 164 1 file changed, 113 insertions(+), 51 deletions(-) diff --git a/samples/sockmap/sockmap_user.c b/samples/sockmap/sockmap_user.c index 7cc9d22..17400d4 100644 --- a/samples/sockmap

[bpf-next PATCH v3 0/7] sockmap sample update

sighandler update, 2/7 free iov in error cases 3/7 fix bogus makefile change, bail out early on errors Thanks Daniel and Martin for the reviews! --- John Fastabend (7): bpf: refactor sockmap sample program update for arg parsing bpf: add sendmsg option for testing BPF programs

Re: [bpf-next PATCH v2 1/7] bpf: refactor sockmap sample program update for arg parsing

On 01/11/2018 08:31 PM, John Fastabend wrote: > On 01/10/2018 05:25 PM, Daniel Borkmann wrote: >> On 01/10/2018 07:39 PM, John Fastabend wrote: >>> sockmap sample program takes arguments from cmd line but it reads them >>> in using offsets into the array. Because we

Re: [bpf-next PATCH v2 3/7] bpf: sockmap sample, use fork() for send and recv

On 01/10/2018 05:31 PM, Daniel Borkmann wrote: > On 01/10/2018 07:39 PM, John Fastabend wrote: >> Currently for SENDMSG tests first send completes then recv runs. This >> does not work well for large data sizes and/or many iterations. So >> fork the recv and send handler so

Re: [bpf-next PATCH v2 1/7] bpf: refactor sockmap sample program update for arg parsing

On 01/10/2018 05:25 PM, Daniel Borkmann wrote: > On 01/10/2018 07:39 PM, John Fastabend wrote: >> sockmap sample program takes arguments from cmd line but it reads them >> in using offsets into the array. Because we want to add more arguments >> in the future lets do pro

Re: [bpf-next PATCH v2 5/7] bpf: sockmap sample add base test without any BPF for comparison

On 01/11/2018 01:10 PM, Martin KaFai Lau wrote: > On Wed, Jan 10, 2018 at 10:40:11AM -0800, John Fastabend wrote: >> Add a base test that does not use BPF hooks to test baseline case. >> >> Signed-off-by: John Fastabend >> --- >> samp

Re: [bpf-next PATCH v2 3/7] bpf: sockmap sample, use fork() for send and recv

On 01/11/2018 01:08 PM, Martin KaFai Lau wrote: > On Wed, Jan 10, 2018 at 10:39:37AM -0800, John Fastabend wrote: >> Currently for SENDMSG tests first send completes then recv runs. This >> does not work well for large data sizes and/or many iterations. So >> fork the recv and

Re: [bpf-next PATCH v2 1/7] bpf: refactor sockmap sample program update for arg parsing

On 01/11/2018 01:05 PM, Martin KaFai Lau wrote: > On Wed, Jan 10, 2018 at 10:39:04AM -0800, John Fastabend wrote: >> sockmap sample program takes arguments from cmd line but it reads them >> in using offsets into the array. Because we want to add more arguments >> in the

[bpf-next PATCH v2 7/7] bpf: sockmap set rlimit

Avoid extra step of setting limit from cmdline and do it directly in the program. Signed-off-by: John Fastabend --- samples/sockmap/sockmap_user.c |7 +++ 1 file changed, 7 insertions(+) diff --git a/samples/sockmap/sockmap_user.c b/samples/sockmap/sockmap_user.c index 9496b2c..16c19c5

[bpf-next PATCH v2 6/7] bpf: sockmap put client sockets in blocking mode

-by: John Fastabend --- samples/sockmap/sockmap_user.c |2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/samples/sockmap/sockmap_user.c b/samples/sockmap/sockmap_user.c index eb19d14..9496b2c 100644 --- a/samples/sockmap/sockmap_user.c +++ b/samples/sockmap/sockmap_user.c

[bpf-next PATCH v2 5/7] bpf: sockmap sample add base test without any BPF for comparison

Add a base test that does not use BPF hooks to test baseline case. Signed-off-by: John Fastabend --- samples/sockmap/sockmap_user.c | 26 +- 1 file changed, 21 insertions(+), 5 deletions(-) diff --git a/samples/sockmap/sockmap_user.c b/samples/sockmap/sockmap_user.c

[bpf-next PATCH v2 3/7] bpf: sockmap sample, use fork() for send and recv

get many GBps of data which helps exercise the sockmap code. Signed-off-by: John Fastabend --- samples/sockmap/Makefile |2 + samples/sockmap/sockmap_user.c | 58 +--- 2 files changed, 43 insertions(+), 17 deletions(-) diff --git a/samples/sockmap

[bpf-next PATCH v2 4/7] bpf: sockmap sample, report bytes/sec

Report bytes/sec sent as well as total bytes. Useful to get rough idea how different configurations and usage patterns perform with sockmap. Signed-off-by: John Fastabend --- samples/sockmap/sockmap_user.c | 37 - 1 file changed, 32 insertions(+), 5

[bpf-next PATCH v2 1/7] bpf: refactor sockmap sample program update for arg parsing

future. Signed-off-by: John Fastabend --- samples/sockmap/sockmap_user.c | 142 +--- 1 file changed, 103 insertions(+), 39 deletions(-) diff --git a/samples/sockmap/sockmap_user.c b/samples/sockmap/sockmap_user.c index 7cc9d22..5cbe7a5 100644 --- a/samples/sockmap

[bpf-next PATCH v2 2/7] bpf: add sendmsg option for testing BPF programs

supported, but more can be added as needed. The new help argument gives the following, Usage: ./sockmap --cgroup options: --help -h --cgroup -c --rate -r --verbose -v --iov_count-i --length -l --test -t Signed-off-by: John Fastabend

[bpf-next PATCH v2 0/7] sockmap sample update

seful, the reporting is bare bones, etc. But, IMO lets push this now rather than sit on it for weeks until I get time to do the above improvements. Additional patches can address the other limitations/issues. v2: removed bogus file added by patch 3/7 --- John Fastabend (7): bpf: refactor sock

Re: [PATCH] ptr_ring: document usage around __ptr_ring_peek

llowing up. Acked-by: John Fastabend

Re: [bpf-next PATCH 4/7] bpf: sockmap sample, report bytes/sec

On 01/09/2018 05:27 AM, Jesper Dangaard Brouer wrote: > On Mon, 08 Jan 2018 10:05:58 -0800 > John Fastabend wrote: > >> Report bytes/sec sent as well as total bytes. Useful to get rough >> idea how different configurations and usage patterns perform with >> sockmap

Re: [bpf-next PATCH 1/7] bpf: refactor sockmap sample program update for arg parsing

On 01/09/2018 05:30 AM, Jesper Dangaard Brouer wrote: > On Mon, 08 Jan 2018 10:05:07 -0800 > John Fastabend wrote: > >> sockmap sample program takes arguments from cmd line but it reads them >> in using offsets into the array. Because we want to add more arguments >> i

[bpf-next PATCH 7/7] bpf: sockmap set rlimit

Avoid extra step of setting limit from cmdline and do it directly in the program. Signed-off-by: John Fastabend --- samples/sockmap/sockmap_user.c |7 +++ 1 file changed, 7 insertions(+) diff --git a/samples/sockmap/sockmap_user.c b/samples/sockmap/sockmap_user.c index 0d8950f..2afbefd

[bpf-next PATCH 6/7] bpf: sockmap put client sockets in blocking mode

-by: John Fastabend --- samples/sockmap/sockmap_user.c |2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/samples/sockmap/sockmap_user.c b/samples/sockmap/sockmap_user.c index bae85f8..0d8950f 100644 --- a/samples/sockmap/sockmap_user.c +++ b/samples/sockmap/sockmap_user.c

[bpf-next PATCH 4/7] bpf: sockmap sample, report bytes/sec

Report bytes/sec sent as well as total bytes. Useful to get rough idea how different configurations and usage patterns perform with sockmap. Signed-off-by: John Fastabend --- samples/sockmap/sockmap_user.c | 37 - 1 file changed, 32 insertions(+), 5

[bpf-next PATCH 3/7] bpf: sockmap sample, use fork() for send and recv

get many GBps of data which helps exercise the sockmap code. Signed-off-by: John Fastabend --- samples/sockmap/sockmap_user. |0 samples/sockmap/sockmap_user.c | 55 2 files changed, 39 insertions(+), 16 deletions(-) create mode 100644 samples

[bpf-next PATCH 5/7] bpf: sockmap sample add base test without any BPF for comparison

Add a base test that does not use BPF hooks to test baseline case. Signed-off-by: John Fastabend --- samples/sockmap/sockmap_user.c | 26 +- 1 file changed, 21 insertions(+), 5 deletions(-) diff --git a/samples/sockmap/sockmap_user.c b/samples/sockmap/sockmap_user.c

[bpf-next PATCH 2/7] bpf: add sendmsg option for testing BPF programs

supported, but more can be added as needed. The new help argument gives the following, Usage: ./sockmap --cgroup options: --help -h --cgroup -c --rate -r --verbose -v --iov_count-i --length -l --test -t Signed-off-by: John Fastabend

[bpf-next PATCH 0/7] sockmap sample updates

seful, the reporting could be better, etc. But, IMO lets push this now rather than sit on it for weeks until I get time to do the above improvements. --- John Fastabend (7): bpf: refactor sockmap sample program update for arg parsing bpf: add sendmsg option for testing BPF programs

[bpf-next PATCH 1/7] bpf: refactor sockmap sample program update for arg parsing

future. Signed-off-by: John Fastabend --- samples/sockmap/sockmap_user.c | 142 +--- 1 file changed, 103 insertions(+), 39 deletions(-) diff --git a/samples/sockmap/sockmap_user.c b/samples/sockmap/sockmap_user.c index 7cc9d22..5cbe7a5 100644 --- a/samples/sockmap

Re: [PATCH v2 bpf] bpf: prevent out-of-bounds speculation

f-by: Alexei Starovoitov > --- LGTM, I'll drop it on my test systems and start running with it. Although I don't have any Variant 1 code to test, but seems that is being covered by others. Thanks! Acked-by: John Fastabend

Re: [bpf-next V4 PATCH 01/14] xdp: base API for new XDP rx-queue info concept

On 01/03/2018 02:25 AM, Jesper Dangaard Brouer wrote: > This patch only introduce the core data structures and API functions. > All XDP enabled drivers must use the API before this info can used. > > There is a need for XDP to know more about the RX-queue a given XDP > frames have arrived on. For

Re: [bpf-next V4 PATCH 13/14] bpf: finally expose xdp_rxq_info to XDP bpf-programs

igned-off-by: Jesper Dangaard Brouer > Acked-by: Alexei Starovoitov > --- LGTM Acked-by: John Fastabend

Re: [Intel-wired-lan] [bpf-next V4 PATCH 03/14] i40e: setup xdp_rxq_info

t; Cc: intel-wired-...@lists.osuosl.org > Cc: Björn Töpel > Cc: Jeff Kirsher > Cc: Paul Menzel > Signed-off-by: Jesper Dangaard Brouer > Reviewed-by: Paul Menzel > --- Same here. LGTM. Acked-by: John Fastabend

Re: [Intel-wired-lan] [bpf-next V4 PATCH 04/14] ixgbe: setup xdp_rxq_info

> > Cc: intel-wired-...@lists.osuosl.org > Cc: Jeff Kirsher > Cc: Alexander Duyck > Signed-off-by: Jesper Dangaard Brouer > --- Looked a bit for reset paths that might be missed but didn't find any. LGTM. Acked-by: John Fastabend

[bpf-net PATCH v2] bpf: sockmap missing NULL psock check

obvious in my opinion. Signed-off-by: John Fastabend --- kernel/bpf/sockmap.c | 11 +-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/kernel/bpf/sockmap.c b/kernel/bpf/sockmap.c index 5ee2e41..1712d31 100644 --- a/kernel/bpf/sockmap.c +++ b/kernel/bpf/sockmap.c @@ -591,8

Re: "lockless" qdisc breaks tx_queue_len change too?

On 01/03/2018 03:41 PM, Cong Wang wrote: > On Wed, Jan 3, 2018 at 10:09 AM, John Fastabend > wrote: >> On 01/02/2018 08:41 PM, Cong Wang wrote: >>> Hi, John >>> >>> While reviewing your ptr_ring fix again today, it looks like your >>> "lockl

[net-next PATCH 1/2] bpf: sockmap remove unused function

This was added for some work that was eventually factored out but the helper call was missed. Remove it now and add it back later if needed. Signed-off-by: John Fastabend --- kernel/bpf/sockmap.c |8 1 file changed, 8 deletions(-) diff --git a/kernel/bpf/sockmap.c b/kernel/bpf

[net-next PATCH 2/2] bpf: only build sockmap with CONFIG_INET

The sockmap infrastructure is only aware of TCP sockets at the moment. In the future we plan to add UDP. In both cases CONFIG_NET should be built-in. So lets only build sockmap if CONFIG_INET is enabled. Signed-off-by: John Fastabend --- include/linux/bpf.h |2 +- include/linux

Re: "lockless" qdisc breaks tx_queue_len change too?

On 01/02/2018 08:41 PM, Cong Wang wrote: > Hi, John > > While reviewing your ptr_ring fix again today, it looks like your > "lockless" qdisc patchset breaks dev->tx_queue_len behavior. > > Before your patchset, dev->tx_queue_len is merely an integer to read, > after your patchset, the skb array h

Re: [net-next PATCH] net: ptr_ring: otherwise safe empty checks can overrun array bounds

On 01/03/2018 07:50 AM, Michael S. Tsirkin wrote: > On Tue, Jan 02, 2018 at 04:25:03PM -0800, John Fastabend wrote: >>> >>> More generally, what makes this usage safe? >>> Is there a way to formalize it at the API level? >>> >> >> Right I think

[bpf PATCH] bpf: sockmap missing NULL psock check

obvious in my opinion. Signed-off-by: John Fastabend --- kernel/bpf/sockmap.c |7 +++ 1 file changed, 7 insertions(+) diff --git a/kernel/bpf/sockmap.c b/kernel/bpf/sockmap.c index 5ee2e41..dfbbde2 100644 --- a/kernel/bpf/sockmap.c +++ b/kernel/bpf/sockmap.c @@ -591,6 +591,13 @@ static

Re: [net-next PATCH] net: ptr_ring: otherwise safe empty checks can overrun array bounds

On 01/02/2018 03:12 PM, Michael S. Tsirkin wrote: > On Tue, Jan 02, 2018 at 01:27:23PM -0800, John Fastabend wrote: >> On 01/02/2018 09:17 AM, Michael S. Tsirkin wrote: >>> On Tue, Jan 02, 2018 at 07:01:33PM +0200, Michael S. Tsirkin wrote: >>>> On Tue, Jan 02, 2

Re: [PATCH net-next v3] net: sched: fix skb leak in dev_requeue_skb()

On 01/02/2018 10:49 AM, David Miller wrote: > From: Wei Yongjun > Date: Wed, 27 Dec 2017 17:05:52 +0800 > >> When dev_requeue_skb() is called with bluked skb list, only the > ^^ > > "bulked" > >> first skb of the list will be requeued to qdisc layer,

Re: [net-next PATCH] net: ptr_ring: otherwise safe empty checks can overrun array bounds

On 01/02/2018 09:17 AM, Michael S. Tsirkin wrote: > On Tue, Jan 02, 2018 at 07:01:33PM +0200, Michael S. Tsirkin wrote: >> On Tue, Jan 02, 2018 at 11:52:19AM -0500, David Miller wrote: >>> From: John Fastabend >>> Date: Wed, 27 Dec 2017 19:50:25 -0800 >>&g

[net-next PATCH] net: ptr_ring: otherwise safe empty checks can overrun array bounds

e normal case checks would suffer some so best to just allocate an extra pointer. Reported-by: Jakub Kicinski Fixes: c5ad119fb6c09 ("net: sched: pfifo_fast use skb_array") Signed-off-by: John Fastabend --- include/linux/ptr_ring.h |7 ++- 1 file changed, 6 insertions(+), 1 delet

Re: [Patch net-next] net_sched: remove the unsafe __skb_array_empty()

On 12/27/2017 10:29 AM, Cong Wang wrote: > On Sat, Dec 23, 2017 at 10:57 PM, John Fastabend > wrote: >> On 12/22/2017 12:31 PM, Cong Wang wrote: >>> I understand why you had it, but it is just not safe. You don't want >>> to achieve performance gain by crash

Re: [PATCH net-next v2] net: sched: fix skb leak in dev_requeue_skb()

On 12/24/2017 07:49 PM, Wei Yongjun wrote: > When dev_requeue_skb() is called with bluked skb list, only the > first skb of the list will be requeued to qdisc layer, and leak > the others without free them. > > TCP is broken due to skb leak since no free skb will be considered > as still in the ho

Re: [Patch net-next] net_sched: remove the unsafe __skb_array_empty()

On 12/22/2017 12:31 PM, Cong Wang wrote: > On Thu, Dec 21, 2017 at 7:06 PM, John Fastabend > wrote: >> On 12/21/2017 04:03 PM, Cong Wang wrote: >>> __skb_array_empty() is only safe if array is never resized. >>> pfifo_fast_dequeue() is called in TX BH context and w

Re: [Patch net-next] net_sched: call qdisc_reset() with qdisc lock

n't help. And it is only a local_bh_disable. > Fixes: 7bbde83b1860 ("net: sched: drop qdisc_reset from dev_graft_qdisc") > Reported-by: Jakub Kicinski > Cc: John Fastabend > Signed-off-by: Cong Wang > --- > net/sched/sch_generic.c | 5 - > 1 file chang

Re: [Patch net-next] net_sched: remove the unsafe __skb_array_empty()

Fixes: c5ad119fb6c0 ("net: sched: pfifo_fast use skb_array") > Reported-by: Jakub Kicinski > Cc: John Fastabend > Signed-off-by: Cong Wang > --- > net/sched/sch_generic.c | 3 --- > 1 file changed, 3 deletions(-) > > diff --git a/net/sched/sch_generic.c b/net/s

Re: RCU callback crashes

On 12/20/2017 11:27 PM, Cong Wang wrote: > On Wed, Dec 20, 2017 at 4:50 PM, Jakub Kicinski wrote: >> On Wed, 20 Dec 2017 16:41:14 -0800, Jakub Kicinski wrote: >>> Just as I hit send... :) but this looks unrelated, "Comm: sshd" - >>> so probably from the management interface. >>> >>> [ 154.604041

Re: [PATCH] net: Revert "net_sched: no need to free qdisc in RCU callback"

On 12/20/2017 01:59 PM, Jakub Kicinski wrote: > On Wed, 20 Dec 2017 12:09:19 -0800, John Fastabend wrote: >> RCU grace period is needed for lockless qdiscs added in the commit >> c5ad119fb6c09 ("net: sched: pfifo_fast use skb_array"). >> >> It is needed now tha

Re: [PATCH] net: Revert "net_sched: no need to free qdisc in RCU callback"

On 12/20/2017 03:23 PM, Cong Wang wrote: > On Wed, Dec 20, 2017 at 3:05 PM, John Fastabend > wrote: >> On 12/20/2017 02:41 PM, Cong Wang wrote: >>> On Wed, Dec 20, 2017 at 12:09 PM, John Fastabend >>> wrote: >>>> RCU grace period is need

Re: [PATCH] net: Revert "net_sched: no need to free qdisc in RCU callback"

On 12/20/2017 02:41 PM, Cong Wang wrote: > On Wed, Dec 20, 2017 at 12:09 PM, John Fastabend > wrote: >> RCU grace period is needed for lockless qdiscs added in the commit >> c5ad119fb6c09 ("net: sched: pfifo_fast use skb_array"). >> >> It is needed now

Re: RCU callback crashes

On 12/20/2017 12:17 PM, Jakub Kicinski wrote: > On Wed, 20 Dec 2017 10:04:17 -0800, John Fastabend wrote: >> On 12/19/2017 10:34 PM, Jakub Kicinski wrote: >>> On Tue, 19 Dec 2017 22:22:27 -0800, Jakub Kicinski wrote: >>>>>> I get this: >>>&

Re: RCU callback crashes

g with the qdisc itself in: >> qdisc_destroy->qdisc_free >> >> Before miniq, tp was checked in the rcu reader path. In case it was not >> null, q was processed. In slow patch, tp is freed after rcu grace period: >> tcf_proto_destroy->kfree_rcu >> >> I assumed that sin

Re: RCU callback crashes

On 12/20/2017 11:59 AM, Jiri Pirko wrote: > Wed, Dec 20, 2017 at 07:17:50PM CET, xiyou.wangc...@gmail.com wrote: >> On Tue, Dec 19, 2017 at 10:34 PM, Jakub Kicinski wrote: >>> Ah, no object debug but KASAN on produces this: >>> >> >> >> I bet it is an ingress qdisc which is being freed? >> >> >> >

[PATCH] net: Revert "net_sched: no need to free qdisc in RCU callback"

to RCU callback. Otherwise we risk the datapath adding skbs during removal. Fixes: c5ad119fb6c09 ("net: sched: pfifo_fast use skb_array") Signed-off-by: John Fastabend --- include/net/sch_generic.h |1 + net/sched/sch_generic.c | 50 - 2 fil

Re: RCU callback crashes

On 12/19/2017 10:34 PM, Jakub Kicinski wrote: > On Tue, 19 Dec 2017 22:22:27 -0800, Jakub Kicinski wrote: I get this: >>> >>> Could you try to run it with kasan on? >> >> I didn't manage to reproduce it with KASAN on so far :( Even enabling >> object debugging to get the second splat in

Re: [Patch net-next] net_sched: properly check for empty skb array on error path

On 12/18/2017 08:31 PM, Cong Wang wrote: > On Mon, Dec 18, 2017 at 7:58 PM, John Fastabend > wrote: >> On 12/18/2017 06:20 PM, Cong Wang wrote: >>> On Mon, Dec 18, 2017 at 5:25 PM, John Fastabend >>> wrote: >>>> On 12/18/2017 02:34 PM, Cong Wang wrote

Re: [Patch net-next] net_sched: properly check for empty skb array on error path

On 12/18/2017 06:20 PM, Cong Wang wrote: > On Mon, Dec 18, 2017 at 5:25 PM, John Fastabend > wrote: >> On 12/18/2017 02:34 PM, Cong Wang wrote: >>> First, the check of &q->ring.queue against NULL is wrong, it >>> is always false. We should check the value rath

Re: [Patch net-next] net_sched: properly check for empty skb array on error path

: pfifo_fast use skb_array") > Reported-by: syzbot > Cc: John Fastabend > Signed-off-by: Cong Wang > --- > net/sched/sch_generic.c | 8 +++- > 1 file changed, 7 insertions(+), 1 deletion(-) >

Re: ixgbe tuning reset when XDP is setup

On 12/15/2017 07:53 AM, David Miller wrote: > From: Eric Leblond > Date: Fri, 15 Dec 2017 11:24:46 +0100 > >> Hello, >> >> When using an ixgbe card with Suricata we are using the following >> commands to get a symmetric hash on RSS load balancing: >> >> ./set_irq_affinity 0-15 eth3 >> ethtool -X

Re: Huge memory leak with 4.15.0-rc2+

On 12/12/2017 09:57 AM, Paweł Staszewski wrote: > > > W dniu 2017-12-11 o 23:27, Paweł Staszewski pisze: >> >> >> W dniu 2017-12-11 o 23:15, John Fastabend pisze: >>> On 12/11/2017 01:48 PM, Paweł Staszewski wrote: >>>> >>>>

Re: Huge memory leak with 4.15.0-rc2+

On 12/11/2017 01:48 PM, Paweł Staszewski wrote: > > > W dniu 2017-12-11 o 22:23, Paweł Staszewski pisze: >> Hi >> >> >> I just upgraded some testing host to 4.15.0-rc2+ kernel >> >> And after some time of traffic processing - when traffic on all ports >> reach about 3Mpps - memleak started. >>

[net-next PATCH 14/14] net: sched: pfifo_fast use skb_array

graft operation occurs. This also removes the logic used to pick the next band to dequeue from and instead just checks a per priority array for packets from top priority to lowest. This might need to be a bit more clever but seems to work for now. Signed-off-by: John Fastabend --- net/sched

[net-next PATCH 12/14] net: sched: add support for TCQ_F_NOLOCK subqueues to sch_mqprio

andle this case add a check when calculating stats and aggregate the per cpu stats if needed. Signed-off-by: John Fastabend --- net/sched/sch_mq.c | 35 +++- net/sched/sch_mqprio.c | 69 +--- 2 files changed, 69 insertions(+), 35

[net-next PATCH 13/14] net: skb_array: expose peek API

This adds a peek routine to skb_array.h for use with qdisc. Signed-off-by: John Fastabend --- include/linux/skb_array.h |5 + 1 file changed, 5 insertions(+) diff --git a/include/linux/skb_array.h b/include/linux/skb_array.h index 8621ffd..c7addf3 100644 --- a/include/linux/skb_array.h

[net-next PATCH 07/14] net: sched: drop qdisc_reset from dev_graft_qdisc

ce period and letting the qdisc_destroy operation clean up the qdisc correctly. Note, a refcnt greater than 1 would cause the destroy operation to be aborted however if this ever happened the reference to the qdisc would be lost and we would have a memory leak. Signed-off-by: John Fastabend --- net/s

[net-next PATCH 11/14] net: sched: add support for TCQ_F_NOLOCK subqueues to sch_mq

this case add a check when calculating stats and aggregate the per cpu stats if needed. Also exports __gnet_stats_copy_queue() to use as a helper function. Signed-off-by: John Fastabend --- include/net/gen_stats.h |3 +++ net/core/gen_stats.c|9 + net/sched/sch_mq.c

[net-next PATCH 10/14] net: sched: helpers to sum qlen and qlen for per cpu logic

-off-by: John Fastabend --- include/net/sch_generic.h | 20 net/sched/sch_api.c |3 ++- 2 files changed, 22 insertions(+), 1 deletion(-) diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index 4717c4b..2fbae2c9 100644 --- a/include/net

[net-next PATCH 09/14] net: sched: check for frozen queue before skb_bad_txq check

I can not think of any reason to pull the bad txq skb off the qdisc if the txq we plan to send this on is still frozen. So check for frozen queue first and abort before dequeuing either skb_bad_txq skb or normal qdisc dequeue() skb. Signed-off-by: John Fastabend --- net/sched/sch_generic.c

[net-next PATCH 08/14] net: sched: use skb list for skb_bad_tx

Similar to how gso is handled use skb list for skb_bad_tx this is required with lockless qdiscs because we may have multiple cores attempting to push skbs into skb_bad_tx concurrently Signed-off-by: John Fastabend --- include/net/sch_generic.h |2 - net/sched/sch_generic.c | 106

[net-next PATCH 06/14] net: sched: explicit locking in gso_cpu fallback

once its possible to have multiple sk_buffs here so we turn gso_skb into a queue. This should be the edge case and if we see this frequently then the netdev/qdisc layer needs to back off. Signed-off-by: John Fastabend --- include/net/sch_generic.h | 20 ++- net/sched/sch_generic.c

[net-next PATCH 05/14] net: sched: a dflt qdisc may be used with per cpu stats

ate the qdisc object so we don't have dangling allocations after qdisc init. Signed-off-by: John Fastabend --- include/net/sch_generic.h |1 + net/sched/sch_generic.c | 16 2 files changed, 17 insertions(+) diff --git a/include/net/sch_generic.h b/include/net/sch_g

[net-next PATCH 03/14] net: sched: remove remaining uses for qdisc_qlen in xmit path

ow it returns true. However in this case all call sites of sch_direct_xmit will implement a dequeue() and get a null skb and abort. This trades tracking qlen in the hotpath for an extra dequeue operation. Overall this seems to be good for performance. Signed-off-by: John Fastabend --- includ

[net-next PATCH 04/14] net: sched: provide per cpu qstat helpers

The per cpu qstats support was added with per cpu bstat support which is currently used by the ingress qdisc. This patch adds a set of helpers needed to make other qdiscs that use qstats per cpu as well. Signed-off-by: John Fastabend --- include/net/sch_generic.h | 35

[net-next PATCH 02/14] net: sched: allow qdiscs to handle locking

doing the enqueue/dequeue operations when tested with pktgen. Signed-off-by: John Fastabend --- include/net/sch_generic.h |1 + net/core/dev.c| 26 ++ net/sched/sch_generic.c | 30 -- 3 files changed, 43 insertions(+), 14

[net-next PATCH 01/14] net: sched: cleanup qdisc_run and __qdisc_run semantics

Currently __qdisc_run calls qdisc_run_end() but does not call qdisc_run_begin(). This makes it hard to track pairs of qdisc_run_{begin,end} across function calls. To simplify reading these code paths this patch moves begin/end calls into qdisc_run(). Signed-off-by: John Fastabend --- include

[net-next PATCH 00/14] lockless qdisc series

m, I left out lockdep annotation for a follow on series to add lockdep more completely, rather than just in code I touched. Comments and feedback welcome. Thanks, John --- John Fastabend (14): net: sched: cleanup qdisc_run and __qdisc_run semantics net: sched: allow qdiscs to handle locking

Re: [PATCH RFC 0/5] Support asynchronous crypto for IPsec GSO.

On 11/20/2017 05:09 AM, David Miller wrote: > From: Steffen Klassert > Date: Mon, 20 Nov 2017 08:37:47 +0100 > >> This patchset implements asynchronous crypto handling >> in the layer 2 TX path. With this we can allow IPsec >> ESP GSO for software crypto. This also merges the IPsec >> GSO and non

Re: [RFC PATCH 06/17] net: sched: explicit locking in gso_cpu fallback

On 11/15/2017 09:51 AM, Willem de Bruijn wrote: > On Wed, Nov 15, 2017 at 10:11 AM, John Fastabend > wrote: >> On 11/14/2017 04:41 PM, Willem de Bruijn wrote: >>>> /* use instead of qdisc->dequeue() for all qdiscs queried with ->peek() */ >>>> static i

Re: [RFC PATCH 10/17] net: sched: qdisc_qlen for per cpu logic

On 11/14/2017 05:16 PM, Willem de Bruijn wrote: > On Mon, Nov 13, 2017 at 3:10 PM, John Fastabend > wrote: >> Add qdisc qlen helper routines for lockless qdiscs to use. >> >> The qdisc qlen is no longer used in the hotpath but it is reported >> via stats query on the

Re: [RFC PATCH 06/17] net: sched: explicit locking in gso_cpu fallback

On 11/14/2017 04:41 PM, Willem de Bruijn wrote: >> /* use instead of qdisc->dequeue() for all qdiscs queried with ->peek() */ >> static inline struct sk_buff *qdisc_dequeue_peeked(struct Qdisc *sch) >> { >> - struct sk_buff *skb = sch->gso_skb; >> + struct sk_buff *skb = skb_peek(&sc

Re: [RFC PATCH 03/17] net: sched: remove remaining uses for qdisc_qlen in xmit path

On 11/14/2017 05:56 PM, Willem de Bruijn wrote: > On Tue, Nov 14, 2017 at 7:11 PM, Willem de Bruijn > wrote: >> On Mon, Nov 13, 2017 at 3:08 PM, John Fastabend >> wrote: >>> sch_direct_xmit() uses qdisc_qlen as a return value but all call sites >>> of the rou

Re: [RFC PATCH 15/17] net: sched: pfifo_fast use skb_array

[...] >> static int pfifo_fast_dump(struct Qdisc *qdisc, struct sk_buff *skb) >> @@ -685,17 +688,40 @@ static int pfifo_fast_dump(struct Qdisc *qdisc, struct >> sk_buff *skb) >> >> static int pfifo_fast_init(struct Qdisc *qdisc, struct nlattr *opt) >> { >> - int prio; >> + unsigned

Re: [RFC PATCH 00/14] Introducing AF_PACKET V4 support

On 11/13/2017 09:33 PM, Björn Töpel wrote: > 2017-11-14 0:50 GMT+01:00 Alexei Starovoitov : >> On 11/13/17 9:07 PM, Björn Töpel wrote: >>> >>> 2017-10-31 13:41 GMT+01:00 Björn Töpel : From: Björn Töpel >>> [...] We'll do a presentation on AF_PACKET V4 in NetDev 2.2 [1

Re: Per-CPU Queueing for QoS

On 11/13/2017 06:18 PM, Michael Ma wrote: > 2017-11-13 16:13 GMT-08:00 Alexander Duyck : >> On Mon, Nov 13, 2017 at 3:08 PM, Eric Dumazet wrote: >>> On Mon, 2017-11-13 at 14:47 -0800, Alexander Duyck wrote: On Mon, Nov 13, 2017 at 10:17 AM, Michael Ma wrote: > 2017-11-12 16:14 GMT-08:00

[RFC PATCH 17/17] net: sched: lock once per bulk dequeue

Signed-off-by: John Fastabend --- 0 files changed diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index 683f6ec..8ab7933 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -206,33 +206,22 @@ static struct sk_buff *dequeue_skb(struct Qdisc *q, bool *validate

[RFC PATCH 16/17] net: skb_array additions for unlocked consumer

Signed-off-by: John Fastabend --- 0 files changed diff --git a/include/linux/skb_array.h b/include/linux/skb_array.h index c7addf3..d0a240e 100644 --- a/include/linux/skb_array.h +++ b/include/linux/skb_array.h @@ -142,6 +142,11 @@ static inline int skb_array_consume_batched_bh(struct

[RFC PATCH 14/17] net: skb_array: expose peek API

This adds a peek routine to skb_array.h for use with qdisc. Signed-off-by: John Fastabend --- 0 files changed diff --git a/include/linux/skb_array.h b/include/linux/skb_array.h index 8621ffd..c7addf3 100644 --- a/include/linux/skb_array.h +++ b/include/linux/skb_array.h @@ -72,6 +72,11

[RFC PATCH 15/17] net: sched: pfifo_fast use skb_array

clever but seems to work for now. Signed-off-by: John Fastabend --- 0 files changed diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index 84c4ea1..683f6ec 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -26,6 +26,7 @@ #include #include #include +#include

[RFC PATCH 13/17] net: sched: add support for TCQ_F_NOLOCK subqueues to sch_mqprio

andle this case add a check when calculating stats and aggregate the per cpu stats if needed. Signed-off-by: John Fastabend --- 0 files changed diff --git a/net/sched/sch_mqprio.c b/net/sched/sch_mqprio.c index b85885a9..24474d0 100644 --- a/net/sched/sch_mqprio.c +++ b/net/sched/sch_mqprio.c

<    7   8   9   10   11   12   13   14   15   16   >