If copy_net_ns() failed after net_alloc(), net->key_domain is leaked.
Fix this, by freeing key_domain in error path.
syzbot report:
BUG: memory leak
unreferenced object 0x8881175007e0 (size 32):
comm "syz-executor902", pid 7069, jiffies 4294944350 (age 28.400s)
hex dump (first 32 bytes):
Fri, Oct 18, 2019 at 10:27:48PM CEST, and...@lunn.ch wrote:
>On Fri, Oct 18, 2019 at 10:08:22PM +0200, Jiri Pirko wrote:
>> Fri, Oct 18, 2019 at 07:43:04PM CEST, and...@lunn.ch wrote:
>> >On Fri, Oct 18, 2019 at 06:07:26PM +0200, Jiri Pirko wrote:
>> >> From: Jiri Pirko
>> >>
>> >> Currently, the
Alexei Starovoitov wrote:
> On Fri, Oct 18, 2019 at 9:52 AM Andrii Nakryiko wrote:
> >
> > On 10/18/19 7:41 AM, John Fastabend wrote:
> > > With commit "libbpf: stop enforcing kern_version,..." we removed the
> > > kernel version section parsing in favor of querying for the kernel
> > > using unam
Greetings
My name is Barrister Hans Erich.
I have a client who is interested to invest in your country, she is a well
known politician in her country and deserve a lucrative investment partnership
with you outside her country without any delay Please can you manage such
investment please Kin
Greetings
My name is Barrister Hans Erich.
I have a client who is interested to invest in your country, she is a well
known politician in her country and deserve a lucrative investment partnership
with you outside her country without any delay Please can you manage such
investment please Kin
On Fri, Oct 18, 2019 at 05:45:26PM -0700, Samudrala, Sridhar wrote:
> On 10/18/2019 5:14 PM, Alexei Starovoitov wrote:
> > On Fri, Oct 18, 2019 at 11:40:07AM -0700, Samudrala, Sridhar wrote:
> > >
> > > Perf report for "AF_XDP default rxdrop" with patched kernel - mitigations
> > > ON
> > > =
On Fri, 18 Oct 2019 19:38:24 +, Saeed Mahameed wrote:
> From: Tariq Toukan
>
> Once the kTLS TX resync function is called, it used to return
> a binary value, for success or failure.
>
> However, in case the TLS SKB is a retransmission of the connection
> handshake, it initiates the resync f
On Fri, Oct 18, 2019 at 9:52 AM Andrii Nakryiko wrote:
>
> On 10/18/19 7:41 AM, John Fastabend wrote:
> > With commit "libbpf: stop enforcing kern_version,..." we removed the
> > kernel version section parsing in favor of querying for the kernel
> > using uname() and populating the version using t
On 10/18/2019 5:14 PM, Alexei Starovoitov wrote:
On Fri, Oct 18, 2019 at 11:40:07AM -0700, Samudrala, Sridhar wrote:
Perf report for "AF_XDP default rxdrop" with patched kernel - mitigations ON
==
Samples: 44K of event 'cyc
From: Alexei Starovoitov
Date: Fri, 18 Oct 2019 16:25:24 -0700
> gmail delivery lags by a day :(
Sorry, I'm working on that.
On Fri, Oct 18, 2019 at 11:40:07AM -0700, Samudrala, Sridhar wrote:
>
> Perf report for "AF_XDP default rxdrop" with patched kernel - mitigations ON
> ==
> Samples: 44K of event 'cycles', Event count (approx.): 38532389541
> O
I was running the updated patches on machines with various workloads, and
have a bunch of different results.
For the following numbers,
Effective = hit / (hit + empty + stall) * 100
In other words, show the hit rate for for every trip to the cache,
and the cache full stat is ignored.
On a webs
On Fri, Oct 18, 2019 at 3:03 PM Jason Baron wrote:
>
> The TCPI_OPT_SYN_DATA bit as part of tcpi_options currently reports whether
> or not data-in-SYN was ack'd on both the client and server side. We'd like
> to gather more information on the client-side in the failure case in order
> to indicate
On 18 Oct 2019, at 14:07, Saeed Mahameed wrote:
> On Thu, 2019-10-17 at 15:06 +0300, Ilias Apalodimas wrote:
>> Hi Saeed,
>>
>> On Wed, Oct 16, 2019 at 03:50:22PM -0700, Jonathan Lemon wrote:
>>> From: Saeed Mahameed
>>>
>>> Add page_pool_update_nid() to be called from drivers when they
>>> de
On 18 Oct 2019, at 14:29, Saeed Mahameed wrote:
> On Wed, 2019-10-16 at 15:50 -0700, Jonathan Lemon wrote:
>> Add statistics to the page pool, providing visibility into its
>> operation.
>>
>> Callers can provide a location where the stats are stored, otherwise
>> the page pool will allocate a
On Wed, Oct 16, 2019 at 5:53 AM wrote:
>
> From: Tonghao Zhang
>
> When creating and inserting flow-mask, if there is no available
> flow-mask, we realloc the mask array. When removing flow-mask,
> if necessary, we shrink mask array.
>
> Signed-off-by: Tonghao Zhang
> Tested-by: Greg Rose
> ---
On 10/18/19 2:02 PM, Vivien Didelot wrote:
> If there are multiple switch trees on the device, only the last one
> will be listed, because the arguments of list_add_tail are swapped.
>
> Fixes: 83c0afaec7b7 ("net: dsa: Add new binding implementation")
> Signed-off-by: Vivien Didelot
Reviewed-by:
On Fri, 18 Oct 2019 16:25:24 -0700, Alexei Starovoitov wrote:
> I'm not sure that cleanup Jakub is proposing is possible or better.
> Not everything is array_size here and in other places
> where cost is computed. u64 is imo much cleaner.
Right, adding multiple components with size_t is going to b
On 18 Oct 2019, at 13:50, Saeed Mahameed wrote:
> On Wed, 2019-10-16 at 15:50 -0700, Jonathan Lemon wrote:
>> This patch combines work from various people:
>> - part of Tariq's work to move the DMA mapping from
>> the mlx5 driver into the page pool. This does not
>> include later patches w
On Wed, Oct 16, 2019 at 5:52 AM wrote:
>
> From: Tonghao Zhang
>
> Port the codes to linux upstream and with little changes.
>
> Pravin B Shelar, says:
> | mask caches index of mask in mask_list. On packet recv OVS
> | need to traverse mask-list to get cached mask. Therefore array
> | is better f
On Wed, Oct 16, 2019 at 5:51 AM wrote:
>
> From: Tonghao Zhang
>
> The idea of this optimization comes from a patch which
> is committed in 2014, openvswitch community. The author
> is Pravin B Shelar. In order to get high performance, I
> implement it again. Later patches will use it.
>
> Pravin
On Wed, Oct 16, 2019 at 5:56 AM wrote:
>
> From: Tonghao Zhang
>
> use the specified functions to init resource.
>
> Signed-off-by: Tonghao Zhang
> Tested-by: Greg Rose
> ---
Looks like this is simply moving code around.
I don't have any opinion.
> net/openvswitch/datapath.c | 60
>
On Fri, Oct 18, 2019 at 01:55:46PM -0700, Pravin Shelar wrote:
> On Sun, Oct 13, 2019 at 3:22 PM Guillaume Nault wrote:
> > The point of my RFC is to know if it's possible to avoid all these
> > gfp_t flags, by allowing ovs_vport_cmd_fill_info() to sleep (at least
> > I'd like to figure out if it'
On Fri, Oct 18, 2019 at 4:25 PM Alexei Starovoitov wrote:
>
> Only raw_tracepoint program type can have bpf_attr.attach_btf_id >= 0.
typo: just > (code is actually correct, though)
> Make sure to reject other program types that accidentally set it to non-zero.
>
> Fixes: ccfe29eb29c2 ("bpf: Add
On Wed, Oct 16, 2019 at 5:56 AM wrote:
>
> From: Tonghao Zhang
>
> Unlocking of a not locked mutex is not allowed.
> Other kernel thread may be in critical section while
> we unlock it because of setting user_feature fail.
>
> Fixes: 95a7233c4 ("net: openvswitch: Set OvS recirc_id from tc chain i
On Wed, Oct 16, 2019 at 5:55 AM wrote:
>
> From: Tonghao Zhang
>
> The most case *index < ma->max, and flow-mask is not NULL.
> We add un/likely for performance.
>
> Signed-off-by: Tonghao Zhang
> Tested-by: Greg Rose
> ---
LGTM
Acked-by: William Tu
> net/openvswitch/flow_table.c | 4 ++--
>
On Wed, Oct 16, 2019 at 5:54 AM wrote:
>
> From: Tonghao Zhang
>
> The full looking up on flow table traverses all mask array.
> If mask-array is too large, the number of invalid flow-mask
> increase, performance will be drop.
>
> One bad case, for example: M means flow-mask is valid and NULL
> o
On Wed, Oct 16, 2019 at 5:54 AM wrote:
>
> From: Tonghao Zhang
>
> Simplify the code and remove the unnecessary BUILD_BUG_ON.
>
> Signed-off-by: Tonghao Zhang
> Tested-by: Greg Rose
> ---
LGTM
Acked-by: William Tu
> net/openvswitch/flow_table.c | 8 ++--
> 1 file changed, 2 insertions(+
On Fri, Oct 18, 2019 at 8:36 AM Toke Høiland-Jørgensen wrote:
>
> Tetsuo pointed out that without an explicit cast, the cost calculation for
> devmap_hash type maps could overflow on 32-bit builds. This adds the
> missing cast.
>
> Fixes: 6f9d451ab1a3 ("xdp: Add devmap_hash map type for looking up
On Fri, 18 Oct 2019 19:38:22 +, Saeed Mahameed wrote:
> From: Tariq Toukan
>
> Do not assume the crypto info is accessible during the
> connection lifetime. Save a copy of it in the private
> TX context.
It should be around as long as the driver knows about the socket, no?
On Fri, 18 Oct 2019 19:38:20 +, Saeed Mahameed wrote:
> From: Tariq Toukan
>
> Cipher type is checked upon connection addition.
> No need to recheck it per every TX resync invocation.
>
> Fixes: d2ead1f360e8 ("net/mlx5e: Add kTLS TX HW offload support")
> Signed-off-by: Tariq Toukan
> Revie
On Fri, 18 Oct 2019 19:38:09 +, Saeed Mahameed wrote:
> From: Tariq Toukan
>
> No Eth segment, so no dynamic inline headers.
> The size of a Dump WQE is fixed, use constants and remove
> unnecessary checks.
>
> Fixes: d2ead1f360e8 ("net/mlx5e: Add kTLS TX HW offload support")
> Signed-off-by
On Fri, 18 Oct 2019 14:13:39 -0700, Jeff Kirsher wrote:
> From: Sasha Neftin
>
> Extend the socket buffer field process and add Rx checksum functionality
> Minor: fix indentation with tab instead of spaces.
>
> Signed-off-by: Sasha Neftin
> Tested-by: Aaron Brown
> Signed-off-by: Jeff Kirsher
On Fri, 18 Oct 2019 06:01:25 -0400, Zhu Yanjun wrote:
> This change adds support for xmit_more based on the igb commit 6f19e12f6230
> ("igb: flush when in xmit_more mode and under descriptor pressure") and
> commit 6b16f9ee89b8 ("net: move skb->xmit_more hint to softnet data") that
> were made to i
On Fri, 18 Oct 2019 12:34:04 +0200, Jiri Olsa wrote:
> The bpftool interface stays the same, but now it's possible
> to run it over BTF raw data, like:
>
> $ bpftool btf dump file /sys/kernel/btf/vmlinux
> libbpf: failed to get EHDR from /sys/kernel/btf/vmlinux
> [1] INT '(anon)' size=4 bits
Intel test robot reported a ~7% regression on TCP_CRR tests
that they bisected to the cited commit.
Indeed, every time a new TCP socket is created or deleted,
the atomic counter net->count is touched (via get_net(net)
and put_net(net) calls)
So cpus might have to reload a contended cache line in
On 10/18/19 3:01 PM, John Fastabend wrote:
> Daniel Borkmann wrote:
>> On Fri, Oct 18, 2019 at 07:54:26AM -0700, John Fastabend wrote:
>>> Following ./Documentation/trace/kprobetrace.rst add support for loading
>>> kprobes programs on older kernels.
>>>
>>> Signed-off-by: John Fastabend
>>> ---
Daniel Borkmann wrote:
> On Fri, Oct 18, 2019 at 07:54:26AM -0700, John Fastabend wrote:
> > Following ./Documentation/trace/kprobetrace.rst add support for loading
> > kprobes programs on older kernels.
> >
> > Signed-off-by: John Fastabend
> > ---
> > tools/lib/bpf/libbpf.c | 81
> > +++
Hi Andrew,
On Thu, 17 Oct 2019 21:20:55 +0200, Andrew Lunn wrote:
> --- a/drivers/net/dsa/mv88e6xxx/global1.h
> +++ b/drivers/net/dsa/mv88e6xxx/global1.h
> @@ -109,6 +109,7 @@
> /* Offset 0x0A: ATU Control Register */
> #define MV88E6XXX_G1_ATU_CTL 0x0a
> #define MV88E6XXX_G1_ATU_CTL_L
On Fri, 18 Oct 2019 22:27:48 +0200, Andrew Lunn wrote:
> I don't really like the all lower case restriction. It makes it hard
> to be consistent. All Marvell Docs refer to the Address Translation
> Unit as ATU. I don't think there is any reference to atu. I would
> prefer to be consistent with the
On Wed, 2019-10-16 at 15:50 -0700, Jonathan Lemon wrote:
> Add statistics to the page pool, providing visibility into its
> operation.
>
> Callers can provide a location where the stats are stored, otherwise
> the page pool will allocate a statistic area.
>
> Signed-off-by: Jonathan Lemon
> ---
On Fri, 2019-10-18 at 13:50 -0700, Saeed Mahameed wrote:
> On Wed, 2019-10-16 at 15:50 -0700, Jonathan Lemon wrote:
> > This patch combines work from various people:
> > - part of Tariq's work to move the DMA mapping from
> > the mlx5 driver into the page pool. This does not
> > include later
On Fri, Oct 18, 2019 at 09:44:18PM +0200, Toke Høiland-Jørgensen wrote:
> It seems I forgot to add handling of devmap_hash type maps to the device
> unregister hook for devmaps. This omission causes devices to not be
> properly released, which causes hangs.
>
> Fix this by adding the missing handl
From: Sasha Neftin
Extend the socket buffer field process and add Rx checksum functionality
Minor: fix indentation with tab instead of spaces.
Signed-off-by: Sasha Neftin
Tested-by: Aaron Brown
Signed-off-by: Jeff Kirsher
---
drivers/net/ethernet/intel/igc/igc_defines.h | 5 ++-
drivers/net
From: Sasha Neftin
VLAN filter table array not implemented yet and shadow_vfta pointer
not used. Clean up the code and remove the unused shadow_vfta pointer.
Signed-off-by: Sasha Neftin
Tested-by: Aaron Brown
Signed-off-by: Jeff Kirsher
---
drivers/net/ethernet/intel/igc/igc.h | 1 -
dr
From: Sasha Neftin
Add multicast addresses list to the MTA table.
Implement basic Rx mode support.
Add option for IPv6 address settings.
Signed-off-by: Sasha Neftin
Tested-by: Aaron Brown
Signed-off-by: Jeff Kirsher
---
drivers/net/ethernet/intel/igc/igc_defines.h | 3 +
drivers/net/ethern
From: Sasha Neftin
Add stream control transmission protocol CRC checksum.
Signed-off-by: Sasha Neftin
Tested-by: Aaron Brown
Signed-off-by: Jeff Kirsher
---
drivers/net/ethernet/intel/igc/igc_main.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/net/ethernet/intel/igc/igc_main.c
From: Sasha Neftin
Implement flow for S0ix support. Modern SoCs support S0ix low power
states during idle periods, which are sub-states of ACPI S0 that increase
power saving while supporting an instant-on experience for providing
lower latency that ACPI S0. The S0ix states shut off parts of the S
This series contains updates to e1000e and igc only.
Sasha adds stream control transmission protocol (SCTP) CRC checksum
support for igc. Also added S0ix support to the e1000e driver. Then
added multicast support by adding the address list to the MTA table and
providing the option for IPv6 addre
On Thu, 2019-10-17 at 15:06 +0300, Ilias Apalodimas wrote:
> Hi Saeed,
>
> On Wed, Oct 16, 2019 at 03:50:22PM -0700, Jonathan Lemon wrote:
> > From: Saeed Mahameed
> >
> > Add page_pool_update_nid() to be called from drivers when they
> > detect
> > numa node changes.
> >
> > It will do:
> > 1)
If there are multiple switch trees on the device, only the last one
will be listed, because the arguments of list_add_tail are swapped.
Fixes: 83c0afaec7b7 ("net: dsa: Add new binding implementation")
Signed-off-by: Vivien Didelot
---
net/dsa/dsa2.c | 2 +-
1 file changed, 1 insertion(+), 1 dele
On Wed, 2019-10-16 at 15:50 -0700, Jonathan Lemon wrote:
> From: Tariq Toukan
>
> After RX page-cache is removed in previous patch, let the
> page_pool be responsible for the DMA mapping.
>
> Issue: 1487631
> Signed-off-by: Tariq Toukan
>
> Signed-off-by: Jonathan Lemon
> ---
> drivers/net/e
On Sun, Oct 13, 2019 at 3:22 PM Guillaume Nault wrote:
>
> On Sun, Oct 13, 2019 at 12:09:43PM -0700, Pravin Shelar wrote:
> > On Thu, Oct 10, 2019 at 12:07 PM Guillaume Nault wrote:
> > >
> > > In rtnl_net_notifyid(), we certainly can't pass a null GFP flag to
> > > rtnl_notify(). A GFP_KERNEL fl
On Wed, 2019-10-16 at 15:50 -0700, Jonathan Lemon wrote:
> From: Tariq Toukan
>
> Obsolete the RX page-cache layer, pages are now recycled
> in page_pool.
>
> This patch introduce a temporary degradation as recycling
> does not keep the pages DMA-mapped. That is fixed in a
> downstream patch.
Ta
On Wed, 2019-10-16 at 15:50 -0700, Jonathan Lemon wrote:
> This patch combines work from various people:
> - part of Tariq's work to move the DMA mapping from
> the mlx5 driver into the page pool. This does not
> include later patches which remove the dma address
> from the driver, as this c
On 17 Oct 2019, at 4:09, Jesper Dangaard Brouer wrote:
On Wed, 16 Oct 2019 15:50:27 -0700
Jonathan Lemon wrote:
Replace the now deprecated inernal cache stats with the page pool
stats.
I can see that the stats you introduced are useful, but they have to
be
implemented in way that does
On Fri, Oct 18, 2019 at 02:00:42PM +0200, Jiri Benc wrote:
> Out of the three nc implementations widely in use, at least two (BSD netcat
> and nmap-ncat) do not support -l combined with -s. Modify the nc invocation
> to be accepted by all of them.
>
> Fixes: 7df5e3db8f63 ("selftests: bpf: tc-bpf f
On 17 Oct 2019, at 1:51, Jesper Dangaard Brouer wrote:
On Wed, 16 Oct 2019 15:50:25 -0700
Jonathan Lemon wrote:
Some drivers may utilize more than one page per RX work entry.
Allow a configurable cache size, with the same defaults if the
size is zero.
Convert magic numbers into descriptiv
From: Russell King
SFF-8472 revision 11.0, 12.1 and 12.2 all have support for 10G Base-ER
so mark this as "rev 10.4 onwards" rather than "rev 10.4 only".
Signed-off-by: Russell King
---
sfpid.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/sfpid.c b/sfpid.c
index 71f0939c
From: Russell King
0x16 and 0x1c are 10G Base-T extended compliance codes. Add these
to our decoding.
Signed-off-by: Russell King
---
sfpid.c | 4
1 file changed, 4 insertions(+)
diff --git a/sfpid.c b/sfpid.c
index 3c50c456f93d..ded3be7337d8 100644
--- a/sfpid.c
+++ b/sfpid.c
@@ -185,1
From: Russell King
A bitrate of 255 is special, it means the bitrate is encoded in
byte 66 in units of 250MBaud. Add support for parsing these bit
rates.
Signed-off-by: Russell King
---
sfpid.c | 19 ---
1 file changed, 16 insertions(+), 3 deletions(-)
diff --git a/sfpid.c b/
On Fri, Oct 18, 2019 at 10:08:22PM +0200, Jiri Pirko wrote:
> Fri, Oct 18, 2019 at 07:43:04PM CEST, and...@lunn.ch wrote:
> >On Fri, Oct 18, 2019 at 06:07:26PM +0200, Jiri Pirko wrote:
> >> From: Jiri Pirko
> >>
> >> Currently, the name format is not required by the code, however it is
> >> requi
On 16 Oct 2019, at 18:30, Jakub Kicinski wrote:
> On Wed, 16 Oct 2019 15:50:19 -0700, Jonathan Lemon wrote:
>> From: Tariq Toukan
>>
>> Obsolete the RX page-cache layer, pages are now recycled
>> in page_pool.
>>
>> This patch introduce a temporary degradation as recycling
>> does not keep the pa
On Wed, Oct 9, 2019 at 8:42 AM Jiri Benc wrote:
>
> On Wed, 9 Oct 2019 10:58:51 -0400, Willem de Bruijn wrote:
> > Yes, this needs a call to gro_cells_receive like geneve_udp_encap_recv
> > and vxlan_rcv, instead of returning a negative value back to
> > udp_queue_rcv_one_skb. Though that's not a
Hi Andrew,
Thanks, review comments for Antoine below.
On Fri, Oct 18, 2019 at 09:08:10PM +0200, Andrew Lunn wrote:
> On Fri, Oct 18, 2019 at 04:39:24PM +0200, Antoine Tenart wrote:
> > This patch converts the MACB Ethernet driver to the Phylink framework.
> > The MAC configuration is moved to the
Fri, Oct 18, 2019 at 07:43:04PM CEST, and...@lunn.ch wrote:
>On Fri, Oct 18, 2019 at 06:07:26PM +0200, Jiri Pirko wrote:
>> From: Jiri Pirko
>>
>> Currently, the name format is not required by the code, however it is
>> required during patch review. All params added until now are in-lined
>> with
On 16 Oct 2019, at 18:33, Jakub Kicinski wrote:
> On Wed, 16 Oct 2019 15:50:21 -0700, Jonathan Lemon wrote:
>> From: Tariq Toukan
>>
>> After RX page-cache is removed in previous patch, let the
>> page_pool be responsible for the DMA mapping.
>>
>> Issue: 1487631
>> Signed-off-by: Tariq Toukan
>
On 10/18/19 9:48 AM, Andrii Nakryiko wrote:
> On 10/18/19 3:34 AM, Jiri Olsa wrote:
>> The bpftool interface stays the same, but now it's possible
>> to run it over BTF raw data, like:
>
> Oh, great, I had similar patch laying around for a while, never got to
> cleaning it up, though, so thanks
It seems I forgot to add handling of devmap_hash type maps to the device
unregister hook for devmaps. This omission causes devices to not be
properly released, which causes hangs.
Fix this by adding the missing handler.
Fixes: 6f9d451ab1a3 ("xdp: Add devmap_hash map type for looking up devices by
On Fri, Oct 18, 2019 at 07:54:26AM -0700, John Fastabend wrote:
> Following ./Documentation/trace/kprobetrace.rst add support for loading
> kprobes programs on older kernels.
>
> Signed-off-by: John Fastabend
> ---
> tools/lib/bpf/libbpf.c | 81
> +++---
From: Tariq Toukan
A call to kTLS completion handler was missing in the TXQSQ release
flow. Add it.
Fixes: d2ead1f360e8 ("net/mlx5e: Add kTLS TX HW offload support")
Signed-off-by: Tariq Toukan
Reviewed-by: Eran Ben Elisha
Signed-off-by: Saeed Mahameed
---
.../mellanox/mlx5/core/en_accel/ktl
From: Navid Emamdoost
In mlx5_fw_fatal_reporter_dump if mlx5_crdump_collect fails the
allocated memory for cr_data must be released otherwise there will be
memory leak. To fix this, this commit changes the return instruction
into goto error handling.
Fixes: 9b1f29823605 ("net/mlx5: Add support f
From: Tariq Toukan
The completion queue consumer index increments upon a call to
mlx5_cqwq_pop().
When dumping an error CQE, the index is already incremented.
Decrease one for the print command.
Fixes: 16cc14d81733 ("net/mlx5e: Dump xmit error completions")
Signed-off-by: Tariq Toukan
Signed-of
From: Navid Emamdoost
In mlx5_fpga_conn_create_cq if mlx5_vector2eqn fails the allocated
memory should be released.
Fixes: 537a50574175 ("net/mlx5: FPGA, Add high-speed connection routines")
Signed-off-by: Navid Emamdoost
Signed-off-by: Saeed Mahameed
---
drivers/net/ethernet/mellanox/mlx5/co
From: Tariq Toukan
Access the record fragments only under the TLS ctx lock.
In the resync flow, save a copy of them to be used when
preparing and posting the required DUMP WQEs.
Fixes: d2ead1f360e8 ("net/mlx5e: Add kTLS TX HW offload support")
Signed-off-by: Tariq Toukan
Reviewed-by: Eran Ben E
From: Tariq Toukan
Not all fields of WQE info are being written in the function,
having some leftovers from previous rounds.
Zero-memset it upon update.
Particularly, not nullifying the wi->resync_dump_frag field
will cause double free of the kTLS DUMPed frags.
Fixes: d2ead1f360e8 ("net/mlx5e:
From: Tariq Toukan
Before posting the context params WQEs, make sure there is enough
contiguous room for them, and fill frag edge if needed.
When posting only a nop, no need for room check, as it needs a single
WQEBB, meaning no contiguity issue.
Fixes: d2ead1f360e8 ("net/mlx5e: Add kTLS TX HW
From: Tariq Toukan
In TX resync flow where DUMP WQEs are posted, keep a pointer to
the fragment page to unref it upon completion, instead of saving
the whole fragment.
In addition, move it the end of the arguments list in tx_fill_wi().
Fixes: d2ead1f360e8 ("net/mlx5e: Add kTLS TX HW offload sup
From: Tariq Toukan
Cited patch removed the assumption only in datapath.
Here we remove it also form control/cleanup flow.
Fixes: 9ab0233728ca ("net/mlx5e: Tx, Don't implicitly assume SKB-less wqe has
one WQEBB")
Signed-off-by: Tariq Toukan
Reviewed-by: Eran Ben Elisha
Signed-off-by: Saeed Mah
From: Tariq Toukan
Once the kTLS TX resync function is called, it used to return
a binary value, for success or failure.
However, in case the TLS SKB is a retransmission of the connection
handshake, it initiates the resync flow (as the tcp seq check holds),
while regular packet handle is expecte
From: Tariq Toukan
Cipher type is checked upon connection addition.
No need to recheck it per every TX resync invocation.
Fixes: d2ead1f360e8 ("net/mlx5e: Add kTLS TX HW offload support")
Signed-off-by: Tariq Toukan
Reviewed-by: Eran Ben Elisha
Signed-off-by: Saeed Mahameed
---
drivers/net/e
Hi Dave,
This series introduces kTLS related fixes to mlx5 driver from Tariq,
and two misc memory leak fixes form Navid Emamdoost.
Please pull and let me know if there is any problem.
I would appreciate it if you queue up kTLS fixes from the list below to
stable kernel v5.3 !
For -stable v4.13:
From: Tariq Toukan
All references for frag pages that are obtained in tx_sync_info_get()
should be released.
Release usually occurs in the corresponding CQE of the WQE.
In error flows, not all fragments have a WQE posted for them, hence
no matching CQE will be generated.
For these pages, release
From: Tariq Toukan
No Eth segment, so no dynamic inline headers.
The size of a Dump WQE is fixed, use constants and remove
unnecessary checks.
Fixes: d2ead1f360e8 ("net/mlx5e: Add kTLS TX HW offload support")
Signed-off-by: Tariq Toukan
Reviewed-by: Eran Ben Elisha
Signed-off-by: Saeed Mahamee
From: Tariq Toukan
Do not assume the crypto info is accessible during the
connection lifetime. Save a copy of it in the private
TX context.
Fixes: d2ead1f360e8 ("net/mlx5e: Add kTLS TX HW offload support")
Signed-off-by: Tariq Toukan
Reviewed-by: Eran Ben Elisha
Signed-off-by: Saeed Mahameed
From: Tariq Toukan
HW expects the data size in DUMP WQEs to be up to MTU.
Make sure they are in range.
We elevate the frag page refcount by 'n-1', in addition to the
one obtained in tx_sync_info_get(), having an overall of 'n'
references. We bulk increments by using a single page_ref_add()
comma
Martin Lau writes:
> On Fri, Oct 18, 2019 at 12:26:55PM +0200, Toke Høiland-Jørgensen wrote:
>> Martin Lau writes:
>>
>> > On Thu, Oct 17, 2019 at 12:52:32PM +0200, Toke Høiland-Jørgensen wrote:
>> >> It seems I forgot to add handling of devmap_hash type maps to the device
>> >> unregister hook
"Samudrala, Sridhar" writes:
> Performance Results
> ===
> Only 1 core is used in all these testcases as the app and the queue irq are
> pinned to the same core.
>
> --
>
> + if (bp->caps & MACB_CAPS_GIGABIT_MODE_AVAILABLE &&
> + (state->interface == PHY_INTERFACE_MODE_NA ||
> + state->interface == PHY_INTERFACE_MODE_GMII ||
> + state->interface == PHY_INTERFACE_MODE_SGMII ||
> + phy_interface_mode_is_rgmii(state->interface)))
On Fri, Oct 18, 2019 at 07:41:26AM -0700, John Fastabend wrote:
> With commit "libbpf: stop enforcing kern_version,..." we removed the
> kernel version section parsing in favor of querying for the kernel
> using uname() and populating the version using the result of the
> query. After this any vers
On Thu, Oct 17, 2019 at 11:09:33PM -0700, Alexei Starovoitov wrote:
> Only raw_tracepoint program type can have bpf_attr.attach_btf_id >= 0.
> Make sure to reject other program types that accidentally set it to non-zero.
>
> Fixes: ccfe29eb29c2 ("bpf: Add attach_btf_id attribute to program load")
On Fri, Oct 18, 2019 at 04:39:24PM +0200, Antoine Tenart wrote:
> This patch converts the MACB Ethernet driver to the Phylink framework.
> The MAC configuration is moved to the Phylink ops and Phylink helpers
> are now used in the ethtools functions.
>
> Signed-off-by: Antoine Tenart
Hi Antoine
The TCPI_OPT_SYN_DATA bit as part of tcpi_options currently reports whether
or not data-in-SYN was ack'd on both the client and server side. We'd like
to gather more information on the client-side in the failure case in order
to indicate the reason for the failure. This can be useful for not only
d
On Fri, Oct 18, 2019 at 5:00 AM Jiri Benc wrote:
>
> Out of the three nc implementations widely in use, at least two (BSD netcat
> and nmap-ncat) do not support -l combined with -s. Modify the nc invocation
> to be accepted by all of them.
>
> Fixes: 7df5e3db8f63 ("selftests: bpf: tc-bpf flow shap
On 10/9/2019 6:06 PM, Alexei Starovoitov wrote:
Will update the patchset with the right performance data and address
feedback from Bjorn.
Hope you are not totally against direct XDP approach as it does provide
value when an AF_XDP socket is bound to a queue and a HW filter can
direct packets
On Fri, 18 Oct 2019 19:09:51 +0200, Jiri Pirko wrote:
> From: Jiri Pirko
>
> Currently, the name format is not required by the code, however it is
> required during patch review. All params added until now are in-lined
> with the following format:
> 1) lowercase characters, digits and underscored
On Fri, 18 Oct 2019 11:15:39 +0200, Toke Høiland-Jørgensen wrote:
> Jakub Kicinski writes:
>
> > On Thu, 17 Oct 2019 12:57:02 +0200, Toke Høiland-Jørgensen wrote:
> >> Tetsuo pointed out that without an explicit cast, the cost calculation for
> >> devmap_hash type maps could overflow on 32-bit
From: Heiner Kallweit
Date: Fri, 18 Oct 2019 20:31:28 +0200
> On 17.10.2019 21:40, David Miller wrote:
>> From: Heiner Kallweit
>> Date: Thu, 17 Oct 2019 21:26:35 +0200
>>
>>> To be on the safe side, let me check with Realtek directly.
>>
>> That's a great idea, let us know what you find out.
Greetings
My name is Barrister Hans Erich.
I have a client who is interested to invest in your country, she is a well
known politician in her country and deserve a lucrative investment partnership
with you outside her country without any delay Please can you manage such
investment please Kin
On 17.10.2019 21:40, David Miller wrote:
> From: Heiner Kallweit
> Date: Thu, 17 Oct 2019 21:26:35 +0200
>
>> To be on the safe side, let me check with Realtek directly.
>
> That's a great idea, let us know what you find out.
>
Realtek suggested to keep the two chip definitions.
Supposedly RTL_
1 - 100 of 163 matches
Mail list logo
