On Wed, Apr 10, 2019 at 9:08 PM Y Song wrote:
>
> On Wed, Apr 10, 2019 at 12:21 AM Magnus Karlsson
> wrote:
> >
> > This patch removes the use of likely and unlikely in xsk.h since they
> > create a dependency on Linux headers as reported by several
> > users. There have also been reports that th
> On 11 Apr 2019, at 03:52, Jakub Kicinski wrote:
>
> On Wed, 10 Apr 2019 20:50:19 +0100, Jiong Wang wrote:
>> Register liveness infrastructure doesn't track register read width at the
>> moment, while the width information will be needed for the later 32-bit
>> safety analysis pass.
>>
>> Thi
> On 11 Apr 2019, at 04:13, Jakub Kicinski wrote:
>
> On Wed, 10 Apr 2019 20:50:20 +0100, Jiong Wang wrote:
>> @@ -2169,6 +2188,12 @@ static int check_mem_access(struct bpf_verifier_env
>> *env, int insn_idx, u32 regn
>> value_regno);
>>
On 10.04.2019 13:32, kavyasree.kotag...@microchip.com wrote:
> The VSC8514 PHY is a 4-ports PHY that is 10/100/1000BASE-T, 100BASE-FX,
> 1000BASE-X, can communicate with the MAC via QSGMII.
> The MAC interface protocol for each port within QSGMII can
> be either 1000BASE-X or SGMII, if the QSGMII M
On Wed, Apr 10, 2019 at 01:43:58PM +0200, Jesper Dangaard Brouer wrote:
> A lot of the performance gain comes from this patch.
>
> While analysing performance overhead it was found that the largest CPU
> stalls were caused when touching the struct page area. It is first read with
> a READ_ONCE fro
From: Simon Horman
Date: Tue, 9 Apr 2019 09:59:07 +0200
> Correct spelling of encapsulation.
> Found by inspection.
>
> Signed-off-by: Simon Horman
Applied, thanks Simon.
From: Jakub Kicinski
Date: Mon, 8 Apr 2019 17:59:50 -0700
> Unlike '&&' operator, the '&' does not have short-circuit
> evaluation semantics. IOW both sides of the operator always
> get evaluated. Fix the wrong operator in
> tls_is_sk_tx_device_offloaded(), which would lead to
> out-of-bounds
On Wed, Apr 10, 2019 at 01:43:47PM +0200, Jesper Dangaard Brouer wrote:
> The function build_skb() also have the responsibility to allocate and clear
> the SKB structure. Introduce a new function build_skb_around(), that moves
> the responsibility of allocation and clearing to the caller. This allo
On 4/10/19 9:38 PM, Peter Oskolkov wrote:
>>
>> Notice the "No route to host" errors.
>
> "No route to host" is OK: there are negative tests, as you requested a
> couple of months ago... :), and these tests correctly trigger "no
> route to host".
>
> This output basically tell me that the test pa
On 4/10/19 1:45 PM, Martin Lau wrote:
> On Wed, Apr 10, 2019 at 11:45:04AM -0700, David Ahern wrote:
@@ -2420,7 +2449,7 @@ static struct rt6_info *__ip6_route_redirect(struct
net *net,
int flags)
{
struct ip6rd_flowi *rdfl = (str
Your test output tells me that everything is OK - see below.
On Wed, Apr 10, 2019 at 9:17 PM David Ahern wrote:
>
> On 4/10/19 6:26 PM, Peter Oskolkov wrote:
> > On Wed, Apr 10, 2019 at 6:19 PM David Ahern wrote:
> >>
> >> On 4/3/19 8:43 AM, Peter Oskolkov wrote:
> >>> This patch adds tests vali
On 4/10/19 6:36 PM, Andrey Ignatov wrote:
> Reported in [1].
>
> With gcc 8.3.0 the following error is issued:
>
>cc -Ibpf@sta -I. -I.. -I.././include -I.././include/uapi
>-fdiagnostics-color=always -fsanitize=address,undefined
> -fno-omit-frame-pointer
>-pipe -D_FILE_OFFSET_BITS=6
On 4/10/19 6:26 PM, Peter Oskolkov wrote:
> On Wed, Apr 10, 2019 at 6:19 PM David Ahern wrote:
>>
>> On 4/3/19 8:43 AM, Peter Oskolkov wrote:
>>> This patch adds tests validating that VRF and BPF-LWT
>>> encap work together well, as requested by David Ahern.
>>>
>>> Signed-off-by: Peter Oskolkov
OK. But it means that you have a very specific encapsulation mode on
top
of the "netdev". I'm still not convinced we should actually make that a
netdev, but I'll elaborate elsewhere.
I recall Daniele also managed to get rmnet working with qmi_wwan
(with an additional patch in which I had made q
On Wed, 10 Apr 2019 20:50:25 +0100, Jiong Wang wrote:
> Enable BPF_F_TEST_RND_HI32 for all existing bpf selftests or other
> independent tests could involve quite a few changes to make sure all bpf
> prog load places has BPF_F_TEST_RND_HI32 set.
>
> Given most of the tests are using libbpf, this p
On Wed, 10 Apr 2019 20:50:20 +0100, Jiong Wang wrote:
> @@ -2169,6 +2188,12 @@ static int check_mem_access(struct bpf_verifier_env
> *env, int insn_idx, u32 regn
> value_regno);
> if (reg_type_may_be_null(reg_type))
>
Hi Leandro,
Thank you for the patch! Yet something to improve:
[auto build test ERROR on net-next/master]
url:
https://github.com/0day-ci/linux/commits/Leandro-Dorileo/net-sched-taprio-fix-picos_per_byte-miscalculation/20190410-083853
config: i386-randconfig-m3-201914 (attached as .config
On Wed, 10 Apr 2019 20:50:19 +0100, Jiong Wang wrote:
> Register liveness infrastructure doesn't track register read width at the
> moment, while the width information will be needed for the later 32-bit
> safety analysis pass.
>
> This patch take the first step to split read liveness into REG_LIV
On Wed, 10 Apr 2019 20:50:18 +0100, Jiong Wang wrote:
> There are a few "regs[regno]" here are there across "check_reg_arg", this
> patch factor it out into a simple "reg" pointer. The intention is to
> simplify code indentation and make the later patches in this set look
> cleaner.
>
> Signed-off
On Wed, 10 Apr 2019 20:50:17 +0100, Jiong Wang wrote:
> After code refactor in previous patches, the propagation logic inside the
> for loop in "propagate_liveness" becomes clear that they are good enough to
> be factored out into a common function "propagate_liveness_reg".
>
> Signed-off-by: Jion
On Wed, 10 Apr 2019 20:50:16 +0100, Jiong Wang wrote:
> Access to reg states were not factored out, the consequence is long code
> for dereferencing them which made the indentation not good for reading.
>
> This patch factor out these code so the core code in the loop could be
> easier to follow.
From: David Miller
Date: Wed, 10 Apr 2019 19:35:16 -0700 (PDT)
> From: Matteo Croce
> Date: Mon, 8 Apr 2019 23:06:45 +0200
>
>> The thunderx driver can't use XDP with all MTU values.
>> This patches sets the right MTU values, and add a check to avoid setting
>> a wrong value which will not fun
From: Matteo Croce
Date: Mon, 8 Apr 2019 23:06:45 +0200
> The thunderx driver can't use XDP with all MTU values.
> This patches sets the right MTU values, and add a check to avoid setting
> a wrong value which will not function.
Series applied and queued up for -stable.
From: Leandro Dorileo
Date: Mon, 8 Apr 2019 10:12:16 -0700
> This set fixes miscalculations based on invalid link speed values.
...
Series applied to net-next, thank you.
From: Hangbin Liu
Date: Mon, 8 Apr 2019 16:45:17 +0800
> After adding a team interface to bridge, the team interface will enter
> promisc mode. Then if we add a new slave to team0, the slave will keep
> promisc off. Fix it by setting slave to promisc on if team master is
> already in promisc mod
Reported in [1].
With gcc 8.3.0 the following error is issued:
cc -Ibpf@sta -I. -I.. -I.././include -I.././include/uapi
-fdiagnostics-color=always -fsanitize=address,undefined
-fno-omit-frame-pointer
-pipe -D_FILE_OFFSET_BITS=64 -Wall -Winvalid-pch -Werror -g -fPIC -g -O2
-Werror -Wall -
On Wed, Apr 10, 2019 at 6:19 PM David Ahern wrote:
>
> On 4/3/19 8:43 AM, Peter Oskolkov wrote:
> > This patch adds tests validating that VRF and BPF-LWT
> > encap work together well, as requested by David Ahern.
> >
> > Signed-off-by: Peter Oskolkov
> > ---
> > .../selftests/bpf/test_lwt_ip_enc
On 4/3/19 8:43 AM, Peter Oskolkov wrote:
> This patch adds tests validating that VRF and BPF-LWT
> encap work together well, as requested by David Ahern.
>
> Signed-off-by: Peter Oskolkov
> ---
> .../selftests/bpf/test_lwt_ip_encap.sh| 134 +++---
> 1 file changed, 86 inserti
From: Jakub Kicinski
Date: Wed, 10 Apr 2019 16:23:39 -0700
> buildbot noticed that TLS_HW is not defined if CONFIG_TLS_DEVICE=n.
> Wrap the cleanup branch into an ifdef, tls_device_free_resources_tx()
> wouldn't be compiled either in this case.
>
> Fixes: 35b71a34ada6 ("net/tls: don't leak parti
On Wed, Apr 10, 2019 at 6:00 AM Jesper Dangaard Brouer
wrote:
>
> This patchset utilize a number of different kernel bulk APIs for optimizing
> the performance for the XDP cpumap redirect feature.
Could you please share some numbers about the optimization?
Thanks,
Song
>
> Patch-1: ptr_ring bat
On Wed, Apr 10, 2019 at 6:02 AM Jesper Dangaard Brouer
wrote:
>
> A lot of the performance gain comes from this patch.
>
> While analysing performance overhead it was found that the largest CPU
> stalls were caused when touching the struct page area. It is first read with
> a READ_ONCE from build_
On Wed, Apr 10, 2019 at 6:02 AM Jesper Dangaard Brouer
wrote:
>
> The function build_skb() also have the responsibility to allocate and clear
> the SKB structure. Introduce a new function build_skb_around(), that moves
> the responsibility of allocation and clearing to the caller. This allows
> ca
On Wed, Apr 10, 2019 at 6:02 AM Jesper Dangaard Brouer
wrote:
>
> As cpumap now batch consume xdp_frame's from the ptr_ring, it knows how many
> SKBs it need to allocate. Thus, lets bulk allocate these SKBs via
> kmem_cache_alloc_bulk() API, and use the previously introduced function
> build_skb_a
On Wed, Apr 10, 2019 at 6:03 AM Jesper Dangaard Brouer
wrote:
>
> Move ptr_ring dequeue outside loop, that allocate SKBs and calls network
> stack, as these operations that can take some time. The ptr_ring is a
> communication channel between CPUs, where we want to reduce/limit any
> cacheline bou
buildbot noticed that TLS_HW is not defined if CONFIG_TLS_DEVICE=n.
Wrap the cleanup branch into an ifdef, tls_device_free_resources_tx()
wouldn't be compiled either in this case.
Fixes: 35b71a34ada6 ("net/tls: don't leak partially sent record in device mode")
Signed-off-by: Jakub Kicinski
---
n
tree: https://git.kernel.org/pub/scm/linux/kernel/git/davem/net.git master
head: 44f5e0480752b356f577aef90a5813299c3c60b1
commit: 35b71a34ada62c9573847a324bf06a133fe11b11 [28/30] net/tls: don't leak
partially sent record in device mode
config: i386-randconfig-x008-201914 (attached as .config)
Now the SKB list implementation assumption can be removed.
And now that we know that the list head is always non-NULL
we can remove the code blocks dealing with that as well.
Signed-off-by: David S. Miller
---
include/net/sctp/ulpqueue.h | 2 +-
net/sctp/stream_interleave.c | 2 +-
net/sct
Pass this, instead of an event. Then everything trickles down and we
always have events a non-empty list.
Then we needs a list creating stub to place into .enqueue_event for
sctp_stream_interleave_1.
Signed-off-by: David S. Miller
---
net/sctp/stream_interleave.c | 49 ++
Inside the loop, we always start with event non-NULL.
Signed-off-by: David S. Miller
---
net/sctp/ulpqueue.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/net/sctp/ulpqueue.c b/net/sctp/ulpqueue.c
index 5dde92101743..0fecc1fb4ab7 100644
--- a/net/sctp/ulpqueue.c
+++ b/net
This way we can make sure events sent this way to
sctp_ulpq_tail_event() are on a list as well. Now all such code paths
are fully covered.
Signed-off-by: David S. Miller
---
net/sctp/stream_interleave.c | 11 ++-
1 file changed, 10 insertions(+), 1 deletion(-)
diff --git a/net/sctp/s
This way we can simplify the logic and remove assumptions
about the implementation of skb lists.
Signed-off-by: David S. Miller
---
net/sctp/ulpqueue.c | 16 ++--
1 file changed, 10 insertions(+), 6 deletions(-)
diff --git a/net/sctp/ulpqueue.c b/net/sctp/ulpqueue.c
index 0fecc1fb
This patch series eliminates the explicit reference to the skb list
implementation via skb->prev dereferences.
The approach used is to pass a non-empty skb list around instead of an
event skb object which may or may not be on a list.
I'd like to thank Marcelo Leitner, Xin Long, and Neil Horman
From: Jakub Kicinski
Date: Wed, 10 Apr 2019 13:18:57 -0700
> Fix comment.
>
> Signed-off-by: Jakub Kicinski
Applied.
On Wed, Apr 10, 2019 at 11:45:04AM -0700, David Ahern wrote:
> >> @@ -2420,7 +2449,7 @@ static struct rt6_info *__ip6_route_redirect(struct
> >> net *net,
> >> int flags)
> >> {
> >>struct ip6rd_flowi *rdfl = (struct ip6rd_flowi *)fl6;
> >> - struct rt
Sync kernel uapi headers.
Signed-off-by: Javier Honduvilla Coto
---
tools/include/uapi/linux/bpf.h | 10 +-
1 file changed, 9 insertions(+), 1 deletion(-)
diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h
index af1cbd951f26..f707b286c21d 100644
--- a/tools/inc
Hi all,
This patch adds the bpf_descendant_of helper which accepts a PID and
returns 1 if the PID of the process currently being executed is a
descendant of it or if it's itself. Returns 0 otherwise.
This is very useful in tracing programs when we want to filter by a
given PID and all the childre
Adding the following test cases:
- bpf_descendant_of(current->pid) == 1
- bpf_descendant_of(current->real_parent->pid) == 1
- bpf_descendant_of(1) == 1
- bpf_descendant_of(0) == 1
- bpf_descendant_of(-1) == 0
- bpf_descendant_of(current->children[0]->pid) == 0
Signed-off-by: Javier Honduvilla Co
This patch adds the bpf_descendant_of helper which accepts a PID and
returns 1 if the PID of the process currently being executed is a
descendant of it or if it's itself. Returns 0 otherwise.
This is very useful in tracing programs when we want to filter by a
given PID and all the children it migh
Fix comment.
Signed-off-by: Jakub Kicinski
---
net/strparser/strparser.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/net/strparser/strparser.c b/net/strparser/strparser.c
index fa6c977b4c41..0ba363624339 100644
--- a/net/strparser/strparser.c
+++ b/net/strparser/strparser
From: Jakub Kicinski
Date: Wed, 10 Apr 2019 11:04:29 -0700
> This series fixes two memory issues and a stack overflow.
> First two patches are fairly simple leaks. Third patch
> partially reverts an optimization made to the strparser
> which causes creation of skb->frag_list->skb->frag_list...
>
After previous patches, verifier has marked those instructions that really
need zero extension on dst_reg.
It is then for all back-ends to decide how to use such information to
eliminate unnecessary zero extension code-gen during JIT compilation.
One approach is:
1. Verifier insert explicit zer
This patch eliminate zero extension code-gen for instructions except
load/store when possible.
Elimination for load/store will be supported in up coming patches.
Reviewed-by: Jakub Kicinski
Signed-off-by: Jiong Wang
---
drivers/net/ethernet/netronome/nfp/bpf/jit.c | 119 +-
Unlike BPF to BPF function call, BPF helper call is calls to native insns
that verifier can't walk. So, verifier needs helper proto type descriptions
for data-flow purpose.
There is such information already, but it is not differentiate sub-register
read with full register read.
This patch split "
Cc: Naveen N. Rao
Cc: Sandipan Das
Signed-off-by: Jiong Wang
---
arch/powerpc/net/bpf_jit_comp64.c | 7 ++-
1 file changed, 6 insertions(+), 1 deletion(-)
diff --git a/arch/powerpc/net/bpf_jit_comp64.c
b/arch/powerpc/net/bpf_jit_comp64.c
index 21a1dcd..d10621b 100644
--- a/arch/powerpc/ne
eBPF ISA specification requires high 32-bit cleared when low 32-bit
sub-register is written. This applies to destination register of ALU32 etc.
JIT back-ends must guarantee this semantic when doing code-gen.
x86-64 and arm64 ISA has the same semantic, so the corresponding JIT
back-end doesn't need
Enable BPF_F_TEST_RND_HI32 for all existing bpf selftests or other
independent tests could involve quite a few changes to make sure all bpf
prog load places has BPF_F_TEST_RND_HI32 set.
Given most of the tests are using libbpf, this patch introduces a new
global variable "libbpf_test_mode" into li
x86_64 and AArch64 perhaps are two arches that running bpf testsuite
frequently, however the zero extension insertion pass is not enabled for
them because of their hardware support.
It is critical to guarantee the pass correction as it is supposed to be
enabled at default for a couple of other arc
v2 changes:
- rebased on top of bpf-next master.
- added comments for what is sub-register def index. (Edward, Alexei)
- removed patch 1 which turns bit mask from enum to macro. (Alexei)
- removed sysctl/bpf_jit_32bit_opt. (Alexei)
- merged sub-register def insn index into reg state. (Ale
CC: Björn Töpel
Signed-off-by: Jiong Wang
---
arch/riscv/net/bpf_jit_comp.c | 32 +++-
1 file changed, 19 insertions(+), 13 deletions(-)
diff --git a/arch/riscv/net/bpf_jit_comp.c b/arch/riscv/net/bpf_jit_comp.c
index 80b12aa..9cba262 100644
--- a/arch/riscv/net/bpf_
This patch randomizes high 32-bit of a definition when BPF_F_TEST_RND_HI32
is set.
It does this once the flag set no matter there is hardware zero extension
support or not. Because this is a test feature and we want to deliver the
most stressful test.
Suggested-by: Alexei Starovoitov
Signed-off-
Cc: Martin Schwidefsky
Cc: Heiko Carstens
Signed-off-by: Jiong Wang
---
arch/s390/net/bpf_jit_comp.c | 13 ++---
1 file changed, 10 insertions(+), 3 deletions(-)
diff --git a/arch/s390/net/bpf_jit_comp.c b/arch/s390/net/bpf_jit_comp.c
index 51dd026..59592d7 100644
--- a/arch/s390/net/b
Register liveness infrastructure doesn't track register read width at the
moment, while the width information will be needed for the later 32-bit
safety analysis pass.
This patch take the first step to split read liveness into REG_LIVE_READ64
and REG_LIVE_READ32.
Liveness propagation code are upd
"test_progs" contains a couple of C based tests, so is a very good
assurance on the pass correctness for this patch set.
After this patch, BPF_F_TEST_RND_HI32 will be set for all of them and
high 32-bit are randomized when possible. This could deliver quite
stressful tests on x86_64.
hi32 randomi
Cc: David S. Miller
Signed-off-by: Jiong Wang
---
arch/sparc/net/bpf_jit_comp_64.c | 8 +++-
1 file changed, 7 insertions(+), 1 deletion(-)
diff --git a/arch/sparc/net/bpf_jit_comp_64.c b/arch/sparc/net/bpf_jit_comp_64.c
index 65428e7..e58f84e 100644
--- a/arch/sparc/net/bpf_jit_comp_64.c
+
Cc: Shubham Bansal
Signed-off-by: Jiong Wang
---
arch/arm/net/bpf_jit_32.c | 22 --
1 file changed, 16 insertions(+), 6 deletions(-)
diff --git a/arch/arm/net/bpf_jit_32.c b/arch/arm/net/bpf_jit_32.c
index c8bfbbf..8cecd06 100644
--- a/arch/arm/net/bpf_jit_32.c
+++ b/arch/ar
Cc: Wang YanQing
Signed-off-by: Jiong Wang
---
arch/x86/net/bpf_jit_comp32.c | 32 ++--
1 file changed, 22 insertions(+), 10 deletions(-)
diff --git a/arch/x86/net/bpf_jit_comp32.c b/arch/x86/net/bpf_jit_comp32.c
index 0d9cdff..8c6cf22 100644
--- a/arch/x86/net/bpf_j
After code refactor in previous patches, the propagation logic inside the
for loop in "propagate_liveness" becomes clear that they are good enough to
be factored out into a common function "propagate_liveness_reg".
Signed-off-by: Jiong Wang
---
kernel/bpf/verifier.c | 30
There are a few "regs[regno]" here are there across "check_reg_arg", this
patch factor it out into a simple "reg" pointer. The intention is to
simplify code indentation and make the later patches in this set look
cleaner.
Signed-off-by: Jiong Wang
---
kernel/bpf/verifier.c | 14 --
1
Propagation for register and stack slot are finished in separate for loop,
while they are perfect to be put into a single loop.
This could also let them share some common variables in later patches.
Signed-off-by: Jiong Wang
---
kernel/bpf/verifier.c | 4 +---
1 file changed, 1 insertion(+), 3
Access to reg states were not factored out, the consequence is long code
for dereferencing them which made the indentation not good for reading.
This patch factor out these code so the core code in the loop could be
easier to follow.
Signed-off-by: Jiong Wang
---
kernel/bpf/verifier.c | 34
On Wed, Apr 10, 2019 at 12:03 PM David Ahern wrote:
>
> From: David Ahern
>
> Govindarajulu reported a regression with Network Manager which sends an
> RTA_GATEWAY attribute with the address set to 0. Fixup the handling of
> RTA_GATEWAY to only set fc_gw_family if the gateway address is actually
On 04/10/2019 11:04 AM, Jakub Kicinski wrote:
> This reverts the first part of commit 4e485d06bb8c ("strparser: Call
> skb_unclone conditionally"). To build a message with multiple
> fragments we need our own root of frag_list. We can't simply
> use the frag_list of orig_skb, because it will l
From: David Ahern
Date: Wed, 10 Apr 2019 10:05:51 -0700
> From: David Ahern
>
> Govindarajulu reported a regression with Network Manager which sends an
> RTA_GATEWAY attribute with the address set to 0. Fixup the handling of
> RTA_GATEWAY to only set fc_gw_family if the gateway address is actua
On 4/10/2019 12:31 PM, Heiner Kallweit wrote:
> On 09.04.2019 22:59, David Miller wrote:
>> From: Heiner Kallweit
>> Date: Sun, 7 Apr 2019 11:57:13 +0200
>>
>>> Recently genphy_read_abilities() has been added that dynamically detects
>>> clause 22 PHY abilities. I *think* this detection should
On 09.04.2019 22:59, David Miller wrote:
> From: Heiner Kallweit
> Date: Sun, 7 Apr 2019 11:57:13 +0200
>
>> Recently genphy_read_abilities() has been added that dynamically detects
>> clause 22 PHY abilities. I *think* this detection should work with all
>> supported PHY's, at least for the ones
From: Thomas Falcon
Date: Wed, 10 Apr 2019 11:06:58 -0500
> In its current state, a driver reset clobbers any feature settings
> a user may have toggled and will disable GRO as it is not explicitly
> enabled in the driver. This patch set enables GRO and tries to retain
> user settings after a res
From: Paolo Abeni
Date: Wed, 10 Apr 2019 14:32:36 +0200
> The commit 46b1c18f9deb ("net: sched: put back q.qlen into a single location")
> introduced some measurable regression in the contended scenarios for
> lock qdisc.
>
> As Eric suggested we could replace q.qlen access with calls to
> qdis
From: Ido Schimmel
Date: Wed, 10 Apr 2019 06:58:10 +
> This patchset contains various small fixes for mlxsw.
>
> Patch #1 fixes a warning generated by switchdev core when the driver
> fails to insert an MDB entry in the commit phase.
>
> Patches #2-#4 fix a warning in check_flush_dependency
On 10/04/2019 12:43, Jesper Dangaard Brouer wrote:
> Reduce BH-disable period further by moving cpu_map_build_skb()
> outside/before invoking the network stack. And build up a
> skb_list that is used for netif_receive_skb_list. This is also
> an I-cache optimization.
>
> When injecting packets int
On 4/5/19 8:57 PM, Antony Antony wrote:
> On Fri, Apr 05, 2019 at 03:07:58PM -0700, Stephen Hemminger wrote:
>> On Thu, 4 Apr 2019 19:07:38 +0300
>> Eyal Birger wrote:
>>
>>> The XFRMA_IF_ID attribute is set in policies/states for them to be
>>> associated with an XFRM interface (4.19+).
>>>
>>>
On 4/10/19 10:36 AM, Martin Lau wrote:
>> diff --git a/net/ipv6/route.c b/net/ipv6/route.c
>> index 0e8becb1e455..d555edaaff13 100644
>> --- a/net/ipv6/route.c
>> +++ b/net/ipv6/route.c
>> @@ -2407,6 +2407,35 @@ void ip6_sk_dst_store_flow(struct sock *sk, struct
>> dst_entry *dst,
>>
On Wed, Apr 10, 2019 at 12:21 AM Magnus Karlsson
wrote:
>
> The use of smp_rmb() and smp_wmb() creates a Linux header dependency
> on barrier.h that is uneccessary in most parts. This patch implements
> the two small defines that are needed from barrier.h. As a bonus, the
> new implementations are
On Wed, Apr 10, 2019 at 09:48:40AM +0300, Leon Romanovsky wrote:
> > > > Mark Bloch (12):
> > > > net/mlx5: E-Switch, don't use hardcoded values for FDB prios
> > > > net/mlx5: E-Switch, add a new prio to be used by the RDMA side
> > > > RDMA/mlx5: Move netdev info into the port struct
> > >
This reverts the first part of commit 4e485d06bb8c ("strparser: Call
skb_unclone conditionally"). To build a message with multiple
fragments we need our own root of frag_list. We can't simply
use the frag_list of orig_skb, because it will lead to linking
all orig_skbs together creating very long
David reports that tls triggers warnings related to
sk->sk_forward_alloc not being zero at destruction time:
WARNING: CPU: 5 PID: 6831 at net/core/stream.c:206
sk_stream_kill_queues+0x103/0x110
WARNING: CPU: 5 PID: 6831 at net/ipv4/af_inet.c:160
inet_sock_destruct+0x15b/0x170
When sender fills
Hi!
This series fixes two memory issues and a stack overflow.
First two patches are fairly simple leaks. Third patch
partially reverts an optimization made to the strparser
which causes creation of skb->frag_list->skb->frag_list...
chains of 100s of skbs, leading to recursive kfree_skb()
filling
Commit f66de3ee2c16 ("net/tls: Split conf to rx + tx") made
freeing of IV and record sequence number conditional to SW
path only, but commit e8f69799810c ("net/tls: Add generic NIC
offload infrastructure") also allocates that state for the
device offload configuration. Remember to free it.
Fixes:
On Wed, Apr 10, 2019 at 12:21 AM Magnus Karlsson
wrote:
>
> This patch removes the use of likely and unlikely in xsk.h since they
> create a dependency on Linux headers as reported by several
> users. There have also been reports that the use of these decreases
> performance as the compiler puts t
On Tue, Apr 09, 2019 at 02:41:19PM -0700, David Ahern wrote:
> From: David Ahern
>
> Move the nexthop evaluation of a fib entry to a helper that can be
> leveraged for each fib6_nh in a multipath nexthop object.
>
> In the move, 'continue' statements means the helper returns false
> (loop should
On Tue, Apr 9, 2019 at 11:55 PM Magnus Karlsson
wrote:
>
> In commit da11b417583e ("libbpf: teach libbpf about log_level bit 2"),
> the BPF_LOG_BUF_SIZE was increased to 16M. The XDP socket part of
> libbpf allocated the log_buf on the stack, but for the new 16M buffer
> size this is not going to
From: David Ahern
Govindarajulu reported a regression with Network Manager which sends an
RTA_GATEWAY attribute with the address set to 0. Fixup the handling of
RTA_GATEWAY to only set fc_gw_family if the gateway address is actually
set.
Fixes: f35b794b3b405 ("ipv4: Prepare fib_config for IPv6 g
On Wed, 10 Apr 2019 16:26:38 +, Vlad Buslov wrote:
> >> Actually, I intended to modify fl_reoffload() to ignore filters with
> >> 'deleted' flag set when adding, but I guess reusing 'reoffload_count' to
> >> retry fl_hw_destroy_filter() would also work.
> >
> > Yeah, I don't see how you can i
On Wed 10 Apr 2019 at 19:09, Jakub Kicinski
wrote:
> On Wed, 10 Apr 2019 16:02:17 +, Vlad Buslov wrote:
>> On Wed 10 Apr 2019 at 18:48, Jakub Kicinski
>> wrote:
>> > On Wed, 10 Apr 2019 14:53:53 +, Vlad Buslov wrote:
>> >> >> For my next patch set that unlocks the offloads API I implem
On 4/10/19 12:58 AM, Daniel Borkmann wrote:
> On 04/10/2019 02:37 AM, Yonghong Song wrote:
>> The issue is reported at https://github.com/libbpf/libbpf/issues/28.
>>
>> Basically, per C standard, for
>>void *memcpy(void *dest, const void *src, size_t n)
>> if "dest" or "src" is NULL, regardle
On Wed, 10 Apr 2019 16:02:17 +, Vlad Buslov wrote:
> On Wed 10 Apr 2019 at 18:48, Jakub Kicinski
> wrote:
> > On Wed, 10 Apr 2019 14:53:53 +, Vlad Buslov wrote:
> >> >> For my next patch set that unlocks the offloads API I implemented the
> >> >> algorithm to track reoffload count for e
In its current state, a driver reset clobbers any feature settings
a user may have toggled and will disable GRO as it is not explicitly
enabled in the driver. This patch set enables GRO and tries to retain
user settings after a reset. If the underlying carrier changes, however,
the driver will disa
Enable Generic Receive Offload in the ibmvnic driver.
Signed-off-by: Thomas Falcon
---
drivers/net/ethernet/ibm/ibmvnic.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/ibm/ibmvnic.c
b/drivers/net/ethernet/ibm/ibmvnic.c
index 5ecbb1a..59362ed 100644
---
While determining offload capabilities of backing hardware during
a device reset, the driver is clobbering current feature settings.
Update hw_features on reset instead of features unless a feature
is enabled that is no longer supported on the current backing device.
Also enable features that were
On Wed 10 Apr 2019 at 18:48, Jakub Kicinski
wrote:
> On Wed, 10 Apr 2019 14:53:53 +, Vlad Buslov wrote:
>> >> For my next patch set that unlocks the offloads API I implemented the
>> >> algorithm to track reoffload count for each tp that works like this:
>> >>
>> >> 1. struct tcf_proto is ex
On Wed, 10 Apr 2019 14:53:53 +, Vlad Buslov wrote:
> >> For my next patch set that unlocks the offloads API I implemented the
> >> algorithm to track reoffload count for each tp that works like this:
> >>
> >> 1. struct tcf_proto is extended with reoffload_count counter that
> >>incremented
1 - 100 of 146 matches
Mail list logo
