Hi Gavin,
Thanks for your comments, I will fix the code style issue in Patch v5, and
the bit operation for
hinic_test_bit and others has beed replaced with common rte_io_XX_bit APIs by
Joyce Kong's pathces, I will check
other remained bit operations and replace it with the common one.
Best
Hi David ,
Thanks for your comments, I check the test-report from
dpdk.org(https://mails.dpdk.org/archives/test-report/2019-October/101893.html),
which shows hinic pmd driver passed meson build on FreeBSD12-64, also I
analysize the building errs, pthread_mutex_consistent is defined
in on Fr
On 10/29, Yang, Qiming wrote:
>> -Original Message-
>> From: Ye, Xiaolong
>> Sent: Tuesday, October 29, 2019 11:04 AM
>> To: Yang, Qiming
>> Cc: Jiang, JunyuX ; dev@dpdk.org; Chas Williams
>> ; sta...@dpdk.org
>> Subject: Re: [dpdk-dev] [PATCH] net/bonding: fix segfault using invalid port
On 10/29, Ying Wang wrote:
>In non-pipeline mode, switch should be used as fdir's backup.
>The patch fiexed rule can't be created when fdir's flow table is full.
>
>Fixes: 7615a6895009 ("net/ice: rework for generic flow enabling")
>
>Signed-off-by: Ying Wang
>---
> drivers/net/ice/ice_generic_flow
> -Original Message-
> From: Ye, Xiaolong
> Sent: Tuesday, October 29, 2019 11:04 AM
> To: Yang, Qiming
> Cc: Jiang, JunyuX ; dev@dpdk.org; Chas Williams
> ; sta...@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH] net/bonding: fix segfault using invalid port
>
> On 10/28, Yang, Qiming wrote:
>
> -Original Message-
> From: Jerin Jacob
> Sent: Tuesday, October 29, 2019 13:47
> To: Wang, Haiyue
> Cc: Thomas Monjalon ; Yigit, Ferruh
> ; dpdk-dev
> ; Ye, Xiaolong ; Kinsella, Ray
> ;
> Iremonger, Bernard ; Sun, Chenmin
> ; Andrew
> Rybchenko ; Slava Ovsiienko
> ; Stephen Hemminge
> -Original Message-
> From: Jerin Jacob
> Sent: Tuesday, October 29, 2019 13:47
> To: Wang, Haiyue
> Cc: Thomas Monjalon ; Yigit, Ferruh
> ; dpdk-dev
> ; Ye, Xiaolong ; Kinsella, Ray
> ;
> Iremonger, Bernard ; Sun, Chenmin
> ; Andrew
> Rybchenko ; Slava Ovsiienko
> ; Stephen Hemminge
>
> 28/10/2019 04:24, Honnappa Nagarahalli:
> > > 23/10/2019 07:03, Jerin Jacob:
> > > > On Wed, Oct 23, 2019 at 2:37 AM Honnappa Nagarahalli
> > > > wrote:
> > > > > > > > On Thu, 2019-08-01 at 07:48 +0800, Gavin Hu wrote:
> > > > > > > > > Arm N1 SDP is an infrastructure segment development
>
On Tue, Oct 29, 2019 at 11:12 AM Wang, Haiyue wrote:
>
> > -Original Message-
> > From: Jerin Jacob
> > Sent: Tuesday, October 29, 2019 13:20
> > To: Wang, Haiyue
> > Cc: Thomas Monjalon ; Yigit, Ferruh
> > ; dpdk-dev
> > ; Ye, Xiaolong ; Kinsella, Ray
> > ;
> > Iremonger, Bernard ; Su
Hi, Hideyuki.
Thanks for providing extra information.
We rechecked the VLAN actions support in OFED 4.7.1, it should be supported.
There are some limitations:
- VLAN pop is supported on ingress direction only
- VLAN push is supported on egress direction only
- not supported in group 0 (this is r
> -Original Message-
> From: Jerin Jacob
> Sent: Tuesday, October 29, 2019 13:20
> To: Wang, Haiyue
> Cc: Thomas Monjalon ; Yigit, Ferruh
> ; dpdk-dev
> ; Ye, Xiaolong ; Kinsella, Ray
> ;
> Iremonger, Bernard ; Sun, Chenmin
> ; Andrew
> Rybchenko ; Slava Ovsiienko
> ; Stephen Hemminge
Hi, Qi
On 10/29, Qi Zhang wrote:
>Not to turn down link during dev_stop, it will cause the device
>can't be bind by kernel driver after DPDK driver quit.
>
>Fixes: e6161345d8a9 ("net/ice: support link status change")
>Cc: sta...@dpdk.org
>
>Signed-off-by: Qi Zhang
>---
> drivers/net/ice/ice_ethde
The link status for 82599eb got from link status register was not correct,
check the enable/disable flag of tx laser when getting the link status,
set the link status down if tx laser disabled since the tx laser flag
could be set correctly when up/down the link status.
Fixes: dc66e5fd01b9 ("net/ix
On Tue, Oct 29, 2019 at 10:14 AM Wang, Haiyue wrote:
>
> > -Original Message-
> > From: Jerin Jacob
> > Sent: Tuesday, October 29, 2019 11:38
> > To: Wang, Haiyue
> > Cc: Thomas Monjalon ; Yigit, Ferruh
> > ; dpdk-dev
> > ; Ye, Xiaolong ; Kinsella, Ray
> > ;
> > Iremonger, Bernard ; Su
VLAN packets with ip length bigger than 1496 will not be received by
i40e/i40evf due to wrong packets size checking. This patch fixes the
issue by correcting the maximum frame size during checking.
Fixes: 43e5488c0ac6 ("net/i40e: support MTU configuration")
Cc: sta...@dpdk.org
Signed-off-by: Xiao
Not to turn down link during dev_stop, it will cause the device
can't be bind by kernel driver after DPDK driver quit.
Fixes: e6161345d8a9 ("net/ice: support link status change")
Cc: sta...@dpdk.org
Signed-off-by: Qi Zhang
---
drivers/net/ice/ice_ethdev.c | 2 --
1 file changed, 2 deletions(-)
On 10/29, Min, JiaqiX wrote:
>Hi Xiaolong,
>
>> -Original Message-
>> From: Ye, Xiaolong
>> Sent: Monday, October 28, 2019 4:46 PM
>> To: Min, JiaqiX
>> Cc: dev@dpdk.org; sta...@dpdk.org
>> Subject: Re: [dpdk-dev] [PATCH v2] net/ice: fix setting max frame size
>>
>> On 10/28, Min JiaqiX w
From: Pavan Nikhilesh
Add command to set supported ptype mask.
Usage:
set port ptype_mask
Disable ptype parsing by default.
Signed-off-by: Pavan Nikhilesh
---
app/test-pmd/cmdline.c | 80 +
app/test-pmd/testpmd.c | 6 ++
From: Pavan Nikhilesh
Disable packet type parsing as l2fwd doesn't rely on packet types.
Signed-off-by: Pavan Nikhilesh
---
examples/l2fwd/Makefile| 1 +
examples/l2fwd/main.c | 2 ++
examples/l2fwd/meson.build | 1 +
3 files changed, 4 insertions(+)
diff --git a/examples/l2fwd/Makef
From: Pavan Nikhilesh
Add new Rx offload flag `DEV_RX_OFFLOAD_RSS_HASH` which can be used to
enable/disable PMDs write to `rte_mbuf::hash::rss`.
PMDs notify the validity of `rte_mbuf::hash:rss` to the applcation
by enabling `PKT_RX_RSS_HASH ` flag in `rte_mbuf::ol_flags`.
Signed-off-by: Pavan Ni
From: Pavan Nikhilesh
Since pipeline_generic uses `rte_mbuf::hash::rss` add the new Rx offload
flag `DEV_RX_OFFLOAD_RSS_HASH` to inform PMD to copy the RSS hash result
into the mbuf.
Signed-off-by: Pavan Nikhilesh
---
examples/eventdev_pipeline/main.c | 128 -
.../p
From: Pavan Nikhilesh
Some PMD can't work when certain offloads are disabled, to work around
this the PMD auto enable the offloads internally and expose it through
dev->data->dev_conf.rxmode.offloads.
After dev_configure is called compare the requested offloads to the
enabled offloads and log any
From: Pavan Nikhilesh
Add DEV_RX_OFFLOAD_RSS_HASH flag for all PMDs that support RSS hash
delivery.
Signed-off-by: Pavan Nikhilesh
Reviewed-by: Andrew Rybchenko
Reviewed-by: Hemant Agrawal
Acked-by: Jerin Jacob
Acked-by: Ajit Khaparde
---
drivers/net/bnxt/bnxt_ethdev.c | 8 +++-
From: Pavan Nikhilesh
Add `rte_eth_dev_set_supported_ptypes` function that will allow the
application to inform the PMD the packet types it is interested in.
Based on the ptypes set PMDs can optimize their Rx path.
-If application doesn’t want any ptype information it can call
`rte_eth_dev_set_s
From: Pavan Nikhilesh
Add new Rx offload flags `DEV_RX_OFFLOAD_RSS_HASH` These flags can be used
to enable/disable PMD writes to rte_mbuf fields `hash.rss` and also
`ol_flags:PKT_RX_RSS` and `ol_flags:PKT_RX_FDIR`.
Add new packet type set function `rte_eth_dev_set_supported_ptypes`,
allows appli
Max frame size is not set to HW, so packets above the MTU
do not get dropped by HW. The patch fixed the issue.
Fixes: 50370662b727 ("net/ice: support device and queue ops")
Cc: sta...@dpdk.org
Signed-off-by: Min JiaqiX
Reviewed-by: Xiaolong Ye
---
v3:
* Changed Fixes line.
v2:
* Changed commit
> -Original Message-
> From: Jerin Jacob
> Sent: Tuesday, October 29, 2019 11:38
> To: Wang, Haiyue
> Cc: Thomas Monjalon ; Yigit, Ferruh
> ; dpdk-dev
> ; Ye, Xiaolong ; Kinsella, Ray
> ;
> Iremonger, Bernard ; Sun, Chenmin
> ; Andrew
> Rybchenko ; Slava Ovsiienko
> ; Stephen Hemminge
> > > > struct rte_eth_burst_mode {
> > > > uint64_t options;
> > > > char dev_specific[128]; /* PMD has specific burst mode
> > > > information */
> > > > };
> > >
> > > I really don't see how we can have generic flags.
> > > The flags which are proposed are just matching
> > > th
Hi, Guinan
Overall the fix looks good to me, a few comments inline.
On 10/25, Sun GuinanX wrote:
>macsec setting is not valid when port is stopped.
>In order to make it valid, the patch changes the setting
>to where port is started.
>
>Fixes: 597f9fafe13b ("app/testpmd: convert to new Tx offloads
Port validation should be prior to getting device data
to avoid segment fault. This patch fixed the segment fault
caused by invalid port using.
Fixes: 6d72657ce379 ("net/bonding: add other aggregator modes")
Fixes: 112891cd27e5 ("net/bonding: add dedicated HW queues for LACP control")
Cc: sta...@d
> -Original Message-
> From: Wang, Ying A
> Sent: Tuesday, October 29, 2019 1:44 AM
> To: Ye, Xiaolong ; Zhang, Qi Z
> Cc: dev@dpdk.org; Yang, Qiming ; Wang, Ying A
>
> Subject: [PATCH] net/ice: fix rte flow framework
>
> In non-pipeline mode, switch should be used as fdir's backup.
Hi Xiaolong,
> -Original Message-
> From: Ye, Xiaolong
> Sent: Monday, October 28, 2019 4:46 PM
> To: Min, JiaqiX
> Cc: dev@dpdk.org; sta...@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v2] net/ice: fix setting max frame size
>
> On 10/28, Min JiaqiX wrote:
> >Max frame size is not set to H
On 10/28, Yang, Qiming wrote:
>Hi, Junyu
>
>> +dev = &rte_eth_devices[port];
>> +internals = dev->data->dev_private;
>Have you build success? I think we need to add (struct bond_dev_private *) for
>force transfer
dev_private is a void *, an explicit type conversion is not needed here.
Th
Port validation should be prior to getting dev data
to avoid segmentation fault. This patch fixed the issue.
Fixes: 6d72657ce379 ("net/bonding: add other aggregator modes")
Fixes: 112891cd27e5 ("net/bonding: add dedicated HW queues for LACP control")
Cc: sta...@dpdk.org
Signed-off-by: Jiang Junyu
On 10/28, Ye Xiaolong wrote:
>Reviewed-by: Xiaolong Ye
>
>Applied to dpdk-next-net-intel. Thanks.
Sorry for the above misleading message, this patch is still under discussion,
not merged yet.
Thanks,
Xiaolong
In non-pipeline mode, switch should be used as fdir's backup.
The patch fiexed rule can't be created when fdir's flow table is full.
Fixes: 7615a6895009 ("net/ice: rework for generic flow enabling")
Signed-off-by: Ying Wang
---
drivers/net/ice/ice_generic_flow.c | 84 ++-
Hi Andrew,
> -Original Message-
> From: Andrew Rybchenko
> Sent: Monday, October 28, 2019 5:16 PM
> To: Ori Kam ; John McNamara
> ; Marko Kovacevic
> ; Thomas Monjalon ;
> Ferruh Yigit
> Cc: dev@dpdk.org; jingjing...@intel.com; step...@networkplumber.org
> Subject: Re: [dpdk-dev] [PATCH
> On Oct 28, 2019, at 10:43 AM, Stephen Hemminger
> wrote:
>
> On Fri, 25 Oct 2019 08:42:25 +0200
> Willy Tarreau wrote:
>
>> Hi Andy,
>>
>>> On Thu, Oct 24, 2019 at 09:45:56PM -0700, Andy Lutomirski wrote:
>>> Hi all-
>>>
>>> Supporting iopl() in the Linux kernel is becoming a maintainability
On Mon, 26 Aug 2019 18:32:35 +0530
Gagandeep Singh wrote:
> --- a/config/common_base
> +++ b/config/common_base
> @@ -224,6 +224,11 @@ CONFIG_RTE_LIBRTE_CXGBE_DEBUG_TX=n
> CONFIG_RTE_LIBRTE_CXGBE_DEBUG_RX=n
> CONFIG_RTE_LIBRTE_CXGBE_TPUT=y
>
> +#
> +# Compile burst-oriented NXP PPFE PMD drive
28/10/2019 13:12, Andrzej Ostruszka:
> On 10/28/19 9:36 AM, Andrzej Ostruszka wrote:
> > On 10/27/19 12:31 PM, Thomas Monjalon wrote:
> [...]
> >> Should we document its use in rte_function_versioning.h
> >> and versioning.rst?
> >
> > Yes, I think so. I'll add that.
>
> One quick notice. There
28/10/2019 18:09, Thomas Monjalon:
> 28/10/2019 15:50, Pavan Nikhilesh Bhagavatula:
> > >21/09/2019 20:16, pbhagavat...@marvell.com:
> > >> From: Pavan Nikhilesh
> > >>
> > >> Some kernel modules have '-' in their name when populated in
> > >> '/sys/bus/pci/drivers/' where as the kernel always pop
28/10/2019 15:50, Pavan Nikhilesh Bhagavatula:
> >21/09/2019 20:16, pbhagavat...@marvell.com:
> >> From: Pavan Nikhilesh
> >>
> >> Some kernel modules have '-' in their name when populated in
> >> '/sys/bus/pci/drivers/' where as the kernel always populates
> >> '/sys/modules/' with '_'.
> >> Repl
Creating a flow is a normal event; should not be logged at error level.
Fixes: 5c1171c97216 ("net/bnxt: refactor filter/flow")
Signed-off-by: Stephen Hemminger
---
drivers/net/bnxt/bnxt_flow.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/bnxt/bnxt_flow.c b/driv
On Fri, 25 Oct 2019 08:42:25 +0200
Willy Tarreau wrote:
> Hi Andy,
>
> On Thu, Oct 24, 2019 at 09:45:56PM -0700, Andy Lutomirski wrote:
> > Hi all-
> >
> > Supporting iopl() in the Linux kernel is becoming a maintainability
> > problem. As far as I know, DPDK is the only major modern user of
>
28/10/2019 12:25, Ali Alnubani:
> OcteonTx was disabled for causing an internal compiler error on old gcc
> versions.
> See commit 4f760550a093 ("mk: disable OcteonTx for buggy compilers").
>
> The condition that was added later to apply disabling OcteonTx
> only on arm64 caused the condition to n
On Fri, 25 Oct 2019, Andy Lutomirski wrote:
> > I'd see an API more or less like this :
> >
> > int ioport(int op, u16 port, long val, long *ret);
>
> Hmm. I have some memory of a /dev/ioport or similar, but now I can't
> find it. It does seem quite reasonable.
crw-r- 1 root kmem 1, 4 Se
>On 10/25/19 7:10 PM, Pavan Nikhilesh Bhagavatula wrote:
>>> On 10/25/19 5:33 PM, pbhagavat...@marvell.com wrote:
From: Pavan Nikhilesh
Add DEV_RX_OFFLOAD_RSS_HASH flag for all PMDs that support
>RSS
>>> hash
delivery.
Signed-off-by: Pavan Nikhilesh
Reviewed-by:
On 10/25/19 7:10 PM, Pavan Nikhilesh Bhagavatula wrote:
On 10/25/19 5:33 PM, pbhagavat...@marvell.com wrote:
From: Pavan Nikhilesh
Add DEV_RX_OFFLOAD_RSS_HASH flag for all PMDs that support RSS
hash
delivery.
Signed-off-by: Pavan Nikhilesh
Reviewed-by: Andrew Rybchenko
Reviewed-by: Hemant
Hi Ori,
On 10/27/19 3:24 PM, Ori Kam wrote:
This commit introduce hairpin queue type.
The hairpin queue in build from Rx queue binded to Tx queue.
It is used to offload traffic coming from the wire and redirect it back
to the wire.
There are 3 new functions:
- rte_eth_dev_hairpin_capability_ge
https://bugs.dpdk.org/show_bug.cgi?id=358
Bug ID: 358
Summary: KNI packets dropped on ARM
Product: DPDK
Version: 19.08
Hardware: ARM
OS: Linux
Status: UNCONFIRMED
Severity: normal
Priority: Norma
>21/09/2019 20:16, pbhagavat...@marvell.com:
>> From: Pavan Nikhilesh
>>
>> Some kernel modules have '-' in their name when populated in
>> '/sys/bus/pci/drivers/' where as the kernel always populates
>> '/sys/modules/' with '_'.
>> Replace '-' in module name with '_' to avoid script complaining
>
During LTO build compiler reports some 'false positive' warnings about
variables being possibly used uninitialized. This patch silences these
warnings.
Exemplary compiler warning to suppress (with LTO enabled):
error: ‘features’ may be used uninitialized in this function
[-Werror=maybe-uninitiali
During LTO build compiler reports some 'false positive' warnings about
variables being possibly used uninitialized. This patch silences these
warnings.
Exemplary compiler warning to suppress (with LTO enabled):
error: ‘link’ may be used uninitialized in this function
[-Werror=maybe-uninitialized]
During LTO build compiler reports some 'false positive' warnings about
variables being possibly used uninitialized. This patch silences these
warnings.
Exemplary compiler warning to suppress (with LTO enabled):
error: ‘transceiver_type’ may be used uninitialized in this function
[-Werror=maybe-un
During LTO build compiler reports some 'false positive' warnings about
variables being possibly used uninitialized. This patch silences these
warnings.
Exemplary compiler warning to suppress (with LTO enabled):
error: ‘filter_idx’ may be used uninitialized in this function
[-Werror=maybe-uninitia
During LTO build compiler reports some 'false positive' warnings about
variables being possibly used uninitialized. This patch silences these
warnings.
Exemplary compiler warning to suppress (with LTO enabled):
error: ‘chunk’ may be used uninitialized in this function
[-Werror=maybe-uninitialized
This patch fixes 'maybe-uninitialized' warnings reported by compiler
when using LTO.
Compiler warning pointing to this error (with LTO enabled):
error: ‘kg_cfg.extracts[0].masks[0].mask’ may be used uninitialized in
this function [-Werror=maybe-uninitialized]
extr->masks[j].mask = cfg->extract
During LTO build compiler reports some 'false positive' warnings about
variables being possibly used uninitialized. This patch silences these
warnings.
Exemplary compiler warning to suppress (with LTO enabled):
error: ‘stats.greatest_free_size’ may be used uninitialized in this
function [-Werror=
Fix the logic for the case of event queue allowing all schedule types.
Compiler warning pointing to this error (with LTO enabled):
error: ‘sched_type’ may be used uninitialized in this function
[-Werror=maybe-uninitialized]
if ((ret < 0 && ret != -EOVERFLOW) ||
Fixes: 6750b21bd6af ("eventdev: a
This patch adds an option to enable link time optimization. In addition
to LTO option itself (-flto) fat-lto-objects are being used. This is
because during the build pmdinfogen scans the generated ELF objects to
find this_pmd_name* symbol in symbol table. Without fat-lto-objects gcc
produces ELF
During LTO build compiler reports some 'false positive' warnings about
variables being possibly used uninitialized. This patch silences these
warnings.
Exemplary compiler warning to suppress (with LTO enabled):
error: ‘service_id’ may be used uninitialized in this function
[-Werror=maybe-uninitia
This patch series adds an option to make use of link time optimization
(if compiler has support for it). It is split as follows:
- 1st patch fixes missing __vsym annotations (needed for LTO)
- 2nd is the LTO enablement
- remaining patches are fixes for the warnings produced by the compiler
and t
Every implementation of a particular version of given symbol needs to be
marked in its declaration as such (using `__vsym` macro). This patch
fixes this and also clarifies the documentation about that.
Signed-off-by: Andrzej Ostruszka
---
doc/guides/contributing/versioning.rst| 18 +
Hi Anatoly,
On Fri, Jul 19, 2019 at 03:03:29PM +0100, Burakov, Anatoly wrote:
> On 19-Jul-19 2:38 PM, Olivier Matz wrote:
> > When using iova contiguous memory and objets smaller than page size,
> > ensure that objects are not located across several pages.
> >
> > Signed-off-by: Vamsi Krishna Att
On Wed, Aug 07, 2019 at 06:21:20PM +0300, Andrew Rybchenko wrote:
> On 7/19/19 4:38 PM, Olivier Matz wrote:
> > The size returned by rte_mempool_op_calc_mem_size_default() is aligned
> > to the specified page size. This means that with big pages, the returned
> > amount is more that what we really
On Wed, Aug 07, 2019 at 06:21:58PM +0300, Andrew Rybchenko wrote:
> On 7/19/19 4:38 PM, Olivier Matz wrote:
> > When using iova contiguous memory and objets smaller than page size,
> > ensure that objects are not located across several pages.
>
> It looks like as an attempt to make exception a gen
Hi Andrew,
Better late than never, here are answers to your comments.
I'm sending a new version of this patchset that addresses
them.
On Wed, Aug 07, 2019 at 06:21:01PM +0300, Andrew Rybchenko wrote:
> On 7/19/19 4:38 PM, Olivier Matz wrote:
> > When IOVA mode is VA, a mempool can be created with
On Wed, Aug 07, 2019 at 06:21:41PM +0300, Andrew Rybchenko wrote:
> On 7/19/19 4:38 PM, Olivier Matz wrote:
> > In rte_mempool_populate_default(), we determine the page size,
> > which is needed for calc_size and allocation of memory.
> >
> > Move this in a function and export it, it will be used
On Sat, Oct 26, 2019 at 02:09:51PM +, Vamsi Krishna Attunuru wrote:
> Hi Olivier,
>
> > -Original Message-
> > From: Olivier Matz
> > Sent: Saturday, October 26, 2019 5:55 PM
> > To: Vamsi Krishna Attunuru
> > Cc: Jerin Jacob ; Andrew Rybchenko
> > ; Ferruh Yigit ;
> > tho...@monjalo
When populating a mempool, ensure that objects are not located across
several pages, except if user did not request iova contiguous objects.
Signed-off-by: Vamsi Krishna Attunuru
Signed-off-by: Olivier Matz
---
lib/librte_mempool/rte_mempool.c | 23 +---
lib/librte_mempo
rte_mempool_populate_virt() currently requires that both addr
and length are page-aligned.
Remove this uneeded constraint which can be annoying with big
hugepages (ex: 1GB).
Signed-off-by: Olivier Matz
---
lib/librte_mempool/rte_mempool.c | 18 +++---
lib/librte_mempool/rte_mempool.
In rte_mempool_populate_default(), we determine the page size,
which is needed for calc_size and allocation of memory.
Move this in a function and export it, it will be used in next
commit.
Signed-off-by: Olivier Matz
---
lib/librte_mempool/rte_mempool.c | 51 ++
l
The previous commit reduced the amount of required memory when
populating the mempool with non iova-contiguous memory.
Since there is no big advantage to have a fully iova-contiguous mempool
if it is not explicitly asked, remove this code, it simplifies the
populate function.
Signed-off-by: Olivi
KNI supposes that mbufs are contiguous in kernel virtual memory. This
may not be true when using the IOVA=VA mode. To fix this, a possibility
is to ensure that objects do not cross page boundaries in mempool. This
patchset implements this in the last patch (5/5).
The previous patches prepare the j
The size returned by rte_mempool_op_calc_mem_size_default() is aligned
to the specified page size. Therefore, with big pages, the returned size
can be much more that what we really need to populate the mempool.
For instance, populating a mempool that requires 1.1GB of memory with
1GB hugepages can
Hi Andrew,
> -Original Message-
> From: Andrew Rybchenko
> Subject: Re: [dpdk-dev] [PATCH 1/2] ethdev: add flow action type update as an
> offload
>
> Hi Ori,
>
> On 10/28/19 1:50 PM, Ori Kam wrote:
> > Hi Pavan,
> >
> > Sorry for jumping in late.
> >
> > I don't understand why we need
On Mon, Oct 28, 2019 at 11:49:34AM +0100, Thomas Monjalon wrote:
> The function rte_eth_dev_count() was marked as deprecated in DPDK 18.05
> in commit d9a42a69febf ("ethdev: deprecate port count function").
> It was planned to be removed after 19.11 LTS release,
> but given we must not break ABI be
On 10/28/19 9:36 AM, Andrzej Ostruszka wrote:
> On 10/27/19 12:31 PM, Thomas Monjalon wrote:
[...]
>> Should we document its use in rte_function_versioning.h
>> and versioning.rst?
>
> Yes, I think so. I'll add that.
One quick notice. There is a slight mismatch between documentation of
VERSION_
A vhost-user-blk example that support inflight feature. It uses the
new APIs that introduced in the first patch, so it can show how these
APIs work to support inflight feature.
Signed-off-by: Jin Yu
---
V1 - add the case.
V2 - add the rte_vhost prefix.
V3 - add packed ring support
V4 - fix build,
Hi Ori,
On 10/28/19 1:50 PM, Ori Kam wrote:
Hi Pavan,
Sorry for jumping in late.
I don't understand why we need this feature. If the user didn't set any flow
with MARK
then the user doesn't need to check it.
There is pretty long discussion on the topic already, please, read [1].
[1]
http:
On 10/28/19 1:49 PM, Thomas Monjalon wrote:
The function rte_eth_dev_count() was marked as deprecated in DPDK 18.05
in commit d9a42a69febf ("ethdev: deprecate port count function").
It was planned to be removed after 19.11 LTS release,
but given we must not break ABI between 19.11 and 20.11,
it i
OcteonTx was disabled for causing an internal compiler error on old gcc
versions.
See commit 4f760550a093 ("mk: disable OcteonTx for buggy compilers").
The condition that was added later to apply disabling OcteonTx
only on arm64 caused the condition to never be obeyed because it
compares ["arm64"]
This disables OcteonTx2 for gcc 4.8.5 as the compiler is
emitting "internal compiler error" for aarch64.
Fixes: 9a8864c8b5da ("net/octeontx2: add build and doc infrastructure")
Cc: sta...@dpdk.org
Signed-off-by: Ali Alnubani
---
mk/toolchain/gcc/rte.toolchain-compat.mk | 3 +++
1 file changed,
On Mon, Oct 28, 2019 at 11:50 AM Thomas Monjalon wrote:
>
> The function rte_eth_dev_count() was marked as deprecated in DPDK 18.05
> in commit d9a42a69febf ("ethdev: deprecate port count function").
> It was planned to be removed after 19.11 LTS release,
> but given we must not break ABI between
28/10/2019 11:47, Andrzej Ostruszka:
> On 10/27/19 12:47 PM, Thomas Monjalon wrote:
> >> +turned on. This depends obviously on the capabilities of the compiler
> >
> > Not sure what you are talking about.
> > If it is about fat objects, it is not obvious.
> >
> >> +to do "whole program" optimiza
Ping...
Kindly review this patch.
On Wed, Oct 09, 2019 at 02:25:52PM +, Harman Kalra wrote:
> Ping..
>
> On Fri, Sep 13, 2019 at 11:40:40AM +, Harman Kalra wrote:
> > Ping..
> > Kindly review this patch.
> >
> > On Mon, Aug 19, 2019 at 07:18:21PM +0530, Harman Kalra wrote:
> > > Undefine
Hi Pavan,
Sorry for jumping in late.
I don't understand why we need this feature. If the user didn't set any flow
with MARK
then the user doesn't need to check it.
Also it breaks compatibility.
If my understanding is correct the MARK field is going to be moved to dynamic
field, and this
wil
The function rte_eth_dev_count() was marked as deprecated in DPDK 18.05
in commit d9a42a69febf ("ethdev: deprecate port count function").
It was planned to be removed after 19.11 LTS release,
but given we must not break ABI between 19.11 and 20.11,
it is removed now.
Note the ABI version is not du
Thank you Thomas for the comments. Please see below for replies.
On 10/27/19 12:47 PM, Thomas Monjalon wrote:
> Adding John and Marko for doc review.
>
> 22/10/2019 13:54, Andrzej Ostruszka:
>> --- /dev/null
>> +++ b/doc/guides/prog_guide/lto.rst
>
> This new file is not included in the prog gu
From: Igor Chauskin
Add checking of l4_csum_checked and frag flags before checking the
l4_csum_error flag.
In case of IP fragment/unchecked L4 csum - add PKT_RX_L4_CKSUM_UNKNOWN
flag to the indicated mbuf.
Fixes: 1173fca25af9 ("ena: add polling-mode driver")
Cc: sta...@dpdk.org
Signed-off-by:
Add checking of l4_csum_checked and frag flags before checking the
l4_csum_error flag.
In case of IP fragment/unchecked L4 csum - add PKT_RX_L4_CKSUM_UNKNOWN
flag to the indicated mbuf.
Fixes: 1173fca25af9 ("ena: add polling-mode driver")
Cc: sta...@dpdk.org
Signed-off-by: Igor Chauskin
Reviewe
25/10/2019 11:10, Ray Kinsella:
> Hi Thomas,
>
> QQ - So is there really a 'no png' rule, because we have lots of them in the
> documentation?
Yes, really.
> root@rkinsell-MOBL2:.../rkinsell/dpdk# find doc/ -name "*.png" | wc -l
> 61
> root@rkinsell-MOBL2:.../rkinsell/dpdk# find doc/ -name "*.
From: Pavan Nikhilesh
Use RTE_DIM macro to calculate array size
Suggested-by: David Marchand
Signed-off-by: Pavan Nikhilesh
---
app/test-pmd/cmdline.c| 2 +-
app/test-pmd/icmpecho.c | 2 +-
app/test-pmd/testpmd.c| 2 +-
app/test/test.c |
From: Pavan Nikhilesh
use RTE_DIM to calculate array size.
Suggested-by: David Marchand
Signed-off-by: Pavan Nikhilesh
---
lib/librte_ethdev/rte_ethdev.c| 8 +++-
lib/librte_ip_frag/ip_frag_internal.c | 5 ++---
lib/librte_port/rte_port_eventdev.c | 4 ++--
lib/librte_port/rte_p
From: Pavan Nikhilesh
use RTE_DIM macro to calculate array size.
Suggested-by: David Marchand
Signed-off-by: Pavan Nikhilesh
---
examples/ip_pipeline/parser.c | 2 +-
examples/ipv4_multicast/main.c | 3 +--
examples/l3fwd-power/main.c| 12
examples/l3fwd/l3fwd_em.c |
28/10/2019 09:36, Andrzej Ostruszka:
> On 10/27/19 12:31 PM, Thomas Monjalon wrote:
> > 18/09/2019 15:32, Ray Kinsella:
> >> this is cool, good work.
> >> comments below.
> >>
> >> On 17/09/2019 08:57, Andrzej Ostruszka wrote:
> >>> --- a/lib/librte_distributor/rte_distributor.c
> >>> +++ b/lib/lib
As doc mentioned, promisc and multicast are by-default supported in
virtio pmd. Mac/vlan filter are supported by best effort. These control
messages should return pass.
Fixes: f9b9d1a55775 ("net/virtio-user: add multiple queues in device emulation")
Cc: sta...@dpdk.org
Signed-off-by: Marvin Liu
From: Tianfei zhang
In PAC N3000 card, there is one MAX10 chip in each card, and
all of the sensors are connected to MAX10 chip. To support multiple
cards in one server, we introducing a sensor device list under
intel_max10_device instead of a global list. On the other hand, we
using separate int
From: Tianfei zhang
Introducing new register and unregister API for ifpga interrupt.
1. register FME and AFU interrupt
ifpga_register_msix_irq()
2. unregister FME and AFU interrupt
ifpga_unregister_msix_irq()
On PAC N3000 card, there is one PCIe MSIX interrupt for FME managerment,
like th
From: Tianfei zhang
Add new API to get the board info.
opae_mgr_get_board_info()
Signed-off-by: Tianfei zhang
Signed-off-by: Andy Pei
---
drivers/raw/ifpga/base/ifpga_api.c | 11 +++
drivers/raw/ifpga/base/ifpga_defines.h | 55 ++
drivers/raw/ifpga/base
1 - 100 of 128 matches
Mail list logo
