From: Alexander Duyck
Date: Sun, 10 May 2015 17:01:15 -0700
> The reason for the difference between the two is that in the case of
> netdev_alloc_skb/frag the netdev_alloc_cache can only be accessed with
> IRQs disabled, whereas in the napi_alloc_skb case we can access the
> napi_alloc_cache at a
From: Eric Dumazet
Date: Fri, 08 May 2015 15:22:45 -0700
> From: Eric Dumazet
>
> Wolfgang Walter reported crashes using traceroute6, root caused to this
> commit.
>
> UDP packets can be sent from RAW sockets.
>
> Note that the faulty commit addressed a non problem, as normal
> UDP v6 packets
From: Eric Dumazet
Date: Fri, 08 May 2015 15:05:12 -0700
> From: Eric Dumazet
>
> For DCTCP or similar ECN based deployments on fabrics with shallow
> buffers, hosts are responsible for a good part of the buffering.
>
> This patch adds an optional ce_threshold to codel & fq_codel qdiscs,
> so
From: "John W. Linville"
Date: Fri, 8 May 2015 19:22:36 -0400
> On Fri, May 08, 2015 at 01:55:15PM -0700, Cong Wang wrote:
>> On Fri, May 8, 2015 at 10:20 AM, John W. Linville
>> wrote:
>> > +
>> > +/* Setup stats when device is created */
>> > +static int geneve_init(struct net_device *dev)
>>
From: Bert Vermeulen
Date: Fri, 8 May 2015 16:18:49 +0200
> When the bus id was supplied via a struct platform_device, the driver wasn't
> handling -1 to mean an unspecified id of the only instance of this driver,
> as the platform spec requires.
>
> Signed-off-by: Bert Vermeulen
Applied, tha
From: Mathias Kretschmer
Date: Fri, 8 May 2015 15:44:37 +0200
> This patch fixes an issue where the send(MSG_DONTWAIT) call
> on a TX_RING is not fully non-blocking in cases where the device's sndBuf is
> full. We pass nonblock=true to sock_alloc_send_skb() and return any possibly
> occuring erro
From: Michal Schmidt
Date: Thu, 7 May 2015 20:37:10 +0200
> Commit 12a8541d5c82 "bnx2x: Delay during kdump load" added a 5 seconds
> delay to bnx2x's probe function in the kdump case to let the firmware
> realize the old driver is gone.
>
> The problem with the delay is that it is per-device, s
From: Nicolas Schichan
Date: Thu, 7 May 2015 17:14:21 +0200
> The ARM JIT code emits "ldr rX, [pc, #offset]" to access the literal
> pool. #offset maximum value is 4095 and if the generated code is too
> large, the #offset value can overflow and not point to the expected
> slot in the literal po
From: Nicolas Schichan
Date: Wed, 6 May 2015 18:31:56 +0200
> In that case, emit_udiv() will be called with rn == ARM_R0 (r_scratch)
> and loading rm first into ARM_R0 will result in jit_udiv() function
> being called the same dividend and divisor. Fix that by loading rn
> first into ARM_R1 and
From: Russell King - ARM Linux
Date: Sun, 10 May 2015 20:40:28 +0100
> I think you have taken previous ARM net JIT patches, so I think it
> makes sense if you continue to do so. I'm not knowledgable of the
> JIT interface myself, all I can say about many of these patches is
> that they look okay
From: Alexander Duyck
Date: Wed, 06 May 2015 21:11:34 -0700
> This patch series addresses several things.
>
> First I found an issue in the performance of the pfmemalloc check from
> build_skb. To work around it I have provided a cached copy of pfmemalloc
> to be used in __netdev_alloc_skb and
From: Alexander Duyck
Date: Wed, 06 May 2015 21:11:40 -0700
> + /* use OR instead of assignment to avoid clearing of bits in mask */
> + if (pfmemalloc)
> + skb->pfmemalloc = 1;
> + skb->head_frag = 1;
...
> + /* use OR instead of assignment to avoid clearing of bits
From: Varka Bhadram
Date: Thu, 7 May 2015 09:17:48 +0530
> All spi based drivers have an instance of struct spi_device
> as spi. This patch renames spi_device to spi to synchronize
> with all the drivers.
>
> Signed-off-by: Varka Bhadram
Applied, thank you.
--
To unsubscribe from this list: s
From: "K. Y. Srinivasan"
Date: Wed, 6 May 2015 15:29:05 -0700
> - ret = vmbus_sendpacket_pagebuffer(out_channel,
> - pgbuf,
> - packet->page_buf_cnt,
> -
Hi david,
On 10/05/15 03:01, David Miller wrote:
From: Philippe Reynes
Date: Sun, 10 May 2015 00:16:21 +0200
Hi Fabio,
On 09/05/15 23:59, Fabio Estevam wrote:
Philippe,
On Sat, May 9, 2015 at 6:17 PM, Russell King - ARM Linux
wrote:
Using memcpy_fromio() to copy device registers is no
Sun, May 10, 2015 at 06:48:08PM CEST, sfel...@gmail.com wrote:
>From: Scott Feldman
>
>Well almost clean: ignore the CHECKs for space after cast operator and some
>longer-than-80 char cases where for readability it's better to keep as-is.
>
>Signed-off-by: Scott Feldman
Acked-by: Jiri Pirko
--
Sun, May 10, 2015 at 06:48:01PM CEST, sfel...@gmail.com wrote:
>From: Scott Feldman
>
>Rocker, bonding and team and switch over to the new
>switchdev_port_bridge_dellink to avoid duplicating code in each driver.
>
>Signed-off-by: Scott Feldman
Acked-by: Jiri Pirko
--
To unsubscribe from this li
Sun, May 10, 2015 at 06:48:00PM CEST, sfel...@gmail.com wrote:
>From: Scott Feldman
>
>Same change as setlink. Provide the wrapper op for SELF ndo_bridge_dellink
>and call into the switchdev driver to delete afspec VLANs.
>
>Signed-off-by: Scott Feldman
Acked-by: Jiri Pirko
--
To unsubscribe f
On 5/10/15 12:00 PM, Pablo Neira Ayuso wrote:
The inlined code into the ingress core path seems to have an impact to
people that don't need this, even with the static key.
two emails ago you've accused me of non-technical comments and
now I've posted real numbers that show no impact on users t
On 5/10/15 10:59 AM, Pablo Neira Ayuso wrote:
On Sun, May 10, 2015 at 10:45:42AM -0700, Alexei Starovoitov wrote:
On 5/10/15 9:59 AM, Pablo Neira Ayuso wrote:
The qdisc ingress filtering code is embedded into the core most likely because
at that time we had no RCU in place to define a hook. Thi
This patch improves performance of the super-critical ingress path by moving
the qdisc ingress code to sch_ingress, where this really belongs.
This patchset also removes the Qdisc->enqueue() indirection since this does not
enqueue anything.
This is roughly 92199pps, ~0.42% more performance on my
From: Scott Feldman
New attr-based bridge_setlink can recurse lower devs and recover on err, so
remove old wrapper (including ndo_dflt_switchdev_port_bridge_setlink).
Signed-off-by: Scott Feldman
Acked-by: Jiri Pirko
---
include/net/switchdev.h |9 -
net/switchdev/switchdev.c |
From: Scott Feldman
v7:
Address review comments:
- [Jiri] split the br_setlink and br_dellink reverts into their own patches
- [Jiri] some parameter cleanup of rocker's memory allocators
- [Jiri] pass trans mode as formal parameter rather than hanging off of
rocker_port.
v6:
Address r
Previously RDMA CM relied on the CM module to check the incoming requests
against "compare data" struct to dispatch events for different RDMA CM IDs
based on the request parameters (IP address, address family, etc.). With
namespace support, multiple namespaces in RDMA CM will need to share a
single
Enabling network namespaces for RDMA CM will allow processes on different
namespaces to listen on the same port. In order to leave namespace support
out of the CM layer, this requires that multiple RDMA CM IDs will be able
to share a single CM ID.
This patch adds infrastructure to retrieve an exis
From: Yotam Kenneth
In the case of IPoIB, and maybe in other cases, the network device is
managed by an upper-layer protocol (ULP). In order to expose this
network device to other users of the IB device, let ULPs implement
a callback that returns network device according to connection parameters.
Currently the RDMA subsystem's device list and client list are protected by
a single mutex. This prevents adding user-facing APIs that iterate these
lists, since using them may cause a deadlock. The patch attempts to solve
this problem by adding an SRCU to protect the lists. Readers now don't need
From: Rana Shahout
Introduce mlx5_core_modify_cq_moderation() to be used by the netdev, to
set hardware coalescing.
Signed-off-by: Rana Shahout
Signed-off-by: Saeed Mahameed
Signed-off-by: Amir Vadai
---
drivers/net/ethernet/mellanox/mlx5/core/cq.c | 18 ++
include/linux/mlx5
From: Saeed Mahameed
- Query all supported types of dev caps on driver load.
- Store the Cap data outbox per cap type into driver private data.
- Introduce new Macros to access/dump stored caps (using the auto
generated data types).
- Obsolete SW representation of dev caps (no need for SW copy
This patch contains the resource handling files:
- flow_table.c: This file contains the code to handle the low level API
to configure hardware flow table. It is separated from
the flow_table_en.c, because it will be used in the
future by Raw Ethernet
This is the Ethernet part of the driver for the Mellanox ConnectX(R)-4
Single/Dual-Port Adapter supporting 100Gb/s with VPI. The driver
extends the existing mlx5 driver with Ethernet functionality.
This patch contains the driver entry points but does not include
transmit and receive (see the prev
From: Saeed Mahameed
Those registers will be used by the ethtool to set/get settings.
Signed-off-by: Rana Shahout
Signed-off-by: Saeed Mahameed
Signed-off-by: Amir Vadai
---
drivers/net/ethernet/mellanox/mlx5/core/port.c | 77 ++
include/linux/mlx5/driver.h
en_[rt]x.c contains the data path related code specific to tx or rx.
en_txrx.c contains data path code which is common for both the rx and
tx, this is mainly napi related code.
Below are the objects that are being used by the hardware and the driver
in the data path:
Channel - one channel per IRQ
From: Saeed Mahameed
Introduce mlx5_rename_eq() to rename an EQ. This will be used by a
following commit for the Ethernet functionality of the driver.
Signed-off-by: Saeed Mahameed
Signed-off-by: Amir Vadai
---
drivers/net/ethernet/mellanox/mlx5/core/main.c | 25 --
..
34 matches
Mail list logo
