> -Original Message-
> From: Wodkowski, PawelX
> Sent: Monday, January 19, 2015 5:14 PM
> To: Ouyang, Changchun; dev at dpdk.org
> Cc: Thomas Monjalon; Vlad Zolotarov
> Subject: RE: [dpdk-dev] [PATCH v6 3/6] ixgbe: Get VF queue number
>
>
>
> > -Original Message-
> > From: dev
> -Original Message-
> From: Wodkowski, PawelX
> Sent: Monday, January 19, 2015 6:31 PM
> To: Ouyang, Changchun; dev at dpdk.org
> Cc: Thomas Monjalon; Vlad Zolotarov
> Subject: RE: [dpdk-dev] [PATCH v6 4/6] ether: Check VMDq RSS mode
>
> > -Original Message-
> > From: dev [mailt
> -Original Message-
> From: Olivier MATZ [mailto:olivier.matz at 6wind.com]
> Sent: Monday, January 19, 2015 2:39 PM
> To: Ananyev, Konstantin; Liu, Jijiang
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v3 0/3] enhance TX checksum command and csum
> forwarding engine
>
> Hi Ko
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Pawel Wodkowski
> Sent: Monday, January 19, 2015 9:02 PM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH v2 2/4] ethdev: prevent changing of
> nb_q_per_pool in rte_eth_dev_check_mq_mode()
>
> If SRIOV is us
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Pawel Wodkowski
> Sent: Monday, January 19, 2015 9:03 PM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH v2 3/4] pmd: add support for DCB in SRIOV
> mode for ixgbe driver.
>
> Add support for DCB in SRIOV m
Hi Michael,
On 2015/01/19 23:24, Qiu, Michael wrote:
> Hi, Tetsuya
>
> You see lots of places have below three lines:
>
> if (rte_eth_dev_validate_port(port_id) == DEV_INVALID) {
> PMD_DEBUG_TRACE("Invalid port_id=%d\n", port_id);
> return -EINVAL;
> }
>
> T
> -Original Message-
> From: Olivier MATZ [mailto:olivier.matz at 6wind.com]
> Sent: Tuesday, January 20, 2015 1:27 AM
> To: Neil Horman; Zhang, Helin
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [RFC 01/17] mbuf: add definitions of unified packet
> types
>
> Hi,
>
> On 01/19/2015 05
'null PMD' is a driver of the virtual device particulary designed to measure
performance of DPDK PMDs. When an application call rx, null PMD just allocates
mbufs and returns those. Also tx, the PMD just frees mbufs.
The PMD has following options.
- size: specify packe size allocated by RX. Default
This patch adds port hotplug support to null PMD.
v4:
- Fix commit title.
Signed-off-by: Tetsuya Mukawa
---
lib/librte_pmd_null/rte_eth_null.c | 32
1 file changed, 32 insertions(+)
diff --git a/lib/librte_pmd_null/rte_eth_null.c
b/lib/librte_pmd_null/rte_eth_
> -Original Message-
> From: Neil Horman [mailto:nhorman at tuxdriver.com]
> Sent: Monday, January 19, 2015 9:02 PM
> To: Wang, Zhihong
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH 0/4] DPDK memcpy optimization
>
> On Mon, Jan 19, 2015 at 09:53:30AM +0800, zhihong.wang at intel
> -Original Message-
> From: Ananyev, Konstantin
> Sent: Tuesday, January 20, 2015 12:20 AM
> To: Zhang, Helin; dev at dpdk.org
> Cc: Liang, Cunming; Liu, Jijiang
> Subject: RE: [RFC 01/17] mbuf: add definitions of unified packet types
>
>
>
> > -Original Message-
> > From: Zha
Hi,
2015-01-07 16:39, Reshma Pattan:
> 1)New library to provide reordering of out of ordered
> mbufs based on sequence number of mbuf. Library uses reorder
> buffer structure
> which in tern uses two circular buffers called ready and order
> buffers.
>
Hi Helin,
2014-12-02 10:19, Helin Zhang:
> Hash filter control has been implemented for i40e. It includes
> getting/setting,
> - global hash configurations (hash function type, and symmetric
> hash enable per flow type)
> - symmetric hash enable per port
>
> Signed-off-by: Helin Zhang
> ---
>
Thank you Neil for writing this document.
This is a really important change in DPDK.
It would be very good to have comments or acknowledgement from several
developpers. This policy would be enforced by having several Acked-by lines.
2015-01-16 10:33, Neil Horman:
> Adding a document describing ru
2015-01-19 14:02, Pawel Wodkowski:
> Add support for DCB in SRIOV mode. When no PFC is enabled this feature
> might be used as multiple queues for VF (up to 8 queues if VFs num is
> less or equal 16 or 4 if FVs num is less or equal 32).
>
> The PF must initializes RX in ETH_MQ_RX_VMDQ_DCB and TX i
> > v2 changes:
> > change the return value if adding an existing filter from the filter's
> > index to
> > negative value.
> >
> > The patch set uses new filter_ctrl API to replace old ethertype filter APIs.
> > It uses new functions and structure to replace old ones in igb/ixgbe
> > driver,
> -Original Message-
> From: Ouyang, Changchun
> Sent: Tuesday, January 20, 2015 2:33 AM
> To: Wodkowski, PawelX; dev at dpdk.org
> Cc: Ouyang, Changchun
> Subject: RE: [dpdk-dev] [PATCH v2 2/4] ethdev: prevent changing of
> nb_q_per_pool in rte_eth_dev_check_mq_mode()
>
>
>
> > -Ori
rte_power_freq_min function did not include "extern" keyword,
causing linking errors.
Signed-off-by: Pablo de Lara
Reported-by: Ildar Mustafin
---
lib/librte_power/rte_power.h |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/lib/librte_power/rte_power.h b/lib/librte_pow
On 1/19/2015 6:42 PM, Tetsuya Mukawa wrote:
> The patch adds functions for unmapping igb_uio resources. The patch is only
> for Linux and igb_uio environment. VFIO and BSD are not supported.
>
> v4:
> - Add paramerter checking.
> - Add header file to determine if hotplug can be enabled.
>
> Signed-
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Ouyang Changchun
> Sent: Monday, January 12, 2015 6:59 AM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH v6 5/6] ixgbe: Config VF RSS
>
> It needs config RSS and IXGBE_MRQC and IXGBE_VFPSRTYPE to enable VF R
Hi Helin,
On 01/20/2015 03:28 AM, Zhang, Helin wrote:
>> Another question I've asked several times[1][2] : what does having
>> RTE_PTYPE_TUNNEL_IP mean? What fields are checked by the hardware (or
>> the driver) and what fields should be checked by the application?
>> Are you sure that all the dri
Signed-off-by: Danny Zhou
---
lib/librte_ether/rte_ethdev.c | 45 ++
lib/librte_ether/rte_ethdev.h | 57 +++
2 files changed, 102 insertions(+)
diff --git a/lib/librte_ether/rte_ethdev.c b/lib/librte_ether/rte_ethdev.c
inde
Signed-off-by: Danny Zhou
---
lib/librte_eal/common/include/rte_eal.h| 9 ++
lib/librte_eal/linuxapp/eal/eal_interrupts.c | 120 -
lib/librte_eal/linuxapp/eal/eal_pci_vfio.c | 11 +-
.../linuxapp/eal/include/exec-env/rte_interrupts.h | 3 +
4 fi
DPDK interrupt notification/handling mechanism is based on UIO with
below limitation:
1) It is designed to handle LSC interrupt only with inefficient
suspended pthread wakeup procedure (e.g. UIO wakes up LSC interrupt
handling thread which then wakes up DPDK polling thread). In this way,
it introdu
Signed-off-by: Danny Zhou
---
lib/librte_pmd_ixgbe/ixgbe_ethdev.c | 203
1 file changed, 203 insertions(+)
diff --git a/lib/librte_pmd_ixgbe/ixgbe_ethdev.c
b/lib/librte_pmd_ixgbe/ixgbe_ethdev.c
index 3fc3738..1d694c5 100644
--- a/lib/librte_pmd_ixgbe/ixgbe_
Signed-off-by: Danny Zhou
---
examples/l3fwd-power/main.c | 127 ++--
1 file changed, 98 insertions(+), 29 deletions(-)
diff --git a/examples/l3fwd-power/main.c b/examples/l3fwd-power/main.c
index f6b55b9..71f1d90 100644
--- a/examples/l3fwd-power/main.c
This patch enables rx queue interrupt for ixgbevf with below changes:
1) Configure ixgbevf rx queue interrupts
2) Initialize ixgbevf devices in L3fwd-power appropriately
3) Fix VFIO interrupt vector settings
Signed-off-by: Yong Liu
Signed-off-by: Danny Zhou
---
examples/l3fwd-power/main.c
Hi Thomas,
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Monday, January 19, 2015 5:18 PM
> To: Ananyev, Konstantin
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v2 17/17] libte_acl: remove unused macros.
>
> 2015-01-12 19:16, Konstant
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Monday, January 19, 2015 5:16 PM
> To: Ananyev, Konstantin
> Cc: dev at dpdk.org; Neil Horman
> Subject: Re: [dpdk-dev] [PATCH v2 00/17] ACL: New AVX2 classify method and
> several other enhancemen
2015-01-12 15:16, Jingjing Wu:
> Following commands of ethertype filter are removed:
> - add_ethertype_filter (port_id) ethertype (eth_value)
> - remove_ethertype_filter (port_id) index (idx)
> - get_ethertype_filter (port_id) index (idx)
> New command is added for ethertype filter by using f
Hi again,
I did some further testing and it seems like this issue is linked to
jumbo frames. I think a similar issue has already been reported by
Prashant Upadhyaya with the subject 'Packet Rx issue with DPDK1.8'.
In our application we use the following rxmode port configuration:
.mq_mode= ET
On Tue, Jan 20, 2015 at 08:14:50AM +0100, Thomas Monjalon wrote:
> Thank you Neil for writing this document.
> This is a really important change in DPDK.
> It would be very good to have comments or acknowledgement from several
> developpers. This policy would be enforced by having several Acked-by
> On Jan 20, 2015, at 4:09 AM, Ananyev, Konstantin intel.com> wrote:
>
> Hi Thomas,
>
>> -Original Message-
>> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
>> Sent: Monday, January 19, 2015 5:18 PM
>> To: Ananyev, Konstantin
>> Cc: dev at dpdk.org
>> Subject: Re: [dpdk-de
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Monday, January 19, 2015 5:23 PM
> To: Ananyev, Konstantin
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v2 11/17] librte_acl: add AVX2 as new
> rte_acl_classify() method
>
> 2015-01-12 1
> From: Jim Thompson [mailto:jim at netgate.com]
> Sent: Tuesday, January 20, 2015 10:48 AM
> To: Ananyev, Konstantin
> Cc: Thomas Monjalon; dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v2 17/17] libte_acl: remove unused macros.
>
>
> On Jan 20, 2015, at 4:09 AM, Ananyev, Konstantin intel.c
Hello Sujith,
Any news on the documentation and the performance numbers you said you
would send ?
Thanks.
--
David Marchand
On Thu, Nov 27, 2014 at 4:31 PM, Thomas Monjalon
wrote:
> 2014-11-27 04:27, Sujith Sankar:
> > Thanks Thomas, David and Neil !
> >
> > I shall work on finishing the doc
2015-01-20 10:09, Ananyev, Konstantin:
> Hi Thomas,
>
> > -Original Message-
> > From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> > Sent: Monday, January 19, 2015 5:18 PM
> > To: Ananyev, Konstantin
> > Cc: dev at dpdk.org
> > Subject: Re: [dpdk-dev] [PATCH v2 17/17] libte_acl
Hi,
On 01/20/2015 02:12 AM, Ananyev, Konstantin wrote:
I think a good definition would
be:
Packet is IPv4. This flag must be set when using any offload
feature (TSO, L3 or L4 checksum) to tell the NIC that the packet
is an IPv4 packet.
Th
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Thomas Monjalon
> Sent: Tuesday, January 20, 2015 7:15 AM
> To: Neil Horman
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v5 4/4] docs: Add ABI documentation
>
> Thank you Neil for writing this document
Hi,
I am writing a DPI application on DPDK that needs to communicate with another
non-DPDK Linux application on the same host for configuration/stats exchange.
This needs to be a low-latency communication.
Can the RTE ring be used for communication between DPDK? & non-DPDK
applications? If no, w
2015-01-20 13:37, Iremonger, Bernard:
> Should there be a reference to this document in the programmers guide?
Maybe. You mean that an application developper must be aware of the deprecation
policy? So probably yes.
And I'd add that the release notes should reference the deprecations.
--
Thomas
2015-01-16 10:33, Neil Horman:
> --- /dev/null
> +++ b/doc/abi.txt
> @@ -0,0 +1,36 @@
> +ABI policy:
> + ABI versions are set at the time of major release labeling, and ABI may
> +change multiple times between the last labeling and the HEAD label of the git
> +tree without warning
> +
> + A
On Tue, Jan 20, 2015 at 01:37:35PM +, Iremonger, Bernard wrote:
> > -Original Message-
> > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Thomas Monjalon
> > Sent: Tuesday, January 20, 2015 7:15 AM
> > To: Neil Horman
> > Cc: dev at dpdk.org
> > Subject: Re: [dpdk-dev] [PATCH v
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Neil Horman
> Sent: Tuesday, January 20, 2015 2:24 PM
> To: Iremonger, Bernard
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v5 4/4] docs: Add ABI documentation
>
> On Tue, Jan 20, 2015 at 01:37:35PM
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Neil Horman
> Sent: Tuesday, January 20, 2015 2:24 PM
> To: Iremonger, Bernard
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v5 4/4] docs: Add ABI documentation
>
> On Tue, Jan 20, 2015 at 01:37:35PM +0
On Tue, Jan 20, 2015 at 03:00:01PM +0100, Thomas Monjalon wrote:
> 2015-01-16 10:33, Neil Horman:
> > --- /dev/null
> > +++ b/doc/abi.txt
> > @@ -0,0 +1,36 @@
> > +ABI policy:
> > + ABI versions are set at the time of major release labeling, and ABI may
> > +change multiple times between the last
On Tue, Jan 20, 2015 at 02:29:54PM +, Butler, Siobhan A wrote:
>
>
> > -Original Message-
> > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Neil Horman
> > Sent: Tuesday, January 20, 2015 2:24 PM
> > To: Iremonger, Bernard
> > Cc: dev at dpdk.org
> > Subject: Re: [dpdk-dev]
> -Original Message-
> From: Neil Horman [mailto:nhorman at tuxdriver.com]
> Sent: Tuesday, January 20, 2015 2:42 PM
> To: Butler, Siobhan A
> Cc: Iremonger, Bernard; dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v5 4/4] docs: Add ABI documentation
>
> On Tue, Jan 20, 2015 at 02:29:54
2015-01-20 09:37, Neil Horman:
> On Tue, Jan 20, 2015 at 03:00:01PM +0100, Thomas Monjalon wrote:
> > 2015-01-16 10:33, Neil Horman:
> > > --- /dev/null
> > > +++ b/doc/abi.txt
> > > @@ -0,0 +1,36 @@
> > > +ABI policy:
> > > + ABI versions are set at the time of major release labeling, and ABI may
On Tue, Jan 20, 2015 at 03:01:44AM +, Wang, Zhihong wrote:
>
>
> > -Original Message-
> > From: Neil Horman [mailto:nhorman at tuxdriver.com]
> > Sent: Monday, January 19, 2015 9:02 PM
> > To: Wang, Zhihong
> > Cc: dev at dpdk.org
> > Subject: Re: [dpdk-dev] [PATCH 0/4] DPDK memcpy op
2015-01-20 13:39, Olivier MATZ:
> On 01/20/2015 02:12 AM, Ananyev, Konstantin wrote:
> >> So I will fix that in my coming patch series. Just for information,
> >> I'm pretty sure that having PKT_TX_IPV4 and PKT_TX_IP_CSUM as not
> >> exclusive flag would not require any change anywhere in the PMDs
We had our last community call in December, and then took a break over the
holiday period. I think we should reinstate these, so I've scheduled the next
one for Monday 2nd February. Since our last call was at a time convenient for
Asia, this one is at a time that's more convenient for people bas
On Tue, Jan 20, 2015 at 02:50:43PM +, Butler, Siobhan A wrote:
>
>
> > -Original Message-
> > From: Neil Horman [mailto:nhorman at tuxdriver.com]
> > Sent: Tuesday, January 20, 2015 2:42 PM
> > To: Butler, Siobhan A
> > Cc: Iremonger, Bernard; dev at dpdk.org
> > Subject: Re: [dpdk-de
On Tue, Jan 20, 2015 at 04:06:07PM +0100, Thomas Monjalon wrote:
> 2015-01-20 09:37, Neil Horman:
> > On Tue, Jan 20, 2015 at 03:00:01PM +0100, Thomas Monjalon wrote:
> > > 2015-01-16 10:33, Neil Horman:
> > > > --- /dev/null
> > > > +++ b/doc/abi.txt
> > > > @@ -0,0 +1,36 @@
> > > > +ABI policy:
>
On Tue, Jan 20, 2015 at 10:11:18AM -0500, Neil Horman wrote:
> On Tue, Jan 20, 2015 at 03:01:44AM +, Wang, Zhihong wrote:
> >
> >
> > > -Original Message-
> > > From: Neil Horman [mailto:nhorman at tuxdriver.com]
> > > Sent: Monday, January 19, 2015 9:02 PM
> > > To: Wang, Zhihong
> >
On Tue, 20 Jan 2015 16:18:01 +0100
Thomas Monjalon wrote:
> 2015-01-20 13:39, Olivier MATZ:
> > On 01/20/2015 02:12 AM, Ananyev, Konstantin wrote:
> > >> So I will fix that in my coming patch series. Just for information,
> > >> I'm pretty sure that having PKT_TX_IPV4 and PKT_TX_IP_CSUM as not
>
On Tue, 20 Jan 2015 15:21:40 +
"O'driscoll, Tim" wrote:
> We had our last community call in December, and then took a break over the
> holiday period. I think we should reinstate these, so I've scheduled the next
> one for Monday 2nd February. Since our last call was at a time convenient fo
On Mon, 19 Jan 2015 09:53:34 +0800
zhihong.wang at intel.com wrote:
> Main code changes:
>
> 1. Differentiate architectural features based on CPU flags
>
> a. Implement separated move functions for SSE/AVX/AVX2 to make full
> utilization of cache bandwidth
>
> b. Implement separated co
Hi Olivier,
> -Original Message-
> From: Olivier MATZ [mailto:olivier.matz at 6wind.com]
> Sent: Tuesday, January 20, 2015 12:39 PM
> To: Ananyev, Konstantin; Liu, Jijiang
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v3 0/3] enhance TX checksum command and csum
> forwarding engi
Hi Konstantin,
On 01/20/2015 06:23 PM, Ananyev, Konstantin wrote:
>> Sure, it does not make a big difference in terms of code. But
>> in terms of API, the naming of the flag is coherent to what it is
>> used for. And it's easier to find a simple definition, like:
>>
>>* Packet is IPv4. This fl
Signed-off-by: Konstantin Ananyev
---
lib/librte_acl/rte_acl_osdep_alone.h | 12
1 file changed, 12 insertions(+)
diff --git a/lib/librte_acl/rte_acl_osdep_alone.h
b/lib/librte_acl/rte_acl_osdep_alone.h
index a84b6f9..2a99860 100644
--- a/lib/librte_acl/rte_acl_osdep_alone.h
+++ b/
v3 changes:
Applied review comments from Thomas:
- fix spelling errors reported by codespell.
- split last patch into two:
first to remove unused macros,
second to add some comments about ACL internal layout.
v2 changes:
- When build with the compilers that don't support AVX2 instructions,
Make classify_scalar to behave in the same way as it's vector counterpart:
move match check out of the inner loop, etc.
That makes scalar and vector code look more identical.
Plus it improves scalar code performance.
Signed-off-by: Konstantin Ananyev
---
lib/librte_acl/acl_run_scalar.c | 23
Make data_indexes long enough to survive idle transitions.
That allows to simplify match processing code.
Also fix incorrect size calculations for data indexes.
Signed-off-by: Konstantin Ananyev
---
lib/librte_acl/acl_bld.c | 5 +++--
lib/librte_acl/acl_run.h | 4
2 files changed, 3 inserti
Current rule-wildness based heuristsics can cause unnecessary splits of
the ruleset.
That might have negative performance effect:
more tries to traverse, bigger RT tables.
After removing it, on some test-cases with big rulesets (~10K)
observed ~50% speedup.
No difference for smaller rulesets.
Sign
Signed-off-by: Konstantin Ananyev
---
lib/librte_acl/acl.h | 1 -
lib/librte_acl/acl_run.h | 1 -
2 files changed, 2 deletions(-)
diff --git a/lib/librte_acl/acl.h b/lib/librte_acl/acl.h
index 61b849a..217bab3 100644
--- a/lib/librte_acl/acl.h
+++ b/lib/librte_acl/acl.h
@@ -62,7 +62,6 @@ str
Introduced division of whole 256 child transition enties
into 4 sub-groups (64 kids per group).
So 2 groups within the same node with identical children,
can use one set of transition entries.
That allows to compact some DFA nodes and get space savings in the RT table,
without any negative performa
Right now we allocate indexes for all types of nodes, except MATCH,
at 'gen final RT table' stage.
For MATCH type nodes we are doing it at building temporary tree stage.
This is totally unnecessary and makes code more complex and error prone.
Rework the code and make MATCH indexes being allocated a
Move common check for input parameters up into rte_acl_classify_alg().
Signed-off-by: Konstantin Ananyev
---
lib/librte_acl/acl_run_scalar.c | 4
lib/librte_acl/acl_run_sse.c| 4
lib/librte_acl/rte_acl.c| 19 ---
3 files changed, 12 insertions(+), 15 delet
Vector code reorganisation/deduplication:
To avoid maintaining two nearly identical implementations of calc_addr()
(one for SSE, another for AVX2), replace it with a new macro that suits
both SSE and AVX2 code-paths.
Also remove no needed any more MM_* macros.
Signed-off-by: Konstantin Ananyev
-
Signed-off-by: Konstantin Ananyev
---
lib/librte_acl/acl.h | 38 ++
1 file changed, 38 insertions(+)
diff --git a/lib/librte_acl/acl.h b/lib/librte_acl/acl.h
index 217bab3..4dadab5 100644
--- a/lib/librte_acl/acl.h
+++ b/lib/librte_acl/acl.h
@@ -68,6 +68,44 @@
In test-acl replace command-line option "--scalar" with new one:
"--alg=scalar|sse|avx2".
Allows user manually select preferred classify() method.
Signed-off-by: Konstantin Ananyev
---
app/test-acl/main.c | 93 ++---
1 file changed, 75 insertions(+
If at build phase we don't make any trie splitting,
then temporary build structures and resulting RT structure might be
much bigger than current.
>From other side - having just one trie instead of multiple can speedup
search quite significantly.
>From my measurements on rule-sets with ~10K rules:
R
v2 changes:
When build with the compilers that don't support AVX2 instructions,
make rte_acl_classify_avx2() do nothing and return an error.
Remove unneeded 'ifdef __AVX2__' in acl_run_avx2.*.
Introduce new classify() method that uses AVX2 instructions.
>From my measurements:
On HSW boards when pr
Reorganise SSE code-path a bit by moving lo/hi dwords shuffle
out from calc_addr().
That allows to make calc_addr() for SSE and AVX2 practically identical
and opens opportunity for further code deduplication.
Signed-off-by: Konstantin Ananyev
---
lib/librte_acl/acl_run_sse.h | 38 +++
Make sure that test_acl would not ignore error conditions.
Run classify() with all possible values.
Signed-off-by: Konstantin Ananyev
---
app/test/test_acl.c | 8 ++--
1 file changed, 6 insertions(+), 2 deletions(-)
diff --git a/app/test/test_acl.c b/app/test/test_acl.c
index 356d620..7119a
Signed-off-by: Konstantin Ananyev
---
lib/librte_acl/acl_bld.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/lib/librte_acl/acl_bld.c b/lib/librte_acl/acl_bld.c
index 8bf4a54..22f7934 100644
--- a/lib/librte_acl/acl_bld.c
+++ b/lib/librte_acl/acl_bld.c
@@ -1907,7 +1907,7 @@
Previous improvements made scalar method the fastest one
for tiny bunch of packets (< 4).
That allows us to remove specific vector code-path for small number of packets
(search_sse_2)
and always use scalar method for such cases.
Signed-off-by: Konstantin Ananyev
---
lib/librte_acl/acl_run_avx2.c
New data type to manipulate 256 bit AVX values.
Rename field in the rte_xmm to keep common naming across SSE/AVX fields.
Signed-off-by: Konstantin Ananyev
---
examples/l3fwd/main.c | 2 +-
lib/librte_acl/acl_run_sse.c| 88 -
When using core list argument to define which core to enable (ie -l) the
core_num field of the rte configuration is not updated the same way as using
coremask. This causes rte_lcore_num() to yield different value from the one
using coremask.
Signed-off-by: Remi Pommarel
---
lib/librte_eal/common
Ethernet device's data should contain the virtual device name for pcap port.
This name is correctly set by rte_eth_dev_allocate() at initialization time,
but it is directly lost.
Signed-off-by: Remi Pommarel
---
lib/librte_pmd_pcap/rte_eth_pcap.c | 1 +
1 file changed, 1 insertion(+)
diff --git
On Tue, Jan 20, 2015 at 09:15:38AM -0800, Stephen Hemminger wrote:
> On Mon, 19 Jan 2015 09:53:34 +0800
> zhihong.wang at intel.com wrote:
>
> > Main code changes:
> >
> > 1. Differentiate architectural features based on CPU flags
> >
> > a. Implement separated move functions for SSE/AVX/AVX
the vhost library relies on libfuse, and thats included when we do a normal
shared object build, but when we specify combined libs, its gets left out. Add
it back in
Signed-off-by: Neil Horman
---
mk/rte.app.mk | 4
1 file changed, 4 insertions(+)
diff --git a/mk/rte.app.mk b/mk/rte.app.m
Add initial pass header files to support symbol versioning.
Signed-off-by: Neil Horman
CC: Thomas Monjalon
CC: "Richardson, Bruce"
CC: "Gonzalez Monroy, Sergio"
---
Change Notes:
V2)
Moved ifeq to _INSTALL target
V3)
Undo V2 changes and make librte_compat use the rte.install.
Add linker version script files to each DPDK library to put a stake in the
ground from which we can start cleaning up API's
Signed-off-by: Neil Horman
CC: Thomas Monjalon
CC: "Richardson, Bruce"
---
Change Notes:
v2)
* Updated export map to not require full path
---
lib/librte_acl/Ma
To differentiate libraries that break ABI, we add a library version number
suffix to the library, which must be incremented when a given libraries ABI is
broken. This patch enforces that addition, sets the initial abi soname
extension to 1 for each library and creates a symlink to the base SONAME
Adding a document describing rudimentary ABI policy and adding notice space for
any deprecation announcements
Signed-off-by: Neil Horman
CC: Thomas Monjalon
CC: "Richardson, Bruce"
---
Change notes:
v5) Updated documentation to add notes from Thomas M.
v6) Moved abi.txt to guides/rel_notes/a
On Tuesday, January 20, 2015 03:49:58 PM Neil Horman wrote:
> the vhost library relies on libfuse, and thats included when we do a normal
> shared object build, but when we specify combined libs, its gets left out.
> Add
> it back in
>
> Signed-off-by: Neil Horman
Fixes the build issue here.
R
From: Stephen Hemminger
The rte_eth_stats_get is the only API that should call the device
statistics function directly, and it already does a memset of the
resulting structure. Therefore doing memset() in the driver is
redundant and should be removed.
Signed-off-by: Stephen Hemminger
---
lib/l
89 matches
Mail list logo
