On 11/13/2018 9:06 PM, Ananyev, Konstantin wrote:
>
> -Original Message-
> From: Akhil Goyal [mailto:akhil.go...@nxp.com]
> Sent: Tuesday, November 13, 2018 11:28 AM
> To: dev@dpdk.org
> Cc: tho...@monjalon.net; Ananyev, Konstantin
> ; jerin.ja...@caviumnetworks.com;
> > > > @@ -1301,8 +1346,7 @@ rte_cryptodev_sym_session_free(struct
> > > > rte_cryptodev_sym_session *sess)
> > > >
> > > > /* Check that all device private data has been freed */
> > > > for (i = 0; i < nb_drivers; i++) {
> > > [Fiona] Use the sess.nb_drivers rather than the gl
> > port_id size should be uint16_t,
> > fix where it is defined as uint8_t
> >
> > Fixes: f8244c6399d9 ("ethdev: increase port id range")
> > Cc: zhiyong.y...@intel.com
> >
> > Signed-off-by: Marko Kovacevic
> > ---
> > app/pdump/main.c | 6 +++---
> > 1 file changed, 3 insertions(+), 3 deletion
Hi Mattias,
Sorry for the late reply. Comments inline.
> -Original Message-
> From: Mattias Rönnblom [mailto:mattias.ronnb...@ericsson.com]
> Sent: Tuesday, October 30, 2018 7:38 PM
> To: Zhang, Roy Fan ; dev@dpdk.org
> Cc: maxime.coque...@redhat.com
> Subject: Re: [dpdk-dev] [PATCH] exam
>
> On 11/13/2018 9:06 PM, Ananyev, Konstantin wrote:
> >
> > -Original Message-
> > From: Akhil Goyal [mailto:akhil.go...@nxp.com]
> > Sent: Tuesday, November 13, 2018 11:28 AM
> > To: dev@dpdk.org
> > Cc: tho...@monjalon.net; Ananyev, Konstantin
> > ; jerin.ja...
port_id size should be uint16_t,
fix where it is defined as uint8_t
Fixes: f8244c6399d9 ("ethdev: increase port id range")
Cc: zhiyong.y...@intel.com
Cc: sta...@dpdk.org
Signed-off-by: Marko Kovacevic
--
v2:
Added cc to stable
v3:
Changed two more values to uint16_t (Reshma)
---
app/pdump/
Hi Dong,
> Hi Herakliusz,
>
> Since pkt->ol_flags might set some other bits except IND_ATTACHED_MBUF, how
> about set its value to original pkt->ol_flag
It could be done, but not sure what is the point?
Which flags you think we need to copy from cloned mbuf (segment #2) to new
header?
Konstant
Hi Zhao,
The patch which causes your crash is already reverted (see [1]).
You can find it with head of next-net or in RC3 (once it is issued).
Please re-test your case after applying the reverting patch.
Regards,
Ophir
[1]
Commit 8773db152033 ("app/testpmd: revert setting default RSS")
> -O
When eth_octeontx is initialized before event_octeontx, eth_octeontx
would initiate setting up of the event device with the max ports and
queues available. If number of ports is more than the number of queues,
some ports would be left unused when the ports and queues are mapped
1:1. But even in tha
Power_kvm_vm_autotest should run on kvm vm and
virtio-ports should exist,if not test should skip.
Hence changed return as TEST_SKIPPED.
Fixes: 0ea2dd4409 ("test: skip when required lib not available")
Cc: sta...@dpdk.org
Signed-off-by: Pallantla Poornima
---
test/test/test_power_kvm_vm.c | 2 +-
On 11/12/2018 9:35 AM, Thomas Monjalon wrote:
11/11/2018 08:31, Matan Azrad:
From: Jeff Guo
On 11/9/2018 1:24 PM, Matan Azrad wrote:
From: Jeff Guo
On 11/8/2018 5:35 PM, Matan Azrad wrote:
From: Jeff Guo
On 11/8/2018 3:28 PM, Matan Azrad wrote:
From: Ananyev, Konstantin
From: Guo, Ji
Hi Shahaf,
On Tue, Nov 13, 2018 at 06:39:04PM +, Shahaf Shuler wrote:
> Hi Adrien,
>
> Tuesday, November 13, 2018 7:15 PM, Adrien Mazarguil:
> > Subject: Re: [dpdk-dev] [PATCH v2] ethdev: document RSS default key and
> > types
> >
> > Again a bit late to the party, please see below.
> >
>
Updating platform doc with steps to build when using Cavium OCTEON TX
SDK. SDK would be required for using crypto offload block.
Signed-off-by: Anoob Joseph
---
doc/guides/cryptodevs/octeontx.rst | 29 --
doc/guides/platform/octeontx.rst | 116 +++--
2
Acked-by: Anoob Joseph
> -Original Message-
> From: Akhil Goyal
> Sent: 13 November 2018 16:58
> To: dev@dpdk.org
> Cc: tho...@monjalon.net; konstantin.anan...@intel.com; Jacob, Jerin
> ; Joseph, Anoob
> ; radu.nico...@intel.com;
> declan.dohe...@intel.com; Hemant Agrawal ;
> Akhil Goyal
> -Original Message-
> From: Ophir Munk [mailto:ophi...@mellanox.com]
> Sent: Wednesday, November 14, 2018 5:20 PM
> To: Zhao1, Wei ; Lu, Wenzhuo
> ; Wu, Jingjing ; Iremonger,
> Bernard ; dev@dpdk.org; Peng, Yuan
>
> Cc: Asaf Penso ; Shahaf Shuler
> ; Thomas Monjalon ;
> Olga Shern ; Yo
When queues are stoped release tx buffers.
During start initialize array of empty
tx/rx reqs with default values.
Fixes: df238f84c0a2 ("net/ena: recreate HW IO rings on start and stop")
Signed-off-by: Rafal Kozik
Acked-by: Michal Krawczyk
---
drivers/net/ena/ena_ethdev.c | 7 +++
1 file ch
> -Original Message-
> From: Kovacevic, Marko
> Sent: Wednesday, November 14, 2018 8:39 AM
> To: Pattan, Reshma ; dev@dpdk.org
> Cc: Yigit, Ferruh ; Yang, Zhiyong
>
> Subject: RE: [PATCH] app/pdump: fix port id storage size
>
> > > port_id size should be uint16_t,
> > > fix where it is
Hi Anoob,
>
> Hi Akhil, Konstantin,
>
> Wouldn't the new element, userdata, conflict with the one referred by
>
> rte_cryptodev_sym_session_set_user_data()
> rte_cryptodev_sym_session_get_user_data()
>
> Do you mind a name change for either? Or do you have a clear picture of when
> one should
Hi Konstantin,
In this example, no flags be set by user, so no need to copy flags from
original mbuf.
I'm thinking about some DPDK developers may create their own application based
on this example, they add some flags before packet be cloned.
Regards,
Dong
-Original Message-
From: An
> -Original Message-
> From: Ananyev, Konstantin
> Sent: 14 November 2018 15:38
> To: Joseph, Anoob ; Akhil Goyal
> ; dev@dpdk.org; Ravi Kumar ;
> Jacob, Jerin ; Doherty, Declan
> ; Trahe, Fiona ; Tomasz
> Duszynski ; Dmitri Epshtein ; Natalie
> Samsonov ; Jay Zhou
> Cc: Athreya, Naraya
> Updated the doc on how packets are marked to identify their timestamp as
> valid and considered for latency measurement.
>
> Suggested-by: Bao-Long Tran
> Signed-off-by: Reshma Pattan
>
> +Timestamp and latency calculation
> +~
> +
> +The Latency stats library m
Hi Fiona,
> -Original Message-
> From: Trahe, Fiona
> Sent: Wednesday, November 14, 2018 12:46 AM
> To: Ananyev, Konstantin ; dev@dpdk.org
> Cc: De Lara Guarch, Pablo ; Akhil Goyal
> ; Doherty, Declan ; Ravi
> Kumar ; Jerin Jacob
> ; Zhang, Roy Fan
> ; Tomasz Duszynski ;
> Hemant Agrawal ;
>
> Hi Konstantin,
>
> In this example, no flags be set by user, so no need to copy flags from
> original mbuf.
Then why to bother?
> I'm thinking about some DPDK developers may create their own application
> based on this example, they add some flags before packet be
> cloned.
Hmm you r
On 14-Nov-18 5:45 AM, Sam wrote:
OK, then shortly speaking, DPDK will NOT care about padding.
NIC will care about padding while send and recv with NIC.
kernel will care about while send and recv with vhostuser port.
Is that right?
I cannot speak for virtio/vhost user since i am not terribly fa
On 14-Nov-18 3:44 AM, Thomas Monjalon wrote:
14/11/2018 04:24, Varghese, Vipin:
Tested-by: Vipin Varghese
When creating process data structures, EAL will create many files in
EAL runtime directory. Because we allow multiple secondary processes
to run, each secondary process gets their own
Hi Dong,
I think that if a developer is to manipulate ol_flags, then its onto that
developer to take care of copying ol_flags .
Regards Herakliusz
> -Original Message-
> From: Wang, Dong1
> Sent: Wednesday, November 14, 2018 10:09 AM
> To: Ananyev, Konstantin ; Lipiec, Herakliusz
> ; dev@
> -Original Message-
> From: Kovacevic, Marko
> Sent: Wednesday, November 14, 2018 8:57 AM
> To: dev@dpdk.org
> Cc: Yigit, Ferruh ; Pattan, Reshma
> ; Kovacevic, Marko ;
> Yang, Zhiyong ; sta...@dpdk.org
> Subject: [PATCH v3] app/pdump: fix port id storage size
>
> port_id size should b
On 14-Nov-18 3:10 AM, Thomas Monjalon wrote:
Hi,
07/11/2018 07:10, Li Han:
parse_portmask return type is int,but global variable
"enabled_port_mask" type is uint32_t.so in proc_info_parse_args
function,when parse_portmask return -1,"enabled_port_mask" will
get a huge value and "if (enabled_port
-Original Message-
> Date: Wed, 14 Nov 2018 14:52:03 +0530
> From: "Joseph, Anoob"
> To: Ferruh Yigit , "Jacob, Jerin"
>
> CC: "Joseph, Anoob" , "Athreya, Narayana Prasad"
> , "Bhagavatula, Pavan"
> , "dev@dpdk.org" ,
> "sta...@dpdk.org"
> Subject: [PATCH] net/octeontx: fix failures
Anatoly,
This differs from the Linux kernel's behavior, where padding belongs in the NIC
driver layer, not in the protocol layer. If you pass a runt frame (too short
packet) to a Linux NIC driver's transmission function, the NIC driver (or NIC
hardware) will pad the frame to make it valid. E.g.
Hi Konstantin & Herakliusz,
In mcast_out_pkt(), new header added and to be the first segment. Driver will
check first segment's ol_flags when sending mbuf. So why new first segment
didn't inherit old one's ol_flags? This function already copied other mbuf
fields from old one.
Since this is an
The rte_eal_get_runtime_dir() function is currently being declared in two
header files.
This API was made public in commit 6911c9fd8fbe, adding it to rte_eal.h. To
make it public, the 'rte' prefix was added to the function so it needed to
be modified in the original location of the declaration, ea
This patch fixes the zero copy enable problem for vhost crypto
sample application.
For some Crypto PMDs such as AESNI-MB and AESNI-GCM the data to
be processed will be made a copy in the same buffer but next to the
data. For example, to encrypt 64 bytes data the PMD will copy this
data from offset
> -Original Message-
> From: Wang, Dong1
> Sent: Wednesday, November 14, 2018 11:06 AM
> To: Ananyev, Konstantin ; Lipiec, Herakliusz
> ; dev@dpdk.org
> Cc: tho...@monjalon.net; Burakov, Anatoly ;
> sta...@dpdk.org
> Subject: RE: [PATCH v2] example/ipv4_multicast: fix app hanging when
The rte_eal_get_runtime_dir() function is currently being declared in two
header files.
This API was made public in commit 6911c9fd8fbe ("eal: export function to
get runtime directory"), adding it to rte_eal.h. To make it public, the
'rte' prefix was added to the function so it needed to be modifi
Hi Shahaf,
Great to see such effort! Few comments below.
Note: halfway through writing my comments i realized that i am starting
with an assumption that this API is a replacement for current VFIO DMA
mapping API's. So, if my comments seem out of left field, this is
probably why :)
On 04-Nov
In rte_efd_create() allocated memory for tail queue entry but
not freed.
Added freeing the tail queue entry.
Fixes: 56b6ef874f80 ("efd: new Elastic Flow Distributor library")
Cc: sta...@dpdk.org
Signed-off-by: Hari Kumar Vemula
Acked-by: Reshma Pattan
---
v2: Updated commit message.
---
lib/
On 14-Nov-18 11:14 AM, Kevin Laatz wrote:
The rte_eal_get_runtime_dir() function is currently being declared in two
header files.
This API was made public in commit 6911c9fd8fbe, adding it to rte_eal.h. To
make it public, the 'rte' prefix was added to the function so it needed to
be modified in
Add 'uint64_t opaque_data' inside struct rte_security_session.
That allows upper layer to easily associate some user defined
data with the session.
Proposed new layout for:
struct rte_security_session {
void *sess_private_data;
/**< Private session material */
+ uint64_t opaqu
On 14-Nov-18 11:20 AM, Kevin Laatz wrote:
The rte_eal_get_runtime_dir() function is currently being declared in two
header files.
This API was made public in commit 6911c9fd8fbe ("eal: export function to
get runtime directory"), adding it to rte_eal.h. To make it public, the
'rte' prefix was add
In rte_efd_create() write lock has already been unlocked
before ring creation itself.
So second unlock after the ring creation has been removed.
Fixes: 56b6ef874f80 ("efd: new Elastic Flow Distributor library")
Cc: sta...@dpdk.org
Signed-off-by: Chaitanya Babu Talluri
Acked-by: Reshma Pattan
--
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Konstantin
> Ananyev
> Sent: Wednesday, November 14, 2018 11:24 AM
> To: dev@dpdk.org
> Cc: akhil.go...@nxp.com; Doherty, Declan ;
> Ananyev, Konstantin
> Subject: [dpdk-dev] [PATCH] doc: security deprecation notic
On 14/11/2018 11:23, Konstantin Ananyev wrote:
Add 'uint64_t opaque_data' inside struct rte_security_session.
That allows upper layer to easily associate some user defined
data with the session.
Proposed new layout for:
struct rte_security_session {
void *sess_private_data;
/**
OK, understand now.
Thanks,
Dong
-Original Message-
From: Ananyev, Konstantin
Sent: Wednesday, November 14, 2018 19:19
To: Wang, Dong1 ; Lipiec, Herakliusz
; dev@dpdk.org
Cc: tho...@monjalon.net; Burakov, Anatoly ;
sta...@dpdk.org
Subject: RE: [PATCH v2] example/ipv4_multicast: fix app
Acked-by: Wang Dong
-Original Message-
From: Ananyev, Konstantin
Sent: Tuesday, November 13, 2018 19:52
To: Lipiec, Herakliusz ; dev@dpdk.org
Cc: Wang, Dong1 ; tho...@monjalon.net; Burakov, Anatoly
; sta...@dpdk.org
Subject: RE: [PATCH v2] example/ipv4_multicast: fix app hanging when us
So far only if machine was "native" it did use the re-direction to
not set -march on ppc64 (where -march is not supported).
We have to use mcpu/mtune in any case on ppc for whatever someone using
the build system defines as machine.
Signed-off-by: Christian Ehrhardt
---
config/meson.build | 5 ++
Add the machine definition 'baseline' which is special compared
to 'native' (most optimized for current system) or any explicit
type (external entity has to decide on the type).
It defaults to the per arch agreed common minimal baseline
needed for DPDK to reasonable work.
That might not be the mo
On Wed, 2018-11-14 at 12:34 +0100, Christian Ehrhardt wrote:
> So far only if machine was "native" it did use the re-direction to
> not set -march on ppc64 (where -march is not supported).
> We have to use mcpu/mtune in any case on ppc for whatever someone
> using
> the build system defines as mach
On Wed, 2018-11-14 at 12:34 +0100, Christian Ehrhardt wrote:
> Add the machine definition 'baseline' which is special compared
> to 'native' (most optimized for current system) or any explicit
> type (external entity has to decide on the type).
>
> It defaults to the per arch agreed common minimal
On Tue, Nov 13, 2018 at 11:42:51PM +, Burdick, Cliff wrote:
>
>
> From: Thomas Monjalon [tho...@monjalon.net]
> Sent: Tuesday, November 13, 2018 2:18 PM
> To: Burdick, Cliff
> Cc: Burakov, Anatoly; dev@dpdk.org; bruce.richard...@intel.com
> Subject: Re:
On Wed, Nov 14, 2018 at 11:40:11AM +, Luca Boccassi wrote:
> On Wed, 2018-11-14 at 12:34 +0100, Christian Ehrhardt wrote:
> > Add the machine definition 'baseline' which is special compared
> > to 'native' (most optimized for current system) or any explicit
> > type (external entity has to deci
Fix memory leak for dscp table reported by coverity
Coverity ID: 326961
Fixes: 281eeb8afc55 ("app/testpmd: add commands for metering and policing")
CC: sta...@dpdk.org
Signed-off-by: Jasvinder Singh
---
app/test-pmd/cmdline_mtr.c | 8
1 file changed, 4 insertions(+), 4 deletions(-)
di
ninja does not recognize 'DEST_DIR'.
Fixes: 9c3adc289c5e ("doc: add instructions on build using meson")
Cc: sta...@dpdk.org
Signed-off-by: Ilya Maximets
---
doc/build-sdk-meson.txt | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/doc/build-sdk-meson.txt b/doc/build-sdk-meson.
On Wed, 2018-11-14 at 11:52 +, Bruce Richardson wrote:
> On Wed, Nov 14, 2018 at 11:40:11AM +, Luca Boccassi wrote:
> > On Wed, 2018-11-14 at 12:34 +0100, Christian Ehrhardt wrote:
> > > Add the machine definition 'baseline' which is special compared
> > > to 'native' (most optimized for cu
Adding Koh, see below
Tuesday, November 13, 2018 9:11 PM, Ali Alnubani:
> Subject: [PATCH 2/2] net/mlx5: fix initialization of struct members
>
> This patch fixes compilation errors with meson and the clang compiler caused
> by some of the struct members not being initialized.
>
> ```
> ../driv
The rte_bsf64 in rte_bitmap.h has a global namespace but does not
follow convention of existing rte_bsf32 function in rte_common.h.
Therefore, deprecate the current rte_bsf64 and introduce a new
rte_bitmap_bsf64 function that will do the same thing.
In later release cycles, rte_bsf64 will be remo
Do not start the packet processing threads until all configured
slaves become active.
Signed-off-by: Radu Nicolau
---
examples/bond/main.c | 15 +++
1 file changed, 15 insertions(+)
diff --git a/examples/bond/main.c b/examples/bond/main.c
index b282e68..6623cae 100644
--- a/examples
On 11/14/2018 4:53 PM, Konstantin Ananyev wrote:
> Add 'uint64_t opaque_data' inside struct rte_security_session.
> That allows upper layer to easily associate some user defined
> data with the session.
> Proposed new layout for:
> struct rte_security_session {
> void *sess_private_data;
>
Acked-by: Hemant Agrawal
On 11/13/2018 4:58 PM, Akhil Goyal wrote:
> rte_security has been experimental since DPDK 17.11 release.
> Now the library has matured and expermental tag is removed in
> this patch.
>
>
On 12 Nov 2018, at 12:26, Eelco Chaudron wrote:
On 12 Nov 2018, at 12:18, Alejandro Lucero wrote:
When using large amount of hugepage based memory, doing all the
hugepages mapping can take quite significant time.
The problem is hugepages being initially mmaped to virtual addresses
which wi
> -Original Message-
> From: Shahaf Shuler
> Sent: Tuesday, November 13, 2018 15:22
> To: Slava Ovsiienko ; Yongseok Koh
>
> Cc: dev@dpdk.org
> Subject: RE: [PATCH 2/4] net/mlx5: fix Netlink communication routine
>
> Monday, November 12, 2018 10:02 PM, Slava Ovsiienko:
> > Subject: [PATCH
> -Original Message-
> From: Akhil Goyal [mailto:akhil.go...@nxp.com]
> Sent: Wednesday, November 14, 2018 12:40 PM
> To: Ananyev, Konstantin ; dev@dpdk.org
> Cc: Doherty, Declan
> Subject: Re: [dpdk-dev] [PATCH] doc: security deprecation notice for session
> changes
>
>
>
> On 11/14
On Wed, 2018-11-14 at 12:34 +0100, Christian Ehrhardt wrote:
> Add the machine definition 'baseline' which is special compared
> to 'native' (most optimized for current system) or any explicit
> type (external entity has to decide on the type).
>
> It defaults to the per arch agreed common minimal
On Wed, Nov 14, 2018 at 2:06 PM Luca Boccassi wrote:
>
> On Wed, 2018-11-14 at 12:34 +0100, Christian Ehrhardt wrote:
> > Add the machine definition 'baseline' which is special compared
> > to 'native' (most optimized for current system) or any explicit
> > type (external entity has to decide on t
On Wed, Nov 14, 2018 at 1:05 PM Luca Boccassi wrote:
>
> On Wed, 2018-11-14 at 11:52 +, Bruce Richardson wrote:
> > On Wed, Nov 14, 2018 at 11:40:11AM +, Luca Boccassi wrote:
> > > On Wed, 2018-11-14 at 12:34 +0100, Christian Ehrhardt wrote:
> > > > Add the machine definition 'baseline' wh
Add the machine definition 'default' which is special compared
to 'native' (most optimized for current system) or any explicit
type (external entity has to decide on the type).
It defaults to the per arch agreed common minimal baseline
needed for DPDK to reasonable work.
That might not be the mos
Tuesday, November 13, 2018 9:13 PM, Ali Alnubani:
> Subject: [PATCH 1/2] net/mlx4: fix typo
>
> Fixes: 9797bfcce1c9 ("net/mlx4: add new memory region support")
> Cc: sta...@dpdk.org
>
> Signed-off-by: Ali Alnubani
Series applied to next-net-mlx with minor commit title edit. Please have a
look,
From: Viacheslav Ovsiienko
The flow_dv_query() just returns -ENOTSUP value and does not
set provided error parameter structure, that crashes the
port_flow_query(). The patch fixes flow_db_query(), now it
sets an error parameter structure.
Fixes: 684dafe795d0 ("net/mlx5: add flow query abstractio
Wednesday, November 14, 2018 11:41 AM, Adrien Mazarguil:
> Subject: Re: [dpdk-dev] [PATCH v2] ethdev: document RSS default key and
> types
>
> Hi Shahaf,
>
> On Tue, Nov 13, 2018 at 06:39:04PM +, Shahaf Shuler wrote:
> > Hi Adrien,
> >
> > Tuesday, November 13, 2018 7:15 PM, Adrien Mazarguil:
On Wed, Nov 14, 2018 at 02:18:38PM +0100, Christian Ehrhardt wrote:
> Add the machine definition 'default' which is special compared
> to 'native' (most optimized for current system) or any explicit
> type (external entity has to decide on the type).
>
> It defaults to the per arch agreed common m
On Wed, Nov 14, 2018 at 03:04:38PM +0300, Ilya Maximets wrote:
> ninja does not recognize 'DEST_DIR'.
>
> Fixes: 9c3adc289c5e ("doc: add instructions on build using meson")
> Cc: sta...@dpdk.org
>
> Signed-off-by: Ilya Maximets
> ---
> doc/build-sdk-meson.txt | 2 +-
> 1 file changed, 1 inserti
> -Original Message-
> From: Poornima, PallantlaX
> Sent: Wednesday, November 14, 2018 9:24 AM
> To: dev@dpdk.org
> Cc: Pattan, Reshma ; Hunt, David
> ; Poornima, PallantlaX
> ; sta...@dpdk.org
> Subject: [PATCH] test: fix to skip power kvm vm autotest
>
> Power_kvm_vm_autotest should r
> Patches has to be validated for meson builds. Updating documentation for
> meson build steps in Checking Compilation category.
>
> Signed-off-by: Vipin Varghese
> ---
>
> V3:
> removed extra character - Vipin Varghese
>
> V2:
> updated the meson build options - Bruce Richardson
> ---
> doc/g
Add the machine definition 'default' which is special compared
to 'native' (most optimized for current system) or any explicit
type (external entity has to decide on the type).
It defaults to the per arch agreed common minimal baseline
needed for DPDK to reasonable work.
That might not be the mos
For IOVA as VA mode, we assume that memory is contiguous. However,
for external segments that assumption may not necessarily hold.
Fix the code to not assume that external memory segments are
contiguous even in IOVA as VA mode.
Fixes: 5282bb1c3695 ("mem: allow memseg lists to be marked as external
It may so happen that two memory locations may be adjacent in
virtual memory, but belong to different segment lists. With
current code, such segments will be concatenated. Fix the
adjacency checking code to also check if the adjacent malloc
elements belong to the same memseg list.
Fixes: 66cc45e29
Hi Anatoly,
Wednesday, November 14, 2018 1:19 PM, Burakov, Anatoly:
> Subject: Re: [RFC] ethdev: introduce DMA memory mapping for external
> memory
>
> Hi Shahaf,
>
> Great to see such effort! Few comments below.
>
> Note: halfway through writing my comments i realized that i am starting with
Tuesday, November 13, 2018 3:21 PM, Shahaf Shuler:
> Subject: Re: [dpdk-dev] [PATCH 0/4] net/mlx5: prepare to add E-switch rule
> flags check
>
> Monday, November 12, 2018 10:02 PM, Slava Ovsiienko:
> > Subject: [PATCH 0/4] net/mlx5: prepare to add E-switch rule flags
> > check
> >
> > The tc flow
It may so happen that two memory locations may be adjacent in
virtual memory, but belong to different segment lists. With
current code, such segments will be concatenated. Fix the
adjacency checking code to also check if the adjacent malloc
elements belong to the same memseg list.
Fixes: 66cc45e29
Wednesday, November 14, 2018 9:48 AM, Ali Alnubani:
> Subject: [PATCH 1/2] net/mlx4: fix typo
>
> Fixes: 9797bfcce1c9 ("net/mlx4: add new memory region support")
> Cc: sta...@dpdk.org
>
> Signed-off-by: Ali Alnubani
> ---
>From some reason my previous mail is not shown on patchwork. Resending
Wednesday, November 14, 2018 9:49 AM, Ali Alnubani:
> Subject: [PATCH v2 2/2] net/mlx5: fix typos
>
> Fixes: e1114ff6a5ab ("net/mlx5: support e-switch flow count action")
> Fixes: 974f1e7ef146 ("net/mlx5: add new memory region support")
> Cc: sta...@dpdk.org
>
> Signed-off-by: Ali Alnubani
Appl
> +err_rsa:
> + if (n)
> + BN_free(n);
> + if (e)
> + BN_free(e);
> + if (d)
> + BN_free(d);
> + if (p)
> + BN_free(p);
> + if (q)
> + BN_f
On Wed, Nov 14, 2018 at 01:51:19PM +, Shahaf Shuler wrote:
> IMO, it will make it more clear if the key will *have* to be null, because
> there is no single good reason to have it otherwise.
>
> However it looks like an endless debate between strict and relaxed API. there
> are points to b
Hi Again Tom,
Tuesday, November 13, 2018 12:17 PM, Tom Barbette:
> Subject: [PATCH] mlx5: Report imissed stat
>
> The imissed counters (number of packets dropped because the queues were
> full) were actually reported through xstats as "rx_out_of_buffer"
> but was not reported through stats.
>
>
There will be change in API functions because of mbuf sched field
updates, outlined in deprecation note of mbuf->hash.sched.
Signed-off-by: Jasvinder Singh
Acked-by: Cristian Dumitrescu
---
doc/guides/rel_notes/deprecation.rst | 5 +
1 file changed, 5 insertions(+)
diff --git a/doc/guides/
Worked in my testing.
Tested-by: Gage Eads
Thanks,
Gage
On 14-Nov-18 3:00 PM, Anatoly Burakov wrote:
It may so happen that two memory locations may be adjacent in
virtual memory, but belong to different segment lists. With
current code, such segments will be concatenated. Fix the
adjacency checking code to also check if the adjacent malloc
elements be
Hi Shahaf,
Yes we learned this distinction with rx_discards_phy the hard way. I would
expect imissed to be only rx_out_of_buffer actually. I'd say people look at
imissed to see if they consume packets fast enough. If the problem is pure CPU
power. And that is more the definition of imissed, it
> On Nov 14, 2018, at 4:51 AM, Morten Brørup wrote:
>
> Anatoly,
>
> This differs from the Linux kernel's behavior, where padding belongs in the
> NIC driver layer, not in the protocol layer. If you pass a runt frame (too
> short packet) to a Linux NIC driver's transmission function, the NIC
Rename rte_bsf64 to rte_bsf64_safe (this is a "safe" version in
that it prevents undefined behavior by checking if incoming
parameter is zero) and move it to common header.
Signed-off-by: Anatoly Burakov
---
doc/guides/rel_notes/deprecation.rst | 5 +
lib/librte_eal/common/include/rte
Implement missing functions for 32-bit safe bsf, as well as 64-bit
fls and log2.
Signed-off-by: Anatoly Burakov
---
lib/librte_eal/common/include/rte_common.h | 62 +-
1 file changed, 61 insertions(+), 1 deletion(-)
diff --git a/lib/librte_eal/common/include/rte_common.h
b/
RTE_BITMAP_OPTIMIZATIONS was never set to 0 and makes no sense
anyway, so remove all code related to it. Also, drop the "likely"
for bsf64 code, because it's a generic function and we cannot
make any assumptions about likely values of incoming arguments.
Signed-off-by: Anatoly Burakov
---
lib/li
Remove duplicated code and use library version of 64-bit log2.
Signed-off-by: Anatoly Burakov
---
lib/librte_eal/linuxapp/eal/eal_memalloc.c | 17 +
1 file changed, 1 insertion(+), 16 deletions(-)
diff --git a/lib/librte_eal/linuxapp/eal/eal_memalloc.c
b/lib/librte_eal/linuxapp
Remove duplicated code and use library version of 64-bit log2.
Signed-off-by: Anatoly Burakov
---
app/test-pmd/testpmd.c | 17 +
1 file changed, 1 insertion(+), 16 deletions(-)
diff --git a/app/test-pmd/testpmd.c b/app/test-pmd/testpmd.c
index 9c0edcaed..89c6ffd53 100644
--- a/a
> -Original Message-
> From: Burakov, Anatoly
> Sent: Wednesday, November 14, 2018 4:30 PM
> To: dev@dpdk.org
> Cc: Dumitrescu, Cristian ;
> tho...@monjalon.net; Richardson, Bruce ;
> Yigit, Ferruh ; Singh, Jasvinder
>
> Subject: [PATCH v2 1/5] bitmap: remove useless code
>
> RTE_BITMA
Remove duplicated code and use library version of 64-bit log2.
Signed-off-by: Anatoly Burakov
Acked-by: Cristian Dumitrescu
---
app/test-pmd/testpmd.c | 17 +
1 file changed, 1 insertion(+), 16 deletions(-)
diff --git a/app/test-pmd/testpmd.c b/app/test-pmd/testpmd.c
index 9c0e
RTE_BITMAP_OPTIMIZATIONS was never set to 0 and makes no sense
anyway, so remove all code related to it. Also, drop the "likely"
for bsf64 code, because it's a generic function and we cannot
make any assumptions about likely values of incoming arguments.
Signed-off-by: Anatoly Burakov
Acked-by: C
Implement missing functions for 32-bit safe bsf, as well as 64-bit
fls and log2.
Signed-off-by: Anatoly Burakov
Acked-by: Cristian Dumitrescu
---
Notes:
v3:
- Added clarification that pos is not checked
lib/librte_eal/common/include/rte_common.h | 64 +-
1 file cha
Rename rte_bsf64 to rte_bsf64_safe (this is a "safe" version in
that it prevents undefined behavior by checking if incoming
parameter is zero) and move it to common header.
Signed-off-by: Anatoly Burakov
Acked-by: Cristian Dumitrescu
---
Notes:
v3:
- Added clarification that pos is not
Remove duplicated code and use library version of 64-bit log2.
Signed-off-by: Anatoly Burakov
Acked-by: Cristian Dumitrescu
---
lib/librte_eal/linuxapp/eal/eal_memalloc.c | 17 +
1 file changed, 1 insertion(+), 16 deletions(-)
diff --git a/lib/librte_eal/linuxapp/eal/eal_memall
1 - 100 of 137 matches
Mail list logo
