18/01/2023 18:39, Anoob Joseph:
> > 22/12/2022 10:25, Anoob Joseph:
> > > --- /dev/null
> > > +++ b/lib/net/rte_pdcp_hdr.h
> > > @@ -0,0 +1,93 @@
> > > +/* SPDX-License-Identifier: BSD-3-Clause
> > > + * Copyright(C) 2022 Marvell.
> > > + */
> > > +
> > > +#ifndef _RTE_PDCP_HDR_H_
> > > +#define _R
18/01/2023 18:34, Gregory Etelson:
> > > Current API allows either query or update indirect flow action.
> > > If port hardware allows both update and query in a single operation,
> > > application still has to issue 2 separate hardware requests.
> > >
> > > The patch adds `rte_flow_action_handle_q
On Wed, Jan 18, 2023 at 06:53:33PM +, Ferruh Yigit wrote:
> On 1/18/2023 4:11 PM, Bruce Richardson wrote:
> > When building the example without libbsd and using the DPDK-provided
> > strlcpy function, a compiler warning is emitted by GCC 12 about the copy
> > of the parsed string into the resul
On 12/9/2022 1:35 AM, Chaoyong He wrote:
> From: Peng Zhang
>
> All the queue pairs are shared between VF and PF for NFP,
> and there exist different assignment policy for different
> firmware application.
>
> The user can create VFs up to the max number of queue pairs,
> which will cause the la
Hello Thomas,
> Is this patch related to query/update of indirect flow action?
> It looks like it should be in a separate series.
>
Quota flow action updates relay on rte_flow_action_handle_query_update().
Also, the function is used in the testmd patch section.
> 18/01/2023 11:31, Gregory Etels
On 1/19/2023 1:50 AM, Jesna K E wrote:
> For Multiprocess dpdk applications need to retrieve the offload parameter
> from the Device data directly that is shared between
> primary and secondary processes rather than accessing from
> Port private data (pdata) since this is not shared
> between prima
On 12/1/2022 1:38 AM, Chaoyong He wrote:
> Add a new section 'Supported APIs', inculding the supported
> APIs/items/actions of rte_flow.
>
Isn't this information already exists in
'doc/guides/nics/features/nfp.ini'? Why to duplicate here?
> Signed-off-by: Chaoyong He
> Reviewed-by: Niklas Söder
19/01/2023 10:13, Gregory Etelson:
> Hello Thomas,
>
> > Is this patch related to query/update of indirect flow action?
> > It looks like it should be in a separate series.
> >
>
> Quota flow action updates relay on rte_flow_action_handle_query_update().
> Also, the function is used in the testm
> Acked-by: Kai Ji
>
> > Subject: [PATCH v2] crypto/qat: fix build
> >
> > When building QAT PMD, the following issue comes up:
> >
> > intel-ipsec-mb.h:333: error: "AES_BLOCK_SIZE" redefined
> > 333 | #define AES_BLOCK_SIZE IMB_AES_BLOCK_SIZE
> > In file included from drivers/crypto/q
> -Original Message-
> From: Bruce Richardson
> Sent: Tuesday, January 17, 2023 2:33 PM
> To: Akhil Goyal
> Cc: David Marchand ; Kai Ji ;
> dev@dpdk.org; sta...@dpdk.org; Ciara Power ; Fan
> Zhang
> Subject: Re: [EXT] Re: [PATCH] crypto/qat: fix build for generic x86 with GCC
> 12
>
Hello Thomas,
[]
> > > > + /**
> > > > + * Apply quota verdict - PASS or BLOCK to a flow.
> > >
> > > Is "to" really wanted here?
> > > What about "pass or block a flow"?
> > >
> >
> > The flow action provides information only - it does not affect flow in any
> way.
> > Application needs
在 2023/1/18 16:26, Thomas Monjalon 写道:
20/10/2022 11:31, Huisong Li:
The dev->data->mac_addrs[0] will be changed to a new MAC address when
applications modify the default MAC address by .mac_addr_set(). However,
if the new default one has been added as a non-default MAC address by
.mac_addr_ad
On 1/12/23 20:36, Hernan Vargas wrote:
v3: Remove unused mempool header.
v2: Reordered commits and added one more fix.
v1: Upstreaming ACC100 changes for 23.03.
Hernan Vargas (7):
baseband/acc: acc100 free harq layout pointer
baseband/acc: acc100 fix iteration counter in TB
baseband/
在 2023/1/18 16:38, Thomas Monjalon 写道:
18/01/2023 09:26, Thomas Monjalon:
20/10/2022 11:31, Huisong Li:
The dev->data->mac_addrs[0] will be changed to a new MAC address when
applications modify the default MAC address by .mac_addr_set(). However,
if the new default one has been added as a non
On 11/14/22 04:04, Hao Chen wrote:
When we run dpdk vdpa in the nested virtual machine vm-L1 and ping
test in vm-L2, the ping is not good. The reason for troubleshooting is
that the virtio net in vm-L2 sends control information to the vring,
and the qemu back-end device in vm-L1 cannot obtain
On 12/1/22 08:32, Andy Pei wrote:
"sw_fallback_lm=0" means driver does not provide live migration assistance.
so hardware assistance is needed to support live migration.
However registers for live migration in BAR4 are not implemented.
In the case, we just return error after display some logs.
在 2023/1/18 22:12, Thomas Monjalon 写道:
11/01/2023 11:46, Ferruh Yigit:
On 1/11/2023 10:27 AM, Ferruh Yigit wrote:
On 1/11/2023 12:53 AM, lihuisong (C) wrote:
在 2023/1/11 0:51, Ferruh Yigit 写道:
Hi Huisong,
I haven't checked the patch in detail yet, but I can see it gives some
ABI compatibil
If the node id check failed, graph_lock was not taken before releasing.
Fixes: c59dac2ca14a ("graph: implement node operations")
Cc: sta...@dpdk.org
Signed-off-by: David Marchand
---
lib/graph/node.c | 8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/lib/graph/node.c b/
> -Original Message-
> From: Thomas Monjalon
> Sent: Wednesday, January 18, 2023 4:03 PM
> To: Hanumanth Reddy Pothula
> Cc: Aman Singh ; Yuying Zhang
> ; dev@dpdk.org;
> andrew.rybche...@oktetlabs.ru; viachesl...@nvidia.com; Jerin Jacob
> Kollanukkaran ; Nithin Kumar Dabilpuram
> ; da
On 12/7/2022 2:17 AM, Chaoyong He wrote:
> This patch series mainly add the offload support of the meter for NFP
> cards, includes:
> - Add meter profile
> - Add meter policy
> - Add meter stats
> - Add meter action
> - Add meter offload support
>
> Jin Liu (6):
> net/nfp: correct flower cmsg i
On 9/17/2021 1:43 AM, Min Hu (Connor) wrote:
> 在 2021/9/16 16:21, Andrew Rybchenko 写道:
>> On 9/16/21 11:16 AM, Min Hu (Connor) wrote:
>>> Hi, Andrew,
>>>
>>> 在 2021/9/16 14:22, Andrew Rybchenko 写道:
On 9/16/21 5:56 AM, Min Hu (Connor) wrote:
> Currently, link speed to string only supports s
On 4/20/2022 12:24 PM, madhuker.myt...@oracle.com wrote:
> From: Madhuker Mythri
>
> As per analysis on Tap PMD, the existing RSS algorithm considering
> 4-tuple(Src-IP, Dst-IP, Src-port and Dst-port) and identification of
> fragment packets is not done, thus we are seeing all the fragmented
> ch
This patch addes private dump for vhost and virtio PMDs.
Chengwen Feng (2):
net/virtio: support private dump
net/vhost: support private dump
drivers/net/vhost/rte_eth_vhost.c | 21 +
drivers/net/virtio/virtio_ethdev.c | 19 +++
2 files changed, 40 inserti
This patch implements eth_dev_priv_dump callback which could use for
debugging.
Signed-off-by: Chengwen Feng
---
drivers/net/vhost/rte_eth_vhost.c | 21 +
1 file changed, 21 insertions(+)
diff --git a/drivers/net/vhost/rte_eth_vhost.c
b/drivers/net/vhost/rte_eth_vhost.c
ind
This patch implements eth_dev_priv_dump callback which could use for
debugging.
Signed-off-by: Chengwen Feng
---
drivers/net/virtio/virtio_ethdev.c | 19 +++
1 file changed, 19 insertions(+)
diff --git a/drivers/net/virtio/virtio_ethdev.c
b/drivers/net/virtio/virtio_ethdev.c
in
On 11/22/2022 1:09 PM, Niklas Söderlund wrote:
> From: Jin Liu
>
> The minimum value of TX/RX descriptors for NFP3800 card is not same
> with NFP4000 card, the minimum value of NFP4000 card is 256 while
> NFP3800 card is 512, add the minimum descriptor macro for NFP3800
> card.
>
> Modify the lo
Hello,
ci/iol-testing tried to apply this patchset on the main branch, even though
it's intended for dpdk-next-net-mlx:
https://patches.dpdk.org/project/dpdk/patch/20230112134913.2632849-1-klit...@nvidia.com/
The helper script in dpdk-ci repo reports dpdk-next-net-mlx:
$ ./tools/pw_maintainers_c
Current API allows either query or update indirect flow action.
If indirect action must be conditionally updated according to it's
present state application must first issue action query then
analyze returned data and if needed issue update request.
When the update will be processed, action state c
Quota action limits traffic according to pre-defined configuration.
Quota reflects overall traffic usage regardless bandwidth.
Quota flow action initialized with signed tokens number value.
Quota flow action updates tokens number according to
these rules:
1. if quota was configured to count packet
> -Original Message-
> From: David Marchand
> Sent: 19 January 2023 04:03 PM
> To: dev@dpdk.org
> Cc: sta...@dpdk.org; Jerin Jacob Kollanukkaran ; Kiran
> Kumar Kokkilagadda ; Nithin Kumar Dabilpuram
> ; Pavan Nikhilesh Bhagavatula
>
> Subject: [EXT] [PATCH] graph: fix node shrink
>
>
Hi Abhishek,
On 12/15/22 04:48, Abhishek Maheshwari wrote:
Amending the match table for vdpa/ifcvf driver to work with virtio-blk
devices that have default subsystem IDs.
Cc: sta...@dpdk.org
Is that really a fix? If so, you need to add the Fixes tag so that LTS
maintainers know on which LTS i
Hello,
We are still trying to track this down but want to reach out to the
community and see if there is something obvious we are missing. This
happens to be a CentOS 7 based host with DPDK 22.11.1 in use. We have
obtained the latest compatible Intel drivers and DDP from
https://sourceforge.net/
On 11/22/2022 3:30 PM, ok...@kernel.org wrote:
> From: Sinan Kaya
>
> rte_class_eth: eth_mac_cmp: The status of this call to rte_eth_dev_info_get
> is not checked, potentially leaving dev_info uninitialized.
>
> Signed-off-by: Sinan Kaya
> ---
> lib/ethdev/rte_class_eth.c | 4 +++-
> 1 file ch
Thanks Jerin for the code review and feedback. I will make the suggested
changes in next version of the patch.
> -Original Message-
> From: Jerin Jacob
> Sent: Thursday, January 12, 2023 5:48 PM
> To: Amit Prakash Shukla
> Cc: Jerin Jacob Kollanukkaran ; Kiran Kumar
> Kokkilagadda ; Nit
19/01/2023 11:31, lihuisong (C):
> 在 2023/1/18 22:12, Thomas Monjalon 写道:
> > 11/01/2023 11:46, Ferruh Yigit:
> >> On 1/11/2023 10:27 AM, Ferruh Yigit wrote:
> >>> On 1/11/2023 12:53 AM, lihuisong (C) wrote:
> 在 2023/1/11 0:51, Ferruh Yigit 写道:
> > Hi Huisong,
> >
> > I haven't che
Hi Stephen,
Thanks for the reply. Please find my reply in-line.
> -Original Message-
> From: Stephen Hemminger
> Sent: Thursday, January 12, 2023 10:00 PM
> To: Amit Prakash Shukla
> Cc: Jerin Jacob Kollanukkaran ; Kiran Kumar
> Kokkilagadda ; Nithin Kumar Dabilpuram
> ; dev@dpdk.org
>
Hi,
You missed some questions and comments below.
19/01/2023 10:57, lihuisong (C):
> 在 2023/1/18 16:26, Thomas Monjalon 写道:
> > 20/10/2022 11:31, Huisong Li:
> >> The dev->data->mac_addrs[0] will be changed to a new MAC address when
> >> applications modify the default MAC address by .mac_addr_se
On 10/13/2022 12:47 PM, Ferruh Yigit wrote:
> Driver is disabled by default because its dependencies are not
> upstreamed yet, code is available for development and investigation.
>
> When all dependencies are upstreamed, driver can be enabled back.
>
> Fixes: 517ed6e2d590 ("net/mana: add basic d
On 8/8/2022 4:03 PM, Stephen Hemminger wrote:
> On Mon, 8 Aug 2022 16:49:44 +0200
> Francesco Mancino wrote:
>
>> eth_dev_validate_mtu, introduced in 990912e676e, validates configured
>> MTU plus overhead against max_rx_pktlen.
>> Since TAP is a virtual device, it should support as big MTU as pos
On 12/12/22 08:12, Andy Pei wrote:
In the case using argument "sw-live-migration=1" to enable SW assisted live
migration, we take QEMU as front end for example, after source VM migrates
to destination VM, we keep vdpa process for source VM there, we kill the
QEMU process for source VM, and res
On 12/13/22 08:37, Andy Pei wrote:
For block device, we make sure no IO is inflight before we
stop device. We terminate the notify relay process and wait
for all inflight IOs to be completed. So in this period, we
wait for hardware to update used index, there is no need to
relay new kick.
Sig
This is a follow up on previous work by Kevin Laatz:
http://patches.dpdk.org/project/dpdk/list/?series=24658&state=*
This series is aimed at allowing DPDK applications to expose their CPU
usage stats in the DPDK telemetry under /eal/lcore/info. This is a much
more basic and naive approach which l
Report the same information than rte_lcore_dump() in the telemetry
API into /eal/lcore/list and /eal/lcore/info,ID.
Example:
--> /eal/lcore/info,3
{
"/eal/lcore/info": {
"lcore_id": 3,
"socket": 0,
"role": "RTE",
"cpuset": [
3
]
}
}
Signed-off-
Allow applications to register a callback that will be invoked in
rte_lcore_dump() and when requesting lcore info in the telemetry API.
The callback is expected to return the number of TSC cycles that have
passed since application start and the number of these cycles that were
spent doing busy wor
Allow fetching CPU cycles usage for all lcores with a single request.
This endpoint is intended for repeated and frequent invocations by
external monitoring systems and therefore returns condensed data.
It consists of a single dictionary with three keys: "lcore_ids",
"total_cycles" and "busy_cycle
Add a simple command that calls rte_lcore_dump().
Signed-off-by: Robin Jarry
Acked-by: Morten Brørup
Acked-by: Konstantin Ananyev
---
Notes:
v5 -> v6: No change
app/test-pmd/cmdline.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/app/test-pmd/cmdline.c b/app/test-pmd/cmdline.c
i
Reuse the --record-core-cycles option to account for busy cycles. One
turn of packet_fwd_t is considered "busy" if there was at least one
received or transmitted packet.
Add a new busy_cycles field in struct fwd_stream. Update get_end_cycles
to accept an additional argument for the number of proce
On 11/16/2022 4:23 AM, Singh, Aman Deep wrote:
> Hi Gregory,
>
> Thanks for the patch.
>
> On 11/9/2022 10:39 PM, Gregory Etelson wrote:
>> Testpmd terminated unconditionally if it failed to start all ports.
>>
>> The patch allows testpmd to get into the command line,
>> if the interactive mode w
> -Original Message-
> From: Zhang, Qi Z
> Sent: Wednesday, January 18, 2023 12:10 PM
> To: Koikkara Reeny, Shibin ;
> dev@dpdk.org; Burakov, Anatoly ; Richardson,
> Bruce
> Cc: Loftus, Ciara
> Subject: RE: [PATCH v2] net/af_xdp: AF_XDP PMD CNI Integration
>
>
>
> > -Original Me
On 11/12/2022 5:28 PM, Stephen Hemminger wrote:
> Do a clean shutdown of testpmd when a signal is received;
> instead of having testpmd kill itself.
> This fixes the problem where a signal could be received
> in the middle of a PMD and then the signal handler would call
> PMD's close routine leadin
19/01/2023 15:49, Ferruh Yigit:
> On 10/13/2022 12:47 PM, Ferruh Yigit wrote:
> > Driver is disabled by default because its dependencies are not
> > upstreamed yet, code is available for development and investigation.
> >
> > When all dependencies are upstreamed, driver can be enabled back.
> >
>
Everything looks good to me with the exception of some issues I ran into
with terminal codes. Setting TERM=dumb before running fixed it, but we
might want to set that inside of DTS since I can't think of a reason why we
would need colors or any of the other "fancy" features of the vt220, and
settin
> From: Robin Jarry [mailto:rja...@redhat.com]
> Sent: Thursday, 19 January 2023 16.07
>
> Allow fetching CPU cycles usage for all lcores with a single request.
> This endpoint is intended for repeated and frequent invocations by
> external monitoring systems and therefore returns condensed data.
> Subject: Re: [PATCH] net/mana: disable driver by default
>
> 19/01/2023 15:49, Ferruh Yigit:
> > On 10/13/2022 12:47 PM, Ferruh Yigit wrote:
> > > Driver is disabled by default because its dependencies are not
> > > upstreamed yet, code is available for development and investigation.
> > >
> > >
Morten Brørup, Jan 19, 2023 at 17:21:
> In the future, the rte_lcore_usage structure may contain more fields,
> and some may not be related to the TSC. So consider removing "cycles"
> from the description of the telemetry path.
>
> Don't waste time changing it unless you are providing a new patch
>
On Thu, 19 Jan 2023 08:59:10 +
Bruce Richardson wrote:
> > > struct object {
> > > SLIST_ENTRY(object) next;
> >
> > I confirm it solves the build warning, but what about to get rid of
> > `OBJ_NAME_LEN_MAX` completely if the intentions is to make size same as
> > cmdline library array:
> From: Robin Jarry [mailto:rja...@redhat.com]
> Sent: Thursday, 19 January 2023 17.35
>
> Morten Brørup, Jan 19, 2023 at 17:21:
> > In the future, the rte_lcore_usage structure may contain more fields,
> > and some may not be related to the TSC. So consider removing "cycles"
> > from the descript
Current API allows either query or update indirect flow action.
If indirect action must be conditionally updated according to it's
present state application must first issue action query then
analyze returned data and if needed issue update request.
When the update will be processed, action state c
Quota action limits traffic according to pre-defined configuration.
Quota reflects overall traffic usage regardless bandwidth.
Quota flow action initialized with signed tokens number value.
Quota flow action updates tokens number according to
these rules:
1. if quota was configured to count packet
On Thu, 19 Jan 2023 14:37:49 +
Amit Prakash Shukla wrote:
> >
> > One other option is to add a new custom block type or options in pcapng.
> > Keep comment as a string, since that is what wireshark etc no how to handle.
> > Also, packet id data might be useful for graph.
>
> I gave a try
Testpmd terminated unconditionally if it failed to start all ports.
The patch allows testpmd to get into the command line,
if the interactive mode was requested.
Fixes: 6937d2103e22 ("app/testpmd: add option to not start device")
Cc: step...@networkplumber.org
Cc: sta...@dpdk.org
Signed-off-by:
Testpmd terminated unconditionally if it failed to start all ports.
The patch allows testpmd to get into the command line,
if the interactive mode was requested.
Fixes: 6937d2103e22 ("app/testpmd: add option to not start device")
Cc: step...@networkplumber.org
Cc: sta...@dpdk.org
Signed-off-by:
On Thu, 19 Jan 2023 19:23:12 +0200
Gregory Etelson wrote:
> + if (!no_device_start && start_port(RTE_PORT_ALL) != 0) {
> + fprintf(stderr, "Start ports failed\n");
> + if (!interactive) {
> + rte_eal_cleanup();
> + rte_exit(EXIT_
Testpmd terminated unconditionally if it failed to start all ports.
The patch allows testpmd to get into the command line,
if the interactive mode was requested.
Fixes: 6937d2103e22 ("app/testpmd: add option to not start device")
Cc: step...@networkplumber.org
Cc: sta...@dpdk.org
Signed-off-by:
19/01/2023 17:32, Long Li:
> > Subject: Re: [PATCH] net/mana: disable driver by default
> >
> > 19/01/2023 15:49, Ferruh Yigit:
> > > On 10/13/2022 12:47 PM, Ferruh Yigit wrote:
> > > > Driver is disabled by default because its dependencies are not
> > > > upstreamed yet, code is available for dev
On Thu, Jan 19, 2023 at 08:44:41AM -0800, Stephen Hemminger wrote:
> On Thu, 19 Jan 2023 08:59:10 +
> Bruce Richardson wrote:
>
> > > > struct object {
> > > > SLIST_ENTRY(object) next;
> > >
> > > I confirm it solves the build warning, but what about to get rid of
> > > `OBJ_NAME
vhost internals involves multiple locks to protect data access by
multiple threads.
This series uses clang thread safety checks [1] to catch issues during
compilation: EAL spinlock, seqlock and rwlock are annotated and vhost
code is instrumented so that clang can statically check correctness.
Tho
clang offers some thread safety checks, statically verifying that locks
are taken and released in the code.
To use those checks, the full code leading to taking or releasing locks
must be annotated with some attributes.
Wrap those attributes into our own set of macros.
rwlock, seqlock and the "no
Dedicate send_vhost_slave_message() helper to the case when no reply is
needed.
Add a send_vhost_slave_message_process_reply() helper for the opposite.
This new helper merges both send_vhost_slave_message() and the code
previously in process_slave_message_reply().
The slave_req_lock lock is then o
Be a bit more strict when a programmatic error is detected wrt to the
access_lock not being taken.
Mark the new helper with __rte_assert_exclusive_lock so that clang
understands where locks are expected to be taken.
Signed-off-by: David Marchand
---
lib/vhost/vhost.c | 18 +++---
vhost_user_lock/unlock_all_queue_pairs must be waived since clang
annotations can't express taking a runtime number of locks.
vhost_queue_stats_update() requirement can be expressed with a required
tag.
Signed-off-by: David Marchand
---
Changes since RFC v3:
- removed annotations needed for vhos
vq->async is initialised and must be accessed under vq->access_lock.
Signed-off-by: David Marchand
---
Changes since RFC v3:
- rebased,
- fixed annotations vq->access_lock -> &vq->access_lock,
- reworked free_vq,
---
lib/vhost/vhost.c | 4
lib/vhost/vhost.h | 2 +-
lib/vhost/vh
clang does not support conditionally held locks when statically analysing
taken locks with thread safety checks.
Always take iotlb locks regardless of VIRTIO_F_IOMMU_PLATFORM feature.
Signed-off-by: David Marchand
---
lib/vhost/vhost.c | 8 +++-
lib/vhost/virtio_net.c | 24
The starting point for this is __vhost_iova_to_vva() which requires the
lock to be taken. Annotate all the code leading to a call to it.
vdpa and vhost_crypto code are annotated but they end up not taking
a IOTLB lock and have been marked with a FIXME at the top level.
Signed-off-by: David Marcha
Access to vdpa_device_list must be protected with vdpa_device_list_lock
spinlock.
Signed-off-by: David Marchand
---
Changes since RFC v3:
- rebased,
---
lib/vhost/vdpa.c | 19 ++-
1 file changed, 10 insertions(+), 9 deletions(-)
diff --git a/lib/vhost/vdpa.c b/lib/vhost/vdpa.c
Now that all locks in this library are annotated, we can enable the
check.
Signed-off-by: David Marchand
---
lib/vhost/meson.build | 2 ++
1 file changed, 2 insertions(+)
diff --git a/lib/vhost/meson.build b/lib/vhost/meson.build
index bc7272053b..197a51d936 100644
--- a/lib/vhost/meson.build
+
> From: David Marchand [mailto:david.march...@redhat.com]
> Sent: Thursday, 19 January 2023 19.46
>
> vhost internals involves multiple locks to protect data access by
> multiple threads.
>
> This series uses clang thread safety checks [1] to catch issues during
> compilation: EAL spinlock, seqlo
On 19/01/2023 15:06, Robin Jarry wrote:
Report the same information than rte_lcore_dump() in the telemetry
API into /eal/lcore/list and /eal/lcore/info,ID.
Example:
--> /eal/lcore/info,3
{
"/eal/lcore/info": {
"lcore_id": 3,
"socket": 0,
"role": "RTE",
"cp
On 19/01/2023 15:06, Robin Jarry wrote:
Allow applications to register a callback that will be invoked in
rte_lcore_dump() and when requesting lcore info in the telemetry API.
The callback is expected to return the number of TSC cycles that have
passed since application start and the number of t
On 19/01/2023 15:06, Robin Jarry wrote:
Add a simple command that calls rte_lcore_dump().
Signed-off-by: Robin Jarry
Acked-by: Morten Brørup
Acked-by: Konstantin Ananyev
---
Notes:
v5 -> v6: No change
app/test-pmd/cmdline.c | 3 +++
1 file changed, 3 insertions(+)
Reviewed-by: Kev
On 19/01/2023 15:06, Robin Jarry wrote:
Reuse the --record-core-cycles option to account for busy cycles. One
turn of packet_fwd_t is considered "busy" if there was at least one
received or transmitted packet.
Add a new busy_cycles field in struct fwd_stream. Update get_end_cycles
to accept an a
On 19/01/2023 15:06, Robin Jarry wrote:
Allow fetching CPU cycles usage for all lcores with a single request.
This endpoint is intended for repeated and frequent invocations by
external monitoring systems and therefore returns condensed data.
It consists of a single dictionary with three keys: "
On Thu, 19 Jan 2023 19:46:12 +0100
David Marchand wrote:
> clang offers some thread safety checks, statically verifying that locks
> are taken and released in the code.
> To use those checks, the full code leading to taking or releasing locks
> must be annotated with some attributes.
>
> Wrap th
On Thu, Jan 19, 2023 at 11:42:02AM -0800, Stephen Hemminger wrote:
> On Thu, 19 Jan 2023 19:46:12 +0100
> David Marchand wrote:
>
> > +#ifndef __DOXYGEN__
> > + __rte_exclusive_lock_function(&seqlock->lock)
> > +#endif
> > {
>
> Would be cleaner any required ifdefs was in rte_lock_annotations
On Thu, Jan 19, 2023 at 8:42 PM Stephen Hemminger
wrote:
>
> On Thu, 19 Jan 2023 19:46:12 +0100
> David Marchand wrote:
>
> > +#ifndef __DOXYGEN__
> > + __rte_exclusive_lock_function(&seqlock->lock)
> > +#endif
> > {
>
> Would be cleaner any required ifdefs was in rte_lock_annotations
> rath
On Thu, Jan 19, 2023 at 9:39 PM Tyler Retzlaff
wrote:
>
> On Thu, Jan 19, 2023 at 11:42:02AM -0800, Stephen Hemminger wrote:
> > On Thu, 19 Jan 2023 19:46:12 +0100
> > David Marchand wrote:
> >
> > > +#ifndef __DOXYGEN__
> > > + __rte_exclusive_lock_function(&seqlock->lock)
> > > +#endif
> > >
DAC_READ_SEARCH or DAC_OVERRIDE capability is required to access
/proc/self/pagemap, but the Linux guide mentioned neither one.
Recommend DAC_READ_SEARCH as less impactful.
Fixes: 979bb5d493fb ("doc: add more instructions for running as non-root")
Cc: sta...@dpdk.org
Signed-off-by: Dmitry Kozlyuk
On Thu, Jan 19, 2023 at 10:16:35PM +0100, David Marchand wrote:
> On Thu, Jan 19, 2023 at 9:39 PM Tyler Retzlaff
> wrote:
> >
> > On Thu, Jan 19, 2023 at 11:42:02AM -0800, Stephen Hemminger wrote:
> > > On Thu, 19 Jan 2023 19:46:12 +0100
> > > David Marchand wrote:
> > >
> > > > +#ifndef __DOXYGE
This series adds self monitoring support i.e allows to configure and
read performance measurement unit (PMU) counters in runtime without
using perf utility. This has certain adventages when application runs on
isolated cores with nohz_full kernel parameter.
Events can be read directly using rte_pm
Add support for programming PMU counters and reading their values
in runtime bypassing kernel completely.
This is especially useful in cases where CPU cores are isolated
(nohz_full) i.e run dedicated tasks. In such cases one cannot use
standard perf utility without sacrificing latency and performa
Add support for reading ARM PMU events in runtime.
Signed-off-by: Tomasz Duszynski
---
app/test/test_pmu.c | 4 ++
lib/pmu/meson.build | 7 +++
lib/pmu/pmu_arm64.c | 94 +
lib/pmu/rte_pmu.h | 4 ++
lib/pmu/rte_pmu_pmc_arm64
Add support for reading Intel x86_64 PMU events in runtime.
Signed-off-by: Tomasz Duszynski
---
app/test/test_pmu.c | 2 ++
lib/pmu/meson.build | 1 +
lib/pmu/rte_pmu.h| 2 ++
lib/pmu/rte_pmu_pmc_x86_64.h | 24
4 files changed, 29 inserti
In order to profile app one needs to store significant amount of samples
somewhere for an analysis latern on. Since trace library supports
storing data in a CTF format lets take adventage of that and add a
dedicated PMU tracepoint.
Signed-off-by: Tomasz Duszynski
---
app/test/test_trace_perf.c
On Thu, Jan 19, 2023 at 5:37 AM Edwin Brossette
wrote:
>
> Hello,
>
snip
> I was wondering if patching the code to move this line which sets lsc
> interrupt on somewhere else might be reasonable, or if this could cause
> further trouble. Maybe having a parameter to trigger it ON/OFF m
From: Long Li
The required dependencies for mana from rdma-core and Linux kernel have
been released. Add those information to the doc.
Fixes: 517ed6e2d590 ("net/mana: add basic driver with build environment")
Signed-off-by: Long Li
---
doc/guides/nics/mana.rst | 8 +++-
1 file changed, 3 i
From: Long Li
The dependencies of mana have been released in rdma-core v44 and Linux
kernel 6.2.
Signed-off-by: Long Li
---
MAINTAINERS | 2 +-
drivers/net/meson.build | 1 +
2 files changed, 2 insertions(+), 1 deletion(-)
diff --git a/MAINTAINERS b/MAINTAINERS
index 9a0f416d2e..7
This patchset contains dmadev/ethdev stats reset, and also support
hide zero for ethdev xstats and two telemetry related bugs.
Chengwen Feng (5):
dmadev: support stats reset telemetry command
telemetry: fix repeat display when callback don't init dict
ethdev: add newline to telemetry log str
The stats reset is useful for debugging, so add it to the dmadev
telemetry command lists.
Signed-off-by: Chengwen Feng
Reviewed-by: Bruce Richardson
---
lib/dmadev/rte_dmadev.c | 43 +
1 file changed, 43 insertions(+)
diff --git a/lib/dmadev/rte_dmadev.c
The xstats reset is useful for debugging, so add it to the ethdev
telemetry command lists.
Signed-off-by: Chengwen Feng
---
lib/ethdev/rte_ethdev.c | 27 +++
1 file changed, 27 insertions(+)
diff --git a/lib/ethdev/rte_ethdev.c b/lib/ethdev/rte_ethdev.c
index 9eeae61024.
When a telemetry callback doesn't initialize the telemetry data
structure and returns a non-negative number, the telemetry will repeat
to display the last result. This patch zero the data structure to avoid
the problem.
Fixes: 6dd571fd07c3 ("telemetry: introduce new functionality")
Cc: sta...@dpdk
1 - 100 of 122 matches
Mail list logo
