Adds iterator skb_queue_for_each() to run over a list without modifying it.
Signed-off-by: Edward Cree
---
include/linux/skbuff.h | 16
include/trace/events/net.h | 7 +++
net/core/dev.c | 4 +++-
3 files changed, 26 insertions(+), 1 deletion(-)
diff
netif_receive_skb_list_internal() now processes a list and hands it
on to the next function.
Signed-off-by: Edward Cree
---
net/core/dev.c | 73 ++
1 file changed, 69 insertions(+), 4 deletions(-)
diff --git a/net/core/dev.c b/net/core
First example of a layer splitting the list (rather than merely taking
individual packets off it).
Signed-off-by: Edward Cree
---
net/core/dev.c | 46 ++
1 file changed, 38 insertions(+), 8 deletions(-)
diff --git a/net/core/dev.c b/net/core/dev.c
e one ptype found in ptype_base[hash of skb->protocol].
Signed-off-by: Edward Cree
---
include/trace/events/net.h | 7 +++
net/core/dev.c | 138 -
2 files changed, 105 insertions(+), 40 deletions(-)
diff --git a/include/trace/events/net.h
ccept to cause out-of-order receives, so presumably this is
considered OK.
Signed-off-by: Edward Cree
---
include/linux/netdevice.h | 3 ++
include/linux/netfilter.h | 27 +
include/net/ip.h | 2 ++
net/core/dev.c| 11 +--
net/ipv4/af_inet.c
ing on protocol changes).
Signed-off-by: Edward Cree
---
net/ipv4/ip_input.c | 58 -
1 file changed, 53 insertions(+), 5 deletions(-)
diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c
index 7a8af8ff3f07..63d4dfdb1766 100644
--- a/ne
Generally the check should be very cheap, as the sk_buff_head is in cache.
Signed-off-by: Edward Cree
---
net/core/dev.c | 8 ++--
net/ipv4/ip_input.c | 2 ++
2 files changed, 8 insertions(+), 2 deletions(-)
diff --git a/net/core/dev.c b/net/core/dev.c
index f0eb00e9fb57..11f80d4502b9
Deals with all the pre- and post-amble to the BPF program itself, which is
still called one packet at a time.
Involves some fiddly percpu variables to cope with XDP_REDIRECT handling.
Signed-off-by: Edward Cree
---
include/linux/filter.h | 10 +++
net/core/dev.c | 165
extra
indirect calls.
Signed-off-by: Edward Cree
---
include/linux/filter.h | 38 +-
kernel/bpf/core.c | 26 ++
net/core/dev.c | 19 ---
3 files changed, 63 insertions(+), 20 deletions(-)
diff --git a
list_func as a parameter instead of calculating it
inside the loop. This is safe because BPF execution is not preemptible,
so the percpu variable can't get moved while we're using it.
Signed-off-by: Edward Cree
---
arch/x86/net/bpf_jit_comp.c | 164 +
On 27/06/18 01:06, Eric Dumazet wrote:
> On 06/26/2018 11:17 AM, Edward Cree wrote:
>> Just calls netif_receive_skb() in a loop.
> ...
>
>> +void netif_receive_skb_list(struct sk_buff_head *list)
>
> Please use a standard list_head and standard list operators.
>
>
On 27/06/18 15:36, Willem de Bruijn wrote:
> On Tue, Jun 26, 2018 at 8:19 PM Edward Cree wrote:
>> __netif_receive_skb_taps() does a depressingly large amount of per-packet
>> work that can't easily be listified, because the another_round looping
>> makes it nontrivia
On 27/06/18 17:00, Willem de Bruijn wrote:
> On Wed, Jun 27, 2018 at 10:49 AM Edward Cree wrote:
>> On 27/06/18 15:36, Willem de Bruijn wrote:
>>> Also, this function does more than just process network taps.
>> This is true, but naming things is hard, and I couldn't
istified Generic XDP handling (patches 10-12), though it doesn't
seem to help (see above).
* Extended testing to cover software firewalls / netfilter etc.
[1] http://vger.kernel.org/netconf2018_files/DavidMiller_netconf2018.pdf
[2] http://vger.kernel.org/netconf2018_files/EdwardCree_netcon
Just calls netif_receive_skb() in a loop.
Signed-off-by: Edward Cree
---
include/linux/netdevice.h | 1 +
net/core/dev.c| 19 +++
2 files changed, 20 insertions(+)
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index c6b377a15869..e104b2e4a735
Improves packet rate of 1-byte UDP receives by up to 10%.
Signed-off-by: Edward Cree
---
drivers/net/ethernet/sfc/efx.c| 12
drivers/net/ethernet/sfc/net_driver.h | 3 +++
drivers/net/ethernet/sfc/rx.c | 7 ++-
3 files changed, 21 insertions(+), 1 deletion
Signed-off-by: Edward Cree
---
include/trace/events/net.h | 7 +++
net/core/dev.c | 4 +++-
2 files changed, 10 insertions(+), 1 deletion(-)
diff --git a/include/trace/events/net.h b/include/trace/events/net.h
index 9c886739246a..00aa72ce0e7c 100644
--- a/include/trace/events
netif_receive_skb_list_internal() now processes a list and hands it
on to the next function.
Signed-off-by: Edward Cree
---
net/core/dev.c | 61 +-
1 file changed, 56 insertions(+), 5 deletions(-)
diff --git a/net/core/dev.c b/net/core
First example of a layer splitting the list (rather than merely taking
individual packets off it).
Involves new list.h function, list_cut_before(), like list_cut_position()
but cuts on the other side of the given entry.
Signed-off-by: Edward Cree
---
include/linux/list.h | 30
ccept to cause out-of-order receives, so presumably this is
considered OK.
Signed-off-by: Edward Cree
---
include/linux/netdevice.h | 3 +++
include/linux/netfilter.h | 25 +
include/net/ip.h | 2 ++
net/core/dev.c| 8 +++---
net/ipv4/af_inet.c
ing on protocol changes).
Signed-off-by: Edward Cree
---
net/ipv4/ip_input.c | 54 +++--
1 file changed, 48 insertions(+), 6 deletions(-)
diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c
index 914240830bdf..24b9b0210aeb 100644
--- a/ne
e one ptype found in ptype_base[hash of skb->protocol].
Signed-off-by: Edward Cree
---
net/core/dev.c | 72 +++---
1 file changed, 64 insertions(+), 8 deletions(-)
diff --git a/net/core/dev.c b/net/core/dev.c
index d2454678bc82..edd67b1f1e12 1006
Generally the check should be very cheap, as the sk_buff_head is in cache.
Signed-off-by: Edward Cree
---
net/core/dev.c | 8 ++--
1 file changed, 6 insertions(+), 2 deletions(-)
diff --git a/net/core/dev.c b/net/core/dev.c
index 4c5ebfab9bc8..d6084b0cd9ce 100644
--- a/net/core/dev.c
+++ b
On 29/06/18 23:08, kbuild test robot wrote:
>net//ipv4/ip_input.c: In function 'ip_sublist_rcv':
>>> net//ipv4/ip_input.c:524:14: warning: passing argument 6 of 'NF_HOOK_LIST'
>>> from incompatible pointer type
>head, dev, NULL, ip_rcv_finish);
> ^
>In file inc
_files/DavidMiller_netconf2018.pdf
[2] http://vger.kernel.org/netconf2018_files/EdwardCree_netconf2018.pdf
Edward Cree (9):
net: core: trivial netif_receive_skb_list() entry point
sfc: batch up RX delivery
net: core: unwrap skb list receive slightly further
net: core: Another step of skb receive list proces
Just calls netif_receive_skb() in a loop.
Signed-off-by: Edward Cree
---
include/linux/netdevice.h | 1 +
net/core/dev.c| 19 +++
2 files changed, 20 insertions(+)
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index c6b377a15869..e104b2e4a735
Signed-off-by: Edward Cree
---
include/trace/events/net.h | 7 +++
net/core/dev.c | 4 +++-
2 files changed, 10 insertions(+), 1 deletion(-)
diff --git a/include/trace/events/net.h b/include/trace/events/net.h
index 9c886739246a..00aa72ce0e7c 100644
--- a/include/trace/events
First example of a layer splitting the list (rather than merely taking
individual packets off it).
Involves new list.h function, list_cut_before(), like list_cut_position()
but cuts on the other side of the given entry.
Signed-off-by: Edward Cree
---
include/linux/list.h | 30
Improves packet rate of 1-byte UDP receives by up to 10%.
Signed-off-by: Edward Cree
---
drivers/net/ethernet/sfc/efx.c| 12
drivers/net/ethernet/sfc/net_driver.h | 3 +++
drivers/net/ethernet/sfc/rx.c | 7 ++-
3 files changed, 21 insertions(+), 1 deletion
netif_receive_skb_list_internal() now processes a list and hands it
on to the next function.
Signed-off-by: Edward Cree
---
net/core/dev.c | 61 +-
1 file changed, 56 insertions(+), 5 deletions(-)
diff --git a/net/core/dev.c b/net/core
e one ptype found in ptype_base[hash of skb->protocol].
Signed-off-by: Edward Cree
---
net/core/dev.c | 72 +++---
1 file changed, 64 insertions(+), 8 deletions(-)
diff --git a/net/core/dev.c b/net/core/dev.c
index d2454678bc82..edd67b1f1e12 1006
ccept to cause out-of-order receives, so presumably this is
considered OK.
Signed-off-by: Edward Cree
---
include/linux/netdevice.h | 3 +++
include/linux/netfilter.h | 22 +++
include/net/ip.h | 2 ++
net/core/dev.c| 8 +++---
net/ipv4/af_inet.c| 1 +
Generally the check should be very cheap, as the sk_buff_head is in cache.
Signed-off-by: Edward Cree
---
net/core/dev.c | 8 ++--
1 file changed, 6 insertions(+), 2 deletions(-)
diff --git a/net/core/dev.c b/net/core/dev.c
index 4c5ebfab9bc8..d6084b0cd9ce 100644
--- a/net/core/dev.c
+++ b
ing on protocol changes).
Signed-off-by: Edward Cree
---
net/ipv4/ip_input.c | 54 +++--
1 file changed, 48 insertions(+), 6 deletions(-)
diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c
index 914240830bdf..24b9b0210aeb 100644
--- a/ne
On 02/07/18 16:40, David Ahern wrote:
> Nice work. Have you looked at IPv6 support yet?
I hadn't looked at it yet, no. After a quick glance at ip6_rcv() and
ip6_rcv_finish(), it looks like it'd be basically the same as the IPv4
code in patches 7 and 8. I'll probably add it in a followup when (
On 02/07/18 16:14, Edward Cree wrote:
> +/* Receive a list of IP packets */
> +void ip_list_rcv(struct list_head *head, struct packet_type *pt,
> + struct net_device *orig_dev)
> +{
> + struct net_device *curr_dev = NULL;
> + struct net *curr_net = NULL;
>
ource list before processing, and add it
to the sublist afterwards if it wasn't freed or stolen.
Fixes: 5fa12739a53d net: ipv4: listify ip_rcv_finish
Fixes: 17266ee93984 net: ipv4: listified version of ip_rcv
Signed-off-by: Edward Cree
---
net/ipv4/ip_input.c | 16 +++-
1 file
While working on IPv6 list processing, I found another bug in the IPv4
version. So this patch series has that fix, and the IPv6 version with
both fixes incorporated.
Edward Cree (2):
net: ipv4: fix list processing on L3 slave devices
net: ipv6: listify ipv6_rcv() and ip6_rcv_finish
l can be different in the single packet path (return
NET_RX_SUCCESS) and the list path (forget this packet and continue).
Fixes: 5fa12739a53d net: ipv4: listify ip_rcv_finish
Signed-off-by: Edward Cree
---
net/ipv4/ip_input.c | 23 +++
1 file changed, 15 insertions(+), 8 dele
Essentially the same as the ipv4 equivalents.
Signed-off-by: Edward Cree
---
include/net/ipv6.h | 2 +
net/ipv6/af_inet6.c | 1 +
net/ipv6/ip6_input.c | 131 ---
3 files changed, 118 insertions(+), 16 deletions(-)
diff --git a/include/net
adds a list.h helper macro to refactor code using the dequeue-enqueue
model.
Edward Cree (3):
net: core: fix uses-after-free in list processing
netfilter: fix use-after-free in NF_HOOK_LIST
net: refactor dequeue-model list processing
include/linux/list.h | 15 +++
include
nf_hook() can free the skb, so we need to remove it from the list before
calling, and add passed skbs to a sublist afterwards.
Fixes: 17266ee93984 ("net: ipv4: listified version of ip_rcv")
Reported-by: Dan Carpenter
Signed-off-by: Edward Cree
---
include/linux/netfilter.h | 10 +++
New macro list_for_each_entry_dequeue loops over a list by popping entries
from the head, allowing a more concise expression of the dequeue-enqueue
model of list processing and avoiding the need for a 'next' pointer (as
used in list_for_each_entry_safe).
Signed-off-by: E
afterwards.
In the case of enqueue_to_backlog() we have already decided not to pass
when we call the function, so we do not need a sublist.
Fixes: 7da517a3bc52 ("net: core: Another step of skb receive list processing")
Reported-by: Dan Carpenter
Signed-off-by: Edward Cree
---
net/core/
patch #3 (new list.h helper), per DaveM's request.
Edward Cree (2):
net: core: fix uses-after-free in list processing
netfilter: fix use-after-free in NF_HOOK_LIST
include/linux/netfilter.h | 10 +++---
net/core/dev.c| 21 +
2 files changed, 20 inser
afterwards.
In the case of enqueue_to_backlog() we have already decided not to pass
when we call the function, so we do not need a sublist.
Fixes: 7da517a3bc52 ("net: core: Another step of skb receive list processing")
Reported-by: Dan Carpenter
Signed-off-by: Edward Cree
---
net/core/
nf_hook() can free the skb, so we need to remove it from the list before
calling, and add passed skbs to a sublist afterwards.
Fixes: 17266ee93984 ("net: ipv4: listified version of ip_rcv")
Reported-by: Dan Carpenter
Signed-off-by: Edward Cree
---
include/linux/netfilter.h | 10 +++
another similar bug in __netif_receive_skb_list_core().
v2: dropped patch #3 (new list.h helper), per DaveM's request.
Edward Cree (3):
net: core: fix uses-after-free in list processing
netfilter: fix use-after-free in NF_HOOK_LIST
net: core: fix use-after-free in __netif_receive_skb_list
afterwards.
In the case of enqueue_to_backlog() we have already decided not to pass
when we call the function, so we do not need a sublist.
Fixes: 7da517a3bc52 ("net: core: Another step of skb receive list processing")
Reported-by: Dan Carpenter
Signed-off-by: Edward Cree
---
net/core/
nf_hook() can free the skb, so we need to remove it from the list before
calling, and add passed skbs to a sublist afterwards.
Fixes: 17266ee93984 ("net: ipv4: listified version of ip_rcv")
Reported-by: Dan Carpenter
Signed-off-by: Edward Cree
---
include/linux/netfilter.h | 10 +++
__netif_receive_skb_core can free the skb, so we have to use the dequeue-
enqueue model when calling it from __netif_receive_skb_list_core.
Fixes: 88eb1944e18c ("net: core: propagate SKB lists through packet_type
lookup")
Signed-off-by: Edward Cree
---
net/core/dev.c | 9 +++
o ip_forward.
>
> Fixes: 5fa12739a53d ("net: ipv4: listify ip_rcv_finish")
> Signed-off-by: Jesper Dangaard Brouer
Acked-by: Edward Cree
But it feels weird and asymmetric to only NULL skb->next (not ->prev), and
to have to do this by hand rather than e.g. being able to use
list
On 12/07/18 21:10, Or Gerlitz wrote:
> On Wed, Jul 11, 2018 at 11:06 PM, Jesper Dangaard Brouer
> wrote:
>> One reason I didn't "just" send a patch, is that Edward so-fare only
>> implemented netif_receive_skb_list() and not napi_gro_receive_list().
> sfc does't support gro?! doesn't make sense..
ntext with an equal-weight [0,1]
indirection table could be used to target all three containers by setting
ring_cookie to 1, 3 and 6 on the respective filters.
v2: Initialised ctx in efx_ef10_filter_insert() to avoid (false positive) gcc
warning.
Edward Cree (2):
net: ethtool: extend RXNFC A
-off-by: Edward Cree
---
include/linux/ethtool.h | 5
include/uapi/linux/ethtool.h | 32 +-
net/core/ethtool.c | 64 +---
3 files changed, 80 insertions(+), 21 deletions(-)
diff --git a/include/linux/ethtool.h b
Use a linked list to associate user-facing context IDs with FW-facing
context IDs (since the latter can change after an MC reset).
Signed-off-by: Edward Cree
---
drivers/net/ethernet/sfc/ef10.c | 273 ++
drivers/net/ethernet/sfc/efx.c| 65
This series adds support for the management of extra RSS contexts and RSS
spreading of ntuple/NFC filters introduced in kernel commit 84a1d9c48200
("net: ethtool: extend RXNFC API to support RSS spreading of filter matches").
Edward Cree (3):
ethtool-copy.h: sync with net-next
et
This covers kernel changes up to commit 84a1d9c48200,
net: ethtool: extend RXNFC API to support RSS spreading of filter matches
Signed-off-by: Edward Cree
---
ethtool-copy.h | 32 ++--
1 file changed, 26 insertions(+), 6 deletions(-)
diff --git a/ethtool-copy.h
Signed-off-by: Edward Cree
---
ethtool.8.in | 28
1 file changed, 28 insertions(+)
diff --git a/ethtool.8.in b/ethtool.8.in
index 690e55e..99ac62f 100644
--- a/ethtool.8.in
+++ b/ethtool.8.in
@@ -274,6 +274,7 @@ ethtool \- query or control network driver and
RSS contexts can be created on a device with -X ... context new, modified
with -X ... context N, and deleted with -X ... context N delete.
N-tuple filters can be directed at those contexts with -N ... context N.
Signed-off-by: Edward Cree
---
ethtool.c | 141
On 09/03/18 18:58, Alexei Starovoitov wrote:
> It's not waiting for the whole thing, because once bpfilter starts it
> stays running/sleeping because it's stateful.
So, this has been bugging me a bit.
If bpfilter takes a signal and crashes, all that state goes away.
Does that mean your iptables/net
Implements the ethtool get & set fecparam operations.
Edward Cree (2):
sfc: update MCDI protocol headers
sfc: support FEC configuration through ethtool
drivers/net/ethernet/sfc/ethtool.c| 32 +
drivers/net/ethernet/sfc/mcdi_pcol.h | 2822 +
drivers
le or link partner doesn't support
it. rs|baser (with or without auto bit) means prefer FEC even where
auto wouldn't use it, but let FW choose which encoding to use.
Signed-off-by: Edward Cree
---
drivers/net/ethernet/sfc/ethtool.c| 32
drivers/net/ethernet/sfc/mcdi_port
In do_grfxh(), if rss_context is non-zero, we must not fall back to
do_grxfhindir() if ETHTOOL_GRSSH fails, for that reads the default RSS
context and not the one specified.
Fixes: f5d55b967e0c ("ethtool: add support for extra RSS contexts and RSS
steering filters")
Signed-off-by: E
iveness
marking either.
Edward Cree (3):
bpf/verifier: validate func_calls by marking at do_check() time
bpf/verifier: update selftests
bpf/verifier: per-register parent pointers
include/linux/bpf_verifier.h| 32 +-
kernel/bpf/verifier.c
re determined).
Signed-off-by: Edward Cree
---
tools/testing/selftests/bpf/test_verifier.c | 51 ++---
1 file changed, 32 insertions(+), 19 deletions(-)
diff --git a/tools/testing/selftests/bpf/test_verifier.c
b/tools/testing/selftests/bpf/test_verifier.c
index 3e7718b1a9ae..c
ck is now O(n) in the number of subprogs, rather than having to walk
every insn of every possible call chain.
Signed-off-by: Edward Cree
---
include/linux/bpf_verifier.h | 24 ++-
kernel/bpf/verifier.c| 451 ---
2 files changed, 267 insertions(+)
(r1-r5 into the callee as args, and r0 out as the return value) they also
copy the parent pointer.
Signed-off-by: Edward Cree
---
include/linux/bpf_verifier.h | 8 +-
kernel/bpf/verifier.c| 180 ++-
2 files changed, 45 insertions(+), 143 dele
On 29/03/18 23:44, Edward Cree wrote:
> By storing subprog boundaries as a subprogno mark on each insn, rather than
> a start (and implicit end) for each subprog, we collect a number of gains:
> * More efficient determination of which subprog contains a given insn, and
> thus of
On 03/04/18 02:08, Alexei Starovoitov wrote:
> I like patch 3 and going to play with it.
> How did you test it ?
Just test_verifier and test_progs (the latter has a failure
"test_bpf_obj_id:FAIL:get-prog-info(fd) err 0 errno 2 i 0 type 1(1) info_len
80(40) jit_enabled 0 jited_prog_len 0 xlated_pr
On 04/04/18 00:37, Alexei Starovoitov wrote:
> hmm. that doesn't fail for me and any other bots didn't complain.
> Are you sure you're running the latest kernel and tests?
Ah, test_progs isn't actually rebuilding because __NR_bpf is undeclared;
something must be going wrong with header files.
Neve
On 05/04/18 06:28, Alexei Starovoitov wrote:
> On Thu, Apr 05, 2018 at 12:58:46AM +0100, Edward Cree wrote:
>> On 04/04/18 00:37, Alexei Starovoitov wrote:
>>> hmm. that doesn't fail for me and any other bots didn't complain.
>>> Are you sure you're
On 05/04/18 16:50, Jiong Wang wrote:
> On 03/04/2018 02:08, Alexei Starovoitov wrote:
>> Combining subprog pass with do_check is going into opposite direction
>> of this long term work. Divide and conquer. Combining more things into
>> do_check is the opposite of this programming principle.
>
> Agr
rogs, since that's no longer needed.
Changes from v1->v2:
* No longer allows non-contiguous subprogs.
* No longer allows LD_ABS|IND and pseudo-calls in the same prog.
Edward Cree (5):
bpf/verifier: store subprog info in struct bpf_subprog_info
bpf/verifier: rewrite subprog boundary dete
ALL insns are unreachable. I've
changed it so that they are now reachable, which causes static back-edges
to be detected (since that, like insn reachability, is now tested before
subprog boundaries are determined).
Signed-off-by: Edward Cree
---
tools/testing/self
By storing a subprogno in each insn's aux data, we avoid the need to keep
the list of subprog starts sorted or bsearch() it in find_subprog().
Also, get rid of the weird one-based indexing of subprog numbers.
Signed-off-by: Edward Cree
---
include/linux/bpf_verifier.h | 3 +-
kerne
Per-subprog info is stored in env->subprog_info, an array of structs,
rather than multiple arrays with a common index.
Signed-off-by: Edward Cree
---
include/linux/bpf_verifier.h | 9 +---
kernel/bpf/verifier.c| 49 +++-
2 files changed,
(r1-r5 into the callee as args, and r0 out as the return value) they also
copy the parent pointer.
Signed-off-by: Edward Cree
---
include/linux/bpf_verifier.h | 8 +-
kernel/bpf/verifier.c| 180 ++-
2 files changed, 45 insertions(+), 143 dele
calculating the stack depth it implies for the
subprog's callees. This is O(number of subprogs).
The call graph is populated as part of the check_subprogs() pass.
Signed-off-by: Edward Cree
---
include/linux/bpf_verifier.h | 3 +
kernel/bpf/verifier.c
here's also the issue that we come up in the bad configuration by
default, but that too is a problem for another time.)
Edward Cree (2):
sfc: insert ARFS filters with replace_equal=true
sfc: limit ARFS workitems in flight per channel
drivers/net/ethernet/sfc/net_driver.h | 25 +
Necessary to allow redirecting a flow when the application moves.
Fixes: 3af0f34290f6 ("sfc: replace asynchronous filter operations")
Signed-off-by: Edward Cree
---
drivers/net/ethernet/sfc/rx.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/
-by: Edward Cree
---
drivers/net/ethernet/sfc/net_driver.h | 25 +++
drivers/net/ethernet/sfc/rx.c | 58 ++-
2 files changed, 55 insertions(+), 28 deletions(-)
diff --git a/drivers/net/ethernet/sfc/net_driver.h
b/drivers/net/ethernet/sfc
On 12/04/18 16:11, David Miller wrote:
> From: Edward Cree
> Date: Thu, 12 Apr 2018 15:02:50 +0100
>
>> A misconfigured system (e.g. with all interrupts affinitised to all CPUs)
>> may produce a storm of ARFS steering events. With the existing sfc ARFS
>> implemen
It turns out this may all be moot anyway: I figured out why I was seeing
ARFS storms and it wasn't the configuration issue I originally blamed.
My current ndo_rx_flow_steer() implementation, efx_filter_rfs(), returns
0 for success, but the caller expects a filter ID to be returned (which
we can'
On 13/04/18 13:36, Edward Cree wrote:
> It turns out this may all be moot anyway: I figured out why I was seeing
> ARFS storms and it wasn't the configuration issue I originally blamed.
Hmm, correction, while the fix I mentioned in my previous email is needed,
it doesn't prevent
On 13/04/18 16:03, David Miller wrote:
> Whilst you may not be able to program the filter into the hardware
> synchronously, you should be able to allocate the ID and get all of
> the software state setup.
That's what we were doing before commit 3af0f34290f6 ("sfc: replace
asynchronous filter oper
On 13/04/18 17:14, David Miller wrote:
> Is the issue that you learn about the hardware reset asynchronously,
> and therefore cannot determine if filter insertion programming
> happened afterwards and thus is still in the chip?
Yes, pretty much.
> You must have a table of all the entries, so that
'net'
material.
There's also the issue that we come up in the bad configuration that
triggers #3 by default, but that too is a problem for another time.
Edward Cree (3):
sfc: insert ARFS filters with replace_equal=true
sfc: pass the correctly bogus filter_id to rps_may_expi
Necessary to allow redirecting a flow when the application moves.
Fixes: 3af0f34290f6 ("sfc: replace asynchronous filter operations")
Signed-off-by: Edward Cree
---
drivers/net/ethernet/sfc/rx.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/
f0f34290f6 ("sfc: replace asynchronous filter operations")
Signed-off-by: Edward Cree
---
drivers/net/ethernet/sfc/ef10.c | 3 +--
drivers/net/ethernet/sfc/farch.c | 2 +-
2 files changed, 2 insertions(+), 3 deletions(-)
diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ether
should not negatively impact performance, because it is only
likely to be hit in cases where ARFS will be ineffective anyway.
Signed-off-by: Edward Cree
---
drivers/net/ethernet/sfc/net_driver.h | 25 +++
drivers/net/ethernet/sfc/rx.c | 58
The main object of this patch series is to support verification of eBPF
programs with bounded loops. Only bounds derived from JLT (unsigned <)
with the taken branch in the loop are supported, but it should be clear
how others could be added.
Testing of these changes has consisted only of test_v
Removes a couple of passes from the verifier, one to check subprogs don't
overlap etc., and one to compute max stack depth (which now is done by
topologically sorting the call graph).
Signed-off-by: Edward Cree
---
As noted in the cover letter, I haven't yet integrated the feedback
r than main()". I'm still not 100% sure that's right though.
Signed-off-by: Edward Cree
---
tools/testing/selftests/bpf/test_verifier.c | 46 ++---
1 file changed, 29 insertions(+), 17 deletions(-)
diff --git a/tools/testing/selftests/bpf/test_verifier
(r1-r5 into the callee as args, and r0 out as the return value) they also
copy the parent pointer.
Signed-off-by: Edward Cree
---
include/linux/bpf_verifier.h | 8 +-
kernel/bpf/verifier.c| 180 ++-
2 files changed, 45 insertions(+), 143 dele
This means each entry in the parentage chain can have its insn identified,
which will help to support bounded-loop handling later.
Signed-off-by: Edward Cree
---
include/linux/bpf_verifier.h | 6 --
kernel/bpf/verifier.c| 22 +++---
2 files changed, 15 insertions
ition is known to have been met.
Signed-off-by: Edward Cree
---
kernel/bpf/verifier.c | 53 ++-
1 file changed, 52 insertions(+), 1 deletion(-)
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index e7d898f4fd29..7a8ae633d0c3 100644
---
walked
before reaching the back-edge are valid.
Signed-off-by: Edward Cree
---
include/linux/bpf_verifier.h| 2 +
kernel/bpf/verifier.c | 280 +---
tools/testing/selftests/bpf/test_verifier.c | 12 +-
3 files changed, 97 insertions
hether the recursion was bounded.
Signed-off-by: Edward Cree
---
include/linux/bpf_verifier.h | 5 +
kernel/bpf/verifier.c| 295 +++
2 files changed, 244 insertions(+), 56 deletions(-)
diff --git a/include/linux/bpf_verifier.h b/include/linux/bpf_verifi
Mainly consists of tests that broke (or I expected to break) earlier
versions of the bounded-loop handling.
Also updated some existing tests to deal with changed error messages,
programs now being accepted etc.
Signed-off-by: Edward Cree
---
tools/testing/selftests/bpf/test_verifier.c | 198
201 - 300 of 1233 matches
Mail list logo
