Thank you Yongseok for your comments ! Everything should be integrated in v4.
However, it seems I was in vector mode... So now I cannot get my queue count.
But the count seemed actually more or less correct. Are we speaking of a 4
packet error when in vector mode? Maybe we can actually accept t
Hi Tom,
Adding ethdev maintainers and Oliver as the author of the new API.
Saturday, October 27, 2018 6:11 PM, Tom Barbette:
> Subject: [PATCH v4] mlx5: Support for rte_eth_rx_queue_count
>
I have a more basic question.
The rte_eth_rx_queue_count is a very old API, more or less from the begin
28/10/2018 02:46, Ferruh Yigit:
> RTE_FUNC_PTR_OR_ERR_RET() already does the `ethdev_uninit` NULL check.
>
> Fixes: e489007a411c ("ethdev: add generic create/destroy ethdev APIs")
> Cc: sta...@dpdk.org
>
> Signed-off-by: Ferruh Yigit
Acked-by: Thomas Monjalon
24/08/2018 13:04, Ananyev, Konstantin:
> From: Li Han [mailto:han@zte.com.cn]
> >
> > in struct ip_frag_key,src_dst[] type is uint64_t.
> > but "val" which to store the calc restult ,type is uint32_t.
> > we may lost high 32 bit key. and function return value is int,
> > but it won't return <
27/07/2018 15:52, Cody Doucette:
> Extend rte_ipv6_frag_get_ipv6_fragment_header() to skip over any
> other IPv6 extension headers when finding the fragment header.
>
> According to RFC 8200, there is no guarantee that the IPv6
> Fragment extension header will come before any other extension
> hea
08/07/2018 23:46, Qiaobin Fu:
> Add an iterator over the LPM rules.
> The iterator requires a prefix as a parameter and
> lists all entries as long as the given prefix (or longer).
>
> Signed-off-by: Qiaobin Fu
> Reviewed-by: Cody Doucette
> Reviewed-by: Michel Machado
Any news about this patc
01/08/2018 18:51, Aaron Conole:
> Thomas Monjalon writes:
>
> > 22/07/2018 13:16, Aaron Conole:
> >> Previously, the logid variable would be leaked when the function
> >> returned.
> >>
> >> Signed-off-by: Aaron Conole
> >
> > I am not confortable to merge this kind of patch in RC3.
> > OK to w
22/10/2018 10:00, Thomas Monjalon:
> 12/09/2018 03:31, Ziye Yang:
> > From: Ziye Yang
> >
> > This patch is used to fix the memory leak issue of logid.
> > We use the ASAN test in SPDK when intergrating DPDK and
> > find this memory leak issue.
> >
> > Signed-off-by: Ziye Yang
> > ---
> > - l
A constructor is usually declared with RTE_INIT* macros.
As it is a static function, no need to declare before its definition.
The macro is used directly in the function definition.
Signed-off-by: Thomas Monjalon
---
drivers/bus/vmbus/rte_bus_vmbus.h | 3 +--
drivers/compress/octeontx/otx_
08/10/2018 11:02, Burakov, Anatoly:
> On 07-Oct-18 8:31 PM, Darek Stojaczyk wrote:
> > RTE_MEMZONE_SIZE_HINT_ONLY wasn't checked in any way,
> > causing size hints to be parsed as hard requirements.
> > This resulted in some allocations being failed prematurely.
> >
> > Fixes: 68b6092bd3c7 ("mallo
05/10/2018 17:55, Andrew Rybchenko:
> On 10/5/18 7:26 PM, Ferruh Yigit wrote:
> > --- a/drivers/net/sfc/sfc_rx.c
> > +++ b/drivers/net/sfc/sfc_rx.c
> > -struct sfc_rss_hf_rte_to_efx sfc_rss_hf_map[] = {
> > +static struct sfc_rss_hf_rte_to_efx sfc_rss_hf_map[] = {
>
> I'd prefer "static const" in
Saturday, October 27, 2018 1:54 PM¸ Slava Ovsiienko:
> Subject: [PATCH] net/mlx5: fix flow counters deletion in Verbs
>
> The Flow counters created with Verbs are erroneously destroyed in Flow
> remove function (flow_verbs_remove()). Counter Verbs handles stored in
> the translated rule buffer bec
Saturday, October 27, 2018 2:01 AM, Yongseok Koh:
> Subject: Re: [PATCH 1/2] net/mlx5: fix bit width of flow items
>
>
> > On Oct 25, 2018, at 1:53 AM, Shahaf Shuler
> wrote:
> >
> > Apply the changes from commit c744f6b1b969 ("net/mlx5: fix bit width
> > of item and action flags") in some place
Saturday, October 27, 2018 12:49 AM, Yongseok Koh:
> Subject: Re: [dpdk-dev] [PATCH v3] mlx5: use pkg-config to handle SUSE
> libmnl
>
> On Thu, Oct 25, 2018 at 01:29:38PM -0700, Stephen Hemminger wrote:
> > SUSE decided to install the libmnl include file in a non-standard
> > place: /usr/include/
Thursday, October 25, 2018 9:24 AM, Yongseok Koh:
> Subject: [PATCH] net/mlx5: add 128B padding of Rx completion entry
>
> A PMD parameter (rxq_cqe_pad_en) is added to enable 128B padding of
> CQE on RX side. The size of CQE is aligned with the size of a cacheline of the
> core. If cacheline size
On Sat, 27 Oct 2018 04:42:19 +
"Varghese, Vipin" wrote:
> Hi,
>
>
>
> > ---
> > + struct rte_cryptodev_info dev_info = {0};
> > + struct rte_cryptodev_stats stats = {0};
> > +
> >
> > Memset for initialization as mentioned in other patch.
> >
>
> Yes, I will correc
Hi Thomas,
We are currently working on the patch by incorporating the latest DPDK update,
and will submit a new patch on this.
Thanks,
Qiaobin
> On Oct 28, 2018, at 6:24 AM, Thomas Monjalon wrote:
>
> 08/07/2018 23:46, Qiaobin Fu:
>> Add an iterator over the LPM rules.
>> The iterator require
25/10/2018 05:30, Qi Zhang:
> It's not necessary to insert device argment to devargs_list
> during bus scan, but this happens when we try to attach a
> device on secondary process. The patch fix the issue.
>
> Fixes: cdb068f031c6 ("bus/vdev: scan by multi-process channel")
> Cc: sta...@dpdk.org
>
Flow driver has to provide detected item flags and action flags via
flow_drv_prepare(). DV doesn't return flags at all.
Fixes: 865a0c15672c ("net/mlx5: add Direct Verbs prepare function")
Cc: or...@mellanox.com
Signed-off-by: Yongseok Koh
---
drivers/net/mlx5/mlx5_flow_dv.c | 115 ++
Any review please?
It may require some change after Alejandro patches.
13/08/2018 14:57, Drocula:
> The kernel version 4.14 released with the support of 5-level paging.
> When PML5 enabled, user-space virtual addresses uses up to 56 bits.
> see kernel's Documentation/x86/x86_64/mm.txt.
>
> Signe
11/10/2018 12:47, Jerin Jacob:
> -Original Message-
> > Date: Thu, 11 Oct 2018 11:26:05 +0100
> > From: "Burakov, Anatoly"
> > To: Alejandro Lucero ,
> > dariusz.stojac...@intel.com
> > CC: dev , Santosh Shukla ,
> > Hemant Agrawal , Jerin Jacob
> > , Maxime Coquelin
> > , ch...@att.co
Hi Jerin,
> > > > +
> > > > +/**
> > > > + * Checks that inside given rte_ipsec_session crypto/security fields
> > > > + * are filled correctly and setups function pointers based on these
> > > > values.
> > > > + * @param ss
> > > > + * Pointer to the *rte_ipsec_session* object
> > > > + * @
Garbage in what sense? I would be happy to amend with a little more
information.
The changes to failsafe and rte_net were from previous reviews from
Konstantin:
https://mails.dpdk.org/archives/dev/2018-June/106023.html
https://mails.dpdk.org/archives/dev/2018-July/108701.html
Best,
Cody
On Sun
05/10/2018 14:45, Alejandro Lucero:
> I sent a patchset about this to be applied on 17.11 stable. The memory
> code has had main changes since that version, so here it is the patchset
> adjusted to current master repo.
>
> This patchset adds, mainly, a check for ensuring IOVAs are within a
> restr
05/10/2018 14:45, Alejandro Lucero:
> A device can suffer addressing limitations. This function checks
> memsegs have iovas within the supported range based on dma mask.
>
> PMDs should use this function during initialization if device
> suffers addressing limitations, returning an error if this f
28/10/2018 21:54, Cody Doucette:
> On Sun, Oct 28, 2018 at 6:22 AM Thomas Monjalon wrote:
> > 27/07/2018 15:52, Cody Doucette:
> > > Extend rte_ipv6_frag_get_ipv6_fragment_header() to skip over any
> > > other IPv6 extension headers when finding the fragment header.
> > >
> > > According to RFC 82
07/08/2018 04:35, Takeshi Yoshimura:
> Commit 73a639085938 ("vfio: allow to map other memory regions")
> introduced a bug in sPAPR IOMMU mapping. The commit removed necessary
> ioctl with VFIO_IOMMU_SPAPR_REGISTER_MEMORY. Also, vfio_spapr_map_walk
> should call vfio_spapr_dma_do_map instead of vfio
Some global variables can indeed be static, add static keyword to them.
Signed-off-by: Ferruh Yigit
Acked-by: Jerin Jacob
Acked-by: Shreyansh Jain
---
Series is based on next-net tree
v3:
* Add const keyword where possible
---
app/pdump/main.c| 6 +++---
dr
Various fixes related to the global variable usage.
Fixes: 43e610bb8565 ("compress/octeontx: introduce octeontx zip PMD")
Fixes: c378f084d6e3 ("compress/octeontx: add device setup ops")
Fixes: b43ebc65aada ("compress/octeontx: create private xform")
Fixes: b1ce8ebd97ba ("eventdev: add PMD callback
Some global variables are defined with generic names, add component name
as prefix to variables to prevent collusion with application variables.
Signed-off-by: Ferruh Yigit
Reviewed-by: Maxime Coquelin
Acked-by: Shreyansh Jain
Acked-by: Tianfei zhang
---
drivers/bus/dpaa/include/fsl_fman_crc6
Some global variables can be eliminated, since they are not part of
public interface, it is free to remove them.
Signed-off-by: Ferruh Yigit
Acked-by: Jerin Jacob
Acked-by: Shreyansh Jain
---
lib/librte_compressdev/rte_compressdev.c | 32 --
lib/librte_compressdev/rte_compr
11/10/2018 12:08, Thomas Monjalon:
> +Cc more maintainers in order to collect more reviews
>
> 04/10/2018 11:19, Burakov, Anatoly:
> > On 03-Oct-18 9:53 PM, eric zhang wrote:
> > > This patchset introduces an EAL command line option "--iova-mode"
> > > to give the user a facility to force IOVA mod
17/09/2018 12:33, Burakov, Anatoly:
> On 07-Sep-18 4:58 PM, Darek Stojaczyk wrote:
> > When neither RTE_IOVA_VA nor RTE_IOVA_PA was explicitly
> > requested, DPDK would currently fallback to the default
> > RTE_IOVA_PA mode and possibly encounter a failure later
> > on if running as a non-priviledg
27/10/2018 05:20, Qi Zhang:
> When scanning an already plugged device, the virtual address
> of mapped PCI resource in rte_pci_device will be overridden
> with 0, that may cause driver does not work correctly.
> The fix is not to update any rte_pci_device's field if the being
> scanned device's dri
11/10/2018 15:01, Tiwei Bie:
> On Thu, Oct 11, 2018 at 11:53:12AM +0100, Luca Boccassi wrote:
> > On Thu, 2018-10-11 at 12:27 +0200, Thomas Monjalon wrote:
> > > 28/08/2018 12:12, Luca Boccassi:
> > > > From: Brian Russell
> > > >
> > > > In virtio_read_caps and vtpci_msix_detect, rte_pci_read_co
26/10/2018 16:03, Burakov, Anatoly:
> On 25-Oct-18 11:49 AM, Alejandro Lucero wrote:
> > Invoking the right pci read/write functions is based on interrupt
> > handler type. However, this is not configured for secondary processes
> > precluding to use those functions.
> >
> > This patch fixes the i
26/10/2018 16:23, Burakov, Anatoly:
> On 24-Oct-18 11:11 AM, Darek Stojaczyk wrote:
> > This function is documented to return the number of unregistered
> > callbacks or negative numbers on error, but pci_vfio checks for
> > ret != 0 to detect failures. Not anymore.
> >
> > Fixes: c115fd000c32 ("v
05/10/2018 10:57, Burakov, Anatoly:
> On 03-Oct-18 1:39 PM, Darek Stojaczyk wrote:
> > So far each process in MP used to have a separate container
> > and relied on the primary process to register all memsegs.
> >
> > Mapping external memory via rte_vfio_container_dma_map()
> > in secondary proces
25/09/2018 09:56, Wang, Xiao W:
> Hi,
>
> > From: Dariusz Stojaczyk
> >
> > Always attempt to find already opened fd for an iommu
> > group as subsequent attempts to open it will fail.
> >
> > There's no public API to check if a group was already
> > bound and has a container, so rte_vfio_conta
26/10/2018 16:04, Burakov, Anatoly:
> On 25-Oct-18 11:49 AM, Darek Stojaczyk wrote:
> > We were reading some memory just after freeing it.
> >
> > Fixes: 83a73c5fef66 ("vfio: use generic multi-process channel")
> > Cc: jianfeng@intel.com
> > Cc: anatoly.bura...@intel.com
> > Cc: sta...@dpdk.or
> -Original Message-
> From: Yigit, Ferruh
> Sent: Sunday, October 28, 2018 9:59 AM
> To: Liu, Changpeng
> Cc: Richardson, Bruce ; dpdk-dev
> ; Rafal Kozik
> Subject: Re: [dpdk-dev] [PATCH] pci/uio: enable prefetchable resources mapping
>
> On 10/5/2017 9:33 AM, bruce.richardson at int
26/10/2018 09:54, Darek Stojaczyk:
> In a couple of places we check its error code against -EEXIST,
> but this function returned either -1, 0, or 1.
>
> This gets critical when hotplugging a device in secondary
> process, while the same device is already plugged in the
> primary. Failing to "hotpl
29/10/2018 00:57, Ferruh Yigit:
> Some global variables can be eliminated, since they are not part of
> public interface, it is free to remove them.
>
> Signed-off-by: Ferruh Yigit
> Acked-by: Jerin Jacob
> Acked-by: Shreyansh Jain
Series applied, thanks
> -Original Message-
> From: Zhang, Qi Z
> Sent: Friday, October 26, 2018 11:42 PM
> To: Ananyev, Konstantin ; Xing, Beilei
>
> Cc: dev@dpdk.org; sta...@dpdk.org
> Subject: RE: [dpdk-dev] [PATCH v2] net/i40e: fix Rx instability with vector
> mode
>
>
>
> > -Original Message-
>
24/10/2018 10:50, Wang, Dong1:
> Tested it with 82599 PF.
>
> Acked-by: Wang Dong
The correct tag is probably Tested-by
> From: Zhao1, Wei
>
> Add Wang Dong in the mailing list, he has test this patch.
>
> > > This example has not been enable for receiving multicast packet, so
> > > it will
17/10/2018 16:57, Van Haaren, Harry:
> From: Varghese, Vipin
> >
> > The service core samples has varied profiles created to run on specified
> > lcore count. The patch adds the check before each run, to ensure
> > example has sufficent lcores to be added as service cores on given run
> > profile.
08/10/2018 14:35, Ferruh Yigit:
> On 10/8/2018 12:29 PM, Joseph, Anoob wrote:
> > On 08-10-2018 16:30, Ferruh Yigit wrote:
> >> On 10/8/2018 11:41 AM, Anoob Joseph wrote:
> >>> This patchset fixes multiple issues reported by checkpatch in l2fwd
> >>> code base. These issues would be flagged for any
18/08/2018 11:58, Joseph, Anoob:
> Hi Bruce,
>
> On 14-08-2018 16:03, Bruce Richardson wrote:
> > On Mon, Aug 13, 2018 at 09:29:01PM +0530, Joseph, Anoob wrote:
> >> Hi Bruce,
> >>
> >> The reason why l2fwd was chosen was to allow everyone to chip in their
> >> ideas
> >> while preparing the fram
Hi Stephen,
Please would you like to send a v4?
22/06/2018 11:12, Iremonger, Bernard:
> Hi Stephen,
>
> > -Original Message-
> > From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Stephen Hemminger
> > Sent: Monday, June 18, 2018 10:36 PM
> > To: dev@dpdk.org
> > Cc: Stephen Hemminger
> > Disabled the alarm_autotest UT in FreeBSD Interrupts are not supported in
> > FreeBSD.
> > Alarm API depends on interrupts, so disabled alarm test on FreeBSD.
> >
> > Signed-off-by: Pallantla Poornima
> Acked-by: Reshma Pattan
Applied, thanks
> > The test application didn't call rte_eal_cleanup() on exit, which caused
> > leftover hugepages and memory leaks when running secondary processes. Fix
> > this by calling rte_eal_cleanup() on exit.
> >
> > Signed-off-by: Anatoly Burakov
> > ---
> > test/test/test.c | 33 +
22/09/2018 10:57, Jananee Parthasarathy:
> Reduced test duration for timer_racecond unit test.
> This wil help to receive quicker test results.
>
> Signed-off-by: Jananee Parthasarathy
Applied, thanks
Hi Thomas and Jerin,
The patches were extensively and heavily reviewed by Arm internally. As the 1st
patch was not concluded, so I create a new series(2 patches),
> -Original Message-
> From: Jerin Jacob
> Sent: Saturday, October 27, 2018 11:34 PM
> To: Thomas Monjalon
> Cc: Gavin Hu
Gavin Hu (Arm Technology China) would like to recall the message, "[dpdk-dev]
[PATCH 1/2] ring: synchronize the load and store of the tail".
IMPORTANT NOTICE: The contents of this email and any attachments are
confidential and may also be privileged. If you are not the intended recipient,
please
26/09/2018 15:11, Jananee Parthasarathy:
> From: Chaitanya Babu Talluri
>
> Divided main test to smaller logical tests.
> Registered with UT framework.
> Added cleanup of the resources else ring creation fails
> during consecutive test runs.
> Freed the allocated mempool, rings and uninitalized t
Hi Thomas and Jerin,
The patches were extensively reviewed by Arm internally, as the 1st patch was
not able to be concluded, I created a new patch series(2 patches).
How can I clean up this mess?
1. make all the previous patches Superseded?
2. We have two more new patches, should I submit the 4 p
08/10/2018 15:01, Hari Kumar Vemula:
> Unit testcases are added for metrics library
> Added metrics unit test to autotest list
> Updated meson build file
> Updated MAINTAINERSHIP for metrics unit test
>
> Signed-off-by: Hari Kumar Vemula
> Reviewed-by: Reshma Pattan
> Reviewed-by: Remy Horton
>
Hi,
The command make-guides-pdf fails currently (18.11-rc1).
There is probably a subtle syntax error in one of the RST file.
Please, who wants to look at the issue and fix it?
Thank you
> > >> Subject: Re: [dpdk-stable] [PATCH v2] test/hash: solve unit test
> > >> hash compilation error
> > >>
> > >> +Cc Yipeng
> > >>
> > >> 18/09/2018 21:22, Dharmik Thakkar:
> > >>> Enable print_key_info() function compilation always.
> > >>
> > >> Please see my first comment: you need to show th
29/10/2018 05:16, Honnappa Nagarahalli:
> > > >> Subject: Re: [dpdk-stable] [PATCH v2] test/hash: solve unit test
> > > >> hash compilation error
> > > >>
> > > >> +Cc Yipeng
> > > >>
> > > >> 18/09/2018 21:22, Dharmik Thakkar:
> > > >>> Enable print_key_info() function compilation always.
> > > >>
Hi Thomas,
> -Original Message-
> From: Thomas Monjalon
> Sent: 29 October 2018 07:49
> To: Joseph, Anoob
> Cc: dev@dpdk.org; Ferruh Yigit ; Bruce Richardson
> ; Pablo de Lara
> ; Jacob, Jerin
> ; Athreya, Narayana Prasad
>
> Subject: Re: [dpdk-dev] [PATCH 0/8] examples/l2fwd: fix check
>
> 29/10/2018 05:16, Honnappa Nagarahalli:
> > > > >> Subject: Re: [dpdk-stable] [PATCH v2] test/hash: solve unit
> > > > >> test hash compilation error
> > > > >>
> > > > >> +Cc Yipeng
> > > > >>
> > > > >> 18/09/2018 21:22, Dharmik Thakkar:
> > > > >>> Enable print_key_info() function compilati
A new DPDK release candidate is ready for testing:
https://git.dpdk.org/dpdk/tag/?id=v18.11-rc1
It is one of the biggest -rc1: 1056 patches.
The integration deadline was on October 5th, so we look very late.
The good news is that almost everything planned is integrated.
The release notes
Please let us know , we need this functionality.
Regards
Siddharth
From: Verma, Siddharth
Sent: 26 October 2018 10:53
To: 'dev@dpdk.org'
Subject: Regarding -:[dpdk-dev] net_vmxnet3: link status Interrupt is not
working in DPDK 17.11.2 version
Hi ,
link status Interrupt is not working in DPDK 17
Without this change, virtio-user still works, but it will show
annoying error messages like this on shutdown:
vhost_kernel_set_backend(): VHOST_NET_SET_BACKEND fails, Operation not permitted
vhost_kernel_ioctl(): VHOST_RESET_OWNER failed: Operation not permitted
Fixes: e3b434818bbb ("net/virtio-u
Tiwei Bie (6):
net/virtio-user: do not stop stopped device again
net/virtio-user: do not make vhost user channel nonblock
net/virtio-user: do not reset owner when driver resets
net/virtio-user: fix device features for server mode
net/virtio-user: simplify device features preparation
net
We need to save the supported frontend features (which won't be
announced by vhost backend), otherwise we will lost them when the
connection to vhost-user backend is established in server mode.
Fixes: 201a41651715 ("net/virtio-user: fix multiple queues fail in server mode")
Cc: sta...@dpdk.org
Si
When driver resets the device, virtio-user just needs to send
GET_VRING_BASE messages to stop the vhost backend, and that's
what QEMU does. With this change, we won't need to set owner
when starting virtio-user device anymore. This will help us to
get rid of below error message on startup:
vhost_k
There is no need to make the vhost user channel nonblock, and
making it nonblock will make vhost_user_read() fail with EAGAIN
when vhost messages need a reply.
Fixes: bd8f50a45d0f ("net/virtio-user: support server mode")
Cc: sta...@dpdk.org
Signed-off-by: Tiwei Bie
---
drivers/net/virtio/virtio
Get rid of the duplicated code in device features preparation
which looks awful.
Signed-off-by: Tiwei Bie
---
.../net/virtio/virtio_user/virtio_user_dev.c | 31 +++
1 file changed, 11 insertions(+), 20 deletions(-)
diff --git a/drivers/net/virtio/virtio_user/virtio_user_dev.c
We need to check the status field in virtio net config structure
instead of the bits read from ISR register to know whether we need
to do guest announce.
Fixes: 7365504f77e3 ("net/virtio: support guest announce")
Cc: sta...@dpdk.org
Signed-off-by: Tiwei Bie
---
drivers/net/virtio/virtio_ethdev.
> > > On Wed, Oct 10, 2018 at 02:48:05PM -0700, Yipeng Wang wrote:
> > > > This commit improves the readwrite test to consider extendable
> > > > table feature and add more functional tests to cover more corner cases.
> > > >
> > > > Signed-off-by: Yipeng Wang ---
> > > > test/test/test_hash_readw
Why should DV prepare function return the list of actions?
The only reason I can think of, is if you want to remove the for loop in
dv_translate. And then in flow_dv_create_action change the switch case
to ifs.
Ori
> -Original Message-
> From: Yongseok Koh
> Sent: Sunday, October 28, 2
On Sun, Oct 28, 2018 at 11:57:40PM +, Ferruh Yigit wrote:
> Various fixes related to the global variable usage.
>
> Fixes: 43e610bb8565 ("compress/octeontx: introduce octeontx zip PMD")
> Fixes: c378f084d6e3 ("compress/octeontx: add device setup ops")
> Fixes: b43ebc65aada ("compress/octeontx:
74 matches
Mail list logo
