Hi Wenzhuo,
Thanks for the reply. Yes, i am using ixgbe.
On software front, this is what i am doing.
I am using DPDK 16.04 and pktgen 3.0.00
On my DPDK machine, i have configured RSS rule just for Destination IP
(172.10.10.2).
[image: rss]
The RTE_ETH_VALID_PORTID_OR_ERR_RET macro is used in some places
to check if a port id is valid or not. This commit makes use of it in
some new parts of the code.
Signed-off-by: Mauricio Vasquez B
---
v3:
- use it also in rte_eth_add_rx_callback and rte_eth_add_tx_callback
v2:
- add missed
To avoid multiple stores on fast path, Ethernet drivers
aggregate the writes to data_off, refcnt, nb_segs and port
to an uint64_t data and write the data in one shot
with uint64_t* at >rearm_data address.
Some of the non-IA platforms have store operation overhead
if the store address is not
?Hi All,
It's very basic question, but somehow i am blocked due to this issue.
Please help me out.
I have configured NTUPLE filter in my application with just Destination IP
every ?thing else(SRC IP, S_PORT, D_PORT & proto) is disabled.
But whenever i send packet from any machine, it means
On PPC64, the ioports are mapped in memory. Implement the missing part
of ioport API for PPC64 when using uio. This may also work on other
architectures but it has not been tested.
Signed-off-by: David Marchand
Signed-off-by: Olivier Matz
---
Just resubmitting one patch for this patchset.
v2
Add rte_eth_from_vhost() API to create vhost PMD dynamically from
applications.
Signed-off-by: Ferruh Yigit
---
v2:
* drop rte_ prefix from non-public function
* re-use eth_rx_queue_setup/eth_tx_queue_setup
* pass vdev options as parameter to API
---
drivers/net/vhost/rte_eth_vhost.c
Some architectures (ex: Power8) have a cache line size of 128 bytes,
so the drivers should not expect that prefetching the second part of
the mbuf with rte_prefetch0(>cacheline1) is valid.
This commit add helpers that can be used by drivers to prefetch the
rx or tx part of the mbuf, whatever the
Hi Hiroyuki,
On 05/18/2016 04:41 PM, Hiroyuki Mikita wrote:
> The rte_pktmbuf_detach() function should decrease refcnt on a direct
> buffer.
>
> Signed-off-by: Hiroyuki Mikita
Acked-by: Olivier Matz
Thanks
2016-05-18 17:25, Mauricio V?squez:
> On Wed, May 18, 2016 at 5:01 PM, Thomas Monjalon 6wind.com>
> wrote:
>
> > 2016-05-18 16:41, Mauricio V?squez:
> > > On Wed, May 18, 2016 at 10:15 AM, Thomas Monjalon <
> > thomas.monjalon at 6wind.com
> > > > wrote:
> > >
> > > > 2016-05-17 22:02, Mauricio
On Wed, May 18, 2016 at 07:27:43PM +0530, Jerin Jacob wrote:
> To avoid multiple stores on fast path, Ethernet drivers
> aggregate the writes to data_off, refcnt, nb_segs and port
> to an uint64_t data and write the data in one shot
> with uint64_t* at >rearm_data address.
>
> Some of the non-IA
On Wed, May 18, 2016 at 5:01 PM, Thomas Monjalon
wrote:
> 2016-05-18 16:41, Mauricio V?squez:
> > On Wed, May 18, 2016 at 10:15 AM, Thomas Monjalon <
> thomas.monjalon at 6wind.com
> > > wrote:
> >
> > > 2016-05-17 22:02, Mauricio V?squez:
> > > > On Fri, May 13, 2016 at 6:20 PM, Thomas Monjalon
Modify the compilation makefile to identify C files that export PMD information,
and use that to trigger execution of the pmdinfo binary. If the execution of
pmdinfo is successful, compile the output C file to an object, and use the
linker to do relocatable linking on the resultant object file
Modify the PMD_REGISTER_DRIVER macro, bifurcating it into two
(PMD_REGISTER_DRIVER_PDEV and PMD_REGISTER_DRIVER_VDEV. Both of these do the
same thing the origional macro did, but both add the definition of a string
variable that informs interested parties of the name of the pmd, and the former
pmdinfogen is a tool used to parse object files and build json strings for use
in
later determining hardware support in a dso or application binary. pmdinfo
looks for the non-exported symbol names this_pmd_name and this_pmd_tbl
(where n is a integer counter). It records the name of each of
2016-05-18 16:41, Mauricio V?squez:
> On Wed, May 18, 2016 at 10:15 AM, Thomas Monjalon 6wind.com
> > wrote:
>
> > 2016-05-17 22:02, Mauricio V?squez:
> > > On Fri, May 13, 2016 at 6:20 PM, Thomas Monjalon <
> > thomas.monjalon at 6wind.com>
> > > wrote:
> > > > 2016-04-29 17:23, Mauricio
> > The functions rte_log_cur_msg_loglevel() and rte_log_cur_msg_logtype()
> > return the current log level/type for the message being processed. They
> > are used when implementing a user-defined logging stream.
> >
> > The current log levels and types were stored in a table indexed by the
> >
On Wed, May 18, 2016 at 10:15 AM, Thomas Monjalon wrote:
> 2016-05-17 22:02, Mauricio V?squez:
> > On Fri, May 13, 2016 at 6:20 PM, Thomas Monjalon <
> thomas.monjalon at 6wind.com>
> > wrote:
> > > 2016-04-29 17:23, Mauricio Vasquez B:
> > > > The RTE_ETH_VALID_PORTID_OR_ERR_RET macro is used
2016-05-03 16:03, Ferruh Yigit:
> Add default library output folder to the library search folder.
>
> This is useful for development environment, in production environment
> DPDK libraries already should be in know locations.
>
> Patch removes requirement to set LD_LIBRARY_PATH variable when
2016-05-13 18:41, Santosh Shukla:
> On Fri, May 13, 2016 at 02:50:48PM +0200, Thomas Monjalon wrote:
> > 2016-05-11 19:17, Hemant Agrawal:
> > > IGB_UIO not supported for arm64 arch in kernel so disable.
> >
> > If I understand well, a patch is needed in the kernel to make
> > igb_uio works?
Hi folks.
I want to use stand alone rte_ring and rte_reorder library's.
Without mempool.
For rte_ring I can call rte_ring_get_memsize() function, alloc memory
by self and use rte_ring.
But reorder don't have rte_reorder_get_memsize. Struct
rte_reorder_buffer is private.
Furthermore, code have
On 05/18/2016 03:03 PM, Neil Horman wrote:
> On Wed, May 18, 2016 at 02:48:30PM +0300, Panu Matilainen wrote:
>> On 05/16/2016 11:41 PM, Neil Horman wrote:
>>> This tool searches for the primer sting PMD_DRIVER_INFO= in any ELF binary,
>>> and, if found parses the remainder of the string as a json
On 05/16/2016 11:41 PM, Neil Horman wrote:
> This tool searches for the primer sting PMD_DRIVER_INFO= in any ELF binary,
> and, if found parses the remainder of the string as a json encoded string,
> outputting the results in either a human readable or raw, script parseable
> format
>
>
Hello Jianfeng,
On Wed, Mar 9, 2016 at 2:05 PM, Panu Matilainen wrote:
> On 03/08/2016 07:38 PM, Tan, Jianfeng wrote:
>>
>> Hi Panu,
>>
>> On 3/8/2016 4:54 PM, Panu Matilainen wrote:
>>>
>>> On 03/04/2016 12:05 PM, Jianfeng Tan wrote:
This patch adds option, --avail-cores, to use
2016-05-18 14:48, Panu Matilainen:
> Calling up on the list of requirements from
> http://dpdk.org/ml/archives/dev/2016-May/038324.html, I see a pile of
> technical requirements but perhaps we should stop for a moment to think
> about the use-cases first?
>
> To name some from the top of my
Hi Hiroyuki,
Thanks for submitting a new version.
There are some styling issues in the patch, I highlighted them below
but you can check them by using checkpatch:
DPDK_CHECKPATCH_PATH=/path/to/linux/checkpatch.pl \
scripts/checkpatches.sh file.patch
On 05/17/2016 06:35 PM, Hiroyuki
Signed-off-by: Sergio Gonzalez Monroy
---
doc/guides/sample_app_ug/img/ipsec_endpoints.svg | 850 +
doc/guides/sample_app_ug/ipsec_secgw.rst | 910 ++-
2 files changed, 1400 insertions(+), 360 deletions(-)
create mode 100644
IPSec transport mode support.
Signed-off-by: Sergio Gonzalez Monroy
---
examples/ipsec-secgw/esp.c | 124 ++-
examples/ipsec-secgw/ipsec.h | 1 +
examples/ipsec-secgw/rt.c| 32 +++
examples/ipsec-secgw/sa.c| 39 ++
Support IPSec IPv6 allowing IPv4/IPv6 traffic in IPv4 or IPv6 tunnel.
We need separate Routing (LPM) and SP (ACL) tables for IPv4 and IPv6,
but a common SA table.
Signed-off-by: Sergio Gonzalez Monroy
---
examples/ipsec-secgw/Makefile | 5 +-
examples/ipsec-secgw/esp.c | 128
The application only checks that an SA shoudln't be NULL through ASSERT
(only when debugging is enabled) without properly dealing with the case
of not having an SA for the processed packet.
If no SA is found, drop the packet.
Fixes: d299106e8e31 ("examples/ipsec-secgw: add IPsec sample
Rework implementation moving from function pointers approach, where each
function implements very specific functionality, to a generic function
approach.
Signed-off-by: Sergio Gonzalez Monroy
---
examples/ipsec-secgw/esp.c | 8 +-
examples/ipsec-secgw/esp.h | 9 +-
Add support for building the application with DEBUG=1.
This option adds the compiler stack protection flag and enables extra
output in the application.
Also remove unnecessary VPATH setup.
Signed-off-by: Sergio Gonzalez Monroy
---
examples/ipsec-secgw/Makefile | 5 +++--
1 file changed, 3
Building the application with -O3 and -fstack-protection (default in
Ubuntu) results in the following error:
*** stack smashing detected ***: ./build/ipsec-secgw terminated
The error is caused by storing an 8B value in a 4B variable.
Fixes: d299106e8e31 ("examples/ipsec-secgw: add IPsec sample
Current code fails to correctly check padding sequence for inbound
packets.
Padding sequence starts on 1 but it checks for 0.
Fixes: d299106e8e31 ("examples/ipsec-secgw: add IPsec sample application")
Signed-off-by: Sergio Gonzalez Monroy
---
examples/ipsec-secgw/esp.c | 2 +-
1 file changed,
Update IPSec sample app with IPv6 and Transport mode support.
The series contains some bug fixes to facilitate patch merge.
v2:
- rebase code
- doc improvements
- add missing image file
Sergio Gonzalez Monroy (9):
examples/ipsec-secgw: fix esp padding check
examples/ipsec-secgw: fix
Hi David,
On 05/17/2016 05:54 PM, David Marchand wrote:
> On Tue, May 17, 2016 at 11:59 AM, Olivier Matz
> wrote:
>> + /* mmap the pci resource */
>> + fd = open(filename, O_RDWR);
>> + if (fd < 0) {
>> + RTE_LOG(ERR, EAL, "Cannot open %s: %s\n", filename,
>> +
Remove the deprecation notice and add an entry in the release note
for the changes in mempool allocation.
Signed-off-by: Olivier Matz
---
doc/guides/rel_notes/deprecation.rst | 8
doc/guides/rel_notes/release_16_07.rst | 9 +
2 files changed, 9 insertions(+), 8 deletions(-)
Rework the mempool test to better indicate where it failed,
and, now that this feature is available, add the freeing of the
mempool after the test is done.
Signed-off-by: Olivier Matz
---
app/test/test_mempool.c | 232 +++-
1 file changed, 129
Add a new flag to remove the constraint of having physically contiguous
objects inside a mempool.
Add this flag to the log history mempool to start, but we could add
it in most cases where objects are not mbufs.
Signed-off-by: Olivier Matz
---
lib/librte_eal/common/eal_common_log.c | 2 +-
Check the return value of snprintf to ensure that the name of
the object is not truncated.
By the way, update the test to avoid to trigger an error in
that case.
Signed-off-by: Olivier Matz
---
app/test/test_mempool.c| 12
Now that mempool library provide functions to populate with anonymous
mmap'd memory, we can remove this specific code from test-pmd.
Signed-off-by: Olivier Matz
---
app/test-pmd/Makefile| 4 -
app/test-pmd/mempool_anon.c | 201 ---
Add the following functions to the public mempool API:
- rte_mempool_create_empty()
- rte_mempool_populate_phys()
- rte_mempool_populate_phys_tab()
- rte_mempool_populate_virt()
- rte_mempool_populate_default()
- rte_mempool_populate_anon()
- rte_mempool_free()
Signed-off-by: Olivier Matz
---
Now that we can populate a mempool with any virtual memory,
it is easier to introduce a function to populate a mempool
with memory coming from an anonymous mapping, as it's done
in test-pmd.
The next commit will replace test-pmd anonymous mapping by
this function.
Signed-off-by: Olivier Matz
Instead of creating the internal ring at mempool creation, do
it when populating the mempool with the first memory chunk. The
objective here is to simplify the change of external handler
when it will be introduced.
For instance, this will be possible:
mp = rte_mempool_create_empty(...)
Avoid to have a specific file for that, and remove #ifdefs.
Now that we have introduced a function to populate a mempool
with a virtual area, the support of xen dom0 is much easier.
The only thing we need to do is to convert the guest physical
address into the machine physical address using
The conversion from guest physical address to machine physical address
is fast when the caller knows the memseg corresponding to the gpa.
But in case the user does not know this information, just find it
by browsing the segments. This feature will be used by next commit.
Signed-off-by: Olivier
Introduce a new function rte_mempool_create_empty()
that allocates a mempool that is not populated.
The functions rte_mempool_create() and rte_mempool_xmem_create()
now make use of it, making their code much easier to read.
Currently, they are the only users of rte_mempool_create_empty()
but the
Introduce rte_mempool_free() that:
- unlink the mempool from the global list if it is found
- free all the memory chunks using their free callbacks
- free the internal ring
- free the memzone containing the mempool
Currently this function is only used in error cases when
creating a new mempool,
Storing the pointer to the memzone instead of the physical address
provides more information than just the physical address: for instance,
the memzone flags.
Moreover, keeping the memzone pointer will allow us to free the mempool
(this is done later in the series).
Signed-off-by: Olivier Matz
Introduce a new function rte_mempool_populate_virt() that is now called
by default when hugepages are not supported. This function populate the
mempool with several physically contiguous chunks whose minimum size is
the page size of the system.
Thanks to this, rte_mempool_create() will work
Although the physical address won't be correct in memory segment,
this allows at least to retrieve the physical address using
rte_mem_virt2phy(). Indeed, if the page is not locked, the page
may not be present in physical memory.
With next commit, it allows a mempool to have properly filled
Introduce rte_mempool_populate_default() which allocates
mempool objects in several memzones.
The mempool header is now always allocated in a specific memzone
(not with its objects). Thanks to this modification, we can remove
many specific behavior that was required when hugepages are not
enabled
Update rte_mempool_xmem_size() so that when the page_shift argument is
set to 0, assume that memory is physically contiguous, allowing to
ignore page boundaries. This will be used in the next commits.
By the way, rename the variable 'n' as 'obj_per_page' and avoid the
affectation inside the if().
Introduce a free callback that is passed to the populate* functions,
which is used when freeing a mempool. This is unused now, but as next
commits will populate the mempool with several chunks of memory, we
need a way to free them properly on error.
Later in the series, we will also introduce a
This commit simplifies rte_mempool_xmem_usage().
Since previous commit, the function rte_mempool_xmem_usage() is
now the last user of rte_mempool_obj_mem_iter(). This complex
code can now be moved inside the function. We can get rid of the
callback and do some simplification to make the code more
In the same model than rte_mempool_obj_iter(), introduce
rte_mempool_mem_iter() to iterate the memory chunks attached
to the mempool.
Signed-off-by: Olivier Matz
---
lib/librte_mempool/rte_mempool.c | 16
lib/librte_mempool/rte_mempool.h | 27
Do not use paddr table to store the mempool memory chunks.
This will allow to have several chunks with different virtual addresses.
Signed-off-by: Olivier Matz
---
app/test/test_mempool.c | 2 +-
lib/librte_mempool/rte_mempool.c | 207 ++-
This commit removes MEMPOOL_IS_CONTIG().
The next commits will change the behavior of the mempool library so that
the objects will never be allocated in the same memzone than the mempool
header. Therefore, there is no reason to keep this macro that would
always return 0.
This macro was only used
Store the physical address of the object in its header. It simplifies
rte_mempool_virt2phy() and prepares the removing of the paddr[] table
in the mempool header.
Signed-off-by: Olivier Matz
---
lib/librte_mempool/rte_mempool.c | 17 +++--
lib/librte_mempool/rte_mempool.h | 11
This makes the code of rte_mempool_create() clearer, and it will make
the introduction of external mempool handler easier (in another patch
series). Indeed, this function contains the specific part when a ring is
used, but it could be replaced by something else in the future.
This commit also
Before this patch, the mempool elements were initialized at the time
they were added to the mempool. This patch changes this to do the
initialization of all objects once the mempool is populated, using
rte_mempool_obj_iter() introduced in previous commits.
Thanks to this modification, we are
Use the new rte_mempool_obj_iter() instead the old rte_mempool_obj_iter()
to iterate among objects to audit them (check for cookies).
Signed-off-by: Olivier Matz
---
lib/librte_mempool/rte_mempool.c | 41 ++--
1 file changed, 6 insertions(+), 35 deletions(-)
Now that the mempool objects are chained into a list, we can use it to
browse them. This implies a rework of rte_mempool_obj_iter() API, that
does not need to take as many arguments as before. The previous function
is kept as a private function, and renamed in this commit. It will be
removed in a
In next commits, we will use an iterator to walk through the objects in
mempool in rte_mempool_audit(). This iterator takes a "struct
rte_mempool *" as a parameter because it is assumed that the callback
function can modify the mempool.
The previous approach was to introduce a RTE_DECONST()
This commit removes the const qualifier for the mempool in
rte_mempool_walk() callback prototype.
Indeed, most functions that can be done on a mempool require a non-const
mempool pointer, except the dump and the audit. Therefore, the
mempool_walk() is more useful if the mempool pointer is not
Introduce a list entry in object header so they can be listed and
browsed. The objective is to introduce a more simple way to browse the
elements of a mempool.
The next commits will update rte_mempool_obj_iter() to use this list,
and remove the previous complex implementation.
Signed-off-by:
This commit renames mempool_obj_ctor_t as mempool_obj_cb_t.
In next commits, we will add the ability to populate the
mempool and iterate through objects using the same function.
We will use the same callback type for that. As the callback is
not a constructor anymore, rename it into
Since commits d2e0ca22f and 97e7e685b the headers and trailers
of the mempool are defined as a structure. We can get their
size using a sizeof instead of doing a calculation that will
become wrong at the first structure update.
Signed-off-by: Olivier Matz
---
lib/librte_mempool/rte_mempool.c |
There's no reason to keep this function inlined. Move it to
rte_mempool.c. We need to export the function for when compiling
with shared libraries + debug. We also need to keep the macro,
because we don't want to call an empty function when debug is
disabled.
Signed-off-by: Olivier Matz
---
This commit replaces elt_size by total_elt_size when appropriate.
In some mempool functions, we use the size of the elements as arguments
or variables. There is a confusion between the size including or not
including the header and trailer.
To avoid this confusion:
- update the API documentation
No functional change, just fix some comments and styling issues.
Also avoid to duplicate comments between rte_mempool_create()
and rte_mempool_xmem_create().
Signed-off-by: Olivier Matz
Acked by: Keith Wiles
---
lib/librte_mempool/rte_mempool.c | 17 ++--
This series is a rework of mempool. For those who don't want to read
all the cover letter, here is a sumary:
- it is not possible to allocate large mempools if there is not enough
contiguous memory, this series solves this issue
- introduce new APIs with less arguments: "create, populate,
On 5/14/2016 7:22 PM, Alex Wang wrote:
> From: Alex Wang
>
> Function like 'rte_kni_rx_burst()' keeps allocating
> 'MAX_MBUF_BURST_NUM' mbufs to kni fifo queue unless the
> queue's capacity ('KNI_FIFO_COUNT_MAX') is reached. So,
> if the mempool is under-provisioned, user may run into
> "Out of
On Wed, May 18, 2016 at 10:06 AM, Sergio Gonzalez Monroy
wrote:
> On 17/05/2016 17:40, Thomas Monjalon wrote:
>>
>> 2016-05-12 00:44, Jianfeng Tan:
>>>
>>> This patch adds an option, --huge-trybest, to use a recover mechanism to
>>> the case that there are not so many hugepages (declared in
Hello Sergio,
On Wed, May 18, 2016 at 9:56 AM, Sergio Gonzalez Monroy
wrote:
> On 17/05/2016 17:39, David Marchand wrote:
>>> diff --git a/lib/librte_eal/linuxapp/eal/eal_memory.c
>>> b/lib/librte_eal/linuxapp/eal/eal_memory.c
>>> index 5b9132c..8c77010 100644
>>> ---
On 5/14/2016 7:22 PM, Alex Wang wrote:
> From: Alex Wang
>
> The 'mbufs' alloc/free descriptions for 'rte_kni_tx_burst()'
> and 'rte_kni_rx_burst()' should be inverted.
>
> Signed-off-by: Alex Wang
> ---
> lib/librte_kni/rte_kni.h | 8
> 1 file changed, 4 insertions(+), 4
On Wed, Apr 13, 2016 at 5:03 PM, Thomas Monjalon
wrote:
> After looking at the patches for container support, it appears that
> some changes are needed in the memory management:
> http://thread.gmane.org/gmane.comp.networking.dpdk.devel/32786/focus=32788
>
> I think it is time to collect what
On 16/05/2016 11:42, Tahhan, Maryam wrote:
[..]
>
> Looks Great overall. Is there a need to update prog_guide/poll_mode_drv.rst
> with the new mods?
Yes it will need updating, as the description of rte_eth_xstats is now
incorrect. Will fix in v3..
..Remy
Signed-off-by: Remy Horton
---
doc/guides/rel_notes/release_16_07.rst | 5 +
1 file changed, 5 insertions(+)
diff --git a/doc/guides/rel_notes/release_16_07.rst
b/doc/guides/rel_notes/release_16_07.rst
index f6d543c..bc269b0 100644
--- a/doc/guides/rel_notes/release_16_07.rst
+++
Signed-off-by: Remy Horton
---
examples/Makefile | 1 +
examples/l2fwd-keepalive/Makefile | 4 +-
examples/l2fwd-keepalive/ka-agent/Makefile | 51 ++
examples/l2fwd-keepalive/ka-agent/main.c | 150 +
Signed-off-by: Remy Horton
---
examples/Makefile | 2 +-
lib/librte_eal/bsdapp/eal/rte_eal_version.map | 7
lib/librte_eal/common/include/rte_keepalive.h | 55 -
lib/librte_eal/common/rte_keepalive.c | 48
This patchset adds enhancements to the keepalive core monitoring and
reporting sub-system. The first is support for idled (sleeping and
frequency-stepped) CPU cores, and the second is support for applications
to be notified of active as well as faulted cores. The latter is to allow
core state to
On 05/14/2016 04:06 AM, Rasesh Mody wrote:
>> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
>> Sent: Friday, May 13, 2016 8:38 AM
>>
>> 2016-05-10 13:01, Panu Matilainen:
>>> With gcc >= 6.0, qede base driver fails to build with:
>>> drivers/net/qede/base/ecore_cxt.c: In function
2016-05-17 22:02, Mauricio V?squez:
> On Fri, May 13, 2016 at 6:20 PM, Thomas Monjalon 6wind.com>
> wrote:
> > 2016-04-29 17:23, Mauricio Vasquez B:
> > > The RTE_ETH_VALID_PORTID_OR_ERR_RET macro is used in some places
> > > to check if a port id is valid or not. This commit makes use of it in
>
On 18/05/2016 09:31, Tahhan, Maryam wrote:
[..]
>> +eth_xstats_count_t xstats_count;
>> +/**< Get number of extended statistics. */
>
> Hi Remy
> While reviewing the second patch in this patchset I noticed you aren't
> actually using
> eth_xstats_count_t xstats_count in the
On 17/05/2016 17:40, Thomas Monjalon wrote:
> 2016-05-12 00:44, Jianfeng Tan:
>> This patch adds an option, --huge-trybest, to use a recover mechanism to
>> the case that there are not so many hugepages (declared in sysfs), which
>> can be used. It relys on a mem access to fault-in hugepages, and
On 17/05/2016 17:39, David Marchand wrote:
> Hello Jianfeng,
>
> On Thu, May 12, 2016 at 2:44 AM, Jianfeng Tan
> wrote:
>> This patch adds an option, --huge-trybest, to use a recover mechanism to
>> the case that there are not so many hugepages (declared in sysfs), which
>> can be used. It relys
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Remy Horton
> Sent: Friday, May 6, 2016 12:11 PM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [RFC PATCH v2 3/3] examples/ethtool: add xstats
> display command
>
> Signed-off-by: Remy Horton
> ---
Acked-by: Maryam Tahhan
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Remy Horton
> Sent: Friday, May 6, 2016 12:11 PM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [RFC PATCH v2 2/3] drivers/net/ixgbe: change xstats
> to use integer id
>
> Signed-off-by: Remy Horton
> ---
Acked-by: Maryam Tahhan
>uint8_t stat_idx,
> @@ -1427,6 +1447,10 @@ struct eth_dev_ops {
> eth_stats_reset_t stats_reset; /**< Reset generic device
> statistics. */
> eth_xstats_get_t xstats_get;/**< Get extended device
> statistics. */
>
On Wed, May 18, 2016 at 02:48:30PM +0300, Panu Matilainen wrote:
> On 05/16/2016 11:41 PM, Neil Horman wrote:
> > This tool searches for the primer sting PMD_DRIVER_INFO= in any ELF binary,
> > and, if found parses the remainder of the string as a json encoded string,
> > outputting the results in
The rte_pktmbuf_detach() function should decrease refcnt on a direct
buffer.
Signed-off-by: Hiroyuki Mikita
---
v3:
* fixed rte_pktmbuf_detach() to decrease refcnt.
* free the direct mbuf when refcnt becomes 0.
* added this issue to Resolved Issues in release notes.
v2:
* introduced a new
I think this behavior is not part of the API, it is a bug.
I agree that detach() frees the direct mbuf when refcnt becomes 0,
Konstantin suggests.
It is a right behavior of reference counting.
Regards,
Hiroyuki
2016-05-18 0:45 GMT+09:00 Ananyev, Konstantin :
>
>
>> -Original Message-
>>
92 matches
Mail list logo
