- Only x540 and 82599 devices support LRO.
- Add the appropriate HW configuration.
- Add RSC aware rx_pkt_burst() handlers:
- Implemented bulk allocation and non-bulk allocation versions.
- Add LRO-specific fields to rte_eth_rxmode, to rte_eth_dev_data
and to
- ixgbe_rx_alloc_bufs():
- Reset the rte_mbuf fields only when requested.
- Take the RDT update out of the function.
- Add the stub when RTE_LIBRTE_IXGBE_RX_ALLOW_BULK_ALLOC is not defined.
- ixgbe_recv_scattered_pkts():
- Take the code that updates the fields
- Removed the not needed casting.
- ixgbe_dev_rx_init(): shorten the lines by defining a local alias variable
to access
>data->dev_conf.rxmode.
Signed-off-by: Vlad Zolotarov
---
New in v6:
- Fixed a compilation error caused by a patches recomposition during
This series adds the missing flow for enabling the LRO in the ethdev and
adds a support for this feature in the ixgbe PMD. There is a big hope that this
initiative is going to be picked up by some Intel developer that would add the
LRO support
to other Intel PMDs.
The series starts with some
>
> > Hi Vlad,
> >
> >> -Original Message-
> >> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Vlad Zolotarov
> >> Sent: Monday, March 09, 2015 7:07 PM
> >> To: dev at dpdk.org
> >> Subject: [dpdk-dev] [PATCH v6 3/3] ixgbe: Add LRO support
> >>
> >> - Only x540 and 82599
On 03/10/15 02:30, Ananyev, Konstantin wrote:
> Hi Vlad,
>
>> -Original Message-
>> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Vlad Zolotarov
>> Sent: Monday, March 09, 2015 7:07 PM
>> To: dev at dpdk.org
>> Subject: [dpdk-dev] [PATCH v6 3/3] ixgbe: Add LRO support
>>
>>
> Documentation (with overlaps)
> -
> +M: Siobhan Butler
> F: doc/
Acked-by: Thomas Monjalon
Applied, thanks
> Intel ixgbe
> +M: Helin Zhang
> +M: Konstantin Ananyev
> F: lib/librte_pmd_ixgbe/
Acked-by: Thomas Monjalon
Applied, thanks
> --- a/MAINTAINERS
> +++ b/MAINTAINERS
> Timers
> +M: Robert Sanford
> F: lib/librte_timer/
Applied, thanks
2015-02-05 08:34, Liu, Jijiang:
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> > 2015-02-05 10:31, Jijiang Liu:
> > > I will be a volunteer of reviewing the following files:
> > >lib/librte_eal/linuxapp/xen_dom0/*
> > >lib/librte_eal/linuxapp/eal/eal_xen_memory.c
> > >
> Due the unified flow type changes, this patchset updates the commands for
> flow director
> and correct the flow types ixgbe supported.
>
> Jingjing Wu (2):
> ixgbe: correct the flow types supported
> doc: update flow director commands
Applied, thanks
> When user quits testpmd, and there is traffic being forwarded,
> that may produce a segmentation fault, due to ports being closed,
> while they are still transmitting packets.
>
> This patch prevents the issue from happening,
> by stopping packet forwarding before closing the ports.
>
>
> The contigmem module was using an "int" type for specifying the
> size of blocks of memory to be reserved. A 2GB block was therefore
> overflowing the signed 32-bit value, making 1GB the largest block
> size that could be reserved as a single unit.
> The fix is to change the type used for the
2015-03-09 17:05, Pawel Wodkowski:
> Fix check of returned values during application init phase.
>
> Signed-off-by: Pawel Wodkowski
Applied, thanks
> Mimic UIO/VFIO drivers behavior by not causing errors when a device cannot
> be initialized due to missing or mismatching kernel modules. Display helpful
> messages instead, such as:
>
> [...]
> EAL: PCI device :83:00.0 on NUMA socket 1
> EAL: probe driver: 15b3:1007 librte_pmd_mlx4
>
Hi All,
using my dpdk application on a guest over qemu-kvm (version 1.5.3) I fail
to initialize the hpet.
this returns failure due to "Device or resource busy".
This is also the case when accessing the "/dev/hpet", so obviously this is
not a dpdk issue.
I couldn't resolve this issue at the
Just a bit more on this. We've found that when a link goes down, the TX
descriptor ring appears to fill up with packets fairly quickly, and then
calls to rte_eth_tx_burst() start returning zero. Our application handles
this case, and frees the mbufs that could not be sent.
However, when link is
2015-03-06 18:23, Stephen Hemminger:
> These are things found while reviewing ixgbe driver.
> No bugs or functionality changes, just minor things like naming
> and use of const.
>
> Stephen Hemminger (5):
> ixgbe: make txq_ops const
> ixgbe: make register maps const
> ixgbe: make bulk alloc
On 03/10/15 02:30, Ananyev, Konstantin wrote:
> Hi Vlad,
>
>> -Original Message-
>> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Vlad Zolotarov
>> Sent: Monday, March 09, 2015 7:07 PM
>> To: dev at dpdk.org
>> Subject: [dpdk-dev] [PATCH v6 3/3] ixgbe: Add LRO support
>>
>>
Hi,
You need use dpdkvhost port for VM, not dpdk port.
For example, here is the log
9029690c-9544-4ad0-aa41-b018c1a755e2
Bridge "br1"
Port "dpdkvhost1"
Interface "dpdkvhost1"
type: dpdkvhost
Port "br1"
Interface "br1"
2015-02-24 21:06, Thomas Monjalon:
> 2015-02-24 11:18, Stephen Hemminger:
> > On Mon, 23 Feb 2015 23:51:31 +
> > Thomas Monjalon wrote:
> > > Stephen, this patchset is partially acked and could enter in 2.0.0-rc1.
> > > May you send a v3 addressing comments? Or should I break the serie by
> >
2015-03-05 16:10, Stephen Hemminger:
> From: Stephen Hemminger
>
> Revised version of earlier patches.
> Incorporate the small packet optimization
> Add more cleanups
>
> Stephen Hemminger (10):
> vmxnet3: fix link state handling
> vmxnet3: enable VLAN filtering
> vmxnet3: remove mtu
This patch adds finalization code to free resources allocated by
the PMD. This is based on earlier patches for other PMDs by Tetsuya
Mukawa , with corrections related to data->name.
Signed-off-by: John W. Linville
Cc: Tetsuya Mukawa
---
lib/librte_pmd_af_packet/rte_eth_af_packet.c | 56
2015-03-05 00:16, Thomas Monjalon:
> 2015-01-06 11:44, Thomas Monjalon:
> > 2014-12-27 15:13, Stephen Hemminger:
> > > This is a patch to address the conflict between
> > > and the definitions in . It has two side effects
> > > worth discussion:
> > > 1. It forces inclusion of net/ethernet.h
>
2015-03-05 19:40, Thomas Monjalon:
> The pointer to struct eth_driver is not used and is already set
> in struct rte_eth_dev.
>
> It's a small cleanup in PMD API which probably needs more
> attention to make clear what is a driver, a PCI driver, an
> ethernet driver, etc.
>
> Signed-off-by:
2015-03-05 19:26, Neil Horman:
> On Wed, Mar 04, 2015 at 10:50:00PM +0100, David Marchand wrote:
> > This is a first cleanup at trying to remove references to other dpdk
> > libraries
> > from eal.
> >
> > This cleanup is focused on rte_tailq api which has been marked as "for
> > internal
> >
Hi Arkajit,
I can't find your attachment.
On Tue, Mar 10, 2015 at 10:23 AM, Arkajit Ghosh
wrote:
>
> Hi,
>
> Can any one please suggest how to proceed now.
>
>
> Thanks & Regards
> Arkajit Ghosh
>
>
> -Arkajit Ghosh/DEL/TCS wrote: -
> To: Ngo
On Mon, 9 Mar 2015 23:32:48 +
Yong Wang wrote:
> >
> >I didn?t see where mtu is used to calculate how many rx descriptors will
> >be needed for each packet. Furthermore, as pointed out by the following
> >code comments, the device requires the first rx buffer of a packet be of
> >type
Fix possible memory leak issue: free kvlist before return;
Fix possible resource lost issue: close qssockfd before return;
Signed-off-by: Changchun Ouyang
---
Change in v6:
- Refine exit point;
Change in v5:
- Initialize qsockfd with -1;
Change in v4:
- Check sockfd in
2015-03-10 08:49, Ouyang, Changchun:
> From: Wodkowski, PawelX
> > > + if (kvlist != NULL)
> >
> > No need for if(). This part was fine previous patch.
> >
>
> If kvlist is NULL, no reason to call rte_kvargs_free to free it.
> So, adding this test is better.
No, we don't need to double check.
Hi,
Please find the attachement.
Thanks & Regards
Arkajit Ghosh
-Ngo Doan Lap wrote: -
To: Arkajit Ghosh
From: Ngo Doan Lap
Date: 03/10/2015 10:25AM
Cc: dev at dpdk.org
Subject: Re: #PERSONAL#[dpdk-dev] Guest
Hey Folks,
I'm running into an issue that I hope is obvious and simple. We're running
DPDK 1.6.2 with an 82599 NIC. We find that if, while running traffic, we
disconnect a port and then later reconnect it, we never regain the ability
to transmit packets out of that port after it comes back up.
Sorry for that. Will send the new patch soon
-Original Message-
From: Bruce Richardson [mailto:bruce.richard...@intel.com]
Sent: 09 March 2015 12:28
To: Raz Amir
Cc: dev at dpdk.org
Subject: Re: [PATCH v3] pci: save list of detached devices, and re-probe
during driver unload
On Mon, Mar
Hello Neil,
On Mon, Mar 9, 2015 at 4:21 PM, Neil Horman wrote:
> On Mon, Mar 09, 2015 at 03:56:38PM +0100, David Marchand wrote:
> > Loading shared libraries should be done at the very start of eal init so
> that
> > the code statically built in dpdk and the code loaded from shared
> objects is
>-Original Message-
>From: Mcnamara, John
>Sent: Tuesday, March 10, 2015 8:27 AM
>To: Qiu, Michael; Kavanagh, Mark B; dev at dpdk.org; Panu Matilainen
>Subject: RE: [dpdk-dev] rte_memcpy.h: additional cflags required with OVS
>
>> -Original Message-
>> From: dev
On 2015-03-10 09:49, Ouyang, Changchun wrote:
>
>
>> -Original Message-
>> From: Wodkowski, PawelX
>> Sent: Tuesday, March 10, 2015 4:37 PM
>> To: Ouyang, Changchun; dev at dpdk.org
>> Cc: linville at tuxdriver.com; nhorman at tuxdriver.com
>> Subject: Re: [PATCH v6] af_packet: Fix some
08.03.2015 0:39, Thomas Monjalon ?:
> 2015-03-06 01:39, Qiu, Michael:
>> On 3/6/2015 1:11 AM, Thomas Monjalon wrote:
>>> 2015-03-06 00:55, Michael Qiu:
... skipped ...
+#if defined RTE_ARCH_I686 || defined RTE_ARCH_X86_64
static inline uint32_t
On 03/10/2015 05:04 AM, Qiu, Michael wrote:
> On 3/10/2015 12:44 AM, Kavanagh, Mark B wrote:
>> Hi,
>>
>> Compilation of Open vSwitch fails when linked against current HEAD of DPDK
>> (f2552cd5).
>>
>> The source of this issue appears to be commit ID 9144d6b: "eal/x86: optimize
>> memcpy for SSE
> -
> - return 0;
> +exit:
> + if (kvlist != NULL)
No need for if(). This part was fine previous patch.
> + rte_kvargs_free(kvlist);
> + return ret;
> }
>
> static struct rte_driver pmd_af_packet_drv = {
>
--
Pawel
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Vlad Zolotarov
> Sent: Monday, March 09, 2015 4:29 PM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH v2 1/3] ixgbe: Use the
> rte_le_to_cpu_xx()/rte_cpu_to_le_xx() when reading/setting HW ring descriptor
>-Original Message-
>From: Mcnamara, John
>Sent: Tuesday, March 10, 2015 12:57 AM
>To: Mcnamara, John; Kavanagh, Mark B; dev at dpdk.org
>Subject: RE: rte_memcpy.h: additional cflags required with OVS
>
>> -Original Message-
>> From: dev [mailto:dev-bounces at dpdk.org] On Behalf
On Mon, Mar 09, 2015 at 03:56:39PM +0100, David Marchand wrote:
> Playing with virtio link status triggers a segfault because of an incorrect io
> privilege level.
>
> To reproduce the problem, virtio device must be bound to igb_uio to have lsc
> enabled.
>
> $ lspci |grep Ethernet
> 00:03.0
From: Stephen Hemminger
The old code was doing a floating point divide for each rte_dequeue()
which is very expensive. Change to using fixed point scaled math instead.
This improved performance from 5Gbit/sec to 10 Gbit/sec
Signed-off-by: Stephen Hemminger
---
v2 -- no
From: Stephen Hemminger
The rte_sched statistics API should allow reading statistics without
clearing. Make auto-clear optional.
Signed-off-by: Stephen Hemminger
---
v2 -- change to flag for clear from separate read/clear
app/test/test_sched.c| 4 ++--
From: Stephen Hemminger
Add new statistic to keep track of drops due to RED.
Signed-off-by: Stephen Hemminger
---
v2 -- no changes
lib/librte_sched/rte_sched.c | 28 +++-
lib/librte_sched/rte_sched.h | 6 ++
2 files changed, 29
From: Stephen Hemminger
The QoS subport is limited to 8 bits in original code.
But customers demanded ability to support full number of VLAN's (4096)
therefore use the full part of the tag field of mbuf.
Resize the pipe as well to allow for more pipes in future and
avoid
From: Stephen Hemminger
Want to be able to build with RTE_SCHED_RED enabled but
allow disabling RED on a per-queue basis at runtime.
RED is disabled unless min/max thresholds set.
Signed-off-by: Stephen Hemmminger
---
v2 -- no changes
lib/librte_sched/rte_sched.c | 9
From: Stephen Hemminger
syslog does not like tabs in log messages; tab gets translated to #011
Signed-off-by: Stephen Hemminger
---
v2 -- no changes
lib/librte_sched/rte_sched.c | 14 +++---
1 file changed, 7 insertions(+), 7 deletions(-)
diff --git
These are the bug fixes and enhancements to the RTE sched
code. Based off 2.0 latest git.
Stephen Hemminger (6):
rte_sched: don't put tabs in log messages
rte_sched: make RED optional at runtime
rte_sched: expand scheduler hierarchy for more VLAN's
rte_sched: keep track of RED drops
Hi,
Can any one please suggest how to proceed now.
Thanks & Regards
Arkajit Ghosh
-Arkajit Ghosh/DEL/TCS wrote: -
To: Ngo Doan Lap
From: Arkajit Ghosh/DEL/TCS
Date: 03/09/2015 09:39AM
Cc: dev at dpdk.org
Subject: Re: [dpdk-dev] Guest
> -Original Message-
> From: Wodkowski, PawelX
> Sent: Tuesday, March 10, 2015 4:37 PM
> To: Ouyang, Changchun; dev at dpdk.org
> Cc: linville at tuxdriver.com; nhorman at tuxdriver.com
> Subject: Re: [PATCH v6] af_packet: Fix some klocwork errors
>
> > -
> > - return 0;
> > +exit:
>
On Tue, 10 Mar 2015 14:29:11 +0100
Thomas Monjalon wrote:
> Hi Stephen,
> If, by any chance, you are willing to reply to this thread,
> maybe you would like to send a non-rfc patch with these 2 additions:
> - rename addr_bytes to ether_addr_octet everywhere
> - mark addr_bytes as deprecated in
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Qiu, Michael
> Sent: Tuesday, March 10, 2015 3:05 AM
> To: Kavanagh, Mark B; dev at dpdk.org
> Subject: Re: [dpdk-dev] rte_memcpy.h: additional cflags required with OVS
>
> What's your gcc version? this should
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Stephen Hemminger
> Sent: Sunday, February 15, 2015 11:25 PM
> To: dev at dpdk.org
> Cc: Stephen Hemminger
> Subject: [dpdk-dev] [PATCH 2/5] enic: fix device to work with Xen DOM0
>
> It is possible to
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Stephen Hemminger
> Sent: Sunday, February 15, 2015 11:25 PM
> To: dev at dpdk.org
> Cc: Stephen Hemminger
> Subject: [dpdk-dev] [PATCH 1/5] xen: allow choosing dom0 support at runtime
>
> The previous code
On Tue, Mar 10, 2015 at 10:08:24AM +0100, David Marchand wrote:
> Hello Neil,
>
> On Mon, Mar 9, 2015 at 4:21 PM, Neil Horman wrote:
>
> > On Mon, Mar 09, 2015 at 03:56:38PM +0100, David Marchand wrote:
> > > Loading shared libraries should be done at the very start of eal init so
> > that
> >
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Stephen
> Hemminger
> Sent: Saturday, March 7, 2015 10:23 AM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH 5/5] ixgbe: rename igb_* to ixgbe_*
>
> To avoid any possible confusion or breakage, rename all
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Stephen
> Hemminger
> Sent: Saturday, March 7, 2015 10:23 AM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH 4/5] ixgbe: rename set_tx_function
>
> All global functions in a driver should use the same
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Stephen
> Hemminger
> Sent: Saturday, March 7, 2015 10:23 AM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH 3/5] ixgbe: make bulk alloc static
>
> Only used in this file, make it static.
>
>
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Stephen
> Hemminger
> Sent: Saturday, March 7, 2015 10:23 AM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH 2/5] ixgbe: make register maps const
>
> These are const data structures, just put them in txt
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Stephen
> Hemminger
> Sent: Saturday, March 7, 2015 10:23 AM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH 1/5] ixgbe: make txq_ops const
>
> All virtual function tables should be const so they are put
On 3/10/2015 12:44 AM, Kavanagh, Mark B wrote:
> Hi,
>
> Compilation of Open vSwitch fails when linked against current HEAD of DPDK
> (f2552cd5).
>
> The source of this issue appears to be commit ID 9144d6b: "eal/x86: optimize
> memcpy for SSE and AVX", and can be resolved by passing an
On 3/10/2015 8:57 AM, Mcnamara, John wrote:
>> -Original Message-
>> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Mcnamara, John
>> Sent: Monday, March 9, 2015 5:51 PM
>> To: Kavanagh, Mark B; dev at dpdk.org
>> Subject: Re: [dpdk-dev] rte_memcpy.h: additional cflags required
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Mcnamara, John
> Sent: Monday, March 9, 2015 5:51 PM
> To: Kavanagh, Mark B; dev at dpdk.org
> Subject: Re: [dpdk-dev] rte_memcpy.h: additional cflags required with OVS
>> In the meantime the following might
Hi Vlad,
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Vlad Zolotarov
> Sent: Monday, March 09, 2015 7:07 PM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH v6 3/3] ixgbe: Add LRO support
>
> - Only x540 and 82599 devices support LRO.
> - Add
---
MAINTAINERS | 7 +++
1 file changed, 7 insertions(+)
diff --git a/MAINTAINERS b/MAINTAINERS
index 07fdf5e..b4327d3 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -131,6 +131,12 @@ F: app/test-pmd/mempool_*
F: examples/vhost_xen/
F:
---
MAINTAINERS | 7 +++
1 file changed, 7 insertions(+)
diff --git a/MAINTAINERS b/MAINTAINERS
index 07fdf5e..b4327d3 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -131,6 +131,12 @@ F: app/test-pmd/mempool_*
F: examples/vhost_xen/
F:
On 3/9/15, 4:28 PM, "Yong Wang" wrote:
>On 3/5/15, 4:10 PM, "Stephen Hemminger"
>wrote:
>
>>From: Stephen Hemminger
>>
>>Add support for linking multi-segment buffers together to
>>handle Jumbo packets.
>>
>>Signed-off-by: Stephen Hemminger
>>---
>> lib/librte_pmd_vmxnet3/vmxnet3_ethdev.c |
On 3/5/15, 4:10 PM, "Stephen Hemminger" wrote:
>From: Stephen Hemminger
>
>Add support for linking multi-segment buffers together to
>handle Jumbo packets.
>
>Signed-off-by: Stephen Hemminger
>---
> lib/librte_pmd_vmxnet3/vmxnet3_ethdev.c | 3 +-
> lib/librte_pmd_vmxnet3/vmxnet3_ring.h | 2
69 matches
Mail list logo