On Wednesday, 21 June 2017 18:39:52 IDT Stephen Hemminger wrote:
> On Wed, 21 Jun 2017 16:52:41 +0300
> Gregory Etelson wrote:
>
> > Hello,
> >
> > Ethernet frames sent to multicast destination address 01:ff:ff:ff:ff:01
> > not visible on ixgbe & i40e VF PMDs.
> >
> > PF PMDs receive multicast
Hi Yuanhan,
On Tuesday 20 June 2017 01:52 PM, Yuanhan Liu wrote:
> On Tue, Jun 20, 2017 at 09:35:07AM +0200, Thomas Monjalon wrote:
>> 20/06/2017 06:14, santosh:
>>> On Tuesday 20 June 2017 02:07 AM, Thomas Monjalon wrote:
>>>
08/06/2017 16:28, Santosh Shukla:
> Let test_mbuf alloc and f
On Tuesday 20 June 2017 01:05 PM, Thomas Monjalon wrote:
> 20/06/2017 06:14, santosh:
>> On Tuesday 20 June 2017 02:07 AM, Thomas Monjalon wrote:
>>
>>> 08/06/2017 16:28, Santosh Shukla:
Let test_mbuf alloc and free mempool.
Cc: sta...@dpdk.org
Signed-off-by: Santosh Shukla
>>
> -Original Message-
> From: Wang, Xiao W
> Sent: Thursday, June 22, 2017 7:20 PM
> To: Chen, Jing D
> Cc: dev@dpdk.org; sta...@dpdk.org; Wang, Xiao W
> Subject: [PATCH v2] net/fm10k: initialize link status in device start
>
> Fm10k host driver can't manage PHY directly and provides a
Fm10k host driver can't manage PHY directly and provides a fake link
status by always reporting LINK_UP. We should initialize link status
in device start, otherwise application will get LINK_DOWN status
when LSC configured.
Fixes: 9ae6068c86da ("fm10k: add dev start/stop")
Cc: sta...@dpdk.org
Sig
Hi,
> -Original Message-
> From: Chen, Jing D
> Sent: Wednesday, June 21, 2017 10:36 AM
> To: Wang, Xiao W
> Cc: dev@dpdk.org; sta...@dpdk.org
> Subject: RE: [PATCH] net/fm10k: initialize link status in device start
>
> Hi,
>
> > -Original Message-
> > From: Wang, Xiao W
> > Sen
Enable signature match for rte_flow API.
RTE_FLOW_ITEM_TYPE_FUZZY specify a signature match.
Signed-off-by: Qi Zhang
---
v2:
- will check fuzzy match's last and mask value.
drivers/net/ixgbe/ixgbe_flow.c | 91 --
1 file changed, 79 insertions(+), 12 del
Enable IPv6 support with rte_flow API.
Only support Sigature Match.
Signed-off-by: Qi Zhang
---
v2:
- fix flow type assignment.
drivers/net/ixgbe/ixgbe_flow.c | 118 +
1 file changed, 107 insertions(+), 11 deletions(-)
diff --git a/drivers/net/ixgbe/ixg
Code clean with 2 purposes.
1. No variable "index" needed
2. inline function make it easy and safe when be nest into a loop.
Signed-off-by: Qi Zhang
---
v2:
- Update the commit log
drivers/net/ixgbe/ixgbe_flow.c | 208 +++--
1 file changed, 73 insertions(+),
The patchset is based on
http://dpdk.org/dev/patchwork/patch/25292/
v3:
- will check fuzzy match's last and mask value.
v2:
- add comment to explain macro replacement
- fix wrong flow type assignment
Qi Zhang (3):
net/ixgbe: replace macro with inline function
net/ixgbe: enable signature
> -Original Message-
> From: rsanford2 at gmail.com
> Sent: Tuesday, August 2, 2016 4:43 AM
> Subject: [dpdk-dev] [PATCH 1/4] testpmd: fix LACP ports to work with idle
> links
>
> Problem: When there is little or no TX traffic on an LACP port (bonding mode
> 4),
> we don't call its tx
> -Original Message-
> From: Vasily Philipov [mailto:vasi...@mellanox.com]
> Sent: Wednesday, June 21, 2017 5:44 PM
> To: Wu, Jingjing ; dev@dpdk.org
> Cc: Adrien Mazarguil ; Nélio Laranjeiro
>
> Subject: RE: [dpdk-dev] [PATCH v4 4/4] app/testpmd: add isolated mode
> parameter
>
>
>
>
Hi,
I am resending it again since I did not see any activity with below original
email sent out.
Sorry for the spam if it is a duplicate.
Thanks & Regards,
Leela sankar Gudimetla
From: Gudimetla, Leela Sankar
Sent: Wednesday, June 21, 2017 2:24 PM
To: 'dev@dpdk.org'
Cc: 'hh...@cisco.com'
Subje
Hi,
"Existing thread subject: DPDK 17.02 RC-3 performance degradation of ~10%"
We are in the process of upgrading from dpdk 16.11 to 17.02. We are also seeing
performance degradation in our packet processing application for about ~3% with
17.02.
When we browsed through mailing lists, we found a
* Sam (batmanu...@gmail.com) wrote:
> Thank you~
>
> 1. We have a compare test on qemu-kvm enviroment with huge page and without
> huge page. Qemu start process is much longer in huge page enviromwnt. And I
> write an email titled with '[DPDK-memory] how qemu waste such long time
> under dpdk huge
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Pablo de Lara
> Sent: Wednesday, June 21, 2017 7:42 AM
> To: Doherty, Declan
> Cc: dev@dpdk.org; De Lara Guarch, Pablo ;
> sta...@dpdk.org
> Subject: [dpdk-dev] [PATCH 9/9] doc: fix typo in l2fwd-crypto usage
>
Hi Pablo,
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Pablo de Lara
> Sent: Wednesday, June 21, 2017 7:42 AM
> To: Doherty, Declan
> Cc: dev@dpdk.org; De Lara Guarch, Pablo ;
> sta...@dpdk.org
> Subject: [dpdk-dev] [PATCH 2/9] test/crypto: fix wrong AAD se
Some mempool hw like octeontx/fpa block, demands block size aligned
buffer address.
Introducing an MEMPOOL_F_POOL_BLK_SZ_ALIGNED flag.
If this flag is set:
1) adjust 'off' value to block size aligned value.
2) Allocate one additional buffer. This buffer is used to make sure that
requested 'n' buff
HW pool manager e.g. Octeontx SoC demands s/w to program start and end
address of pool. Currently, there is no such handle in external mempool.
Introducing rte_mempool_update_range handle which will let HW(pool manager)
know when common layer selects hugepage:
For each hugepage - update its start/e
HW mempool blocks may need physical contiguous obj in a pool.
Introducing MEMPOOL_F_POOL_CONTIG flag for such use-case. The flag
useful to detect whether all buffer fits within a hugepage or not. If
not then return -ENOSPC. This way, we make sure that all object within a
pool is contiguous.
Signe
In order to support octeontx HW mempool manager, the common mempool layer must
meet below condition.
- buffer start address should be block size aligned.
- buffers must have the physically contiguous address within the pool.
And right now mempool doesn't support both.
Patchset adds infrastrucure
Allow external mempool to advertise its capability.
A handler been introduced called rte_mempool_ops_get_hw_cap.
- Upon ->get_hw_cap call, mempool driver will advertise
capability by returning flag.
- Common layer updates flag value in 'mp->flags'.
Signed-off-by: Santosh Shukla
Signed-off-by: Jer
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Pablo de Lara
> Sent: Wednesday, June 21, 2017 7:42 AM
> To: Doherty, Declan
> Cc: dev@dpdk.org; De Lara Guarch, Pablo
> Subject: [dpdk-dev] [PATCH 4/9] test/crypto: remove offsets in wireless
> algorithm vecto
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Pablo de Lara
> Sent: Wednesday, June 21, 2017 7:42 AM
> To: Doherty, Declan
> Cc: dev@dpdk.org; De Lara Guarch, Pablo ;
> sta...@dpdk.org
> Subject: [dpdk-dev] [PATCH 6/9] examples/l2fwd-crypto: fix application
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Pablo de Lara
> Sent: Wednesday, June 21, 2017 7:42 AM
> To: Doherty, Declan
> Cc: dev@dpdk.org; De Lara Guarch, Pablo ;
> sta...@dpdk.org
> Subject: [dpdk-dev] [PATCH 7/9] examples/l2fwd-crypto: fix auth info d
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Pablo de Lara
> Sent: Wednesday, June 21, 2017 7:42 AM
> To: Doherty, Declan
> Cc: dev@dpdk.org; De Lara Guarch, Pablo
> Subject: [dpdk-dev] [PATCH 5/9] test/crypto: use existing algorithm strings
>
> Instead o
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Pablo de Lara
> Sent: Wednesday, June 21, 2017 7:42 AM
> To: Doherty, Declan
> Cc: dev@dpdk.org; De Lara Guarch, Pablo ;
> sta...@dpdk.org
> Subject: [dpdk-dev] [PATCH 8/9] app/crypto-perf: fix error message
>
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Pablo de Lara
> Sent: Wednesday, June 21, 2017 7:42 AM
> To: Doherty, Declan
> Cc: dev@dpdk.org; De Lara Guarch, Pablo
> Subject: [dpdk-dev] [PATCH 3/9] test/crypto: remove unnecessary offset
>
> For AES/DES te
The PCI domain in Azure maybe 32 bits. When device is passed through
the domain is synthesize from the internal GUID.
Signed-off-by: Stephen Hemminger
---
drivers/net/mlx5/mlx5_ethdev.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/mlx5/mlx5_ethdev.c b/drivers/n
The function strtoul returns unsigned long and can be directly
assigned to a smaller type. Removing the casts allows easier
expansion of PCI domain.
Signed-off-by: Stephen Hemminger
---
lib/librte_eal/linuxapp/eal/eal_pci.c | 8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --gi
In some environments, the PCI domain can be larger than 16 bits.
For example, a PCI device passed through in Azure gets a synthetic domain
id which is internally generated based on GUID. The PCI standard does
not restrict domain to be 16 bits.
This change breaks ABI for API's that expose PCI addr
These patches address a potential future issue when devices are passed through
to DPDK applications on Azure. Since PCI bus is changing in current release
it makes sense to change now.
Stephen Hemminger (3):
pci: remove unnecessary casts from strtoul
eal: PCI domain should be 32 bits
mlx5: h
> -Original Message-
> From: De Lara Guarch, Pablo
> Sent: Wednesday, June 21, 2017 8:47 AM
> To: De Lara Guarch, Pablo
> Cc: dev@dpdk.org
> Subject: [PATCH 01/22] cryptodev: move session type to generic crypto op
>
> Session type (operation with or without session) is not something speci
Looking at some other issues, I noticed that both the TAP and MLX5 device
drivers are looking at kernel version through uname. Although this may
seem like a good way to deal with kernel API changes, it is not reliable.
Enterprise kernel distro vendors never change kernel version but do backport
f
Hi,
I'm working with fm10k 100G card, fm10k driver is 0.21.7, testpoint using SDK
version- 4.3.2.
I have 2 vms, each one attached to 1 VF- binded with dpdk.
One of the VFs receives a frame of size 1514, it encapsulates it with vxlan and
sends the new frame (size 1584) to the 2nd VF.
The result is
> On Jun 21, 2017, at 10:14 AM, Gaëtan Rivet wrote:
>
> Hi,
>
> On Wed, Jun 21, 2017 at 02:27:49PM +, Wiles, Keith wrote:
>>
>>> On Jun 21, 2017, at 5:27 AM, Bruce Richardson
>>> wrote:
>>>
>>> On Tue, Jun 20, 2017 at 11:21:39PM +0200, Thomas Monjalon wrote:
If a library or a build
On 18/06/2017 16:52, Gregory Etelson wrote:
Use get_mempolicy(2) to find numa socket on Linux
instead of parsing /proc/self/numa_maps.
When process maps around 1K hugepages
numa_maps file can miss huge records in older Linux kernels.
get_mempolicy() proved more reliable
Requires numactl dev pack
Now that AAD is only used in AEAD algorithms,
there is no need to keep AAD in the authentication
structure.
Signed-off-by: Pablo de Lara
---
app/test-crypto-perf/cperf_ops.c | 2 --
drivers/crypto/openssl/rte_openssl_pmd.c | 1 -
lib/librte_cryptodev/rte_crypto_sym.h| 24 --
AEAD algorithms such as AES-GCM needed to be
used as a concatenation of a cipher transform and
an authentication transform.
Instead, a new transform and functions to handle it
are created to support these kind of algorithms,
making their use easier.
Signed-off-by: Pablo de Lara
---
lib/librte_c
AEAD operation parameters can be set in the new
aead structure, in the crypto operation.
This structure is within an union with the cipher
and authentication parameters, since operations can be:
- AEAD: using the aead structure
- Cipher only: using only the cipher structure
- Auth only: using onl
AES-GMAC is an authentication algorithm, based on AES-GCM
without encryption. To simplify its usage, now it can be used
setting the authentication parameters, without requiring
to concatenate a ciphering transform.
Therefore, it is not required to set AAD, but authentication
data length and offset
Additional authenticated data (AAD) information was duplicated
in the authentication transform and in the crypto
operation structures.
Since AAD length is not meant to be changed in a same session,
it is removed from the crypto operation structure.
Signed-off-by: Pablo de Lara
---
app/test-cryp
Digest length was duplicated in the authentication transform
and the crypto operation structures.
Since digest length is not expected to change in a same
session, it is removed from the crypto operation.
Also, the length has been shrunk to 16 bits,
which should be sufficient for any digest.
Sign
For wireless algorithms (SNOW3G, KASUMI, ZUC),
the IV for the authentication algorithms (F9, UIA2 and EIA3)
was taken from the AAD parameter, as there was no IV parameter
in the authentication structure.
Now that IV is available for all algorithms, there is need
to keep doing this, so AAD is not u
Authentication algorithms, such as AES-GMAC or the wireless
algorithms (like SNOW3G) use IV, like cipher algorithms.
So far, AES-GMAC has used the IV from the cipher structure,
and the wireless algorithms have used the AAD field,
which is not technically correct.
Therefore, authentication IV param
Since IV parameters (offset and length) should not
change for operations in the same session, these parameters
are moved to the crypto transform structure, so they will
be stored in the sessions.
Signed-off-by: Pablo de Lara
---
app/test-crypto-perf/cperf_ops.c | 22 ++--
app/
Since IV now is copied after the crypto operation, in
its private size, IV can be passed only with offset
and length.
Signed-off-by: Pablo de Lara
---
app/test-crypto-perf/cperf_ops.c| 21 ++
drivers/crypto/aesni_gcm/aesni_gcm_pmd.c| 80 +++--
drivers/crypto
Usually, IV will change for each crypto operation.
Therefore, instead of pointing at the same location,
IV is copied after each crypto operation.
This will let the IV to be passed as an offset from
the beginning of the crypto operation, instead of
a pointer.
Signed-off-by: Pablo de Lara
---
exa
Usually, IV will change for each crypto operation.
Therefore, instead of pointing at the same location,
IV is copied after each crypto operation.
This will let the IV to be passed as an offset from
the beginning of the crypto operation, instead of
a pointer.
Signed-off-by: Pablo de Lara
---
exa
Usually, IV will change for each crypto operation.
Therefore, instead of pointing at the same location,
IV is copied after each crypto operation.
This will let the IV to be passed as an offset from
the beginning of the crypto operation, instead of
a pointer.
Signed-off-by: Pablo de Lara
---
app
Usually, IV will change for each crypto operation.
Therefore, instead of pointing at the same location,
IV is copied after each crypto operation.
This will let the IV to be passed as an offset from
the beginning of the crypto operation, instead of
a pointer.
Signed-off-by: Pablo de Lara
---
tes
Usually, IV will change for each crypto operation.
Therefore, instead of pointing at the same location,
IV is copied after each crypto operation.
This will let the IV to be passed as an offset from
the beginning of the crypto operation, instead of
a pointer.
Signed-off-by: Pablo de Lara
---
tes
In order to facilitate the access to the private data,
after the crypto operation, two new macros have been
implemented:
- rte_crypto_op_ctod_offset(c,t,o), which returns a pointer
to "o" bytes after the start of the crypto operation
- rte_crypto_op_ctophys_offset(c, o), which returns
the phys
QAT PMD was assuming that cipher IV was always prepended,
before the input buffer, but it is not necessary to have it
there, only the auth IV and the input buffer needs to be
contiguous.
It was also assuming that the IV was starting at offset 0,
which is not always the case.
Fixes: d4f2745300e0 (
Instead of storing a pointer to operation specific parameters,
such as symmetric crypto parameters, use a zero-length array,
to mark that these parameters will be stored after the
generic crypto operation structure, which was already assumed
in the code.
Signed-off-by: Pablo de Lara
---
examples
Instead of storing some crypto operation flags,
such as operation status, as enumerations,
store them as uint8_t, for memory efficiency.
Also, reserve extra 5 bytes in the crypto operation,
for future additions.
Signed-off-by: Pablo de Lara
---
lib/librte_cryptodev/rte_crypto.h | 9 +
1
Storing a pointer to the user data is unnecessary,
since user can store additional data, after the crypto operation.
Signed-off-by: Pablo de Lara
---
app/test-crypto-perf/cperf_test_latency.c | 30 +++---
lib/librte_cryptodev/rte_crypto.h | 5 -
2 files chang
Session type (operation with or without session) is not
something specific to symmetric operations.
Therefore, the variable is moved to the generic crypto operation
structure.
Signed-off-by: Pablo de Lara
---
drivers/crypto/aesni_gcm/aesni_gcm_pmd.c| 15 ---
drivers/crypto/aesni_
On Wed, 21 Jun 2017 16:52:41 +0300
Gregory Etelson wrote:
> Hello,
>
> Ethernet frames sent to multicast destination address 01:ff:ff:ff:ff:01
> not visible on ixgbe & i40e VF PMDs.
>
> PF PMDs receive multicast frames normally.
>
> Tested with DPDK v17.05.0 and
> 3566af6be7f968c678a867399ddc6
On Wed, 21 Jun 2017 12:06:51 +0100
Ferruh Yigit wrote:
> To have the support corresponding kernel module (UNCI) needs to be
> inserted. If kernel module is not there, application will run as
> it is without kernel control path support.
>
> When UNCI module inserted, running application creates
On Wed, 21 Jun 2017 12:06:49 +0100
Ferruh Yigit wrote:
> +static struct ethtool_input_buffer {
> + int magic;
> + void *buffer;
> + size_t length;
> + struct completion *msg_received;
> + int *err;
> + u32 in_use;
> +} ethtool_input_buffer
Naming in kernel is important. T
Hi,
On Wed, Jun 21, 2017 at 02:27:49PM +, Wiles, Keith wrote:
>
> > On Jun 21, 2017, at 5:27 AM, Bruce Richardson
> > wrote:
> >
> > On Tue, Jun 20, 2017 at 11:21:39PM +0200, Thomas Monjalon wrote:
> >> If a library or a build tool uses a definition from a driver,
> >> there is a build ord
Sorry for not getting these out last week.
Here are the meeting notes for the DPDK technical board meeting held on
June 7th, 2017.
Attendees:
- Stephen Hemminger
- Thomas Monjalon
- Jerin Jacob
- Hemant Agrawal
- Bruce Richardson
- Olivier Matz
- Yuanhan Liu
- Konstantin Ananyev
- Kevin Traynor
-
Hi,
During the last techboard meeting, it has been decided to call
more frequently for reviews by giving a list of hot patches.
For this first burst, I would like to highlight some important
patches in EAL and ethdev. The first EAL patches are all related
to NUMA. We have also some work around ph
08/06/2017 01:58, Gaetan Rivet:
> Signed-off-by: Gaetan Rivet
> ---
> lib/librte_eal/common/eal_common_dev.c | 50
> --
> 1 file changed, 30 insertions(+), 20 deletions(-)
It's difficult to understand this patch.
Please add some explanations.
And it may be better
Fixes: ba7b86b1419b ("doc: add l2fwd-crypto sample app guide")
CC: sta...@dpdk.org
Signed-off-by: Pablo de Lara
---
doc/guides/sample_app_ug/l2_forward_crypto.rst | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/doc/guides/sample_app_ug/l2_forward_crypto.rst
b/doc/guides/samp
Fixes: f6cefe253cc8 ("app/crypto-perf: add range/list of sizes")
CC: sta...@dpdk.org
Signed-off-by: Pablo de Lara
---
app/test-crypto-perf/cperf_options_parsing.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/app/test-crypto-perf/cperf_options_parsing.c
b/app/test-crypto-p
Instead of passing the authentication and cipher offset
from the test vectors, just get them from the IV and AAD lengths.
Signed-off-by: Pablo de Lara
---
test/test/test_cryptodev.c | 59 ++---
.../test/test_cryptodev_kasumi_hash_test_vectors.h | 25 --
Fixes: 4790f99d2d31 ("examples/l2fwd-crypto: use cryptodev algorithm parser")
CC: sta...@dpdk.org
Signed-off-by: Pablo de Lara
---
examples/l2fwd-crypto/main.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/examples/l2fwd-crypto/main.c b/examples/l2fwd-crypto/main.c
index 84
Chain parameter can be CIPHER_HASH, HASH_CIPHER,
CIPHER_ONLY or HASH_ONLY, but only the first two
were shown in the application help.
Fixes: 1a75e9f3fadb ("examples/l2fwd-crypto: add cipher/hash only cases")
CC: sta...@dpdk.org
Signed-off-by: Pablo de Lara
---
examples/l2fwd-crypto/main.c | 3 +
Instead of using duplicated functions to get the algorithm
strings, use the functions from the cryptodev library.
Signed-off-by: Pablo de Lara
---
test/test/test_cryptodev_perf.c | 83 -
1 file changed, 16 insertions(+), 67 deletions(-)
diff --git a/test/
For AES/DES tests, there is no need to leave out
the first block from the crypto operation.
Signed-off-by: Pablo de Lara
---
test/test/test_cryptodev_perf.c | 14 ++
1 file changed, 6 insertions(+), 8 deletions(-)
diff --git a/test/test/test_cryptodev_perf.c b/test/test/test_cryptod
AAD should not point at IV for AES algorithms.
For AES-GCM, AAD will point at additional data in the mbuf.
For the other algorithms (such as AES CBC), AAD is not used.
Fixes: ffbe3be0d4b5 ("app/test: add libcrypto")
CC: sta...@dpdk.org
Signed-off-by: Pablo de Lara
---
test/test/test_cryptodev_p
Some assert checks in the driver were
incorrect, but they are not necessary anyway,
as application will panic in any case.
Fixes: 0f548b50a160 ("crypto/aesni_mb: process crypto op on dequeue")
CC: sta...@dpdk.org
Signed-off-by: Pablo de Lara
---
drivers/crypto/aesni_mb/rte_aesni_mb_pmd.c | 4 --
Prior to new crypto patches for this release,
there were some fixes and some cleanup to do in
some crypto drivers, tests, sample apps and documents.
Pablo de Lara (9):
crypto/aesni_mb: remove assert check
test/crypto: fix wrong AAD setting
test/crypto: remove unnecessary offset
test/crypto
08/06/2017 01:58, Gaetan Rivet:
> +RTE_REGISTER_BUS(VIRTUAL_BUS_NAME, rte_vdev_bus);
nice :)
08/06/2017 01:54, Gaetan Rivet:
> Signed-off-by: Gaetan Rivet
> ---
> This patch depends on:
>
> bus: attach / detach API
> http://dpdk.org/ml/archives/dev/2017-May/066330.html
> http://dpdk.org/dev/patchwork/patch/24489/
Please insert this patch the related series above.
21/06/2017 01:29, Gaetan Rivet:
> + return !!dev->device->devargs->bus->unplug;
lib/librte_ether/rte_ethdev.c:364:33: fatal error:
no member named 'bus' in 'struct rte_devargs'
Please move this patch alone, specifying the right dependency.
21/06/2017 01:29, Gaetan Rivet:
> --- a/lib/librte_ether/rte_ethdev.c
> +++ b/lib/librte_ether/rte_ethdev.c
> @@ -354,26 +354,14 @@ rte_eth_dev_get_port_by_name(const char *name, uint8_t
> *port_id)
> static int
> rte_eth_dev_is_detachable(uint8_t port_id)
> {
> - uint32_t dev_flags;
> +
Signed-off-by: Pablo de Lara
Acked-by: Declan Doherty
---
lib/librte_cryptodev/rte_cryptodev.h | 2 -
lib/librte_cryptodev/rte_cryptodev_pmd.h | 64
2 files changed, 66 deletions(-)
diff --git a/lib/librte_cryptodev/rte_cryptodev.h
b/lib/librte_cryptodev/r
Signed-off-by: Pablo de Lara
Acked-by: Declan Doherty
---
drivers/crypto/qat/qat_crypto.c| 1 +
drivers/crypto/qat/qat_qp.c| 8 +---
drivers/crypto/qat/rte_qat_cryptodev.c | 32
3 files changed, 26 insertions(+), 15 deletions(-)
diff --
Signed-off-by: Pablo de Lara
Acked-by: Declan Doherty
---
lib/librte_cryptodev/rte_cryptodev.c | 103 -
lib/librte_cryptodev/rte_cryptodev_pmd.h | 13
lib/librte_cryptodev/rte_cryptodev_version.map | 10 +--
3 files changed, 1 insertion(+), 125 dele
Signed-off-by: Pablo de Lara
---
drivers/crypto/dpaa2_sec/dpaa2_sec_dpseci.c | 9 +++--
1 file changed, 3 insertions(+), 6 deletions(-)
diff --git a/drivers/crypto/dpaa2_sec/dpaa2_sec_dpseci.c
b/drivers/crypto/dpaa2_sec/dpaa2_sec_dpseci.c
index 4e01fe8..e32b27e 100644
--- a/drivers/crypto/d
Add PCI probe/remove/init/uninit functions in a separate
file rte_cryptodev_pci.h, which do not use cryptodev driver,
in order to be removed in next commits.
Signed-off-by: Pablo de Lara
Acked-by: Declan Doherty
---
lib/librte_cryptodev/Makefile| 1 +
lib/librte_cryptodev/rte_cryp
Call rte_cryptodev_pmd_release_device() if probing a
PCI crypto device, instead of accessing the variables
directly. This will be useful when rte_cryptodev_pci_probe()
gets moved to a separate file.
Signed-off-by: Pablo de Lara
Acked-by: Declan Doherty
---
lib/librte_cryptodev/rte_cryptodev.c |
Move all functions handling virtual devices to a separate
header file "rte_cryptodev_vdev.h", in order to leave only
generic functions for any device in the rest of the files.
Signed-off-by: Pablo de Lara
---
drivers/crypto/aesni_gcm/aesni_gcm_pmd.c | 1 +
drivers/crypto/aesni_mb/rte_aes
rte_cryptodev_devices_get() function returns an array of devices
sharing the same driver.
Instead of having two different paths depending on the device being
virtual or physical, retrieve the driver name from rte_device structure.
Signed-off-by: Pablo de Lara
Acked-by: Declan Doherty
---
lib/l
Do not set PCI information in the device information structure
for any crypto device, just for the ones that are PCI, so
this is set internally in the PCI crypto PMDs (only QAT now).
Signed-off-by: Pablo de Lara
Acked-by: Declan Doherty
---
drivers/crypto/qat/qat_crypto.c | 5 +++--
lib/li
When retrieving device information for a crypto driver,
driver name was only set when it was a PCI driver.
Getting the driver name from rte_device structure
allows rte_cryptodev_get_info() function to return it
regardless they are virtual or physical devices.
Signed-off-by: Pablo de Lara
Acked-by
rte_cryptodev_devices_get() function was parsing a crypto
device name as an argument, but the function actually
returns device identifiers of devices that share the
same crypto driver, so the argument should be driver name, instead.
Fixes: 38227c0e3ad2 ("cryptodev: retrieve device info")
CC: sta..
Only non virtual devices were storing the pointer to
rte_device structure in rte_cryptodev, which will be needed
to retrieve the driver name for any device.
Signed-off-by: Pablo de Lara
Acked-by: Declan Doherty
---
drivers/crypto/aesni_gcm/aesni_gcm_pmd.c | 3 ++-
drivers/crypto/aesni_mb/rte_
Following the same approach taken for ethdev, this patch series
removes the PCI specific structure cryptodev driver from rte_cryptodev.
The patchset makes the following changes:
- Moves the virtual and PCI device specific functions to separate files,
so other cryptodev header files contain only
> On Jun 21, 2017, at 5:27 AM, Bruce Richardson
> wrote:
>
> On Tue, Jun 20, 2017 at 11:21:39PM +0200, Thomas Monjalon wrote:
>> If a library or a build tool uses a definition from a driver,
>> there is a build ordering issue, like seen when moving PCI code
>> into a bus driver.
>>
>> One opti
Hi Stephen,
> -Original Message-
> From: Stephen Hemminger [mailto:step...@networkplumber.org]
> Sent: Wednesday, May 31, 2017 11:52 PM
> To: De Lara Guarch, Pablo
> Cc: Doherty, Declan ; dev@dpdk.org;
> sta...@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH] crypto/aesni_mb: fix assert check
>
Hello,
Ethernet frames sent to multicast destination address 01:ff:ff:ff:ff:01
not visible on ixgbe & i40e VF PMDs.
PF PMDs receive multicast frames normally.
Tested with DPDK v17.05.0 and
3566af6be7f968c678a867399ddc644fb762873b (current master)
Regards,
Gregory
The following changes since commit 3566af6be7f968c678a867399ddc644fb762873b:
app/testpmd: always build VF and MACsec functions (2017-06-20 09:28:34 +0200)
are available in the git repository at:
http://dpdk.org/git/next/dpdk-next-eventdev
for you to fetch changes up to c687cebe095b2ea325aa
Document NIC features, add more information about them and add more
implementation related support.
Signed-off-by: Ferruh Yigit
---
TODO:
- Not all features all fully documented, need help from community
- Instead of having a new file, it would be nice to auto generate this
file from defaults.i
When creating a virtual pmd to test link bonding,
the socket id was checked, if it was in the range
of available sockets.
This check is unnecessary, as the socket specified
might not have memory anyway, so it will fail
at memory allocation.
Therefore, the best solution is to remove this check.
Si
Socket id parsed from the user was checked
if it was in the range of available sockets.
This check is unnecessary, as the socket specified
might not have memory anyway, so it will fail
at memory allocation.
Therefore, the best solution is to remove this check.
Signed-off-by: Pablo de Lara
---
d
Socket id parsed from the user was checked
if it was in the range of available sockets.
This check is unnecessary, as the socket specified
might not have memory anyway, so it will fail
at memory allocation.
Therefore, the best solution is to remove this check.
Signed-off-by: Pablo de Lara
---
d
1 - 100 of 165 matches
Mail list logo
