18.11.2014 23:46, Neil Horman ?:
> On Tue, Nov 18, 2014 at 11:13:17PM +0600, Yerden Zhumabekov wrote:
>> 18.11.2014 22:00, Neil Horman ?:
>>>
>>> You need to edit the makefile so that the compiler gets passed the option
>>> -msse42. That way it will know to emit sse42 instructions. It wil
18.11.2014 22:00, Neil Horman ?:
> On Tue, Nov 18, 2014 at 09:06:35PM +0600, Yerden Zhumabekov wrote:
>> 18.11.2014 20:41, Neil Horman ?:
>>> On Tue, Nov 18, 2014 at 08:03:40PM +0600, Yerden Zhumabekov wrote:
/**
* Use single crc32 instruction to perform a hash on a 4 byte val
18.11.2014 20:41, Neil Horman ?:
> On Tue, Nov 18, 2014 at 08:03:40PM +0600, Yerden Zhumabekov wrote:
>> Initially, SSE4.2 support is detected via CPUID instruction.
>>
>> Added rte_hash_crc_set_alg() function to detect and set CRC32
>> implementation if necessary. SSE4.2 is allowed by default
Since rte_hash_crc() can now be run regardless of SSE4.2 support,
we can safely remove compile checks for RTE_MACHINE_CPUFLAG_SSE4_2
in test utilities.
Signed-off-by: Yerden Zhumabekov
---
app/test/test_hash.c |7 ---
app/test/test_hash_perf.c | 11 ---
2 files changed, 18
Calculating hash for data of variable length is more efficient
when that data is sliced into 8-byte pieces. The rest part of data
is hashed using CRC32 functions with either 8 and 4 byte operands.
Signed-off-by: Yerden Zhumabekov
---
lib/librte_hash/rte_hash_crc.h | 33
Initially, SSE4.2 support is detected via CPUID instruction.
Added rte_hash_crc_set_alg() function to detect and set CRC32
implementation if necessary. SSE4.2 is allowed by default. If it's
not available, fall back to sw implementation.
Best available algorithm is detected upon application startu
SSE4.2 provides _mm_crc32_u64 intrinsic with 8-byte operand.
Signed-off-by: Yerden Zhumabekov
---
lib/librte_hash/rte_hash_crc.h | 16
1 file changed, 16 insertions(+)
diff --git a/lib/librte_hash/rte_hash_crc.h b/lib/librte_hash/rte_hash_crc.h
index 4d7532a..15f687a 100644
-
Add lookup tables for CRC32 algorithm, crc32c_1word() and
crc32c_2words() functions returning hash of 32-bit and
64-bit operand.
Signed-off-by: Yerden Zhumabekov
---
lib/librte_hash/rte_hash_crc.h | 316
1 file changed, 316 insertions(+)
diff --git a/li
This is a rework of my previous patches improving performance of rte_hash_crc.
In addition, this revision brings a fallback mechanism to ensure that CRC32
hash is calculated regardless of hardware support from CPU (i.e. SSE4.2
intrinsics). Performance of software CRC32 implementation is also imp
18.11.2014 19:33, Neil Horman ?:
> On Tue, Nov 18, 2014 at 10:56:24AM +0600, Yerden Zhumabekov wrote:
>> Sorry, maybe I made a mistake here.
>>
>> Accoring to lib/librte_eal/common/eal_common_cpuflags.c code, it seemed
>> to me that constructor attribute is not supported by intel compiler. So
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Michal Jastrzebski
> Sent: Tuesday, November 18, 2014 4:31 PM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH v4 2/2] testpmd: add mode 4 support
>
> From: Pawel Wodkowski
>
>
> Signed-off-by: Pawel Wodk
Hi Folks,
I may not be able to attend todays conference call because it will be
midnight in my country.
So I just collect up current status of this function.
- I've received a email that let me know some bugs and suggestions in my
RFC. Now I am fixing it.
- Anyway, to use this function, we need t
On Tue, Nov 18, 2014 at 12:46:19PM -0500, Neil Horman wrote:
> On Tue, Nov 18, 2014 at 11:13:17PM +0600, Yerden Zhumabekov wrote:
> >
> > 18.11.2014 22:00, Neil Horman ?:
> > > On Tue, Nov 18, 2014 at 09:06:35PM +0600, Yerden Zhumabekov wrote:
> > >> 18.11.2014 20:41, Neil Horman ?:
> > >>
I have a general question about using CPUID to detect supported instruction set.
What if we are compiling the software with some old hardware which does not
support SSE4.2, but run it on new hardware which does support SSE4.2. Is there
still a static way to force the compiler to turn on the SSE4.
Signed-off-by: Pablo de Lara
---
doc/guides/testpmd_app_ug/testpmd_funcs.rst | 10 ++
1 files changed, 6 insertions(+), 4 deletions(-)
diff --git a/doc/guides/testpmd_app_ug/testpmd_funcs.rst
b/doc/guides/testpmd_app_ug/testpmd_funcs.rst
index 8030f3b..abb3f54 100644
--- a/doc/guides/
Moved commands in testpmd UG to set queue rates
to match order in testpmd command help.
Created new section "Filters" to match that
same section in testpmd UG, and moved all
commands related to it there.
Signed-off-by: Pablo de Lara
---
doc/guides/testpmd_app_ug/testpmd_funcs.rst | 1019 +++
tx_checksum set mask function now allows 4 extra bits in the mask
for TX checksum offload
Signed-off-by: Pablo de Lara
---
doc/guides/testpmd_app_ug/testpmd_funcs.rst | 10 +-
1 files changed, 9 insertions(+), 1 deletions(-)
diff --git a/doc/guides/testpmd_app_ug/testpmd_funcs.rst
b/
Added info in testpmd functions section for the following commands:
- tunnel_filter add
- tunnel_filter rm
- rx_vxlan_port add
- rx_vxlan_port rm
- port stop/start queue
- set port mac address filter (for VF)
Signed-off-by: Pablo de Lara
---
doc/guides/testpmd_app_ug/testpmd_funcs.rst | 47 ++
TestPMD app has been added new commands, which should be reflected
in the user guide. Also, this patchset fixes several typos and
changes the location of the description of some commands.
Changes in v2:
Title underline in added commands was too short.
Pablo de Lara (4):
doc: Added new commands
On Tue, Nov 18, 2014 at 05:52:27PM +, Bruce Richardson wrote:
> On Tue, Nov 18, 2014 at 12:46:19PM -0500, Neil Horman wrote:
> > On Tue, Nov 18, 2014 at 11:13:17PM +0600, Yerden Zhumabekov wrote:
> > >
> > > 18.11.2014 22:00, Neil Horman ?:
> > > > On Tue, Nov 18, 2014 at 09:06:35PM +0600,
From: Pawel Wodkowski
Signed-off-by: Pawel Wodkowski
---
app/test-pmd/cmdline.c | 28 +++--
app/test-pmd/csumonly.c |9
app/test-pmd/icmpecho.c | 17 ++-
app/test-pmd/iofwd.c|9
app/test-pmd/macfwd-retry.c |
From: Pawel Wodkowski
Signed-off-by: Pawel Wodkowski
---
lib/librte_ether/rte_ether.h |1 +
lib/librte_pmd_bond/Makefile |2 +
lib/librte_pmd_bond/rte_eth_bond.h|5 +
lib/librte_pmd_bond/rte_eth_bond_8023ad.c | 1218
From: Pawel Wodkowski
v2 changes:
New version handles race issues with setting/cancelin callbacks,
fixes promiscus mode setting in mode 4 and some other minor errors in mode 4
implementation.
v3 changes:
This is a rework of previous patchset. Basic functionality is the same but
contain followin
On Tue, Nov 18, 2014 at 04:04:26PM +, Bruce Richardson wrote:
> On Tue, Nov 18, 2014 at 11:00:05AM -0500, Neil Horman wrote:
> > On Tue, Nov 18, 2014 at 09:06:35PM +0600, Yerden Zhumabekov wrote:
> > >
> > > 18.11.2014 20:41, Neil Horman ?:
> > > > On Tue, Nov 18, 2014 at 08:03:40PM +0600,
On Tue, Nov 18, 2014 at 11:00:05AM -0500, Neil Horman wrote:
> On Tue, Nov 18, 2014 at 09:06:35PM +0600, Yerden Zhumabekov wrote:
> >
> > 18.11.2014 20:41, Neil Horman ?:
> > > On Tue, Nov 18, 2014 at 08:03:40PM +0600, Yerden Zhumabekov wrote:
> > >> Initially, SSE4.2 support is detected via C
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Ananyev, Konstantin
> Sent: Tuesday, November 18, 2014 3:26 PM
> To: Zhang, Helin; Liu, Jijiang; dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH 0/4] Translate packet types for i40e
>
>
>
> > -Original
Change the codes in testpmd due to introducing abstract packet type.
Signed-off-by: Jijiang Liu
---
app/test-pmd/csumonly.c | 12 ++--
app/test-pmd/rxonly.c | 20 +---
2 files changed, 15 insertions(+), 17 deletions(-)
diff --git a/app/test-pmd/csumonly.c b/app/tes
Translate i40e packet types to abstract packet types, and keep the usage of the
PKT_RX_IPV4_HDR and the PKT_RX_IPV4_HDR as before in i40e driver.
Signed-off-by: Jijiang Liu
---
lib/librte_pmd_i40e/i40e_rxtx.c | 604 +--
1 files changed, 332 insertions(+), 27
The PKT_RX_TUNNEL_IPV4_HDR and the PKT_RX_TUNNEL_IPV6_HDR are removed, they are
useless now.
Signed-off-by: Jijiang Liu
---
lib/librte_mbuf/rte_mbuf.h |2 --
1 files changed, 0 insertions(+), 2 deletions(-)
diff --git a/lib/librte_mbuf/rte_mbuf.h b/lib/librte_mbuf/rte_mbuf.h
index 678db0d.
This patch abstracts packet types of L2 packet, Non Tunneled IPv4/6, IP in IP,
IP in GRE, MAC in GRE and MAC in UDP, and add 4 MACROS to check packet IP
header.
Signed-off-by: Jijiang Liu
---
lib/librte_mbuf/rte_mbuf.h | 223
1 files changed, 223 i
The i40e NIC can recognize many packet types, including ordinary L2 packet
format and tunneling packet format such as IP in IP, IP in GRE, MAC in GRE and
MAC in UDP.
This patch set provides abstract definitions of packet types,
which can help user to use these packet types directly in their appl
> -Original Message-
> From: Richardson, Bruce
> Sent: Tuesday, November 18, 2014 1:08 PM
> To: Ananyev, Konstantin
> Cc: Liu, Jijiang; dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH 0/4] Translate packet types for i40e
>
> On Tue, Nov 18, 2014 at 11:33:24AM +, Ananyev, Konstantin
> -Original Message-
> From: Zhang, Helin
> Sent: Tuesday, November 18, 2014 2:12 PM
> To: Ananyev, Konstantin; Liu, Jijiang; dev at dpdk.org
> Subject: RE: [dpdk-dev] [PATCH 0/4] Translate packet types for i40e
>
>
>
> > -Original Message-
> > From: dev [mailto:dev-bounces at
Hi Thomas,
On 11/17/2014 11:05 PM, Thomas Monjalon wrote:
> 2014-11-14 18:03, Olivier Matz:
>> The tx mbuf flags are ordered from the highest value to the
>> the lowest. Move the PKT_TX_VXLAN_CKSUM at the right place.
>
> Please, could you reorder them from the lowest to the highest?
> It will be
> Bernard Iremonger (1):
> doc: programmers guide
Applied
Thanks a lot Bernard.
We have the complete documentation of DPDK 1.7!
Now it's time to update it for DPDK 1.8.
--
Thomas
2014-11-18 08:33, Neil Horman:
> On Tue, Nov 18, 2014 at 10:56:24AM +0600, Yerden Zhumabekov wrote:
> > Sorry, maybe I made a mistake here.
> >
> > Accoring to lib/librte_eal/common/eal_common_cpuflags.c code, it seemed
> > to me that constructor attribute is not supported by intel compiler. So
>
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Ananyev, Konstantin
> Sent: Tuesday, November 18, 2014 7:33 PM
> To: Liu, Jijiang; dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH 0/4] Translate packet types for i40e
>
> Hi Frank,
>
> > -Original Messa
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Jingjing Wu
> Sent: Thursday, November 13, 2014 8:50 PM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH v3 0/2] support ethertype filter on fortville
>
> From: "jingjing.wu"
>
> The patch set supports eth
On Tue, Nov 18, 2014 at 11:33:24AM +, Ananyev, Konstantin wrote:
> Hi Frank,
>
> > -Original Message-
> > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Jijiang Liu
> > Sent: Tuesday, November 18, 2014 7:37 AM
> > To: dev at dpdk.org
> > Subject: [dpdk-dev] [PATCH 0/4] Transla
This is just a reminder of our call later today, which is at 17:00 GMT. The
time in a variety of timezones is included below.
We're going to use GoToMeeting this time. If you follow the meeting link
(https://global.gotomeeting.com/join/843960205), the GoToMeeting web viewer
will start. You then
On Tue, Nov 18, 2014 at 11:13:17PM +0600, Yerden Zhumabekov wrote:
>
> 18.11.2014 22:00, Neil Horman ?:
> > On Tue, Nov 18, 2014 at 09:06:35PM +0600, Yerden Zhumabekov wrote:
> >> 18.11.2014 20:41, Neil Horman ?:
> >>> On Tue, Nov 18, 2014 at 08:03:40PM +0600, Yerden Zhumabekov wrote:
> >>
On Tue, Nov 18, 2014 at 04:04:26PM +, Bruce Richardson wrote:
> On Tue, Nov 18, 2014 at 11:00:05AM -0500, Neil Horman wrote:
> > On Tue, Nov 18, 2014 at 09:06:35PM +0600, Yerden Zhumabekov wrote:
> > >
> > > 18.11.2014 20:41, Neil Horman ?:
> > > > On Tue, Nov 18, 2014 at 08:03:40PM +0600,
Hi Frank,
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Jijiang Liu
> Sent: Tuesday, November 18, 2014 7:37 AM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH 0/4] Translate packet types for i40e
>
> The i40e NIC can recognize many packet types, includi
On Tue, Nov 18, 2014 at 09:06:35PM +0600, Yerden Zhumabekov wrote:
>
> 18.11.2014 20:41, Neil Horman ?:
> > On Tue, Nov 18, 2014 at 08:03:40PM +0600, Yerden Zhumabekov wrote:
> >> Initially, SSE4.2 support is detected via CPUID instruction.
> >>
> >> Added rte_hash_crc_set_alg() function to de
Sorry, maybe I made a mistake here.
Accoring to lib/librte_eal/common/eal_common_cpuflags.c code, it seemed
to me that constructor attribute is not supported by intel compiler. So
in that case here I decided to leave the code for autodetection. Am I
correct?
18.11.2014 9:21, Yerden Zhumabekov ???
Test application to transmit 32-packet bursts of 220-byte UDP packets every
50 us, approximating 240,000 pps. We found it useful for testing hypervisor
performance for a transmit-heavy but bursty workload in a VM with DPDK.
Signed-off-by: Bhavesh Davda
---
examples/Makefile |1 +
ex
Hi Konstantin,
On 11/17/2014 08:00 PM, Ananyev, Konstantin wrote:
>> +/*
>> + * Get the name of a RX offload flag
>> + */
>> +const char *rte_get_rx_ol_flag_name(uint64_t mask)
>> +{
>> +switch (mask) {
>> +case PKT_RX_VLAN_PKT: return "PKT_RX_VLAN_PKT";
>> +case PKT_RX_RSS_HASH: retur
Hi Konstantin,
On 11/17/2014 07:26 PM, Ananyev, Konstantin wrote:
> Just one thing - double semicolon - looks like a typo:
>> +/* check if TCP segmentation required for this packet */
>> +if (ol_flags & PKT_TX_TCP_SEG) {
>> +/* implies IP cksum and TCP cksum */
>> +
Hi Konstantin,
On 11/17/2014 07:15 PM, Ananyev, Konstantin wrote:
> Just 2 nits from me:
>
> 1)
>> +static inline uint16_t
>> +rte_raw_cksum(const char *buf, size_t len)
>> +{
> ...
>> +while (len >= 8) {
>> +sum += u16[0]; sum += u16[1]; sum += u16[2]; sum += u16[3];
>
> Can you p
On Tue, Nov 18, 2014 at 08:03:40PM +0600, Yerden Zhumabekov wrote:
> Initially, SSE4.2 support is detected via CPUID instruction.
>
> Added rte_hash_crc_set_alg() function to detect and set CRC32
> implementation if necessary. SSE4.2 is allowed by default. If it's
> not available, fall back to sw
Since rte_hash_crc() can now be run regardless of SSE4.2 support,
we can safely remove compile checks for RTE_MACHINE_CPUFLAG_SSE4_2
in test utilities.
Signed-off-by: Yerden Zhumabekov
---
app/test/test_hash.c |7 ---
app/test/test_hash_perf.c | 11 ---
2 files changed, 18
Calculating hash for data of variable length is more efficient
when that data is sliced into 8-byte pieces. The rest part of data
is hashed using CRC32 functions with either 8 and 4 byte operands.
Signed-off-by: Yerden Zhumabekov
---
lib/librte_hash/rte_hash_crc.h | 33
Initially, SSE4.2 support is detected via CPUID instruction.
Added rte_hash_crc_set_alg() function to detect and set CRC32
implementation if necessary. SSE4.2 is allowed by default. If it's
not available, fall back to sw implementation.
Depending on compiler attributes support, best available alg
SSE4.2 provides _mm_crc32_u64 intrinsic with 8-byte operand.
Signed-off-by: Yerden Zhumabekov
---
lib/librte_hash/rte_hash_crc.h | 16
1 file changed, 16 insertions(+)
diff --git a/lib/librte_hash/rte_hash_crc.h b/lib/librte_hash/rte_hash_crc.h
index 4d7532a..15f687a 100644
-
Add lookup tables for CRC32 algorithm, crc32c_1word() and
crc32c_2words() functions returning hash of 32-bit and
64-bit operand.
Signed-off-by: Yerden Zhumabekov
---
lib/librte_hash/rte_hash_crc.h | 316
1 file changed, 316 insertions(+)
diff --git a/li
This is a rework of my previous patches improving performance of rte_hash_crc.
In addition, this revision brings a fallback mechanism to ensure that CRC32
hash is calculated regardless of hardware support from CPU (i.e. SSE4.2
intrinsics). Performance of software CRC32 implementation is also imp
On Tue, Nov 18, 2014 at 10:56:24AM +0600, Yerden Zhumabekov wrote:
> Sorry, maybe I made a mistake here.
>
> Accoring to lib/librte_eal/common/eal_common_cpuflags.c code, it seemed
> to me that constructor attribute is not supported by intel compiler. So
> in that case here I decided to leave the
57 matches
Mail list logo
