There is an error with clang (tested with 3.6.1):
examples/ip_pipeline/config_check.c:63:26: error: comparison of integers of
different signs: 'uint32_t' (aka 'unsigned int') and 'int'
[-Werror,-Wsign-compare]
APP_CHECK((app->n_links == __builtin_popcountll(app->port_mask)),
2015-07-03 16:56, Bruce Richardson:
> On Fri, Jul 03, 2015 at 05:45:34PM +0200, Thomas Monjalon wrote:
> > Hi Bruce,
> >
> > 2015-07-03 16:40, Bruce Richardson:
> > > As well as the fast-path functions in the rxtx code, there are also
> > > functions which set up and tear down the descriptor rings
2015-07-03 17:04, Bruce Richardson:
> On Fri, Jul 03, 2015 at 05:46:55PM +0200, Thomas Monjalon wrote:
> > 2015-07-03 16:40, Bruce Richardson:
> > > The function to clear the TX ring when a port was being closed, e.g. on
> > > exit in testpmd, was not checking the mbuf refcnt before freeing it.
> >
* Forward declaration of enum in C++ requires explicit underlying
type definitions.
* This fixes the issue at:
http://dpdk.org/ml/archives/dev/2015-April/017065.html
Signed-off-by: Joongi Kim
---
lib/librte_eal/common/include/arch/x86/rte_cpuflags.h | 4 ++--
lib/librte_eal/common/incl
* C++ requires explicit conversion of void pointer types to other
pointer types.
* This issue was introduced by previous commits 6cf14ce4 and 7755baae8.
Two subsequent commits 2f935c12 and 7621d6a also have the same issue,
I did not fix them because they are NOT headers potentially incl
Nevermind on this issue.
Apparently this was related with a physical issue on the PCIe port. The testpmd
and pktgen tool are now working fine.
EP
From: Posadas, Emerson
Sent: Friday, July 3, 2015 11:45 AM
To: 'dev at dpdk.org'
Cc: Richardson, Bruce
Subject: Re: [dpdk-dev] Failed to initialize
Thanks for the clarification Bruce. But I find this link below in the
documentation which says it should not be used in cases where the
scheduling policy is SCHED_RR because guess it can lead to an endless
spin-lock-wait by the SCHED_RR thread waiting for the lower prio
thread which got pre-empted.
On 2015/07/02 20:32, Bruce Richardson wrote:
> On Tue, Jun 30, 2015 at 05:24:16PM +0900, Tetsuya Mukawa wrote:
>> Currently Linux implementation and BSD implementation have almost same
>> code about pci uio. This patch series cleans up it.
>>
> Overall, patchset looks a good idea. I've made some co
On 2015/07/02 20:11, Bruce Richardson wrote:
> On Tue, Jun 30, 2015 at 05:24:25PM +0900, Tetsuya Mukawa wrote:
>> From: "Tetsuya.Mukawa"
>>
>> The patch consolidates below functions, and implemented in common
>> eal code.
>> - pci_map_resource()
>> - pci_unmap_resource()
>>
>> Signed-off-by: Tet
On 2015/07/02 19:46, Bruce Richardson wrote:
> On Tue, Jun 30, 2015 at 05:24:22PM +0900, Tetsuya Mukawa wrote:
>> From: "Tetsuya.Mukawa"
>>
>> This patch adds a new function called pci_uio_alloc_resource().
>> The function hides how to prepare uio resource in linuxapp and bsdapp.
>> With the funct
On 2015/07/02 19:20, Bruce Richardson wrote:
> On Tue, Jun 30, 2015 at 05:24:21PM +0900, Tetsuya Mukawa wrote:
>> From: "Tetsuya.Mukawa"
>>
>> This patch fixes below.
>> - bsdapp
>> - Use map_id in pci_uio_map_resource().
>> - Fix interface of pci_map_resource().
>> - Move path variable of mapp
On 2015/07/02 18:57, Bruce Richardson wrote:
> On Tue, Jun 30, 2015 at 05:24:19PM +0900, Tetsuya Mukawa wrote:
>> ---
>> lib/librte_eal/bsdapp/eal/eal_pci.c | 14 ++--
>> lib/librte_eal/linuxapp/eal/eal_pci_uio.c | 56
>> ---
>> 2 files changed, 48 insertions
2015-07-03 16:40, Bruce Richardson:
> The function to clear the TX ring when a port was being closed, e.g. on
> exit in testpmd, was not checking the mbuf refcnt before freeing it.
> Since the function in the vector driver to clear the ring after TX does
> not set the pointer to NULL post-free, thi
Hi Bruce,
2015-07-03 16:40, Bruce Richardson:
> As well as the fast-path functions in the rxtx code, there are also
> functions which set up and tear down the descriptor rings. Since these
> are not performance critical functions, there is no need to have them
> extensively optimized, so we add __
On Fri, Jul 03, 2015 at 05:46:55PM +0200, Thomas Monjalon wrote:
> 2015-07-03 16:40, Bruce Richardson:
> > The function to clear the TX ring when a port was being closed, e.g. on
> > exit in testpmd, was not checking the mbuf refcnt before freeing it.
> > Since the function in the vector driver to
On Fri, Jul 03, 2015 at 05:45:34PM +0200, Thomas Monjalon wrote:
> Hi Bruce,
>
> 2015-07-03 16:40, Bruce Richardson:
> > As well as the fast-path functions in the rxtx code, there are also
> > functions which set up and tear down the descriptor rings. Since these
> > are not performance critical f
The significant ABI changes are planned for unified packet type
which will be supported from release 2.2. Here announces that ABI
changes in detail.
Signed-off-by: Helin Zhang
---
doc/guides/rel_notes/abi.rst | 1 +
1 file changed, 1 insertion(+)
v2 changes:
* Added 'struct rte_kni_mbuf' to the
This is what I'm getting from the DEBUG_INIT and DEBUG_DRIVER flags turned on
while running testpmd:
PMD: e1000_setup_link_82571(): e1000_setup_link_82571
PMD: e1000_setup_link_generic(): e1000_setup_link_generic
PMD: e1000_check_reset_block_generic(): e1000_check_reset_block
PMD: e1000_setup_lin
The function to clear the TX ring when a port was being closed, e.g. on
exit in testpmd, was not checking the mbuf refcnt before freeing it.
Since the function in the vector driver to clear the ring after TX does
not set the pointer to NULL post-free, this caused crashes if mbuf
debugging was turne
As well as the fast-path functions in the rxtx code, there are also
functions which set up and tear down the descriptor rings. Since these
are not performance critical functions, there is no need to have them
extensively optimized, so we add __attribute__((cold)) to their
definitions. This has the
When mbuf debug support is turned on in the build time config, crashes will
occur when clearing up the RX/TX rings, if the 10G vector PMD is in use. This
error can be reproduced using testpmd.
This patchset makes the setup/teardown code easier to debug by marking "cold"
code paths as such, and then
As unified packet types are used instead, those old bit masks and
the relevant macros for packet type indication need to be removed.
To avoid breaking ABI compatibility, all the changes would be
enabled by RTE_NEXT_ABI, which is disabled by default.
Signed-off-by: Helin Zhang
---
lib/librte_mbuf
To unify packet types among all PMDs, bit masks of packet type for
'ol_flags' are replaced by unified packet type.
To avoid breaking ABI compatibility, all the changes would be enabled
by RTE_NEXT_ABI, which is disabled by default.
Signed-off-by: Helin Zhang
---
examples/tep_termination/vxlan.c
To unify packet types among all PMDs, bit masks of packet type for
'ol_flags' are replaced by unified packet type.
To avoid breaking ABI compatibility, all the changes would be
enabled by RTE_NEXT_ABI, which is disabled by default.
Signed-off-by: Helin Zhang
---
examples/l3fwd/main.c | 123 +
To unify packet types among all PMDs, bit masks of packet type for
'ol_flags' are replaced by unified packet type.
To avoid breaking ABI compatibility, all the changes would be
enabled by RTE_NEXT_ABI, which is disabled by default.
Signed-off-by: Helin Zhang
---
examples/l3fwd-power/main.c | 8 +
To unify packet types among all PMDs, bit masks of packet type for
'ol_flags' are replaced by unified packet type.
To avoid breaking ABI compatibility, all the changes would be
enabled by RTE_NEXT_ABI, which is disabled by default.
Signed-off-by: Helin Zhang
---
examples/l3fwd-acl/main.c | 29 ++
To unify packet types among all PMDs, bit masks of packet type for
'ol_flags' are replaced by unified packet type.
To avoid breaking ABI compatibility, all the changes would be
enabled by RTE_NEXT_ABI, which is disabled by default.
Signed-off-by: Helin Zhang
---
examples/ip_reassembly/main.c | 9
To unify packet types among all PMDs, bit masks of packet type for
'ol_flags' are replaced by unified packet type.
To avoid breaking ABI compatibility, all the changes would be
enabled by RTE_NEXT_ABI, which is disabled by default.
Signed-off-by: Helin Zhang
---
examples/ip_fragmentation/main.c
Severl useless code lines are added accidently, which blocks packet
type unification. They should be deleted at all.
To avoid breaking ABI compatibility, all the changes would be
enabled by RTE_NEXT_ABI, which is disabled by default.
Signed-off-by: Helin Zhang
---
app/test/packet_burst_generator
To unify packet types among all PMDs, bit masks of packet type for
'ol_flags' are replaced by unified packet type.
To avoid breaking ABI compatibility, all the changes would be
enabled by RTE_NEXT_ABI, which is disabled by default.
Signed-off-by: Helin Zhang
Signed-off-by: Jijiang Liu
---
app/t
To unify packet types among all PMDs, bit masks of packet type for
'ol_flags' are replaced by unified packet type.
To avoid breaking ABI compatibility, all the changes would be
enabled by RTE_NEXT_ABI, which is disabled by default.
Signed-off-by: Helin Zhang
---
app/test-pipeline/pipeline_hash.c
To unify packet types among all PMDs, bit masks of packet type for
'ol_flags' are replaced by unified packet type.
To avoid breaking ABI compatibility, all the changes would be enabled
by RTE_NEXT_ABI, which is disabled by default.
Signed-off-by: Helin Zhang
---
drivers/net/cxgbe/sge.c | 8 +
To unify packet types among all PMDs, bit masks of packet type for
'ol_flags' are replaced by unified packet type.
To avoid breaking ABI compatibility, all the changes would be
enabled by RTE_NEXT_ABI, which is disabled by default.
Signed-off-by: Helin Zhang
---
drivers/net/fm10k/fm10k_rxtx.c |
To unify packet types among all PMDs, bit masks of packet type for
'ol_flags' are replaced by unified packet type.
To avoid breaking ABI compatibility, all the changes would be
enabled by RTE_NEXT_ABI, which is disabled by default.
Signed-off-by: Helin Zhang
---
drivers/net/vmxnet3/vmxnet3_rxtx.
To unify packet types among all PMDs, bit masks of packet type for
'ol_flags' are replaced by unified packet type.
To avoid breaking ABI compatibility, all the changes would be
enabled by RTE_NEXT_ABI, which is disabled by default.
Signed-off-by: Helin Zhang
---
drivers/net/enic/enic_main.c | 26
To unify packet types among all PMDs, bit masks of packet type for
'ol_flags' are replaced by unified packet type.
To avoid breaking ABI compatibility, all the changes would be
enabled by RTE_NEXT_ABI, which is disabled by default.
Signed-off-by: Helin Zhang
---
drivers/net/i40e/i40e_rxtx.c | 55
To unify packet type among all PMDs, bit masks of packet type for
'ol_flags' are replaced by unified packet type.
To avoid breaking ABI compatibility, all the changes would be
enabled by RTE_NEXT_ABI, which is disabled by default.
Note that around 2.5% performance drop (64B) was observed of doing
4
To unify packet types among all PMDs, bit masks of packet type for
'ol_flags' are replaced by unified packet type.
To avoid breaking ABI compatibility, all the changes would be
enabled by RTE_NEXT_ABI, which is disabled by default.
Signed-off-by: Helin Zhang
---
drivers/net/e1000/igb_rxtx.c | 10
As there are only 6 bit flags in ol_flags for indicating packet
types, which is not enough to describe all the possible packet
types hardware can recognize. For example, i40e hardware can
recognize more than 150 packet types. Unified packet type is
composed of L2 type, L3 type, L4 type, tunnel type
In order to unify the packet type, the field of 'packet_type' in 'struct
rte_mbuf'
needs to be extended from 16 to 32 bits. Accordingly, some fields in 'struct
rte_mbuf'
are re-organized to support this change for Vector PMD. As 'struct
rte_kni_mbuf' for
KNI should be right mapped to 'struct rte
Currently only 6 bits which are stored in ol_flags are used to indicate the
packet types. This is not enough, as some NIC hardware can recognize quite
a lot of packet types, e.g i40e hardware can recognize more than 150 packet
types. Hiding those packet types hides hardware offload capabilities whi
On Fri, Jul 03, 2015 at 02:59:15PM +, Emerson Posadas wrote:
> Hello
>
> I?m trying to use the pktgen tool to generate some traffic, but my
> device (Intel
> Corporation 82571EB Gigabit Ethernet) is failing to initialize. Is this
> related with the resolved issue 7.19, even when is not the sam
On 07/03/2015 03:32 PM, Olivier MATZ wrote:
>
>
> On 07/01/2015 11:03 AM, Zoltan Kiss wrote:
>> The current way has a few problems:
>>
>> - if cache->len < n, we copy our elements into the cache first, then
>> into obj_table, that's unnecessary
>> - if n >= cache_size (or the backfill fails),
add new functions igb_dev_free_queues() and em_dev_free_queues()
Signed-off-by: Bernard Iremonger
---
drivers/net/e1000/e1000_ethdev.h |2 ++
drivers/net/e1000/em_ethdev.c|1 +
drivers/net/e1000/em_rxtx.c |6 ++
drivers/net/e1000/igb_ethdev.c |3 ++-
drivers/net/e1
This patch depends on the Port Hotplug Framework.
It implements the eth_dev_uninit functions for rte_em_pmd,
rte_igb_pmd and rte_igbvf_pmd.
Signed-off-by: Bernard Iremonger
---
drivers/net/e1000/e1000_ethdev.h |8 +++-
drivers/net/e1000/em_ethdev.c| 46 ++-
drivers/net/
Changes in v7:
Rebase to latest code.
Remove freeing of queue memory from uninit functions
Add patch 2, free queue memory in close functions
Changes in v6:
Set nb_rx_queues and nb_tx_queues to 0 in uninit functions.
Rebase to latest code.
Changes in v5:
Moved adapter stopped boolean to struct e10
Hi:
If INTX fails, igb_uio falls back to running without IRQ ( refer to the
implementation in igbuio_pci_probe).
On QEMU 0.12.0, the INTX seems to have broken, and the intr_mode falls to
RTE_INTR_MODE_NONE
However this sets the udev->info.irq = 0;
Setting of udev->info.irq to ?0? does not
On 07/01/2015 11:03 AM, Zoltan Kiss wrote:
> The current way has a few problems:
>
> - if cache->len < n, we copy our elements into the cache first, then
> into obj_table, that's unnecessary
> - if n >= cache_size (or the backfill fails), and we can't fulfil the
> request from the ring alone
Hi Maryam,
On 06/26/2015 02:59 PM, Maryam Tahhan wrote:
> Extend rte_eth_xstats_get to retrieve additional stats from the device
> driver as well the ethdev generic stats.
>
> Signed-off-by: Maryam Tahhan
> ---
> drivers/net/ixgbe/ixgbe_ethdev.c | 2 +-
> lib/librte_ether/rte_ethdev.c| 20
On 07/03/2015 03:16 PM, Olivier MATZ wrote:
> Hi Maryam,
>
> On 06/26/2015 02:59 PM, Maryam Tahhan wrote:
>> Implement ixgbe_dev_xstats_reset and ixgbe_dev_xstats_get.
>>
>> Signed-off-by: Maryam Tahhan
>> ---
>> drivers/net/ixgbe/ixgbe_ethdev.c | 85
>> +++
Hi Maryam,
On 06/26/2015 02:59 PM, Maryam Tahhan wrote:
> Implement ixgbe_dev_xstats_reset and ixgbe_dev_xstats_get.
>
> Signed-off-by: Maryam Tahhan
> ---
> drivers/net/ixgbe/ixgbe_ethdev.c | 85
>
> 1 file changed, 85 insertions(+)
>
> diff --git a/d
add i40e_dev_free_queues() function and call it from close() functions.
Signed-off-by: Bernard Iremonger
---
drivers/net/i40e/i40e_ethdev.c|1 +
drivers/net/i40e/i40e_ethdev_vf.c |1 +
drivers/net/i40e/i40e_rxtx.c | 20
drivers/net/i40e/i40e_rxtx.h |
There is a segmentation fault if rxq is NULL
Signed-off-by: Bernard Iremonger
---
drivers/net/i40e/i40e_rxtx.c |5 +
1 files changed, 5 insertions(+), 0 deletions(-)
diff --git a/drivers/net/i40e/i40e_rxtx.c b/drivers/net/i40e/i40e_rxtx.c
index 2de0ac4..2a89e84 100644
--- a/drivers/net/
Signed-off-by: Bernard Iremonger
---
drivers/net/i40e/i40e_ethdev_vf.c |1 +
1 files changed, 1 insertions(+), 0 deletions(-)
diff --git a/drivers/net/i40e/i40e_ethdev_vf.c
b/drivers/net/i40e/i40e_ethdev_vf.c
index b2c32d6..d2b86a7 100644
--- a/drivers/net/i40e/i40e_ethdev_vf.c
+++ b/driver
_clear_cmd() was not being called in failure situations,
resulting in the next command also failing.
Fix several typos.
Signed-off-by: Bernard Iremonger
---
drivers/net/i40e/i40e_ethdev_vf.c | 11 +++
1 files changed, 7 insertions(+), 4 deletions(-)
diff --git a/drivers/net/i40e/i40e_
Increase delay from 50 * 10 to 100 * 20 to avoid i40evf_read_pfmsg() failures.
Signed-off-by: Bernard Iremonger
---
drivers/net/i40e/i40e_ethdev_vf.c |4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/net/i40e/i40e_ethdev_vf.c
b/drivers/net/i40e/i40e_ethdev_vf.c
Signed-off-by: Bernard Iremonger
---
drivers/net/i40e/i40e_ethdev.c |9 +
1 files changed, 9 insertions(+), 0 deletions(-)
diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40e_ethdev.c
index 449785b..bc1bab2 100644
--- a/drivers/net/i40e/i40e_ethdev.c
+++ b/drivers/net
This patch depends on the Port Hotplug Framework.
It implements the eth_dev_uninit functions for rte_i40e_pmd and
rte_i40evf_pmd.
Signed-off-by: Bernard Iremonger
---
drivers/net/i40e/i40e_ethdev.c| 68 -
drivers/net/i40e/i40e_ethdev_vf.c | 45
Changes in V6:
Rebased to latest code.
Removed release of rx and tx queues from uninit() functions.
added patch 7, add function i40e_dev_free_queues() and call from close()
functions.
Changes in V5:
Increased timeout in i40evf_wait_cmd_done()
Set nb_rx_queues and nb_tx_queues to 0 in uninit funct
Hello
I?m trying to use the pktgen tool to generate some traffic, but my
device (Intel
Corporation 82571EB Gigabit Ethernet) is failing to initialize. Is this
related with the resolved issue 7.19, even when is not the same device?
http://dpdk.org/doc/guides/rel_notes/resolved_issues.html
Here is
Add a VXLAN sample guide in the sample_app_ug directory.
It includes:
- Add the overlay networking picture with svg format.
- Add the TEP termination framework picture with svg format.
- Add the tep_termination.rst file
- Change the index.rst file for the above pictures index.
Hi,
I have downloaded DPDK release 2.0 and was looking at how i can install
bifurcated driver support. Can somebody point me towards it?
I dont plan to use the KNI to push the slow path traffic through the kernel
space networking stack as i end up paying some penalty. I intend to use the
bifurcat
On Fri, 3 Jul 2015 16:40:05 +0100
Bruce Richardson wrote:
> --- a/drivers/net/ixgbe/ixgbe_rxtx.c
> +++ b/drivers/net/ixgbe/ixgbe_rxtx.c
> @@ -1757,7 +1757,7 @@ ixgbe_recv_pkts_lro_bulk_alloc(void *rx_queue, struct
> rte_mbuf **rx_pkts,
> * needed. If the memzone is already created, then this
On Fri, 3 Jul 2015 15:37:40 +0530
Prathap T wrote:
> Hi:
>
>
>
> If INTX fails, igb_uio falls back to running without IRQ ( refer to the
> implementation in igbuio_pci_probe).
>
> On QEMU 0.12.0, the INTX seems to have broken, and the intr_mode falls to
> RTE_INTR_MODE_NONE
>
> However this
Feature Add: Rx/Tx flow control support for the i40e
All the Rx/Tx enable/disable operation is done by the F/W, so device driver
need to use the Set PHY Config AD command to trigger the PHY to do the
auto-negotiation, after the tx/rx pause ablility is negotiated, the F/W will
help us to set the
2015-07-03 10:57, Bruce Richardson:
> On Thu, Jul 02, 2015 at 09:04:48PM +, Sanford, Robert wrote:
> > When one adds multiple post-RX-burst callbacks to a queue, their execution
> > order is the opposite of the order in which they are added. For example, we
> > add callback A( ), and then we
Hello, folks!
We have found root of issue.
Intel do not offer wire speed for 64b packets in XL710 at all.
As mentioned in data sheet
http://www.intel.ru/content/dam/www/public/us/en/documents/product-briefs/xl710-10-40-gbe-controller-brief.pdf
we have:
Small packet performance: Maintains wire-r
On Fri, Jul 03, 2015 at 11:05:16AM +0100, Iremonger, Bernard wrote:
> > -Original Message-
> > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Bruce Richardson
> > Sent: Friday, July 3, 2015 10:54 AM
> > To: Thomas Monjalon
> > Cc: dev at dpdk.org
> > Subject: Re: [dpdk-dev] [PATCH
2015-07-03 08:53, Liu, Yong:
> Acked-by: Marvin Liu
Please strip the patch when ack'ing.
It will avoid useless scrolling.
Thanks
2015-07-03 09:16, Gonzalez Monroy, Sergio:
> On 02/07/2015 13:14, Thomas Monjalon wrote:
> > 2015-06-26 16:29, Sergio Gonzalez Monroy:
> >> --- a/MAINTAINERS
> >> +++ b/MAINTAINERS
> >> @@ -73,6 +73,7 @@ F: lib/librte_eal/common/*
> >> F: lib/librte_eal/common/include/*
> >> F: lib/librte_eal/c
Flow classification pipeline implementation is split to two files.
pipeline_flow_classification.c file handles front-end functions (cli
commands parsing) pipeline_flow_classification_ops.c contains
implementation of functions done by pipeline (back-end).
Signed-off-by: Maciej Gajdzica
---
exampl
Routing pipeline implementation is split to two files.
pipeline_routing.c file handles front-end functions (cli commands
parsing) pipeline_routing_ops.c contains implementation of functions
done by pipeline (back-end).
Signed-off-by: Pawel Wodkowski
---
examples/ip_pipeline/Makefile
From: Daniel Mrzyglod
Firewall pipeline implementation is split to two files.
pipeline_firewall.c file handles front-end functions (cli commands
parsing) pipeline_firewall_ops.c contains implementation of functions
done by pipeline (back-end).
Signed-off-by: Daniel Mrzyglod
---
examples/ip_pip
From: Jasvinder Singh
Passthrough pipeline implementation is split to two files.
pipeline_passthrough.c file handles front-end functions (cli commands
parsing) pipeline_passthrough_ops.c contains implementation of functions
done by pipeline (back-end).
Signed-off-by: Jasvinder Singh
---
exampl
Created new folder for config(.cfg) and script(.sh) files.
Signed-off-by: Maciej Gajdzica
---
examples/ip_pipeline/config/ip_pipeline.cfg |9 +++
examples/ip_pipeline/config/ip_pipeline.sh |5 ++
examples/ip_pipeline/config/tm_profile.cfg | 105 +++
examples/ip
Application thread runs pipelines on assigned cores.
Signed-off-by: Maciej Gajdzica
---
examples/ip_pipeline/Makefile |1 +
examples/ip_pipeline/main.c |6 +++
examples/ip_pipeline/thread.c | 110 +
3 files changed, 117 insertions(+)
create mod
From: Jasvinder Singh
Master pipeline is responsible for command line handling and
communicationg with all other pipelines via message queues. Removed
cmdline.c file, as its functionality will be split over multiple
pipeline files.
Signed-off-by: Jasvinder Singh
---
examples/ip_pipeline/Makefi
Moved pipelines to separate folder, removed not needed pipelines and
modified Makefile to match that change.
Signed-off-by: Maciej Gajdzica
---
examples/ip_pipeline/Makefile |9 +-
examples/ip_pipeline/pipeline/pipeline_firewall.c | 313 +
.../pipeline/pipe
After changes in config parser, app params struct is changed and
requires modifications in initialization procedures.
Signed-off-by: Maciej Gajdzica
---
examples/ip_pipeline/Makefile |1 +
examples/ip_pipeline/init.c | 1632 ++---
examples/ip_pipeline/ma
From: Jasvinder Singh
After loading configuration from a file, data integrity is checked.
Signed-off-by: Jasvinder Singh
---
examples/ip_pipeline/Makefile |1 +
examples/ip_pipeline/config_check.c | 396 +++
examples/ip_pipeline/main.c |2
From: Pawel Wodkowski
New syntax of config files is needed for ip_pipeline example
enhancements. Some old files are temporarily disabled in the Makefile.
It is part of a bigger change.
Signed-off-by: Pawel Wodkowski
---
examples/ip_pipeline/Makefile | 17 +-
examples/ip_pipeline/app
This patchset enhances functionality of ip_pipeline application. New config
file syntax is introduced, so parser is changed. Changed structure of the
application. Now every global variable is stored in app_struct in app.h.
Syntax of pipeline cli commands was changed. Implementation of cli commands
On Thu, Jul 02, 2015 at 09:04:48PM +, Sanford, Robert wrote:
> When one adds multiple post-RX-burst callbacks to a queue, their execution
> order is the opposite of the order in which they are added. For example, we
> add callback A( ), and then we add callback B( ). When we call
> rte_eth_r
Update malloc documentation to reflect new implementation details.
Signed-off-by: Sergio Gonzalez Monroy
---
doc/guides/prog_guide/env_abstraction_layer.rst | 220 +-
doc/guides/prog_guide/img/malloc_heap.png | Bin 81329 -> 80952 bytes
doc/guides/prog_guide/index.rst
Announce the creation of dummy malloc library for 2.1 and removal of
such library, now integrated in librte_eal, for 2.2 release.
Signed-off-by: Sergio Gonzalez Monroy
---
doc/guides/rel_notes/abi.rst | 1 +
1 file changed, 1 insertion(+)
diff --git a/doc/guides/rel_notes/abi.rst b/doc/guides/r
Add new unit test for rte_memzone_free API.
Signed-off-by: Sergio Gonzalez Monroy
---
app/test/test_memzone.c | 82 +++--
1 file changed, 80 insertions(+), 2 deletions(-)
diff --git a/app/test/test_memzone.c b/app/test/test_memzone.c
index 6934eee..c3
Implement rte_memzone_free which, as its name implies, would free a
memzone.
Currently memzone are tracked in an array and cannot be free.
To be able to reuse the same array to track memzones, we have to
change how we keep track of reserved memzones.
With this patch, any memzone with addr NULL is
Remove free_memseg field from internal mem config structure as it is
not used anymore.
Also remove code in ivshmem that was setting up free_memseg on init.
Signed-off-by: Sergio Gonzalez Monroy
---
lib/librte_eal/common/include/rte_eal_memconfig.h | 3 ---
lib/librte_eal/linuxapp/eal/eal_ivshmem
During initializaio malloc sets all available memory as part of the heaps.
CONFIG_RTE_MALLOC_MEMZONE_SIZE was used to specify the default memory
block size to expand the heap. The option is not used/relevant anymore,
so we remove it.
Signed-off-by: Sergio Gonzalez Monroy
---
config/common_bsdap
Some unit test are not relevant anymore. It is the case of those malloc
UTs that checked corner cases when allocating MALLOC_MEMZONE_SIZE
chunks, and the case of those memzone UTs relaying of specific free
memsegs of rhte reserved memzone.
Other UTs just need to be update, for example, to calculat
In the current memory hierarchy, memsegs are groups of physically
contiguous hugepages, memzones are slices of memsegs and malloc further
slices memzones into smaller memory chunks.
This patch modifies malloc so it partitions memsegs instead of memzones.
Thus memzones would call malloc internally
Move malloc inside eal and create a new section in MAINTAINERS file for
Memory Allocation in EAL.
Create a dummy malloc library to avoid breaking applications that have
librte_malloc in their DT_NEEDED entries.
This is the first step towards using malloc to allocate memory directly
from memsegs.
Current implemetation allows reserving/creating memzones but not the opposite
(unreserve/free). This affects mempools and other memzone based objects.
>From my point of view, implementing free functionality for memzones would look
like malloc over memsegs.
Thus, this approach moves malloc inside e
On Thu, Jul 02, 2015 at 06:20:39PM +0200, Thomas Monjalon wrote:
> 2015-07-02 14:50, John McNamara:
> > +* Use one sentence per line in a paragraph, i.e., put a newline character
> > + after each period/full stop.
>
> What about adding this?
> Only blank line will generate a newline.
>
> I think
On Fri, Jul 03, 2015 at 05:52:11PM +0900, Tetsuya Mukawa wrote:
> On 2015/07/02 19:46, Bruce Richardson wrote:
> > On Tue, Jun 30, 2015 at 05:24:22PM +0900, Tetsuya Mukawa wrote:
> >> From: "Tetsuya.Mukawa"
> >>
> >> This patch adds a new function called pci_uio_alloc_resource().
> >> The function
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Bruce Richardson
> Sent: Friday, July 3, 2015 10:54 AM
> To: Thomas Monjalon
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH 2/3] doc: added guidelines on dpdk
> documentation
>
> On Thu, Jul 02, 2015 at
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Maciej Gajdzica
> Sent: Friday, July 3, 2015 9:58 AM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH v4 00/11] ip_pipeline: ip_pipeline application
> enhancements
>
>
> Changes in v4:
> - fixed build issue
On 02/07/2015 13:14, Thomas Monjalon wrote:
> 2015-06-26 16:29, Sergio Gonzalez Monroy:
>> --- a/MAINTAINERS
>> +++ b/MAINTAINERS
>> @@ -73,6 +73,7 @@ F: lib/librte_eal/common/*
>> F: lib/librte_eal/common/include/*
>> F: lib/librte_eal/common/include/generic/
>> F: doc/guides/prog_guide/env_
Hi,
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Jijiang Liu
> Sent: Friday, July 03, 2015 2:58 PM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH v2] doc/sample_app_ug:add a VXLAN sample guide
>
> Add a VXLAN sample guide in the sample_app_ug director
Hi,
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of John McNamara
> Sent: Thursday, July 2, 2015 11:16 PM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH v3 0/7] ethdev: add support for ieee1588
> timestamping
>
> This patchset adds ethdev API to enable a
1 - 100 of 108 matches
Mail list logo
