> -Original Message-
> From: Damjan Marion (damarion)
> Sent: Friday, October 25, 2019 2:14 PM
> To: Thomas Monjalon
> Cc: David Marchand ; Anatoly Burakov
> ; Shahaf Shuler ; Ray
> Kinsella ; dev ; Neil Horman
> ; John McNamara ;
> Marko Kovacevic ; Bruce Richardson
>
> Subject: Re: [
On Fri, Oct 25, 2019 at 5:38 PM Reshma Pattan wrote:
>
> 1)pdump, latency, bitrate, ring PMD and test_event_eth_tx_adapter
> unit tests are dependent on ring PMD, so compile those
> tests only when ring PMD is enabled else ignore.
>
> 2)get rid of make file error which was added by bond unit test
27/10/2019 07:14, David Marchand:
> Let's stick to the current model of per library ABI version until the
> new model is in place.
>
> Fixes: 4f25d7d2252f ("ethdev: add return code to device info get function")
>
> Signed-off-by: David Marchand
In order to make it more obvious, it would be nice
Hi All
From: Andrew Rybchenko
> On 10/18/19 7:35 PM, Ferruh Yigit wrote:
> > On 10/2/2019 2:58 PM, Thomas Monjalon wrote:
> >> 24/09/2019 14:03, Matan Azrad:
> >>> From: Ferruh Yigit
> On 9/15/2019 8:48 AM, Matan Azrad wrote:
> > Hi Ferruh
> >
> > From: Ferruh Yigit
> >> On 8
Hi Pavan,
> -Original Message-
> From: Pavan Nikhilesh Bhagavatula
> Sent: Saturday, October 26, 2019 1:06 AM
> To: Gavin Hu (Arm Technology China) ;
> jer...@marvell.com
> Cc: dev@dpdk.org; nd ; nd
> Subject: RE: [dpdk-dev] [PATCH] event/octeontx2: use wfe while waiting for
> head
>
>
On Sun, Oct 27, 2019 at 10:02 AM Thomas Monjalon wrote:
>
> 27/10/2019 07:14, David Marchand:
> > Let's stick to the current model of per library ABI version until the
> > new model is in place.
> >
> > Fixes: 4f25d7d2252f ("ethdev: add return code to device info get function")
> >
> > Signed-off-
07/10/2019 17:45, Bruce Richardson:
> The compat.h header file provided macros for two purposes:
> 1. it provided the macros for marking functions as rte_experimental
> 2. it provided the macros for doing function versioning
[...]
> lib/librte_eal/common/include/rte_compat.h| 70 --
On Sun, Oct 27, 2019 at 7:26 AM David Marchand
wrote:
>
> On Sat, Oct 26, 2019 at 9:18 PM David Marchand
> wrote:
> >
> > On Fri, Oct 25, 2019 at 3:56 PM David Marchand
> > wrote:
> > >
> > > Let's prepare for the ABI freeze.
> > >
> > > The first patches are about changes that had been announce
On Wed, Aug 14, 2019 at 8:37 PM David Christensen
wrote:
>
> gcc 4.8.5 used on RHEL 7.6 can identify a Power 9 CPU but cannot generate
> Power 9 code when the "-mcpu=native" command line argument is used. Test
> whether the compiler can generate Power 9 code and adjust the machine
> setting approp
On Fri, Oct 25, 2019 at 6:41 AM Tiwei Bie wrote:
>
> On Wed, Aug 14, 2019 at 01:36:41PM -0500, David Christensen wrote:
> > Use of %llx print formatting causes meson build error on Power systems with
> > RHEL 7.6 and gcc 4.8.5. Replace with PRIx64 macro.
> >
> > Fixes: 9b62e2da1844 (vhost: regist
On Fri, Oct 25, 2019 at 8:31 PM Ferruh Yigit wrote:
>
> From: Igor Ryzhov
>
> Starting with kernel version 4.10, there are new min/max MTU values in
> net_device structure, which are set to ETH_MIN_MTU and ETH_DATA_LEN by
> default. We should be able to change these values to allow MTU more than
23/10/2019 12:19, Andrzej Ostruszka:
> On 10/7/19 5:45 PM, Bruce Richardson wrote:
> > Adding support for LTO has exposed some issues with how the functions
> > versioning was supported by meson, which was always set to build both
> > shared and static libraries.
> [...]
> > V3: following feedback,
On Thu, Sep 26, 2019 at 4:44 PM Bruce Richardson
wrote:
>
> On Thu, Sep 26, 2019 at 04:11:59PM +0100, Ciara Power wrote:
> > Replaced strncpy and strcpy with strlcpy.
> > Also replaced snprintf with strlcpy where applicable.
> > Using strlcpy is safe practice when copying strings,
> > as it will i
On Fri, Oct 4, 2019 at 1:26 PM Nipun Gupta wrote:
>
> IOVA_PHYS flag is not required in the DPAA config, thus disable it.
> Also, disable the 1588 timer support by default on DPAA2 platform
> due to the performance impact
With this patch, we don't have a single target using the 1588 timer support
18/09/2019 15:32, Ray Kinsella:
> this is cool, good work.
> comments below.
>
> On 17/09/2019 08:57, Andrzej Ostruszka wrote:
> > --- a/lib/librte_distributor/rte_distributor.c
> > +++ b/lib/librte_distributor/rte_distributor.c
> > @@ -32,7 +32,7 @@ EAL_REGISTER_TAILQ(rte_dist_burst_tailq)
> >
Adding John and Marko for doc review.
22/10/2019 13:54, Andrzej Ostruszka:
> --- /dev/null
> +++ b/doc/guides/prog_guide/lto.rst
This new file is not included in the prog guide index.
> @@ -0,0 +1,36 @@
You are missing the mandatory SPDX and Copyright header.
> +Link Time Optimization
> +=
24/09/2019 17:19, Bruce Richardson:
> On Tue, Sep 24, 2019 at 01:04:14PM +0100, Bruce Richardson wrote:
> > The meson build was missing the define to enable pcap port support if
> > libpcap (development) package was found on the build platform. Rather than
> > duplicating the checks for libpcap fou
The queue state defines are internal to the DPDK.
This commit moves them to a private header file.
Signed-off-by: Ori Kam
Reviewed-by: Andrew Rybchenko
---
V6:
- add rte_ethdev_driver include to nfb driver.
---
drivers/net/nfb/nfb_tx.h | 1 +
lib/librte_ethdev/rte_ethdev.h
This commit query and store the hairpin capabilities from the device.
Those capabilities will be used when creating the hairpin queue.
Signed-off-by: Ori Kam
Acked-by: Viacheslav Ovsiienko
---
drivers/net/mlx5/mlx5.h | 4
drivers/net/mlx5/mlx5_devx_cmds.c | 7 +++
2 files ch
This patch set implements the hairpin feature.
The hairpin feature was introduced in RFC[1]
The hairpin feature (different name can be forward) acts as "bump on the wire",
meaning that a packet that is received from the wire can be modified using
offloaded action and then sent back to the wire wit
This commit introduce hairpin queue type.
The hairpin queue in build from Rx queue binded to Tx queue.
It is used to offload traffic coming from the wire and redirect it back
to the wire.
There are 3 new functions:
- rte_eth_dev_hairpin_capability_get
- rte_eth_rx_hairpin_queue_setup
- rte_eth_tx
This commit adds the support for creating Tx hairpin queues.
Hairpin queue is a queue that is created using DevX and only used
by the HW.
Signed-off-by: Ori Kam
Acked-by: Viacheslav Ovsiienko
---
drivers/net/mlx5/mlx5.c | 36 +-
drivers/net/mlx5/mlx5.h | 46
d
This commits adds the hairpin get capabilities function.
Signed-off-by: Ori Kam
Acked-by: Viacheslav Ovsiienko
---
V6:
- add MLX5 hairpin support to release note.
---
doc/guides/rel_notes/release_19_11.rst | 1 +
drivers/net/mlx5/mlx5.c| 2 ++
drivers/net/mlx5/mlx5.h
Currenlty all Tx queues are created using Verbs.
This commit modify the naming so it will not include verbs,
since in next commit a new type will be introduce (hairpin)
Signed-off-by: Ori Kam
Acked-by: Viacheslav Ovsiienko
---
drivers/net/mlx5/mlx5.c | 2 +-
drivers/net/mlx5/mlx5.h
This commit introduce the hairpin queues to the testpmd.
the hairpin queue is configured using --hairpinq=
the hairpin queue adds n queue objects for both the total number
of TX queues and RX queues.
The connection between the queues are 1 to 1, first Rx hairpin queue
will be connected to the first
This commit adds the support for creating Rx hairpin queues.
Hairpin queue is a queue that is created using DevX and only used
by the HW. This results in that all the data part of the RQ is not being
used.
Signed-off-by: Ori Kam
Acked-by: Viacheslav Ovsiienko
---
drivers/net/mlx5/mlx5.c
Since the encap action is not supported in RX, we need to split the
hairpin flow into RX and TX.
Signed-off-by: Ori Kam
Acked-by: Viacheslav Ovsiienko
---
drivers/net/mlx5/mlx5.c| 10 ++
drivers/net/mlx5/mlx5.h| 10 ++
drivers/net/mlx5/mlx5_flow.c | 281 +
When starting the port, in addition to creating the queues
we need to bind the hairpin queues.
Signed-off-by: Ori Kam
Acked-by: Viacheslav Ovsiienko
---
drivers/net/mlx5/mlx5.h | 1 +
drivers/net/mlx5/mlx5_devx_cmds.c | 1 +
drivers/net/mlx5/mlx5_prm.h | 6 +++
drivers/net/ml
Add support for rss on hairpin queues.
Signed-off-by: Ori Kam
Acked-by: Viacheslav Ovsiienko
---
drivers/net/mlx5/mlx5.h | 3 ++
drivers/net/mlx5/mlx5_ethdev.c | 102 ++--
drivers/net/mlx5/mlx5_rss.c | 1 +
drivers/net/mlx5/mlx5_rxq.c |
When using hairpin all traffic from TX hairpin queues should jump
to dedecated table where matching can be done using regesters.
Signed-off-by: Ori Kam
Acked-by: Viacheslav Ovsiienko
---
drivers/net/mlx5/mlx5.h | 2 ++
drivers/net/mlx5/mlx5_flow.c| 60 ++
This commit introduce the setting and matching on regiters.
This item and and action will be used with number of different
features like hairpin, metering, metadata.
Signed-off-by: Ori Kam
Acked-by: Viacheslav Ovsiienko
---
drivers/net/mlx5/mlx5_flow.c| 52 +
drivers/net/mlx5/m
When splitting flows for example in hairpin / metering, there is a need
to combine the flows. This is done using ID.
This commit introduce a simple way to generate such IDs.
The reason why bitmap was not used is due to fact that the release and
allocation are O(n) while in the chosen approch the a
While using ticket lock, cores repeatedly poll the lock variable.
This is replaced by rte_wait_until_equal API.
Running ticketlock_autotest on ThunderX2, Ampere eMAG80, and Arm N1SDP[1],
there were variances between runs, but no notable performance gain or
degradation were seen with and without th
V11:
-add rte_ prefix to the __atomic_load_ex_x funtions(Ananyev Konstantin)
-define the above rte_atomic_load_ex_x funtions even if not
RTE_WAIT_UNTIL_EQUAL_ARCH_DEFINED for future non-wfe usages(Ananyev Konstantin)
-use the above functions for arm specific rte_wait_until_equal_x
functions(Anany
Use the new API to wait in low power state instead of continuous
polling to save CPU cycles and power.
Signed-off-by: Gavin Hu
Reviewed-by: Ruifeng Wang
Reviewed-by: Jerin Jacob
---
drivers/event/opdl/Makefile| 1 +
drivers/event/opdl/meson.build | 1 +
drivers/event/opdl/opdl_ring.c | 5 +
There are two definitions conflicting each other, for more
details, refer to [1].
include/rte_atomic_64.h:19: error: "dmb" redefined [-Werror]
drivers/bus/fslmc/mc/fsl_mc_sys.h:36: note: this is the location of the
previous definition
#define dmb() {__asm__ __volatile__("" : : : "memory"); }
The
The rte_wait_until_equal_xx APIs abstract the functionality of
'polling for a memory location to become equal to a given value'.
Add the RTE_ARM_USE_WFE configuration entry for aarch64, disabled
by default. When it is enabled, the above APIs will call WFE instruction
to save CPU cycles and power.
Use the new API to wait in low power state instead of continuous
polling to save CPU cycles and power.
Signed-off-by: Gavin Hu
Reviewed-by: Ruifeng Wang
Acked-by: Jerin Jacob
---
drivers/net/thunderx/Makefile | 1 +
drivers/net/thunderx/meson.build | 1 +
drivers/net/thunderx/nicvf_rxtx.c
On Fri, Oct 25, 2019 at 9:03 AM Xiaoyun Li wrote:
> This patch adds return value checking and error handling for
> rte_rawdev_en/dequeue_buffers() and rte_eth_link_get().
Different class of issues are fixed in this patch, but the initial
commit mixed different features too...
Please split your pa
Hi Konstantin,
I think all your comments are addressed in v11, you can have a check, thanks
for your review and really appreciate.
/Gavin
On Thu, Oct 10, 2019 at 2:38 PM Ciara Power wrote:
>
> This patchset enables the ioat_rawdev_autotest to be run multiple times. To
> make this possible, changes were made to stop the rawdev on completion of
> the test, and to ensure the initial ring size is as expected.
>
> Ciara Power (2):
> ra
On Wed, Oct 16, 2019 at 3:36 PM Bruce Richardson
wrote:
>
> On Wed, Oct 16, 2019 at 02:16:26PM +0100, Ciara Power wrote:
> > The rawdev xstats_reset function is now enabled. It is called when the
> > ioat autotest completes, to reset all xstat values after they have been
> > modified during testi
On Thu, Oct 17, 2019 at 2:10 PM wrote:
>
> From: Pavan Nikhilesh
>
> Use RTE_DIM instead of re-defining ARRAY_SIZE.
>
> Signed-off-by: Pavan Nikhilesh
> Acked-by: Gage Eads
Applied, thanks.
I had a similar patch in store :-).
If you want to continue on this topic, look at other files in app/
On Thu, Oct 17, 2019 at 2:10 PM wrote:
>
> From: Pavan Nikhilesh
>
> Use RTE_DIM instead of re-defining ARRAY_SIZE.
>
> Signed-off-by: Pavan Nikhilesh
Reviewed-by: David Marchand
Applied, thanks.
--
David Marchand
Glue function mlx5_glue_devx_qp_query() uses wrong directive
HAVE_IBV_DEVX_ASYNC by mistake.
This patch replaces the directive to HAVE_IBV_DEVX_OBJ, as required.
Fixes: 62d6f70f30f4 ("net/mlx5: add glue for queue query via DevX")
Cc: sta...@dpdk.org
Signed-off-by: Dekel Peled
---
drivers/net/ml
> -Original Message-
> From: Dekel Peled
> Sent: Sunday, October 27, 2019 17:33
> To: Matan Azrad ; Slava Ovsiienko
> ; Shahaf Shuler
> Cc: Ori Kam ; dev@dpdk.org; sta...@dpdk.org
> Subject: [PATCH] net/mlx5: fix use of wrong compilation directive
>
> Glue function mlx5_glue_devx_qp_quer
> Bruce Richardson (9):
> examples/bpf: remove from list of examples to build
> examples/ethtool: allow building as part of meson build
> examples/performance-thread: remove warning disabling
> examples/performance-thread: add l3fwd-thread to meson
> examples/performance-thread: add pthre
> -Original Message-
> From: dev On Behalf Of Viacheslav Ovsiienko
> Sent: Thursday, October 24, 2019 4:12 PM
> To: dev@dpdk.org
> Cc: Thomas Monjalon ; arybche...@solarflare.com;
> Matan Azrad ; Raslan Darawsheh
> ; Yongseok Koh
> Subject: [dpdk-dev] [PATCH v3] ethdev: add flow tag
>
Hi Slava,
Some small comments inline.
Please also add deprication for the META cap on TX.
> -Original Message-
> From: dev On Behalf Of Viacheslav Ovsiienko
> Subject: [dpdk-dev] [PATCH v3] ethdev: extend flow metadata
>
> Currently, metadata can be set on egress path via mbuf tx_metad
07/10/2019 13:08, Bruce Richardson:
> From: Pawel Modrak
> --- a/examples/ioat/ioatfwd.c
> +++ b/examples/ioat/ioatfwd.c
> @@ -161,7 +161,6 @@ ioat_enqueue_packets(struct rte_mbuf **pkts,
> rte_mempool_put_bulk(ioat_pktmbuf_pool, (void *)&pkts_copy[i],
> nb_rx - i);
>
> -
>
07/10/2019 13:08, Bruce Richardson:
> From: Marcin Baran
>
> Added guide for IOAT sample app usage and code description.
>
> Signed-off-by: Marcin Baran
> Acked-by: Bruce Richardson
> ---
> doc/guides/sample_app_ug/index.rst | 1 +
> doc/guides/sample_app_ug/intro.rst | 4 +
> doc/guides/
07/10/2019 13:08, Bruce Richardson:
> Marcin Baran (4):
> examples/ioat: add software copy support
> examples/ioat: add rawdev copy mode support
> examples/ioat: add stats printing for each port
> doc/guides/: provide IOAT sample app guide
>
> Pawel Modrak (2):
> examples/ioat: create sa
10/10/2019 12:46, Bruce Richardson:
> The ether header does not need to be packed since that makes no sense for
> structures with only bytes in them, but it should be aligned to a two-byte
> boundary to simplify access to it from code. Other packed structures that
> use this also need to be updated
+Cc Konstantin
14/08/2019 10:54, Feifei Wang:
> The cache-misses problem is very serious when the function
> lpm_cb_parse_ptype is called to read the content of packets. That is
> because the contents of packages previously stored in the cache are
> overwritten by the following instructions or var
24/10/2019 10:07, Hunt, David:
>
> On 23/10/2019 15:32, Thomas Monjalon wrote:
> > +Cc Dave, the maintainer
> >
> >>> Interrupt will not be received when disabling RX interrupt without
> >>> synchronization mechanism sometimes which leads to wake up issue,
> >>> add spinlock to fix it.
> >>>
> >>>
27/09/2019 10:42, Marcin Hajkowski:
> Marcin Hajkowski (4):
> [PATCH v6 1/4] power: fix invalid socket indicator value
> [PATCH v6 2/4] power: extend guest channel API for reading
> [PATCH v6 3/4] power: process incoming confirmation cmds
> [PATCH v6 4/4] power: send confirmation cmd to vm guest
A
24/10/2019 17:07, Bruce Richardson:
> On Thu, Oct 24, 2019 at 03:26:10PM +0100, David Hunt wrote:
> > channel_monitor.c was dependent on i40e driver being available.
> > This is only necessary for the TRAFFIC policy, so use #ifdef's
> > to not call these when i40e not available.
> >
> > Fixes: f14
06/08/2019 13:18, Thomas Monjalon:
> 26/07/2019 12:15, Burakov, Anatoly:
> > So it's biased towards scaling up quickly, but it's doing that over a
> > period. Please correct me if i'm wrong as i'm not really familiar with
> > this codebase, but, assuming the window size is long enough, you could
Currently, metadata can be set on egress path via mbuf tx_metadata field
with PKT_TX_METADATA flag and RTE_FLOW_ITEM_TYPE_META matches metadata.
This patch extends the metadata feature usability.
1) RTE_FLOW_ACTION_TYPE_SET_META
When supporting multiple tables, Tx metadata can also be set by a r
A tag is a transient data which can be used during flow match. This can be
used to store match result from a previous table so that the same pattern
need not be matched again on the next table. Even if outer header is
decapsulated on the previous match, the match result can be kept.
Some device ex
> -Original Message-
> From: dev On Behalf Of Viacheslav Ovsiienko
> Subject: [dpdk-dev] [PATCH v4] ethdev: extend flow metadata
>
> Currently, metadata can be set on egress path via mbuf tx_metadata field
> with PKT_TX_METADATA flag and RTE_FLOW_ITEM_TYPE_META matches
> metadata.
>
>
> -Original Message-
> From: Viacheslav Ovsiienko
> Sent: Sunday, October 27, 2019 8:42 PM
> To: dev@dpdk.org
> Cc: Thomas Monjalon ; Matan Azrad
> ; Ori Kam ; Yongseok Koh
>
> Subject: [PATCH v4] ethdev: add flow tag
>
> A tag is a transient data which can be used during flow match.
On Thu, Oct 24, 2019 at 6:07 PM Burakov, Anatoly
wrote:
>
> On 12-Aug-19 11:03 AM, David Marchand wrote:
> > On Wed, Jul 24, 2019 at 6:08 PM Anatoly Burakov
> > wrote:
> >>
> >> Currently, primary process holds an exclusive lock on the config
> >> file, thereby preventing other primaries from spi
Hi David, Ferruh,
Sorry I was on vacation. Thank you for dealing with this.
Best regards,
Igor
On Sun, Oct 27, 2019 at 1:12 PM David Marchand
wrote:
> On Fri, Oct 25, 2019 at 8:31 PM Ferruh Yigit
> wrote:
> >
> > From: Igor Ryzhov
> >
> > Starting with kernel version 4.10, there are new min/
27/09/2019 14:15, David Hunt:
> Marcin Hajkowski (4):
> 1/4 power: extend guest channel for query freq
> 2/4 power: process cpu freq query
> 3/4 power: add mechanism to disable queries
> 4/4 power: add cmd to query CPU freq
This first patch needs a fix to be able to compile (missing size parameter
30/09/2019 15:00, David Hunt:
> Extend guest channel and sample apps to query CPU capabilities. Useful
> when a guest wants to see if there is a high frequency/priority
> core available for heavier workloads, i.e. Intel SST-BF.
>
> The guest_cli implements an example of this via the query_cpu_caps
Hi Ferruh, Dan,
Sure, I remember last year discussion but now I see the problem in current
implementation.
Ferruh, here is an example:
We have a thread in the application that processes KNI commands from the
kernel.
It receives config_network_if command to set interface up, calls
rte_eth_dev_sta
21/09/2019 20:16, pbhagavat...@marvell.com:
> From: Pavan Nikhilesh
>
> Some kernel modules have '-' in their name when populated in
> '/sys/bus/pci/drivers/' where as the kernel always populates
> '/sys/modules/' with '_'.
> Replace '-' in module name with '_' to avoid script complaining
> that
15/10/2019 14:39, Robin Jarry:
> Running dpdk-pmdinfo.py on Ubuntu 18.04 (bionic) with python 3 and
> pyelftools installed produces no output but no error is reported
> neither:
>
> ~$ python3 usertools/dpdk-pmdinfo.py -r build/app/testpmd
> ~$ echo $?
> 0
>
> While with python 2, it works:
17/10/2019 19:37, Stephen Hemminger:
> On Thu, 17 Oct 2019 19:32:12 +0200
> Robin Jarry wrote:
>
> > When running the dpdk-telemetry-client.py with python 3, we get the
> > following syntax errors:
> >
> > File "usertools/dpdk-telemetry-client.py", line 70
> > print "\nResponse: \n", str
On Sun, Oct 27, 2019 at 1:53 PM Gavin Hu wrote:
[snip]
> diff --git a/lib/librte_eal/common/include/arch/arm/rte_pause_64.h
> b/lib/librte_eal/common/include/arch/arm/rte_pause_64.h
> index 93895d3..1680d7a 100644
> --- a/lib/librte_eal/common/include/arch/arm/rte_pause_64.h
> +++ b/lib/librte_
23/10/2019 07:03, Jerin Jacob:
> On Wed, Oct 23, 2019 at 2:37 AM Honnappa Nagarahalli
> wrote:
> > > > > On Thu, 2019-08-01 at 07:48 +0800, Gavin Hu wrote:
> > > > > > Arm N1 SDP is an infrastructure segment development platform based
> > > > > > on armv8.2-a Neoverse N1 CPU. For more information,
22/10/2019 02:19, Dharmik Thakkar:
> Some of the internal header files have 'rte_' prefix
> and some don't.
> Remove 'rte_' prefix from all internal header files.
>
> Suggested-by: Thomas Monjalon
> Signed-off-by: Dharmik Thakkar
> Reviewed-by: Phil Yang
> Reviewed-by: Ruifeng Wang
Applied, t
> The rte_wait_until_equal_xx APIs abstract the functionality of
> 'polling for a memory location to become equal to a given value'.
>
> Add the RTE_ARM_USE_WFE configuration entry for aarch64, disabled
> by default. When it is enabled, the above APIs will call WFE instruction
> to save CPU cycl
25/10/2019 03:48, Di, ChenxuX:
> From: Yigit, Ferruh
> > On 10/24/2019 2:03 AM, Di ChenxuX wrote:
> > > Hotplug remove cause infinite loops. Fix by canceling port_close
> > > before port_detach function when rmv_port_callback.
> >
> > Can you please give more details/backtrace of how loop happens?
Running ld with '-r' switches the linker to a very special mode where
some other linker options don't make sense.
In particular, '--export-dynamic' normally requires that all global
symbols be included in the dynamic symbol table, but a .o file doesn't
even have a dynamic symbol table.
When given
Thanks Anatoly.
I think now we 're close to sync. You understood it right: the problem is
dma/iommu mappings are not getting created.
1)vfio_mem_event_callback() is getting called later when memory is
allocated from heap but not when external memory is added to heap
2)vfio_dma_mem_map() is called
On Fri, Oct 25, 2019 at 07:45:47AM -0700, Andy Lutomirski wrote:
> But, for uses like DPDK, /sys/.../resource0 seems like a *far* better
> API, since it actually uses the kernel's concept of which io range
> corresponds to which device instead of hoping that the mappings don't
> change out from und
Hi Andy,
On Thu, Oct 24, 2019 at 09:45:56PM -0700, Andy Lutomirski wrote:
> Hi all-
>
> Supporting iopl() in the Linux kernel is becoming a maintainability
> problem. As far as I know, DPDK is the only major modern user of
> iopl().
>
> After doing some research, DPDK uses direct io port access
On Fri, 25 Oct 2019, Stephen Hemminger wrote:
> On Thu, 24 Oct 2019 21:45:56 -0700
> Andy Lutomirski wrote:
> > 3. Use ioperm() instead of iopl().
>
> Ioperm has the wrong thread semantics. All DPDK applications have
> multiple threads and the initialization logic needs to work even
> if the thre
Hi all,
Here is a new LTS release:
https://fast.dpdk.org/rel/dpdk-18.11.3.tar.xz
The git tree is at:
https://dpdk.org/browse/dpdk-stable/?h=18.11
It has 300+ bugfixes since the previous release.
Thanks to the authors who helped with backports and to
the following who helped with
A new DPDK release candidate is ready for testing:
https://git.dpdk.org/dpdk/tag/?id=v19.11-rc1
There are 1045 patches accumulated in the master branch
since the last release. It is more than enough to have a new tag.
Note that this -rc1 is exceptionnally not feature-complete.
We are expec
> -Original Message-
> From: Thomas Monjalon [mailto:tho...@monjalon.net]
> Sent: Saturday, October 26, 2019 7:01 AM
> To: Yu, Jin
> Cc: dev@dpdk.org; Maxime Coquelin ; Liu,
> Changpeng ; Bie, Tiwei ;
> Wang, Zhihong
> Subject: Re: [dpdk-dev] [PATCH v11 0/9] vhost: support inflight sha
> -Original Message-
> From: Adrian Moreno [mailto:amore...@redhat.com]
> Sent: Friday, October 25, 2019 8:21 PM
> To: Liu, Yong ; maxime.coque...@redhat.com; Bie, Tiwei
> ; Wang, Zhihong
> Cc: dev@dpdk.org
> Subject: Re: [PATCH] vhost: fix vhost user virtqueue not accessable
>
> Hi Mar
Compile librte_vhost/vhost_crypto.c needs the rte_hash.h
So we need the librte_hash to be compiled before vhost.
Add the DEPDIRs to make sure this.
Bugzilla ID: 356
Cc: sta...@dpdk.org
Signed-off-by: Jin Yu
---
lib/Makefile | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/lib
Hi david,
> -Original Message-
> From: David Marchand
> Sent: Monday, October 28, 2019 4:50 AM
> To: Gavin Hu (Arm Technology China)
> Cc: dev ; nd ; Ananyev, Konstantin
> ; tho...@monjalon.net; Stephen
> Hemminger ; hemant.agra...@nxp.com;
> jer...@marvell.com; Pavan Nikhilesh ;
> Honnap
Hi Konstantin,
> -Original Message-
> From: Ananyev, Konstantin
> Sent: Monday, October 28, 2019 6:20 AM
> To: Gavin Hu (Arm Technology China) ; dev@dpdk.org
> Cc: nd ; david.march...@redhat.com; tho...@monjalon.net;
> step...@networkplumber.org; hemant.agra...@nxp.com;
> jer...@marvell.c
Hi, Jiaqi
On 10/25, Min, JiaqiX wrote:
>Hi Xiaolong,
>
>> -Original Message-
>> From: Ye, Xiaolong
>> Sent: Friday, October 25, 2019 11:06 AM
>> To: Min, JiaqiX
>> Cc: dev@dpdk.org; Lu, Wenzhuo ; Yang, Qiming
>>
>> Subject: Re: [dpdk-dev] [PATCH] net/ice: set max frame size
>>
>> Hi,
>
Hi, Junyu
> -Original Message-
> From: Jiang, JunyuX
> Sent: Friday, October 25, 2019 4:56 AM
> To: dev@dpdk.org
> Cc: Chas Williams ; Yang, Qiming ;
> Jiang, JunyuX ; sta...@dpdk.org
> Subject: [PATCH] net/bonding: fix segfault using invalid port
>
> Port validation should be prior to ge
Hi Konstantin,
Please see inline.
Thanks,
Anoob
> -Original Message-
> From: Ananyev, Konstantin
> Sent: Friday, October 25, 2019 3:10 PM
> To: Anoob Joseph ; Akhil Goyal
> ; Nicolau, Radu
> Cc: Thomas Monjalon ; Jerin Jacob Kollanukkaran
> ; Narayana Prasad Raju Athreya
> ; Lukas Bart
Hi, Chenxu
> -Original Message-
> From: Di, ChenxuX
> Sent: Friday, October 25, 2019 12:53 PM
> To: dev@dpdk.org
> Cc: Lu, Wenzhuo ; Yang, Qiming
> ; Di, ChenxuX
> Subject: [PATCH] net/ixgbe: fix port close in FreeBSD OS
>
> FreeBSD OS don't support igb_uio interrupt, so it will fail th
On 10/25, Thomas Monjalon wrote:
>25/10/2019 13:47, David Marchand:
>> On Fri, Oct 25, 2019 at 12:30 PM Ferruh Yigit wrote:
>> > > Reviewed-by: Xiaolong Ye
>> > >
>> > > Applied to dpdk-next-net-intel. Thanks.
>> > >
>> >
>> > Should we update the MAINTAINERS file to say that "drivers/raw/ifpga"
Add raw/ifpga to next-net-intel tree.
Signed-off-by: Xiaolong Ye
---
MAINTAINERS | 1 +
1 file changed, 1 insertion(+)
diff --git a/MAINTAINERS b/MAINTAINERS
index 550b7b324..8fb5319ae 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -1134,6 +1134,7 @@ Rawdev Drivers
Intel FPGA
M: Rosen Xu
M:
Hi,
> -Original Message-
> From: Di, ChenxuX
> Sent: Thursday, October 24, 2019 9:03 AM
> To: dev@dpdk.org
> Cc: Lu, Wenzhuo ; Yang, Qiming
> ; Di, ChenxuX
> Subject: [PATCH] app/testpmd: fix device hotplug remove
>
> Hotplug remove cause infinite loops. Fix by canceling port_close befo
On 10/25, Xiao Wang wrote:
>There's a corner case that all the Rx queue mbufs are allocated but none
>of them is used, this patch fixes mbuf free for this case.
>
>Fixes: b6719f8a04bb ("fm10k: release mbuf for vector Rx")
>Cc: sta...@dpdk.org
>
>Signed-off-by: Xiao Wang
>Signed-off-by: Anna Lukin
On 10/25, Ying Wang wrote:
>The original packet type table's hw-ptype index to sw-ptype
>mapping is wrong. This patch fixed the issue.
>
>Fixes: ab9dccf09af9 ("net/ice: support packet type getting")
>Cc: sta...@dpdk.org
>
>Signed-off-by: Ying Wang
>---
> drivers/net/ice/ice_rxtx.c | 312 ++
https://bugs.dpdk.org/show_bug.cgi?id=342
ruifeng.w...@arm.com changed:
What|Removed |Added
Resolution|--- |FIXED
CC|
Virtio spec only set rule that packed ring maximum size is up to 2^15
entries. Do not limit packed ring size to power of two.
Signed-off-by: Marvin Liu
---
lib/librte_vhost/vhost_user.c | 15 ++-
1 file changed, 14 insertions(+), 1 deletion(-)
diff --git a/lib/librte_vhost/vhost_use
Virtio spec only set rule that packed ring maximum size is up to 2^15
entries. Do not limit packed ring size to power of two.
Signed-off-by: Marvin Liu
---
drivers/net/virtio/virtio_ethdev.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/net/virtio/virtio_ethdev.
On 05/16, Shweta Choudaha wrote:
>From: Shweta Choudaha
>
>When PHY reset is blocked as is the case when BMC is connected via NC-SI
>do not set GO_LINKD bit in PHY power management register in dev_stop as
>this will disconnect the PHY. Also, in dev_close clear the GO_LINKD
>bit only if PHY reset i
1 - 100 of 101 matches
Mail list logo
