2024-10-15 16:04 (UTC-0700), Stephen Hemminger:
> This looks ok, but I know nothing about windows drivers.
> Could we get a review by Dmitry?
v3 got Reviewed-by almost a year ago, v4 only did a bit of renaming:
http://inbox.dpdk.org/dev/20231130071001.4c7931c8@sovereign/
There were no ot
On 2024-10-16 06:36, Mattias Rönnblom wrote:
On 2024-10-16 00:00, Stephen Hemminger wrote:
On Tue, 15 Oct 2024 20:25:35 +0200
Mattias Rönnblom wrote:
Remove the single event enqueue and dequeue, since they did not
provide any noticable performance benefits.
This is a change of the ABI, previ
On Tue, 15 Oct 2024 17:11:43 +
Vladimir Medvedkin wrote:
> Fixes the behavior of the rte_fib_rcu_qsbr_add() function regarding its
> return value to align with the existing rte_fib API.
>
> Fixes: 96c3d06a3547 ("fib: implement RCU rule reclamation")
>
> Signed-off-by: Vladimir Medvedkin
>
On 16-10-2024 10:43, Stephen Hemminger wrote:
On Wed, 16 Oct 2024 10:39:40 +0530
Hemant Agrawal wrote:
diff --git a/devtools/build-dict.sh b/devtools/build-dict.sh
index a8cac49029..d503fb52fc 100755
--- a/devtools/build-dict.sh
+++ b/devtools/build-dict.sh
@@ -17,6 +17,8 @@ sed '/^..->/d' |
On Wed, 16 Oct 2024 10:39:40 +0530
Hemant Agrawal wrote:
> >
> > diff --git a/devtools/build-dict.sh b/devtools/build-dict.sh
> > index a8cac49029..d503fb52fc 100755
> > --- a/devtools/build-dict.sh
> > +++ b/devtools/build-dict.sh
> > @@ -17,6 +17,8 @@ sed '/^..->/d' |
> > sed '/^uint->/d' |
>
On 16-10-2024 03:41, Stephen Hemminger wrote:
On Tue, 15 Oct 2024 12:43:50 +0530
Gagandeep Singh wrote:
From: Jun Yang
refactor the driver code with changes in:
- multiple HW queues
- SMA single copy and SG copy
- silent mode
Signed-off-by: Jun Yang
---
To prevent false positives from c
On 2024-10-16 00:00, Stephen Hemminger wrote:
On Tue, 15 Oct 2024 20:25:35 +0200
Mattias Rönnblom wrote:
Remove the single event enqueue and dequeue, since they did not
provide any noticable performance benefits.
This is a change of the ABI, previously announced as a deprecation
notice. These
On 2024-10-16 00:35, Stephen Hemminger wrote:
On Tue, 15 Oct 2024 11:33:38 +0200
Mattias Rönnblom wrote:
+/**
+ * Allocate space in the per-lcore id buffers for an lcore variable.
+ *
+ * The pointer returned is only an opaque identifier of the variable. To
+ * get an actual pointer to a parti
On 2024-10-16 00:33, Stephen Hemminger wrote:
On Tue, 15 Oct 2024 11:33:38 +0200
Mattias Rönnblom wrote:
+ * Lcore variables
+ *
+ * This API provides a mechanism to create and access per-lcore id
+ * variables in a space- and cycle-efficient manner.
+ *
+ * A per-lcore id variable (or lcor
Hi Ferruh Yigit,
The patch is in 'Changes Requested' status. Can you review it for me?
Regards,
Mingjin
> -Original Message-
> From: Ferruh Yigit
> Sent: Wednesday, October 16, 2024 1:58 AM
> To: Stephen Hemminger ; Ye, MingjinX
>
> Cc: dev@dpdk.org; Su, Simei ; Wenjun Wu
> ; Rybalchen
> -Original Message-
> From: Stephen Hemminger
> Sent: Wednesday, October 16, 2024 1:44 AM
> To: Ye, MingjinX
> Cc: dev@dpdk.org; Su, Simei ; Wenjun Wu
> ; Rybalchenko, Kirill
> Subject: Re: [PATCH v7] examples/ptpclient: add frequency adjustment
>
> On Tue, 15 Oct 2024 08:22:01 +000
> >> Add PM QoS configuration to declease the delay after sleep in case of
> >> entering deeper idle state.
> >>
> >> Signed-off-by: Huisong Li
> >> Acked-by: Morten Brørup
> >> ---
> >> examples/l3fwd-power/main.c | 24
> >> 1 file changed, 24 insertions(+)
> >>
> >
On Mon, 25 Mar 2024 17:01:41 +
Ian Stokes wrote:
> From: Ananth S
>
> This patch supports rte flow with mask for FDIR, including
> eth/ipv4/ipv6/tcp/udp flow items, where src/dst for ipv4/ipv6
> and sport/dport for tcp/udp are realized by switch filter.
>
> This patch additionally contains
On Mon, 4 Dec 2023 18:32:14 +0800
Ric Li wrote:
> The virt2phys_translate function previously scanned existing blocks,
> returning the physical address from the stored MDL info if present.
> This method was problematic when a virtual address pointed to a freed
> and reallocated memory segment, p
On Tue, 15 Oct 2024 11:33:38 +0200
Mattias Rönnblom wrote:
> +/**
> + * Allocate space in the per-lcore id buffers for an lcore variable.
> + *
> + * The pointer returned is only an opaque identifier of the variable. To
> + * get an actual pointer to a particular instance of the variable use
> +
On Tue, 15 Oct 2024 11:33:38 +0200
Mattias Rönnblom wrote:
> + * Lcore variables
> + *
> + * This API provides a mechanism to create and access per-lcore id
> + * variables in a space- and cycle-efficient manner.
> + *
> + * A per-lcore id variable (or lcore variable for short) holds a
> + * uniq
On Fri, 11 Oct 2024 18:17:00 +
Vladimir Medvedkin wrote:
> +
> +uint32_t
> +rte_thash_get_rand_poly(uint32_t poly_degree)
> +{
> + uint32_t ret_poly;
> +
> + if (poly_degree > 32)
> + return 0;
> +
> + do
> + ret_poly = __thash_get_rand_poly(poly_degree);
>
On Tue, 15 Oct 2024 12:43:50 +0530
Gagandeep Singh wrote:
> From: Jun Yang
>
> refactor the driver code with changes in:
> - multiple HW queues
> - SMA single copy and SG copy
> - silent mode
>
> Signed-off-by: Jun Yang
> ---
To prevent false positives from checkpatch spell check, it would
b
On Tue, 15 Oct 2024 20:25:35 +0200
Mattias Rönnblom wrote:
> Remove the single event enqueue and dequeue, since they did not
> provide any noticable performance benefits.
>
> This is a change of the ABI, previously announced as a deprecation
> notice. These functions were not directly called by
> From: Thomas Monjalon [mailto:tho...@monjalon.net]
> Sent: Tuesday, 15 October 2024 21.58
>
> 15/10/2024 16:44, Morten Brørup:
> > > From: Thomas Monjalon [mailto:tho...@monjalon.net]
> > > Sent: Tuesday, 15 October 2024 16.13
> > >
> > > 15/10/2024 14:53, Morten Brørup:
> > > > > From: David Ma
> From: Mattias Rönnblom [mailto:hof...@lysator.liu.se]
> Sent: Tuesday, 15 October 2024 21.03
>
> On 2024-10-15 12:13, Morten Brørup wrote:
> >> +void *
> >> +rte_lcore_var_alloc(size_t size, size_t align)
> >> +{
> >> + /* Having the per-lcore buffer size aligned on cache lines
> >> + * assur
15/10/2024 16:44, Morten Brørup:
> > From: Thomas Monjalon [mailto:tho...@monjalon.net]
> > Sent: Tuesday, 15 October 2024 16.13
> >
> > 15/10/2024 14:53, Morten Brørup:
> > > > From: David Marchand [mailto:david.march...@redhat.com]
> > > > @@ -255,7 +255,13 @@ __rte_experimental
> > > > static
test suite for verifying layer 3/4 checksum offload
features on poll mode driver.
Signed-off-by: Dean Marx
---
dts/framework/config/conf_yaml_schema.json | 3 +-
dts/tests/TestSuite_checksum_offload.py| 274 +
2 files changed, 276 insertions(+), 1 deletion(-)
create mo
Add csum_set_hw method to testpmd shell class.
Signed-off-by: Dean Marx
---
dts/framework/remote_session/testpmd_shell.py | 58 +++
1 file changed, 58 insertions(+)
diff --git a/dts/framework/remote_session/testpmd_shell.py
b/dts/framework/remote_session/testpmd_shell.py
index
Port over checksum hardware offload testing suite from old DTS. The
suite verifies the ability of the PMD to recognize whether an incoming
packet has valid or invalid L4/IP checksum values.
-
v1:
* In the original test plan, there were two Tx checksum test cases. I
removed them due to the
On 2024-10-15 12:13, Morten Brørup wrote:
+void *
+rte_lcore_var_alloc(size_t size, size_t align)
+{
+ /* Having the per-lcore buffer size aligned on cache lines
+* assures as well as having the base pointer aligned on cache
+* size assures that aligned offsets also translat
On 2024-10-15 19:07, Stephen Hemminger wrote:
On Tue, 15 Oct 2024 10:49:33 +0200
Mattias Rönnblom wrote:
Remove the single-event enqueue and dequeue functions from the
eventdev "ops" struct, to reduce complexity, leaving performance
unaffected.
This ABI change has been announced as a DPDK dep
Signed-off-by: Mattias Rönnblom
---
drivers/event/cnxk/cn10k_eventdev.c | 2 --
drivers/event/cnxk/cn10k_worker.c | 49 +
drivers/event/cnxk/cn10k_worker.h | 9 --
drivers/event/cnxk/cn9k_eventdev.c | 1 -
drivers/event/cnxk/cn9k_worker.c| 26 +-
Signed-off-by: Mattias Rönnblom
--
RFC v2:
* Remove unused functions.
---
drivers/event/skeleton/skeleton_eventdev.c | 29 --
1 file changed, 29 deletions(-)
diff --git a/drivers/event/skeleton/skeleton_eventdev.c
b/drivers/event/skeleton/skeleton_eventdev.c
index 848b3be
Signed-off-by: Mattias Rönnblom
---
drivers/event/sw/sw_evdev.c| 2 --
drivers/event/sw/sw_evdev.h| 2 --
drivers/event/sw/sw_evdev_worker.c | 12
3 files changed, 16 deletions(-)
diff --git a/drivers/event/sw/sw_evdev.c b/drivers/event/sw/sw_evdev.c
index 7913bc54
Signed-off-by: Mattias Rönnblom
---
drivers/event/opdl/opdl_evdev.c | 2 --
1 file changed, 2 deletions(-)
diff --git a/drivers/event/opdl/opdl_evdev.c b/drivers/event/opdl/opdl_evdev.c
index 25853166bf..ffa65ef930 100644
--- a/drivers/event/opdl/opdl_evdev.c
+++ b/drivers/event/opdl/opdl_evdev.
Signed-off-by: Mattias Rönnblom
---
drivers/event/dsw/dsw_evdev.c | 2 --
drivers/event/dsw/dsw_evdev.h | 2 --
drivers/event/dsw/dsw_event.c | 12
3 files changed, 16 deletions(-)
diff --git a/drivers/event/dsw/dsw_evdev.c b/drivers/event/dsw/dsw_evdev.c
index 1df3121b21..e819412
Signed-off-by: Mattias Rönnblom
---
drivers/event/dpaa/dpaa_eventdev.c | 27 ++-
1 file changed, 2 insertions(+), 25 deletions(-)
diff --git a/drivers/event/dpaa/dpaa_eventdev.c
b/drivers/event/dpaa/dpaa_eventdev.c
index 85ccb586ef..853cc1ecf9 100644
--- a/drivers/event/
Remove the single event enqueue and dequeue, since they did not
provide any noticable performance benefits.
This is a change of the ABI, previously announced as a deprecation
notice. These functions were not directly called by the application,
so the API remains unaffected.
Signed-off-by: Mattias
Signed-off-by: Mattias Rönnblom
--
RFCv2:
* Remove unused function. (Stephen Hemminger)
---
drivers/event/dlb2/dlb2.c | 40 ++-
1 file changed, 2 insertions(+), 38 deletions(-)
diff --git a/drivers/event/dlb2/dlb2.c b/drivers/event/dlb2/dlb2.c
index 09e4107
Signed-off-by: Mattias Rönnblom
--
RFC v2:
* Remove unused functions.
---
drivers/event/dpaa2/dpaa2_eventdev.c | 15 ---
1 file changed, 15 deletions(-)
diff --git a/drivers/event/dpaa2/dpaa2_eventdev.c
b/drivers/event/dpaa2/dpaa2_eventdev.c
index f0b2c7de99..6c8ed3ff6b 100644
--
Remove the single-event enqueue and dequeue functions from the
eventdev "ops" struct, to reduce complexity, leaving performance
unaffected.
This ABI change has been announced as a DPDK deprication notice,
originally scheduled for DPDK 23.11.
Mattias Rönnblom (10):
event/dsw: remove single event
Signed-off-by: Mattias Rönnblom
---
drivers/event/octeontx/ssovf_evdev.h | 1 -
drivers/event/octeontx/ssovf_worker.c | 40 +++
2 files changed, 4 insertions(+), 37 deletions(-)
diff --git a/drivers/event/octeontx/ssovf_evdev.h
b/drivers/event/octeontx/ssovf_evdev.h
in
On 10/15/2024 6:43 PM, Stephen Hemminger wrote:
> On Tue, 15 Oct 2024 08:22:01 +
> Mingjin Ye wrote:
>
>> This patch adds PI servo controller to support frequency
>> adjustment API for IEEE1588 PTP.
>>
>> For example, the command for starting ptpclient with PI controller is:
>> dpdk-ptpclient
On Tue, 15 Oct 2024 08:22:01 +
Mingjin Ye wrote:
> This patch adds PI servo controller to support frequency
> adjustment API for IEEE1588 PTP.
>
> For example, the command for starting ptpclient with PI controller is:
> dpdk-ptpclient -a :81:00.0 -c 1 -n 3 -- -T 0 -p 0x1 -c 1
>
> Signed
On Fri, 6 Sep 2024 05:19:39 +
Mingjin Ye wrote:
> This patch adds PI servo algorithm to support frequency
> adjustment API for IEEE1588 PTP.
>
> For example, the command for starting ptpclient with PI algorithm is:
> ./build/examples/dpdk-ptpclient -a :81:00.0 -c 1 -n 3 -- -T 0 -p 0x1
>
Fixes the behavior of the rte_fib_rcu_qsbr_add() function regarding its
return value to align with the existing rte_fib API.
Fixes: 96c3d06a3547 ("fib: implement RCU rule reclamation")
Signed-off-by: Vladimir Medvedkin
---
lib/fib/dir24_8.c | 21 -
lib/fib/rte_fib.h | 8 +++
On Tue, 15 Oct 2024 10:49:33 +0200
Mattias Rönnblom wrote:
> Remove the single-event enqueue and dequeue functions from the
> eventdev "ops" struct, to reduce complexity, leaving performance
> unaffected.
>
> This ABI change has been announced as a DPDK deprication notice,
> originally scheduled
On Tue, 15 Oct 2024 17:35:38 +0530
Hemant Agrawal wrote:
> This patch remove the new line char from several logs in the
> fmlib base driver.
>
> Signed-off-by: Hemant Agrawal
> ---
Acked-by: Stephen Hemminger
On Tue, Oct 01, 2024 at 02:14:36PM +0100, Bruce Richardson wrote:
> On Tue, Oct 01, 2024 at 11:12:54AM +0200, David Marchand wrote:
> > Following two upstream Linux kernel changes (see links), the mac address
> > of a iavf port, serviced by a i40e PF driver, is lost when the DPDK iavf
> > driver pr
Sorry, I should have flagged that the series is called "untitled series #"
which is why I think there is a problem with how the series was submitted.
It looks like this did not get processed correctly by patchwork, or there
is an issue with the patchfiles.
https://patchwork.dpdk.org/project/dpdk/list/?series=&submitter=&state=&q=gdtc&archive=&delegate=
Can you resubmit? It was not picked up by 3/4 of the CI labs.
On Mon, Oct 14, 2024 at 4:16
Implement rte_flow_async_create_by_index_with_pattern() function.
Rework the driver implementation to reduce code duplication by
providing a single flow insertion routine, that can be called with
different parameters depending on the insertion type.
Signed-off-by: Alexander Kozyrev
---
drivers/n
Implement RTE_FLOW_ACTION_TYPE_JUMP_TO_TABLE_INDEX action.
Create the hardware steering jump to matcher action,
associated with the template matcher. Use this action and
provide the rule index as an offset in the matcher.
Note that it is only supported by the isolated matcher,
i.e. the table insert
From: Hamdan Igbaria
Introduce jump to matcher action, this action will allow
jumping to another matcher.
For now this jump restricted to STE array matchers and
matchers of size 1.
Signed-off-by: Hamdan Igbaria
---
drivers/net/mlx5/hws/mlx5dr.h| 29 ++
drivers/net/mlx5/hws/mlx5
Create an array STE matcher for a template table
in case of insertion by index with pattern is selected.
Packets will be matched on a pattern at the index.
This table is isolated from any other tables in a group.
That means packets missed the rule won't go to a lower
priority tables, but proceed wi
From: Hamdan Igbaria
introduce STE array matcher, where this matcher can only
be isolated under a parent table and not chained to the
table matchers chain.
Signed-off-by: Hamdan Igbaria
---
drivers/net/mlx5/hws/mlx5dr.h | 13 +-
drivers/net/mlx5/hws/mlx5dr_debug.c | 12 +-
dr
Implement RTE_FLOW_ACTION_TYPE_JUMP_TO_TABLE_INDEX action.
Create the hardware steering jump to matcher action,
assosiated with the template matcher. Use this action and
provided rule index as an offset in the matcher.
Note that it is only supported by the isolated matcher,
i.e. the table insertion
Create an array STE matcher for a template table
in case of insertion by index with pattern is selected.
Packets will be matched on a pattern at the index.
This table is isolated from any other tables in a group.
That means packets missed the rule won't go to a lower
priority tables, but proceed wi
Implement rte_flow_async_create_by_index_with_pattern() function.
Rework the driver implementaion to reduce the code duplication by
providing a signle flow insertion routine, that can be called with
different parameters depending on the insertion type.
Signed-off-by: Alexander Kozyrev
---
driver
From: Hamdan Igbaria
introduce STE array matcher, where this matcher can only
be isolated under a parent table and not chained to the
table matchers chain.
Signed-off-by: Hamdan Igbaria
---
drivers/net/mlx5/hws/mlx5dr.h | 13 +-
drivers/net/mlx5/hws/mlx5dr_debug.c | 12 +-
dr
From: Hamdan Igbaria
Introduce jump to matcher action, this action will allow
jumping to another matcher.
For now this jump restricted to STE array matchers and
matchers of size 1.
Signed-off-by: Hamdan Igbaria
---
drivers/net/mlx5/hws/mlx5dr.h| 29 ++
drivers/net/mlx5/hws/mlx5
On Thu, Aug 01, 2024 at 04:40:15AM +, Soumyadeep Hore wrote:
> E822 PHY does not support switching from bypass to Vernier mode.
> Remove ice_phy_exit_bypass_e822() and program fixed offsets for bypass
> mode.
>
> Fixes: ce9ad8c5bc6d ("net/ice/base: remove PHY port timer bypass mode")
> Cc: sta
On Tue, Oct 01, 2024 at 11:15:07AM +0200, David Marchand wrote:
> Caught by code review.
> The mailbox message maximum size is larger than the biggest message
> to update mac addresses.
> Remove the double loop and add some build / assert checks.
>
> This cleanup also fixes a (never hit) issue whe
On Mon, 14 Oct 2024 10:00:26 +
Mingjin Ye wrote:
> Without using allow (-a) or block (-b), EAL loads all devices by default.
> Unexpected devices may be loaded when running test cases in sub-processes.
>
> This patch fixes the issue by copying the parameters of the master process
> if the al
On Thu, 3 Oct 2024 18:52:19 +0530
wrote:
> From: Pavan Nikhilesh
>
> Add platform specific event device probe and remove, also add
> event device info get function.
>
> Signed-off-by: Pavan Nikhilesh
> ---
> doc/guides/eventdevs/cnxk.rst | 30 +
> doc/guides/rel_notes/releas
On Tue, 15 Oct 2024 14:01:05 +0100
Konstantin Ananyev wrote:
> From: Konstantin Ananyev
>
> NOTE UPFRONT: this version is still not ready for merging.
> Missing items:
> - ARM/PPC tests passing
> - PG update
>
> v4 -> v5
> - fix public API/doc comments from Jerin
> - update devtools/build-dict
On Tue, 15 Oct 2024 14:01:05 +0100
Konstantin Ananyev wrote:
> From: Konstantin Ananyev
>
> NOTE UPFRONT: this version is still not ready for merging.
> Missing items:
> - ARM/PPC tests passing
> - PG update
>
> v4 -> v5
> - fix public API/doc comments from Jerin
> - update devtools/build-dict
On Tue, 15 Oct 2024 13:44:35 +0800
Junlong Wang wrote:
> provided zxdh dev configure ops for queue
> check,reset,alloc resources,etc.
>
> Signed-off-by: Junlong Wang
These build failures need to be addressed.
Probably as simple as adding
struct bar_msg_header msg_header = { 0 };
On Tue, 15 Oct 2024 11:09:10 +0800
Howard Wang wrote:
> R8169 pmd supports Realtek 2.5G and 5G ethernet nics.
>
> Howard Wang (18):
> net/r8169: add PMD driver skeleton
> net/r8169: add logging structure
> net/r8169: add hardware registers access routines
> net/r8169: implement core logi
On Tue, 15 Oct 2024 17:41:39 +0800
"lihuisong (C)" wrote:
> Hi Stephen,
>
> Can you take a look at this reply so as to send out the next version ASAP?
> Thanks.😁
>
> /Huisong
> 在 2024/10/14 20:19, lihuisong (C) 写道:
The biggest issue is that lcore is not the same as cpu as far as kernel is
con
On Tue, 15 Oct 2024 13:44:35 +0800
Junlong Wang wrote:
> diff --git a/drivers/net/zxdh/meson.build b/drivers/net/zxdh/meson.build
> index 9aec47e68f..cde96d8111 100644
> --- a/drivers/net/zxdh/meson.build
> +++ b/drivers/net/zxdh/meson.build
> @@ -14,8 +14,9 @@ if not dpdk_conf.has('RTE_ARCH_X86_
On Tue, 15 Oct 2024 08:30:19 -0700
Stephen Hemminger wrote:
> On Tue, 15 Oct 2024 11:09:22 +0800
> Howard Wang wrote:
>
> > +rtl_tx_queue_release_mbufs(struct rtl_tx_queue *txq)
> > +{
> > + int i;
> > +
> > + PMD_INIT_FUNC_TRACE();
> > +
> > + if (txq != NULL) {
> > + if (txq->
On Tue, 15 Oct 2024 11:09:22 +0800
Howard Wang wrote:
> + /* Allocate memory for the software ring */
> + txq->sw_ring = rte_zmalloc_socket("r8169 sw tx ring",
> + nb_tx_desc * sizeof(struct
> rtl_tx_entry),
> +
On Tue, 15 Oct 2024 11:09:22 +0800
Howard Wang wrote:
> +rtl_tx_queue_setup(struct rte_eth_dev *dev, uint16_t queue_idx,
> + uint16_t nb_tx_desc, unsigned int socket_id,
> + const struct rte_eth_txconf *tx_conf)
> +{
> + struct rtl_tx_queue *txq;
> + co
On Tue, 15 Oct 2024 11:09:22 +0800
Howard Wang wrote:
> +rtl_tx_queue_release_mbufs(struct rtl_tx_queue *txq)
> +{
> + int i;
> +
> + PMD_INIT_FUNC_TRACE();
> +
> + if (txq != NULL) {
> + if (txq->sw_ring != NULL) {
> + for (i = 0; i < txq->nb_tx_desc;
On Tue, 15 Oct 2024 11:09:25 +0800
Howard Wang wrote:
> +static int
> +rtl_dev_mtu_set(struct rte_eth_dev *dev, uint16_t mtu)
> +{
> + struct rte_eth_dev_info dev_info;
> + struct rtl_adapter *adapter = RTL_DEV_PRIVATE(dev);
> + struct rtl_hw *hw = &adapter->hw;
> + int ret;
> +
Rather than assuming that each VSI can hold up to 256 queue pairs,
or the reported device limit, query the available nodes in the scheduler
tree to check that we are not overflowing the limit for number of
child scheduling nodes at each level. Do this by multiplying
max_children for each level beyo
The DDP package file being loaded at init time may contain an
alternative Tx Scheduler topology in it. Add driver option to load this
topology at init time.
Signed-off-by: Bruce Richardson
---
V2: use uint8_t rather than int for boolean flag for consistency.
---
drivers/net/ice/base/ice_ddp.c |
The "Dynamic Device Personalization" package is loaded at initialization
time by the driver, but the specific package file loaded depends upon
what package file is found first by searching through a hard-coded list
of firmware paths. To enable greater control over the package loading,
we can add a
The function to dump the TX scheduler topology only adds to the chart
nodes connected to TX queues or for the flow director VSI. Change the
function to work recursively from the root node and thereby include all
scheduler nodes, whether in use or not, in the dump.
Also, improve the output of the T
If the flow-director queue is stopped at some point during the running
of an application, the shutdown procedure for the port issues an error
as it tries to stop the queue a second time, and fails to do so. We can
eliminate this error by setting the tail-register pointer to NULL on
stop, and checki
This patchset contains a set of updates for the ice driver, a number of
which are in the area of the "rte_tm" APIs for Tx scheduling.
These patches were previously submitted as part of a larger set[1], but
separating them out here for easier review and merge ahead of the more
substantial changes f
> > > > > > > > > 1. rte_ring_generic_pvt.h:
> > > > > > > > > =
> > > > > > > > >
> > > > > > > > > pseudo-c-code //
> > > > > > > > > related armv8 instructions
> > > > > > > > >
> From: Konstantin Ananyev [mailto:konstantin.v.anan...@yandex.ru]
> Sent: Tuesday, 15 October 2024 15.01
Cleaning up is good. :-)
Only a few initial comments; haven't reviewed in depth.
> +__rte_ring_headtail_move_head(struct rte_ring_headtail *d,
> + const struct rte_ring_headtail
> From: Thomas Monjalon [mailto:tho...@monjalon.net]
> Sent: Tuesday, 15 October 2024 16.13
>
> 15/10/2024 14:53, Morten Brørup:
> > > From: David Marchand [mailto:david.march...@redhat.com]
> > > @@ -255,7 +255,13 @@ __rte_experimental
> > > static inline bool
> > > rte_bitset_test(const uint64
> -Original Message-
> From: Alexander Kozyrev
> Sent: Thursday, August 22, 2024 17:32
> To: dev@dpdk.org
> Cc: Dariusz Sosnowski ; Ori Kam ;
> nithind1...@gmail.com; olivier.m...@6wind.com; NBU-Contact-Thomas
> Monjalon (EXTERNAL) ; Matan Azrad
> ; jer...@marvell.com; rbhans...@marvell.co
On Wed, Oct 09, 2024 at 10:45:43AM -0700, Stephen Hemminger wrote:
> On Wed, 9 Oct 2024 18:08:19 +0100
> Bruce Richardson wrote:
>
> > The function to dump the TX scheduler topology only adds to the chart
> > nodes connected to TX queues or for the flow director VSI. Change the
> > function to w
15/10/2024 14:53, Morten Brørup:
> > From: David Marchand [mailto:david.march...@redhat.com]
> > @@ -255,7 +255,13 @@ __rte_experimental
> > static inline bool
> > rte_bitset_test(const uint64_t *bitset, size_t bit_num)
> > {
> > +#ifdef ALLOW_EXPERIMENTAL_API
> > return __RTE_BITSET_DELEGAT
From: Konstantin Ananyev
Add both functional and stess test-cases for soring API.
Stress test serves as both functional and performance test of soring
enqueue/dequeue/acquire/release operations under high contention
(for both over committed and non-over committed scenarios).
Signed-off-by: Eimea
From: Konstantin Ananyev
Add a new test suite which purpose is to run 'stress' tests:
main purpose is put a pressure to dpdk sync algorithms
to flag their misbehaving/slowdown/etc.
Right now it consists from just 2 test-cases:
meson test --suite stress-tests --list
DPDK:stress-tests / ring_stress
From: Konstantin Ananyev
Note upfront: that change doesn't introduce any functional
or performance changes.
It is just a code-reordering for:
- improve code modularity and re-usability
- ability in future to re-use the same code to introduce new functionality
There is no real need for enqueue_
From: Konstantin Ananyev
Staged-Ordered-Ring (SORING) provides a SW abstraction for 'ordered' queues
with multiple processing 'stages'.
It is based on conventional DPDK rte_ring, re-uses many of its concepts,
and even substantial part of its code.
It can be viewed as an 'extension' of rte_ring fu
From: Eimear Morrissey
The current rte_ring_dump function uses the generic rte_ring_headtail
structure to access head/tail positions. This is incorrect for the RTS
case where the head is stored in a different offset in the union of
structs. Switching to a separate function for each sync type allo
From: Konstantin Ananyev
Note upfront: that change doesn't introduce any functional or
performance changes.
It is just a code-reordering for:
- code deduplication
- ability in future to re-use the same code to introduce new functionality
For each sync mode corresponding move_prod_head() and
mo
From: Konstantin Ananyev
NOTE UPFRONT: this version is still not ready for merging.
Missing items:
- ARM/PPC tests passing
- PG update
v4 -> v5
- fix public API/doc comments from Jerin
- update devtools/build-dict.sh (Stephen)
- fix MSVC warnings
- introduce new test-suite for meson (stress) wit
> From: David Marchand [mailto:david.march...@redhat.com]
> Sent: Tuesday, 15 October 2024 14.11
>
> For a reason similar to the change on bitops header, hide bitset
> implementation relying on experimental API.
>
> Fixes: 99a1197647d8 ("eal: add bitset type")
>
> Signed-off-by: David Marchand
> From: David Marchand [mailto:david.march...@redhat.com]
> Sent: Tuesday, 15 October 2024 14.11
>
> Building OVS against current DPDK fails with following warnings:
>
> In file included from .../ovs/dpdk-dir/include/rte_memory.h:18,
> from .../ovs/dpdk-dir/include/rte_ring_core.
An exported header should be usable w/ and w/o ALLOW_EXPERIMENTAL_API so
that an application that only wants stable API may include it.
Plus, the widely common case is that an application will not use
internal API.
Cover those cases but keep the original test.
Signed-off-by: David Marchand
---
For a reason similar to the change on bitops header, hide bitset
implementation relying on experimental API.
Fixes: 99a1197647d8 ("eal: add bitset type")
Signed-off-by: David Marchand
---
lib/eal/include/rte_bitset.h | 123 +++
1 file changed, 123 insertions(+)
Building OVS against current DPDK fails with following warnings:
In file included from .../ovs/dpdk-dir/include/rte_memory.h:18,
from .../ovs/dpdk-dir/include/rte_ring_core.h:29,
from .../ovs/dpdk-dir/include/rte_ring.h:37,
from .../ovs/dpdk-dir/i
We currently check that exported headers are fine with
-DALLOW_EXPERIMENTAL_API and -DALLOW_INTERNAL_API.
Such a check won't catch issues like the one fixed in patch 1, where OVS
compilation is broken by the additional of experimental API in a header
commonly included in other parts of DPDK.
Idea
This patch remove the new line char from several logs in the
fmlib base driver.
Signed-off-by: Hemant Agrawal
---
drivers/net/dpaa/fmlib/fm_ext.h | 21 +++--
drivers/net/dpaa/fmlib/fm_lib.c | 76 -
drivers/net/dpaa/fmlib/fm_vsp.c | 20 -
3 files change
I already did, though to my personal one. This patch is from a way
back, I did forget about it.
On Thu, Oct 3, 2024 at 2:29 PM Radoslaw Biernacki wrote:
>
> Staszek, can you update your email? (same with up...@semihalf.com?)
>
> On Wed, Oct 2, 2024 at 6:51 PM 'Stephen Hemminger' via upstream
> w
> +void *
> +rte_lcore_var_alloc(size_t size, size_t align)
> +{
> + /* Having the per-lcore buffer size aligned on cache lines
> + * assures as well as having the base pointer aligned on cache
> + * size assures that aligned offsets also translate to alipgned
> + * pointers acro
1 - 100 of 169 matches
Mail list logo
