For test-pipeline, there is dev start operation, but when thread need to
exit, there is no dev close operation. This is not safe, to fix this,
add dev close operation.
Signed-off-by: Feifei Wang
Reviewed-by: Ruifeng Wang
---
app/test-pipeline/main.c | 19 +++
1 file changed, 19
For test-pipeline, if the main core receive SIGINT signal, it will kill
all the threads immediately and not wait other threads to finish their
jobs.
To fix this, add 'signal_handler' function.
Fixes: 48f31ca50cc4 ("app/pipeline: packet framework benchmark")
Cc: cristian.dumitre...@intel.com
Cc: s
For some drivers which can not support the configured RSS hash functions,
the thread reports 'invalid rss_hf' when doing device configure.
For example, i40e driver can not support 'RTE_ETH_RSS_IPV4',
'RTE_ETH_RSS_IPV6' and 'RTE_ETH_RSS_NONFRAG_IPV6_OTHER', thus it can not
run successfully in test-
For test-pipeline application, there are some problems with the normal
operation of the program and security issues. These patches can fix
these issues.
v3: fix SIGINT handling issue and add dev close operation
Feifei Wang (3):
app/test-pipeline: relax RSS hash requirement
app/test-pipeline:
This patch still have problem, I'll fix next version.
Stephen Hemminger 于2023年9月6日周三 23:08写道：
>
> On Thu, 31 Aug 2023 19:19:37 +0800
> Fengnan Chang wrote:
>
> > +#ifndef RTE_MALLOC_DEBUG
> > + if (internal_conf->legacy_mem) {
> > + /* If orig_elem is dirty, the joint element is
On 2023/9/8 20:01, Andrew Rybchenko wrote:
On 9/8/23 14:28, Jie Hai wrote:
The DPDK framework reports the queue state, which is stored in
dev->data->tx_queue_state and dev->data->rx_queue_state. The
state is maintained by the driver. Users may determine whether
a queue participates in packet for
Hi, Thomas,
Kindly ping for review.
Thanks,
Jie Hai
On 2023/7/4 17:04, Jie Hai wrote:
This patch set supports telemetry cmd to list rings and dump information
of a ring by its name.
v1->v2:
1. Add space after "switch".
2. Fix wrong strlen parameter.
v2->v3:
1. Remove prefix "rte_" for static
On 2023/9/9 4:58, Stephen Hemminger wrote:
On Fri, 8 Sep 2023 17:28:08 +0800
Jie Hai wrote:
Hi, Stephen Hemminger
On 2023/9/6 23:10, Stephen Hemminger wrote:
On Sat, 26 Aug 2023 15:46:03 +0800
Jie Hai wrote:
Currently, rte_eth_rss_conf supports configuring and querying
rss hash functio
Add some about document about the how to use
rte_flow on ice PMD.
Signed-off-by: Qi Zhang
---
doc/guides/nics/ice.rst | 45 +
1 file changed, 45 insertions(+)
diff --git a/doc/guides/nics/ice.rst b/doc/guides/nics/ice.rst
index 5a47109c3f..b36a4c260a 1006
Unified the supported patten array name to flow
ice__supported_pattern.
Signed-off-by: Qi Zhang
---
drivers/net/ice/ice_acl_filter.c| 6 +++---
drivers/net/ice/ice_fdir_filter.c | 6 +++---
drivers/net/ice/ice_switch_filter.c | 6 +++---
3 files changed, 9 insertions(+), 9 deletions(-)
di
Mapping rte_flow_attr->group to a specific hardware stage.
Group 0 -> switch filter
Group 1 -> acl filter (dcf mode only)
Group 2 -> fdir filter (pf mode only)
For RSS, it will only be selected if there is a RTE_FLOW_ACTION_RSS
action target no queue group and the group ID is ignored.
Since each
Only use "disable_engine_mask" for flow engine disabling
In PF mode, only ACL engine will be disabled.
In DCF mode, FDIR and HASH engine will be disabled.
In DCF mode with "acl=off", ACL engine will also be
disabled.
Signed-off-by: Qi Zhang
---
drivers/net/ice/ice_acl_filter.c | 3 ---
drivers
This marks the initial phase of refactoring the ice rte_flow
implementation.
The combination of switch and fdir rules within the same syntax has led
to inconvenient user experiences. Naturally, the switch filter and fdir
filter represent distinct pipeline stages with differing hardware
capabilitie
1. remove pipeline mode
2. apply group to hardware pipeline stage static mapping
3. add genenic flow document
v3:
- fix the issue when acl=off on dcf mode
- refine the disabled engine handling
- unified the supported pattern variable name
- add document
v2:
- fix segment fault when an unintialize
On Mon, Sep 11, 2023 at 06:13:48PM +0200, Thomas Monjalon wrote:
> 08/09/2023 06:22, Tyler Retzlaff:
> > On Wed, Sep 06, 2023 at 06:12:27PM +0200, Thomas Monjalon wrote:
> > > --- a/lib/eal/common/eal_common_thread.c
> > > +++ b/lib/eal/common/eal_common_thread.c
> > > @@ -248,7 +248,7 @@ struct rt
Hi Ric,
2023-09-11 21:09 (UTC+0800), Ric Li:
> 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, potent
> From: Thomas Monjalon [mailto:tho...@monjalon.net]
> Sent: Monday, 11 September 2023 17.26
>
> 07/09/2023 13:10, Morten Brørup:
> > > From: David Marchand [mailto:david.march...@redhat.com]
> > > Sent: Thursday, 7 September 2023 10.55
> > >
> > > On Thu, Sep 7, 2023 at 10:53 AM David Marchand
>
> -Original Message-
> From: Ori Kam
> Sent: Friday, September 1, 2023 7:58 AM
> To: Jerin Jacob
> Cc: Ferruh Yigit ; Dumitrescu, Cristian
> ; Morten Brørup
> ; Zhang, Qi Z ; NBU-
> Contact-Thomas Monjalon (EXTERNAL) ;
> david.march...@redhat.com; Richardson, Bruce ;
> jer...@marvell.com;
11/09/2023 09:41, Michael Baum:
> Add support for a new item type "RTE_FLOW_ITEM_TYPE_RANDOM".
> This item enables to match on some random value as a part of flow rule.
>
> Signed-off-by: Michael Baum
> ---
> +Item: ``RANDOM``
> +
> +
> +Matches a random value.
> +
> +This value i
08/09/2023 06:22, Tyler Retzlaff:
> On Wed, Sep 06, 2023 at 06:12:27PM +0200, Thomas Monjalon wrote:
> > --- a/lib/eal/common/eal_common_thread.c
> > +++ b/lib/eal/common/eal_common_thread.c
> > @@ -248,7 +248,7 @@ struct rte_thread_ctrl_params {
> > enum __rte_ctrl_thread_status ctrl_thread_st
07/09/2023 13:10, Morten Brørup:
> > From: David Marchand [mailto:david.march...@redhat.com]
> > Sent: Thursday, 7 September 2023 10.55
> >
> > On Thu, Sep 7, 2023 at 10:53 AM David Marchand
> > wrote:
> > >
> > > On Thu, Sep 7, 2023 at 10:50 AM Morten Brørup
> > wrote:
> > > > > This 10 value i
On Mon, 11 Sep 2023 18:55:45 +0200
Morten Brørup wrote:
> > From: Michael Baum [mailto:michae...@nvidia.com]
> > Sent: Monday, 11 September 2023 08.42
> >
> > Add support for matching random value using new
> > "rte_flow_item_random" structure.
>
> I get it. It can be used for things like sto
> From: Michael Baum [mailto:michae...@nvidia.com]
> Sent: Monday, 11 September 2023 08.42
>
> Add support for matching random value using new "rte_flow_item_random"
> structure.
I get it. It can be used for things like stochastic sampling.
However, it doesn't provide support for e.g. 1/100 or 1
On 2023-09-11 18:06, Stephen Hemminger wrote:
On Fri, 8 Sep 2023 09:04:29 +0200
Mattias Rönnblom wrote:
Also, right now the array is sized at 129 entries to allow for the
maximum number of lcores. When the maximum is increased to 512 or
1024 the problem will get worse.
Using TLS will penaliz
> From: Stephen Hemminger [mailto:step...@networkplumber.org]
> Sent: Monday, 11 September 2023 17.04
>
> On Wed, 6 Sep 2023 19:54:32 +0200
> Morten Brørup wrote:
>
> > >
> > > - idx = rte_lcore_id();
> > > + seed = __atomic_load_n(&rte_rand_seed, __ATOMIC_RELAXED);
> > > + if (unlikely(seed !=
On Mon, Sep 11, 2023 at 3:09 AM Thomas Monjalon wrote:
>
> 09/09/2023 02:01, Ajit Khaparde:
> > On Fri, Sep 8, 2023 at 1:44 AM Jie Hai wrote:
> > >
> > > Hi, Thomas
> > > Thanks for your review.
> > >
> > > On 2023/9/4 15:45, Thomas Monjalon wrote:
> > > > 04/09/2023 09:10, Jie Hai:
> > > >> On 2
On Fri, 8 Sep 2023 09:04:29 +0200
Mattias Rönnblom wrote:
> > Also, right now the array is sized at 129 entries to allow for the
> > maximum number of lcores. When the maximum is increased to 512 or
> > 1024 the problem will get worse.
>
> Using TLS will penalize every thread in the process, n
On Wed, 6 Sep 2023 19:54:32 +0200
Morten Brørup wrote:
> >
> > - idx = rte_lcore_id();
> > + seed = __atomic_load_n(&rte_rand_seed, __ATOMIC_RELAXED);
> > + if (unlikely(seed != rand_state->seed)) {
>
> Please note that rte_rand_seed lives in a completely different cache
> line than RTE
Hi Anoob,
Thank you for that feedback - I was on extended leave so only just getting back
to it now.
See replies below.
Regards,
David
> -Original Message-
> From: Anoob Joseph
> Sent: Friday, August 11, 2023 12:09 PM
> To: Coyle, David ; dev@dpdk.org
> Cc: Ji, Kai ; O'Sullivan, Kevin
On Mon, 11 Sep 2023 11:00:46 +0200
Mattias Rönnblom wrote:
> > uint64_t rte_rand_r(struct rte_rand_state * const state);
> > void rte_srand_r(struct rte_rand_state * const state, uint64_t
> > seed); uint64_t rte_rand_max_r(struct rte_rand_state * const state,
> > uint64_t upper_bound); double rte
Mattias,
Yes that’s correct.
-Original Message-
From: Mattias Rönnblom
Sent: Friday, September 8, 2023 12:28 AM
To: Sevincer, Abdullah ; Stephen Hemminger
; Thomas Monjalon
Cc: dev@dpdk.org; Tyler Retzlaff
Subject: Re: quick thread in DLB2
On 2023-09-08 00:09, Sevincer, Abdullah wro
On 9/11/2023 2:06 PM, Andrew Rybchenko wrote:
> On 9/11/23 13:47, Viacheslav Galaktionov wrote:
>> Maximum packet length must be set to configure representor devices.
>>
>> Fixes: a62ec90522a6 ("net/sfc: add port representors infrastructure")
>> Cc: sta...@dpdk.org
>>
>> Signed-off-by: Viacheslav G
On 9/6/23 19:12, Thomas Monjalon wrote:
Calls to rte_ctrl_thread_create() are replaced with
rte_thread_create_internal_control().
Other pthread-related functions are replaced with the rte_thread API.
Only pthread_cancel() has no replacement.
The mlx5 vDPA control threads were real-time threads
w
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, potentially changing the physical
address mapping. Yet, virt
On 9/11/23 13:47, Viacheslav Galaktionov wrote:
Memory buffers passed by the application may contain private data,
which must not be sent to the NIC. Omit this private data by using
the actual data address instead of the default one.
Fixes: 3037e6cf3dde ("net/sfc: support regioned NIC DMA memory
On 9/11/23 13:47, Viacheslav Galaktionov wrote:
Maximum packet length must be set to configure representor devices.
Fixes: a62ec90522a6 ("net/sfc: add port representors infrastructure")
Cc: sta...@dpdk.org
Signed-off-by: Viacheslav Galaktionov
Reviewed-by: Andy Moreton
Acked-by: Andrew Rybc
Memory buffers passed by the application may contain private data,
which must not be sent to the NIC. Omit this private data by using
the actual data address instead of the default one.
Fixes: 3037e6cf3dde ("net/sfc: support regioned NIC DMA memory mapping type")
Cc: sta...@dpdk.org
Signed-off-by
Maximum packet length must be set to configure representor devices.
Fixes: a62ec90522a6 ("net/sfc: add port representors infrastructure")
Cc: sta...@dpdk.org
Signed-off-by: Viacheslav Galaktionov
Reviewed-by: Andy Moreton
---
drivers/net/sfc/sfc_repr.c | 1 +
1 file changed, 1 insertion(+)
di
07/09/2023 15:00, Anoob Joseph:
> Description for rte_cryptodev_get_sec_ctx is missing. Add the same.
>
> Signed-off-by: Anoob Joseph
Looks like a fix, please add "Fixes:" line.
09/09/2023 02:01, Ajit Khaparde:
> On Fri, Sep 8, 2023 at 1:44 AM Jie Hai wrote:
> >
> > Hi, Thomas
> > Thanks for your review.
> >
> > On 2023/9/4 15:45, Thomas Monjalon wrote:
> > > 04/09/2023 09:10, Jie Hai:
> > >> On 2023/8/30 19:46, Thomas Monjalon wrote:
> > >>> 26/08/2023 09:46, Jie Hai:
>
Update both high and low section of the ESN.
Fixes: 6bc987ecb860 ("net/iavf: support IPsec inline crypto")
Cc: sta...@dpdk.org
Signed-off-by: Radu Nicolau
---
drivers/net/iavf/iavf_ipsec_crypto.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/net/iavf/iavf_ipsec_crypto.c
b/drivers
On 2023-09-09 13:23, Morten Brørup wrote:
From: Mattias Rönnblom [mailto:hof...@lysator.liu.se]
Sent: Saturday, 9 September 2023 08.45
On 2023-09-09 02:13, Konstantin Ananyev wrote:
06/09/2023 21:02, Mattias Rönnblom пишет:
On 2023-09-06 19:20, Stephen Hemminger wrote:
Move the random number
Mapping rte_flow_attr->group to a specific hardware stage.
Group 0 -> switch filter
Group 1 -> acl filter
Group 2 -> fdir filter
For RSS, it will only be selected if there is a RTE_FLOW_ACTION_RSS
action target no queue group and the group ID is ignored.
Since each flow parser will be selected b
This marks the initial phase of refactoring the ice rte_flow
implementation.
The combination of switch and fdir rules within the same syntax has led
to inconvenient user experiences. Naturally, the switch filter and fdir
filter represent distinct pipeline stages with differing hardware
capabilitie
1. remove pipeline mode
2. apply group to hardware pipeline stage static mapping
v2:
- fix segment fault when an unintialized engine has been selected.
Qi Zhang (2):
net/ice: remove pipeline mode
net/ice: map group to pipeline stage
doc/guides/nics/ice.rst | 19 ---
drivers/net
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, potentially changing the physical
address mapping. Yet, virt
Allocation would fail with ASAN enabled if the size and alignment was
equal to half of the page size, e.g.:
size_t pg_sz = 2 * (1 << 20);
rte_malloc(NULL, pg_sz / 2, pg_sz / 2);
In such case, try_expand_heap_primary() only allocated one page but it
is not enough to fit this allocation with such a
Add support for random item, usage example:
pattern random spec value 0x1 mask value 0x3 / eth / end
Flow rule with above pattern matching 25% of the traffic, it hits only
when random value suffix is "01" and miss the others ("00", "10", "11").
Signed-off-by: Michael Baum
---
app/test-
Add support for a new item type "RTE_FLOW_ITEM_TYPE_RANDOM".
This item enables to match on some random value as a part of flow rule.
Signed-off-by: Michael Baum
---
doc/guides/nics/features/default.ini | 1 +
doc/guides/prog_guide/rte_flow.rst | 11 +
doc/guides/rel_notes/release_
Add support for matching random value using new "rte_flow_item_random"
structure.
This random value is not based on the packet data/headers.
Application shouldn't assume that this value is kept during the life
time of the packet.
v2:
- Rabase.
- Fix copy-paste mistake in release notes.
Michael
50 matches
Mail list logo
