Signed-off-by: Xiao Wang
---
MAINTAINERS | 7 +++
1 file changed, 7 insertions(+)
diff --git a/MAINTAINERS b/MAINTAINERS
index fe7c59da1..8bd5bf75c 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -517,6 +517,13 @@ T: git://dpdk.org/next/dpdk-next-net-intel
F: drivers/net/avf/
F: doc/guides/n
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Didier Pallard
> Sent: Wednesday, March 28, 2018 8:44 AM
> To: dev@dpdk.org
> Subject: [dpdk-dev] [PATCH 0/8] net/vmxnet3: fix offload issues
>
> This patchset fixes several issues found in vmxnet3 driver
> when en
Hi Vipin,
I missed your point:
You claim that TAP should work regardless of any pi.proto values.
Can you confirm that for ALL kernels versions (past and future)?
> -Original Message-
> From: Ophir Munk
> Sent: Saturday, April 21, 2018 12:49 AM
> To: Varghese, Vipin ; dev@dpdk.org;
> pascal
Hi Vipin,
Please find comments inline.
> -Original Message-
> From: Varghese, Vipin [mailto:vipin.vargh...@intel.com]
> Sent: Friday, April 13, 2018 6:18 AM
> To: Ophir Munk ; dev@dpdk.org;
> pascal.ma...@6wind.com; Yigit, Ferruh ; Thomas
> Monjalon ; Olga Shern ;
> Shahaf Shuler
> Subje
We introduced private rte_eth_dev_data to allow vdev to be created
both in primary process and secondary process(es). This is not
friendly to multi-process model, for example, it leads to port id
contention issue if two processes both find the data entry is free.
And to get stats of primary vdev i
dpdk-procinfo, as a secondary process, cannot fetch stats for vdev.
This patch enables that by attaching the port from the shared data.
We also fill the eth dev ops, with only some ops works in secondary
process, for example, stats_get().
Note that, we still cannot Rx/Tx packets on the ports whic
To scan the vdevs in primary, we send request to primary process
to obtain the names for vdevs.
Only the name is shared from the primary. In probe(), the device
driver is supposed to locate (or request more) the detail
information from the primary.
Signed-off-by: Jianfeng Tan
Reviewed-by: Qi Zha
As we could add virtual devices from different threads now, we
add a spin lock to protect the vdev device list.
Suggested-by: Anatoly Burakov
Signed-off-by: Jianfeng Tan
Reviewed-by: Qi Zhang
---
drivers/bus/vdev/vdev.c | 95 +++--
1 file changed, 69
v4:
- Change the lock code style as suggested by Anatoly.
- Add function note as suggested by Anatoly.
v3:
- Update doc.
- Rebase on master.
v2:
- Add spinlock for vdev device list as suggested by Anatoly.
- Add ring, cxgbe and remove the free in each PMDs as suggested by Matan.
- R
Adjust the init sequence: put mp channel init before bus scan
so that we can init the vdev bus through mp channel in the
secondary process before the bus scan.
Signed-off-by: Jianfeng Tan
Reviewed-by: Qi Zhang
---
lib/librte_eal/bsdapp/eal/eal.c | 23 +--
lib/librte_eal/li
> > +
> > +static void sigbus_handler(int signum __rte_unused, siginfo_t *info,
> > + void *ctx __rte_unused)
> > +{
> > + int ret;
> > +
> > + RTE_LOG(ERR, EAL, "SIGBUS error, fault address:%p\n", info->si_addr);
> > + ret = dev_uev_failure_process(NULL, info->si_a
> -Original Message-
> From: Hemant Agrawal [mailto:hemant.agra...@nxp.com]
> Sent: Thursday, April 19, 2018 5:53 PM
> To: dev@dpdk.org
> Cc: De Lara Guarch, Pablo ;
> akhil.go...@nxp.com
> Subject: [PATCH 1/4] crypto/dpaa_sec: support dynamic logging
>
> This patch adds the support for
> -Original Message-
> From: Akhil Goyal [mailto:akhil.go...@nxp.com]
> Sent: Friday, April 20, 2018 11:40 AM
> To: Hemant Agrawal ; dev@dpdk.org; De Lara
> Guarch, Pablo
> Cc: Nipun Gupta
> Subject: Re: [PATCH 4/4] crypto/dpaa_sec: update maximum queue pairs to 2
>
> Hi Hemant,
> On 4
Kevin Traynor writes:
> On 04/18/2018 02:28 PM, Thomas Monjalon wrote:
>> 18/04/2018 14:28, Ferruh Yigit:
>>> On 4/18/2018 10:14 AM, Thomas Monjalon wrote:
18/04/2018 11:05, Ferruh Yigit:
> On 4/11/2018 12:28 AM, Thomas Monjalon wrote:
>> -Typically a new stable release versi
On 20-Apr-18 3:01 PM, Arnon Warshavsky wrote:
I don't think rte_panic should be considered forbidden. Rather their
use should be flagged (as this patch does). However, the 'exit 1'
(which will return a failure for the automatic checkpatch script bot)
might end up problematic a
On 20-Apr-18 4:32 PM, Tan, Jianfeng wrote:
On 4/20/2018 11:19 PM, Burakov, Anatoly wrote:
On 20-Apr-18 3:28 PM, Tan, Jianfeng wrote:
On 4/20/2018 4:41 PM, Burakov, Anatoly wrote:
On 19-Apr-18 5:50 PM, Jianfeng Tan wrote:
To scan the vdevs in primary, we send request to primary process
to
On 4/20/2018 11:19 PM, Burakov, Anatoly wrote:
On 20-Apr-18 3:28 PM, Tan, Jianfeng wrote:
On 4/20/2018 4:41 PM, Burakov, Anatoly wrote:
On 19-Apr-18 5:50 PM, Jianfeng Tan wrote:
To scan the vdevs in primary, we send request to primary process
to obtain the names for vdevs.
Only the name i
Hi Shally,
> -Original Message-
> From: Verma, Shally [mailto:shally.ve...@cavium.com]
> Sent: Friday, April 20, 2018 6:31 PM
> To: Gujjar, Abhinandan S ; Jacob, Jerin
> ; hemant.agra...@nxp.com;
> akhil.go...@nxp.com; dev@dpdk.org
> Cc: De Lara Guarch, Pablo ; Doherty, Declan
> ; Vangati,
For non-legacy mode, we are preallocating space for hugepages, so
we know in advance which pages we will be able to allocate, and
which we won't. However, the init procedure was using hugepage
counts gathered from sysfs and paid no attention to hugepage
sizes that were actually available for reserv
Previously, if we couldn't preallocate VA space on 32-bit for
one page size, we simply bailed out, even though we could've
tried allocating VA space with other page sizes.
For example, if user had both 1G and 2M pages enabled, and
has asked DPDK to allocate memory on both sockets, DPDK
would've tr
32-bit mode has an upper limit on amount of VA space it can preallocate,
but the original implementation used the wrong constant, resulting in
failure to initialize due to integer overflow. Fix it by using the
correct constant.
Fixes: 66cc45e293ed ("mem: replace memseg with memseg lists")
Cc: anat
On 4/20/2018 11:16 PM, Burakov, Anatoly wrote:
On 20-Apr-18 3:19 PM, Tan, Jianfeng wrote:
On 4/20/2018 4:26 PM, Burakov, Anatoly wrote:
On 19-Apr-18 5:50 PM, Jianfeng Tan wrote:
As we could add virtual devices from different threads now, we
add a spin lock to protect the vdev device list.
On 20-Apr-18 4:10 PM, Xiao Wang wrote:
A previously mapped region is skipped during the search, leading to
DMA unmap fails.
This patch fixes it and rewords the comment.
Fixes: 73a639085938 ("vfio: allow to map other memory regions")
Signed-off-by: Xiao Wang
---
Acked-by: Anatoly Burakov
On Fri, 20 Apr 2018 16:15:40 +0530
Sunil Kumar Kori wrote:
> dpaa_eth_dev_info(dev, &dev_info);
> if (((~(dev_info.rx_offload_capa) & rx_offloads) != 0)) {
> - DPAA_PMD_ERR("Some Rx offloads are not supported "
> + DPAA_PMD_WARN("Some Rx offloads are not suppor
On 20-Apr-18 3:28 PM, Tan, Jianfeng wrote:
On 4/20/2018 4:41 PM, Burakov, Anatoly wrote:
On 19-Apr-18 5:50 PM, Jianfeng Tan wrote:
To scan the vdevs in primary, we send request to primary process
to obtain the names for vdevs.
Only the name is shared from the primary. In probe(), the device
> -Original Message-
> From: Verma, Shally [mailto:shally.ve...@cavium.com]
> Sent: Friday, April 20, 2018 6:18 PM
> To: Gujjar, Abhinandan S ; Jacob, Jerin
> ; hemant.agra...@nxp.com;
> akhil.go...@nxp.com; dev@dpdk.org
> Cc: De Lara Guarch, Pablo ; Doherty, Declan
> ; Vangati, Narender
In original implementation, timeout event for an async request
will be ignored. As a result, an async request will never
trigger the action if it cannot receive any reply any more.
We fix this by counting timeout as a processed reply.
Fixes: f05e26051c15 ("eal: add IPC asynchronous request")
Sig
v2:
- Fix typo.
- Rebase on below patches:
http://dpdk.org/dev/patchwork/patch/38358/
http://dpdk.org/dev/patchwork/patch/38359/
http://dpdk.org/dev/patchwork/patch/38360/
Patch 1 is to rename some variable names for better readability.
Patch 2 is a bug fix.
Jianfeng Tan (2):
i
Following below commit, we change some internal function and variable
names:
commit ce3a7312357b ("eal: rename IPC request as synchronous one")
Also use calloc to supersede malloc + memset for code clean up.
Signed-off-by: Jianfeng Tan
Acked-by: Anatoly Burakov
---
lib/librte_eal/common/eal_
On 20-Apr-18 3:19 PM, Tan, Jianfeng wrote:
On 4/20/2018 4:26 PM, Burakov, Anatoly wrote:
On 19-Apr-18 5:50 PM, Jianfeng Tan wrote:
As we could add virtual devices from different threads now, we
add a spin lock to protect the vdev device list.
Suggested-by: Anatoly Burakov
Signed-off-by: Jia
Driver should return error and report error message rather than call
rte_panic().
Signed-off-by: Xiao Wang
---
drivers/net/ifc/ifcvf_vdpa.c | 30 ++
1 file changed, 22 insertions(+), 8 deletions(-)
diff --git a/drivers/net/ifc/ifcvf_vdpa.c b/drivers/net/ifc/ifcvf_vdp
On Thu, 19 Apr 2018 14:45:01 +0530
Anoob Joseph wrote:
> > +/**
> >* The eth device event type for interrupt, and maybe others in the
> > future.
> >*/
> > enum rte_eth_event_type {
> > @@ -2446,6 +2486,7 @@ enum rte_eth_event_type {
> > RTE_ETH_EVENT_INTR_RESET,
> >
Besides the fix, this patch adds a toggle parameter to ifcvf_dma_map to
avoid code duplication.
Fixes: 7737386d1c22 ("net/ifcvf: add ifcvf vdpa driver")
Signed-off-by: Xiao Wang
---
drivers/net/ifc/ifcvf_vdpa.c | 55 +++-
1 file changed, 19 insertions(+),
On 20-Apr-18 3:41 PM, Anatoly Burakov wrote:
For non-legacy mode, we are preallocating space for hugepages, so
we know in advance which pages we will be able to allocate, and
which we won't. However, the init procedure was using hugepage
counts gathered from sysfs and paid no attention to hugepag
A previously mapped region is skipped during the search, leading to
DMA unmap fails.
This patch fixes it and rewords the comment.
Fixes: 73a639085938 ("vfio: allow to map other memory regions")
Signed-off-by: Xiao Wang
---
lib/librte_eal/linuxapp/eal/eal_vfio.c | 8
1 file changed, 4
Ferruh Yigit writes:
> On 4/19/2018 7:05 AM, Alejandro Lucero wrote:
>> On Wed, Apr 18, 2018 at 1:32 PM, Aaron Conole wrote:
>>
>>> Alejandro Lucero writes:
>>>
On Tue, Apr 17, 2018 at 8:19 PM, Aaron Conole
>>> wrote:
Alejandro Lucero writes:
> I was just wondering
On 4/20/2018 3:21 PM, Ajit Khaparde wrote:
> Apologize for the multiple iterations.
> Please apply.
>
> Ajit Khaparde (3):
> net/bnxt: cache address of doorbell to subsequent access
> net/bnxt: avoid invalid vnic id in set L2 Rx mask
> net/bnxt: fix mbuf data offset initialization
Series ap
Arnon Warshavsky writes:
> Thanks Aaron
>
> Previously, it wasn't possible for mem_cfg_fd to be reused after a
>
> failure. Now it is - please reset it to -1. in these close conditions.
>
> Will do.
>
> Again, previously this would have aborted on a failure. So it needs to
> be reset to a
Coverity issue: 272575
Fixes: 133c2c6565d6 ("examples/ip_pipeline: add link object")
Signed-off-by: Fan Zhang
---
v2:
- removed unecessary link.c update
examples/ip_pipeline/cli.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/examples/ip_pipeline/cli.c b/examples/ip_pipeline/cli.c
index
> -Original Message-
> From: Ferruh Yigit
> Sent: Friday, April 20, 2018 10:35 PM
> To: Xueming(Steven) Li ; Shahaf Shuler
> ; Nelio Laranjeiro
> ; Wenzhuo Lu ; Jingjing Wu
> ;
> Thomas Monjalon ; Adrien Mazarguil
>
> Cc: dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v5 1/2] ethdev: a
On 4/20/2018 3:55 PM, Burakov, Anatoly wrote:
On 20-Apr-18 2:21 AM, Jianfeng Tan wrote:
Following below commit, we change some internal function and variable
names:
commit ce3a7312357b ("eal: rename IPC request as synchronous one")
Also use calloc to supersede malloc + memset for code clea
For non-legacy mode, we are preallocating space for hugepages, so
we know in advance which pages we will be able to allocate, and
which we won't. However, the init procedure was using hugepage
counts gathered from sysfs and paid no attention to hugepage
sizes that were actually available for reserv
Previously, if we couldn't preallocate VA space on 32-bit for
one page size, we simply bailed out, even though we could've
tried allocating VA space with other page sizes.
For example, if user had both 1G and 2M pages enabled, and
has asked DPDK to allocate memory on both sockets, DPDK
would've tr
32-bit mode has an upper limit on amount of VA space it can preallocate,
but the original implementation used the wrong constant, resulting in
failure to initialize due to integer overflow. Fix it by using the
correct constant.
Fixes: 66cc45e293ed ("mem: replace memseg with memseg lists")
Cc: anat
On 4/17/2018 11:46 PM, Anatoly Burakov wrote:
This patchset fixes a few Coverity issues introduced
when various parts of DPDK IPC were added, and explains
away other reported issues.
Coverity issues fixed:
- 272595 - return without mutex unlock
- 272609 - fd leak
Coverity issues intentionally
On 4/20/2018 3:30 PM, Xueming Li wrote:
> Add supported RSS hash function check in device configuration to
> have better error verbosity for application developers.
>
> Signed-off-by: Xueming Li
> Acked-by: Adrien Mazarguil
Hi Xueming,
Can you please update the patchwork to only keep latest ve
This patches add "default" parameter to "port config all rss" command.
"default" means all supported hash types reported by device info.
Signed-off-by: Xueming Li
Acked-by: Adrien Mazarguil
---
app/test-pmd/cmdline.c | 13 +
doc/guides/testpmd_app_ug/testpmd_fun
Add supported RSS hash function check in device configuration to
have better error verbosity for application developers.
Signed-off-by: Xueming Li
Acked-by: Adrien Mazarguil
---
lib/librte_ether/rte_ethdev.c | 23 +++
1 file changed, 23 insertions(+)
diff --git a/lib/librte
Wrong patch, please ignore this one.
> -Original Message-
> From: Xueming Li
> Sent: Friday, April 20, 2018 9:07 PM
> To: Shahaf Shuler ; Nelio Laranjeiro
> ; Wenzhuo Lu
> ; Jingjing Wu ; Thomas Monjalon
> ;
> Adrien Mazarguil
> Cc: Xueming(Steven) Li ; dev@dpdk.org
> Subject: [PATCH v
On 4/20/2018 4:41 PM, Burakov, Anatoly wrote:
On 19-Apr-18 5:50 PM, Jianfeng Tan wrote:
To scan the vdevs in primary, we send request to primary process
to obtain the names for vdevs.
Only the name is shared from the primary. In probe(), the device
driver is supposed to locate (or request mor
Hi Ferruh,
> -Original Message-
> From: Ferruh Yigit
> Sent: Friday, April 20, 2018 9:48 PM
> To: Xueming(Steven) Li ; Shahaf Shuler
> ; Nelio Laranjeiro
> ; Wenzhuo Lu ; Jingjing Wu
> ;
> Thomas Monjalon ; Adrien Mazarguil
>
> Cc: dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v5 1/2]
On 4/20/2018 2:48 PM, Ferruh Yigit wrote:
> On 4/20/2018 2:06 PM, Xueming Li wrote:
>> This patch introduce new TX offload flags for device that supports
>> IP or UDP tunneled packet L3/L4 checksum and TSO offload.
>> It will be used for non-standard tunnels.
>>
>> The support from the device is fo
While creating TX, Rx, CQ rings use cached DB address instead of
getting it from the PCI memory resource.
Signed-off-by: Ajit Khaparde
---
drivers/net/bnxt/bnxt.h| 1 +
drivers/net/bnxt/bnxt_cpr.c| 2 +-
drivers/net/bnxt/bnxt_ethdev.c | 12
drivers/net/bnxt/bnxt_ring.c
Apologize for the multiple iterations.
Please apply.
Ajit Khaparde (3):
net/bnxt: cache address of doorbell to subsequent access
net/bnxt: avoid invalid vnic id in set L2 Rx mask
net/bnxt: fix mbuf data offset initialization
drivers/net/bnxt/bnxt.h| 1 +
drivers/net/bnxt/bnxt_cpr.
In some cases bnxt_hwrm_cfa_l2_set_rx_mask is being called before
VNICs are allocated. The FW returns an error in such cases.
Move bnxt_init_nic to bnxt_dev_init such that the ids are initialized
to an invalid id.
Prevent sending the command to the FW only with a valid vnic id.
Fixes: 244bc98b0da7
Initialize mbuf->data_off to RTE_PKTMBUF_HEADROOM after allocation.
Without this, it might be possible that the DMA address provided
to the HW may not be in sync to what is indicated to the application
in bnxt_rx_pkt.
Fixes: 2eb53b134aae ("net/bnxt: add initial Rx code")
Cc: sta...@dpdk.org
Signe
On 4/20/2018 4:26 PM, Burakov, Anatoly wrote:
On 19-Apr-18 5:50 PM, Jianfeng Tan wrote:
As we could add virtual devices from different threads now, we
add a spin lock to protect the vdev device list.
Suggested-by: Anatoly Burakov
Signed-off-by: Jianfeng Tan
Reviewed-by: Qi Zhang
---
<...
On 4/19/2018 7:05 AM, Alejandro Lucero wrote:
> On Wed, Apr 18, 2018 at 1:32 PM, Aaron Conole wrote:
>
>> Alejandro Lucero writes:
>>
>>> On Tue, Apr 17, 2018 at 8:19 PM, Aaron Conole
>> wrote:
>>>
>>> Alejandro Lucero writes:
>>>
>>> > I was just wondering, if device device PCI sysfs resour
On 4/20/2018 4:16 PM, Burakov, Anatoly wrote:
On 19-Apr-18 5:50 PM, Jianfeng Tan wrote:
Adjust the init sequence: put mp channel init before bus scan
so that we can init the vdev bus through mp channel in the
secondary process before the bus scan.
Signed-off-by: Jianfeng Tan
Reviewed-by: Qi
>
> I don't think rte_panic should be considered forbidden. Rather their
> use should be flagged (as this patch does). However, the 'exit 1'
> (which will return a failure for the automatic checkpatch script bot)
> might end up problematic as maintainers might consider it a patch that
> is not re
https://dpdk.org/tracker/show_bug.cgi?id=27
Bug ID: 27
Summary: Can't build mlx5 from 18.02 on CentOS Linux release
7.4.1708 with kernel 3.10.0-693.11.1.el7.x86_64
Product: DPDK
Version: 18.02
Hardware: x86
20/04/2018 15:23, Arnon Warshavsky:
> >
> > Lots of "!= 0"'s - you might gather by now that I don't like them :-)
> >
>
> No way. Would have never guessed that :)
> Sure. When in Rome..
It is a matter of taste. I like the explicit "!= 0".
At least, explicit NULL comparisons are recommended in the
Thanks Aaron
Previously, it wasn't possible for mem_cfg_fd to be reused after a
> failure. Now it is - please reset it to -1. in these close conditions.
>
> Will do.
>
>
> Again, previously this would have aborted on a failure. So it needs to
> be reset to a value that allows retry.
>
Same h
On 4/20/2018 2:06 PM, Xueming Li wrote:
> This patch introduce new TX offload flags for device that supports
> IP or UDP tunneled packet L3/L4 checksum and TSO offload.
> It will be used for non-standard tunnels.
>
> The support from the device is for inner and outer checksums on
> IPV4/TCP/UDP an
Agree. As I wrote below - I will put this instance back in place for this
patchset and handle it on a different one
On Thu, Apr 19, 2018 at 8:31 PM, Kevin Traynor wrote:
> On 04/19/2018 03:57 PM, Burakov, Anatoly wrote:
> > On 19-Apr-18 3:48 PM, Arnon Warshavsky wrote:
> >> Copy on the commit me
On Thu, Apr 19, 2018 at 5:57 PM, Burakov, Anatoly wrote:
> On 19-Apr-18 3:48 PM, Arnon Warshavsky wrote:
>
>> Copy on the commit message and volatile.
>>
>> Regarding the new function defunct_and_remain_in_endless_loop ()
>> I don't think I can put that in a separate patch without breaking the
>
> --- /dev/null
> +++ b/drivers/net/ixgbe/ixgbe_vf_representor.c
> @@ -0,0 +1,217 @@
> +/* SPDX-License-Identifier: BSD-3-Clause
> + * Copyright(c) 2018 Intel Corporation.
> + */
> +
> +#include
> +#include
> +#include
> +
> +#include "base/ixgbe_type.h"
> +#include "base/ixgbe_vf.h"
> +#includ
>
> Lots of "!= 0"'s - you might gather by now that I don't like them :-)
>
No way. Would have never guessed that :)
Sure. When in Rome..
>
> hmm, I'm wondering should void __rte_experimental
> rte_eth_dev_owner_delete change to return an int, now that there is a
> fail case and it is still expe
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Declan Doherty
> Sent: Monday, April 16, 2018 2:06 PM
> To: dev@dpdk.org
> Cc: Adrien Mazarguil ; Yigit, Ferruh
> ; Thomas Monjalon ;
> Shahaf Shuler ; Doherty, Declan
>
> Subject: [dpdk-dev] [PATCH v7 7/9] eth
Not deliberate.Thanks
On Thu, Apr 19, 2018 at 8:27 PM, Kevin Traynor wrote:
> On 04/19/2018 07:01 AM, Arnon Warshavsky wrote:
> > replace panic calls with log and retrun value.
> > Thread function removes the noretrun attribute.
> >
> > Signed-off-by: Arnon Warshavsky
> > ---
> > lib/librte_ea
Ok. Thanks
On Thu, Apr 19, 2018 at 8:26 PM, Kevin Traynor wrote:
> On 04/19/2018 07:01 AM, Arnon Warshavsky wrote:
> > replace panic calls with log and retrun value.
> >
> > --
> > v4 - fix split literal strings in log messages
> >
> > Signed-off-by: Arnon Warshavsky
> > ---
> > lib/librte_eve
> Introduces a new structure, rte_eth_devargs, to support generic
> ethdev arguments common across NET PMDs, with a new API
> rte_eth_devargs_parse API to support PMD parsing these arguments.
>
> Signed-off-by: Remy Horton
> Signed-off-by: Declan Doherty
> ---
> lib/Makefile
> On 04/19/2018 07:01 AM, Arnon Warshavsky wrote:
> > replace panic calls with log and retrun value.
>
> typo return, seems to be in a few commit msgs
>
Yup. They were not even copy pasted so at least I am consistent :)
> + return -1;
>
> similar comment as previous patch about using
Now clear.Thanks
On Thu, Apr 19, 2018 at 5:50 PM, Burakov, Anatoly wrote:
> On 19-Apr-18 3:45 PM, Burakov, Anatoly wrote:
>
>> On 19-Apr-18 3:09 PM, Arnon Warshavsky wrote:
>>
>>> Thanks Anatoly. Will fix that in v5.
>>> Is it preferred to keep all version notes in the cover letter alone?
>>>
>>
-Original Message-
> Date: Fri, 20 Apr 2018 17:04:36 +0530
> From: Akhil Goyal
> To: "Gujjar, Abhinandan S" ,
> "jerin.ja...@caviumnetworks.com" ,
> "hemant.agra...@nxp.com" , "dev@dpdk.org"
>
> CC: "De Lara Guarch, Pablo" , "Doherty,
> Declan" , "Vangati, Narender"
> , "Rao, Nikhil"
Same as in the other patches. Will do. Thanks
On Thu, Apr 19, 2018 at 8:25 PM, Kevin Traynor wrote:
> On 04/19/2018 07:01 AM, Arnon Warshavsky wrote:
> > replace panic calls with log and retrun value.
> > Local function to this file,
> > changing from void to int is non-abi-breaking
> > --
> > v
Will do. Thanks
On Thu, Apr 19, 2018 at 8:25 PM, Kevin Traynor wrote:
> On 04/19/2018 07:01 AM, Arnon Warshavsky wrote:
> > replace panic calls with log and retrun value.
> > Local functions to this file,
> > changing from void to int are non-abi-breaking
> > --
> > v4 - fix split literal string
Thanks Kevin
Will address that
On Thu, Apr 19, 2018 at 5:36 PM, Kevin Traynor wrote:
> On 04/19/2018 07:01 AM, Arnon Warshavsky wrote:
> > replace panic calls with log and retrun value.
> >
> > v4
> > static size calculation function changed to return success/fail code
> > in addition to filling
"show port cap" and "csum parse tunnel" command support TX generic
tunnel offloads
Signed-off-by: Xueming Li
---
app/test-pmd/cmdline.c | 14 --
app/test-pmd/config.c | 17 +
2 files changed, 29 insertions(+), 2 deletions(-)
diff --git a/app/test-pmd/cmdline.c b/app
This patch introduce new TX offload flags for device that supports
IP or UDP tunneled packet L3/L4 checksum and TSO offload.
It will be used for non-standard tunnels.
The support from the device is for inner and outer checksums on
IPV4/TCP/UDP and TSO for *any packet with the following format*:
Hi Declan,
> Add new bus generic ethdev create/destroy APIs which are bus independent
> and provide hooks for bus specific initialisation.
>
> Signed-off-by: Declan Doherty
> ---
> lib/librte_ether/rte_ethdev.c | 95
> -
> lib/librte_ether/rte_ethdev_d
HI
>-Original Message-
>From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Abhinandan Gujjar
>Sent: 04 April 2018 12:27
>To: Jacob, Jerin ;
>hemant.agra...@nxp.com; akhil.go...@nxp.com; dev@dpdk.org
>Cc: pablo.de.lara.gua...@intel.com; declan.dohe...@intel.com;
>narender.vang...@intel.
This patch introduce new TX offload flags for device that supports
IP or UDP tunneled packet L3/L4 checksum and TSO offload.
It will be used for non-standard tunnels.
The support from the device is for inner and outer checksums on
IPV4/TCP/UDP and TSO for *any packet with the following format*:
V7:
- Removed PKT_TX_OUTER_IP_CKSUM deprecation suggestion from commit log
V6:
- Comments update according to maillist discussion.
V5:
- Removed duplicated testpmd patch in other pathset.
- More comments on PKT_TX_TUNNEL_IP and PKT_TX_TUNNEL_UDP
V4:
- Removed DEV_TX_OFFLOAD_GENERIC_TNL_CKSUM and
"show port cap" and "csum parse tunnel" command support TX generic
tunnel offloads
Signed-off-by: Xueming Li
---
app/test-pmd/cmdline.c | 14 --
app/test-pmd/config.c | 17 +
2 files changed, 29 insertions(+), 2 deletions(-)
diff --git a/app/test-pmd/cmdline.c b/app
HI,
>-Original Message-
>From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Abhinandan Gujjar
>Sent: 04 April 2018 12:27
>To: Jacob, Jerin ;
>hemant.agra...@nxp.com; akhil.go...@nxp.com; dev@dpdk.org
>Cc: pablo.de.lara.gua...@intel.com; declan.dohe...@intel.com;
>narender.vang...@intel
This patch cleanup tunnel checksum offloads.
Once tunnel packet type(RTE_PTYPE_TUNNEL_xxx) identified,
PKT_RX_IP_CKSUM_XXX and PKT_RX_L4_CKSUM_XXX represent checksum result of
inner headers, outer L3 and L4 header checksum are always valid as soon
as tunnel identified. If no tunnel identified, PKT
On 4/20/2018 11:45 AM, Sunil Kumar Kori wrote:
> Currently DPAA/DPAA2 driver returns error if non-supported offloads
> (Rx or Tx) are requested from user which causes init failures in
> multiple application.
>
> Some offloads may be enabled by default in many hardware and cannot
> be disabled. The
Remove tunnel limitations, add new hardware tunnel offload features.
Signed-off-by: Xueming Li
---
doc/guides/nics/features/default.ini | 1 +
doc/guides/nics/features/mlx5.ini| 3 +++
doc/guides/nics/mlx5.rst | 4 ++--
3 files changed, 6 insertions(+), 2 deletions(-)
diff --git
This patch split out flow RSS hash field handling logic to dedicate
function.
Signed-off-by: Xueming Li
Acked-by: Nelio Laranjeiro
---
drivers/net/mlx5/mlx5_flow.c | 126 +++
1 file changed, 68 insertions(+), 58 deletions(-)
diff --git a/drivers/net/mlx5
Tunnel w/o tunnel id pattern could match any non-tunneled packet,
this patch allowed tunnel w/o tunnel id pattern after proper outer spec.
Signed-off-by: Xueming Li
Acked-by: Nelio Laranjeiro
---
drivers/net/mlx5/mlx5_flow.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/
Signed-off-by: Xueming Li
Acked-by: Nelio Laranjeiro
---
doc/guides/nics/mlx5.rst | 8 ++--
drivers/net/mlx5/mlx5_flow.c | 107 ++-
drivers/net/mlx5/mlx5_rxtx.c | 3 +-
3 files changed, 111 insertions(+), 7 deletions(-)
diff --git a/doc/guides/nic
Dump verb flow detail including flow spec type and size for debugging
purpose.
Signed-off-by: Xueming Li
---
drivers/net/mlx5/mlx5_flow.c | 68 ---
drivers/net/mlx5/mlx5_rxq.c | 26 ++---
drivers/net/mlx5/mlx5_utils.h | 6
3 files chan
Tunnel RSS level of flow RSS action offers user a choice to do RSS hash
calculation on inner or outer RSS fields. Testpmd flow command examples:
GRE flow inner RSS:
flow create 0 ingress pattern eth / ipv4 proto is 47 / gre / end
actions rss queues 1 2 end level 1 / end
GRE tunnel flow outer RS
Signed-off-by: Xueming Li
---
drivers/net/mlx5/mlx5_flow.c | 101 ---
1 file changed, 94 insertions(+), 7 deletions(-)
diff --git a/drivers/net/mlx5/mlx5_flow.c b/drivers/net/mlx5/mlx5_flow.c
index 5402cb148..b365f9868 100644
--- a/drivers/net/mlx5/mlx5_fl
This patch introduced tunnel type identification based on flow rules.
If flows of multiple tunnel types built on same queue,
RTE_PTYPE_TUNNEL_MASK will be returned, user application could use bits
in flow mark as tunnel type identifier.
Signed-off-by: Xueming Li
---
drivers/net/mlx5/mlx5_flow.c
This patch support L3 VXLAN, no inner L2 header comparing to standard
VXLAN protocol. L3 VXLAN using specific overlay UDP destination port to
discriminate against standard VXLAN, device parameter and FW has to be
configured to support it:
sudo mlxconfig -d -y s IP_OVER_VXLAN_EN=1
sudo mlxconfi
Important note:
please note that this patchset relies on Adrien's patchset of flow API
overhaul: http://www.dpdk.org/ml/archives/dev/2018-April/098047.html
v5:
- Removed %lx prints
- Per review request, clear mbuf tunnel type in case of multiple tunnel types.
- Rebase on Adriens fl
This patch supports new 16 Verbs flow priorities by trying to create a
simple flow of priority 15. If 16 priorities not available, fallback to
traditional 8 priorities.
Verb priority mapping:
8 priorities>=16 priorities
Control flow: 4-7 8-15
User
//snip
>> >+int __rte_experimental
>> >+rte_cryptodev_sym_session_set_private_data(
>> >+ struct rte_cryptodev_sym_session
>> *sess,
>> >+ void *data,
>> >+ uint16_t size)
>> >+{
>> >+ uint16_t
1 - 100 of 164 matches
Mail list logo
