On 20.09.2018 15:50, Anatoly Burakov wrote:
> When NUMA-aware hugepages config option is set, we rely on
> libnuma to tell the kernel to allocate hugepages on a specific
> NUMA node. However, we allocate node mask before we check if
> NUMA is available in the first place, which, according to
> the
> > > > > >
> > > > > > @@ -69,5 +89,13 @@ kni_fifo_get(struct rte_kni_fifo *fifo,
> > > > > > void **data, unsigned num) static inline uint32_t
> > > > > > kni_fifo_count(struct rte_kni_fifo *fifo) {
> > > > > > +#ifdef RTE_USE_C11_MEM_MODEL
> > > > > > + unsigned fifo_write = __atomic_loa
Adding 3DES-ECB & AES-XTS to the list of ciphers supported
Signed-off-by: Anoob Joseph
---
doc/guides/cryptodevs/features/default.ini | 4
1 file changed, 4 insertions(+)
diff --git a/doc/guides/cryptodevs/features/default.ini
b/doc/guides/cryptodevs/features/default.ini
index 92a7ccf..81
-Original Message-
> Date: Fri, 21 Sep 2018 05:48:44 +
> From: Honnappa Nagarahalli
> To: Jerin Jacob
> CC: "Phil Yang (Arm Technology China)" , "dev@dpdk.org"
> , nd , "kkokkilaga...@caviumnetworks.com"
> , "Gavin Hu (Arm Technology China)"
> , "ferruh.yi...@intel.com"
> Subject:
> > > -Original Message-
> > > > Date: Wed, 19 Sep 2018 21:42:39 +0800
> > > > From: Phil Yang
> > > > To: dev@dpdk.org
> > > > CC: n...@arm.com, jerin.ja...@caviumnetworks.com,
> > > > kkokkilaga...@caviumnetworks.com, honnappa.nagaraha...@arm.com,
> > > > gavin...@arm.com
> > > > Subject
Hi Xiaolong,
> -Original Message-
> From: Ye, Xiaolong
> Sent: Friday, September 21, 2018 6:28 AM
> To: dev@dpdk.org; Maxime Coquelin ; Bie,
> Tiwei ; Wang, Zhihong
> Cc: Wang, Xiao W ; Rami Rosen
> ; Wang, Haiyue ; Ye,
> Xiaolong
> Subject: [PATCH v3 2/2] examples/vdpa: introduce a new
-Original Message-
> Date: Thu, 20 Sep 2018 23:11:12 +0530
> From: Nikhil Rao
> To: jerin.ja...@caviumnetworks.com, olivier.m...@6wind.com,
> marko.kovace...@intel.com, john.mcnam...@intel.com
> CC: dev@dpdk.org, Nikhil Rao
> Subject: [PATCH v4 1/5] eventdev: add eth Tx adapter APIs
> X-
If the devices used for DPDK are bound to the "uio_pci_generic" kernel
module, the IOMMU should be disabled in order not to break the IO
transmission because of the virtual / physical address mapping.
The patch clarifies the IOMMU configurations on both x86_64 and arm64
systems.
Signed-off-by: to
Hi Luca and Bruce,
I appreciate for the comments. I ran more test cases in my test bed, and I will
update the patch and push v4 soon.
Thanks a lot for your help!
Br,
Tone
-Original Message-
From: Luca Boccassi
Sent: Wednesday, September 19, 2018 7:54 PM
To: Bruce Richardson ; Tone Zh
Hi Reshma,
> -Original Message-
> From: reshma.pat...@intel.com [mailto:reshma.pat...@intel.com]
> Sent: Thursday, September 20, 2018 8:09 PM
> To: long...@viettel.com.vn
> Cc: dev@dpdk.org
> Subject: RE: Incorrect latencystats implementation
>
>
>
> > -Original Message-
> > Fro
On 9/20/2018 7:37 AM, Yogev Chaimovich wrote:
> 'stats_reset()' callback was missing because the device backend doesn’t
> support it.
> This commit adds a workaround to this and implements the callback by taking a
> snapshot of the stats (SNAPSHOT) each time 'stats_reset()' is called.
> When gett
On 9/20/2018 3:53 PM, Ananyev, Konstantin wrote:
>
>
>>
>> Current Intel tx prepare function does not properly handle the
>> case where only IP checksum is requested, without requesting
>> any L4 checksum or TSO: IP checksum is not properly reset to 0
>> and output packet may contain invalid IP c
On 9/20/2018 5:11 PM, Iremonger, Bernard wrote:
>> -Original Message-
>> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Reshma Pattan
>> Sent: Wednesday, September 19, 2018 3:01 PM
>> To: adrien.mazarg...@6wind.com; dev@dpdk.org
>> Cc: Pattan, Reshma
>> Subject: [dpdk-dev] [PATCH] ap
On 9/10/2018 5:55 PM, Iremonger, Bernard wrote:
>
>> -Original Message-
>> From: Laatz, Kevin
>> Sent: Monday, September 10, 2018 5:29 PM
>> To: dev@dpdk.org
>> Cc: Iremonger, Bernard ; Laatz, Kevin
>>
>> Subject: [PATCH v2] app/testpmd: add check for Rx offload security flag
>>
>> Add a
On 9/11/2018 12:36 PM, Ferruh Yigit wrote:
> On 9/10/2018 4:52 PM, Cian Ferriter wrote:
>> The MAC addresses are generated in a similar manner as in the TAP PMD,
>> where the address is based on the number of PCAP ports created.
>>
>> This is useful for the purposes of debugging DPDK applications u
On 9/19/2018 4:02 PM, Gaëtan Rivet wrote:
> Hi,
>
> for the series,
> Acked-by: Gaetan Rivet
Series applied to dpdk-next-net/master, thanks.
>
> On Fri, Aug 31, 2018 at 05:09:36PM +0100, Andrew Rybchenko wrote:
>> The patch series should be applied on top of [1] and [2].
>>
>> [1] https://patc
On 9/20/2018 2:55 PM, Andrew Rybchenko wrote:
> Since the topic is raised in multicast address list patchset, I'd like
> to highlight it here as well. Current version uses failsafe ops directly
> on sync to synchronize queues state which iterates over all sub-devices.
> For already in sync sub-devi
On 9/19/2018 4:14 PM, Gaëtan Rivet wrote:
> On Fri, Aug 31, 2018 at 05:16:32PM +0100, Andrew Rybchenko wrote:
>> From: Igor Romanov
>>
>> Releasing a queue that is already released by slave may cause a
>> segmentation fault. For example, after a successfull device
>> configuration a queue is set u
On 9/10/2018 12:01 PM, Xiao Wang wrote:
> IFCVF can help to log dirty page in live migration stage,
> each queue's index can be read and configured to support
> VHOST_USER_GET_VRING_BASE and VHOST_USER_SET_VRING_BASE.
>
> Signed-off-by: Xiao Wang
<...>
> +static void
> +ifcvf_used_ring_log(stru
On 9/8/2018 9:30 PM, Rasesh Mody wrote:
> From: Shahed Shaikh
>
> - rename qede_fdir.c to qede_filter.c
> - move all filter code to qede_filter.c
>
> Signed-off-by: Shahed Shaikh
> ---
> drivers/net/qede/Makefile |2 +-
> drivers/net/qede/qede_ethdev.c | 687 +--
Hi Anatoly
>
> Introduce an example application demonstrating the use of
> external memory support. This is a simple application based on
> skeleton app, but instead of using internal DPDK memory, it is
> using externally allocated memory.
>
> The RX/TX and init path is a carbon-copy of skelet
On Thu, Sep 20, 2018 at 3:16 PM Stephen Hemminger <
step...@networkplumber.org> wrote:
> On Thu, 20 Sep 2018 15:02:53 -0500
> Jay Rolette wrote:
>
> > On Thu, Sep 20, 2018 at 1:11 PM Stephen Hemminger <
> > step...@networkplumber.org> wrote:
> >
> > > I wonder if KNI is claiming performance that
The awk code previously read inline in checkpatches.pl
was using -d which is a bash option,
while bash is not the default shell in all distributions.
Now moved to be read from a separate file.
Signed-off-by: Arnon Warshavsky
---
devtools/check-forbidden-tokens.awk | 68 ++
Hello DPDK dev,
I was struggling with the packet fragmentation,
since the dpdk application(l3fwd, ip-fragmentation, ip-assemble) did not
recognized the received packet as the ipv4 header.
At first, I have crafted my own packets when sending ipv4 headers so I
assumed that I was crafting the pack
On Thu, 20 Sep 2018 15:02:53 -0500
Jay Rolette wrote:
> On Thu, Sep 20, 2018 at 1:11 PM Stephen Hemminger <
> step...@networkplumber.org> wrote:
>
> > I wonder if KNI is claiming performance that was never measured on current
> > CPU, OS, DPDK.
> >
> > With single stream and TCP testing on IXG
On Thu, 20 Sep 2018 15:02:53 -0500
Jay Rolette wrote:
> On Thu, Sep 20, 2018 at 1:11 PM Stephen Hemminger <
> step...@networkplumber.org> wrote:
>
> > I wonder if KNI is claiming performance that was never measured on current
> > CPU, OS, DPDK.
> >
> > With single stream and TCP testing on IXG
On Thu, Sep 20, 2018 at 1:11 PM Stephen Hemminger <
step...@networkplumber.org> wrote:
> I wonder if KNI is claiming performance that was never measured on current
> CPU, OS, DPDK.
>
> With single stream and TCP testing on IXGBE (DPDK), I see lowest
> performance with KNI.
>
> Rx
I wonder if KNI is claiming performance that was never measured on current CPU,
OS, DPDK.
With single stream and TCP testing on IXGBE (DPDK), I see lowest performance
with KNI.
Rx Tx
KNI 3.2 Gbit/sec1.3 Gbit/sec
TAP 4.9 4.7
-Original Message-
From: dev on behalf of Yogev Chaimovich
Date: Wednesday, September 19, 2018 at 11:37 PM
To: "dev@dpdk.org"
Cc: Yogev Chaimovich
Subject: [dpdk-dev] [PATCH] vmxnet3: stats_reset implementation
'stats_reset()' callback was missing because the device backend doesn’
> -Original Message-
> From: Thomas Monjalon
> Sent: Thursday, September 20, 2018 5:49 AM
> To: Ferruh Yigit ; Rahul Lakkireddy
> ; Wenzhuo Lu ; Qi
> Zhang ; Xiao Wang ;
> Konstantin Ananyev ; Mody, Rasesh
> ; Patil, Harish ; Shaikh,
> Shahed ; Yong Wang
> Cc: dev@dpdk.org
> Subject: [PAT
Add programmer's guide doc to explain the use of the
Event Ethernet Tx Adapter library.
Signed-off-by: Nikhil Rao
Acked-by: Jerin Jacob
---
MAINTAINERS| 1 +
.../prog_guide/event_ethernet_tx_adapter.rst | 165 +
doc/guides/prog
This patch adds tests for the eth Tx adapter APIs. It also
tests the data path for the rte_service function based
implementation of the APIs.
Signed-off-by: Nikhil Rao
Acked-by: Jerin Jacob
---
test/test/test_event_eth_tx_adapter.c | 699 ++
MAINTAINERS
The caps API allows the application to query if the transmit
stage is implemented in the eventdev PMD or uses the common
rte_service function. The PMD callbacks support the
eventdev PMD implementation of the adapter.
Signed-off-by: Nikhil Rao
Acked-by: Jerin Jacob
---
lib/librte_eventdev/rte_ev
This patch implements the Tx adapter APIs by invoking the
corresponding eventdev PMD callbacks and also provides
the common rte_service function based implementation when
the eventdev PMD support is absent.
Signed-off-by: Nikhil Rao
---
config/rte_config.h|1 +
li
The ethernet Tx adapter abstracts the transmit stage of an
event driven packet processing application. The transmit
stage may be implemented with eventdev PMD support or use a
rte_service function implemented in the adapter. These APIs
provide a common configuration and control interface and
an tra
On 09/18/2018 10:35 AM, Jens Freimann wrote:
> This adds a new forwarding mode to testpmd to simulate
> more realistic behavior of a guest machine engaged in receiving
> and sending packets performing Virtual Network Function (VNF).
>
> The goal is to enable a simple way of measuring performance i
On 9/8/2018 9:30 PM, Rasesh Mody wrote:
> This patchset adds enhancements and fixes for QEDE PMD.
>
> Rasesh Mody (8):
> net/qede/base: fix to handle stag update event
> net/qede/base: add support for OneView APIs
> net/qede/base: get pre-negotiated values for stag and bw
> net/qede: fix t
19/09/2018 18:03, Gaetan Rivet:
> The eth device class can now parse a field name,
> matching the eth_dev name with one passed as
>
>"class=eth,name=xx"
I am not sure what is the purpose of the "name" property.
I think we should not need it to choose a port by its ethdev name.
If you are
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Reshma Pattan
> Sent: Wednesday, September 19, 2018 3:01 PM
> To: adrien.mazarg...@6wind.com; dev@dpdk.org
> Cc: Pattan, Reshma
> Subject: [dpdk-dev] [PATCH] app/testpmd: fix missing jump action in flow
> action
>
19/09/2018 18:03, Gaetan Rivet:
> The vdev bus parses a field "driver", matching
> a vdev driver name with one passed as follows:
>
>"bus=vdev,driver="
I think the property should be "name".
We already have a "driver" category.
So it may be "bus=vdev,name=mytap/driver=tap"
Until now we w
On 19-Sep-18 3:39 PM, Pallantla Poornima wrote:
Disabled the alarm_autotest UT in FreeBSD
Interrupts are not supported in FreeBSD.
Alarm API depends on interrupts, so disabled alarm test on FreeBSD.
Signed-off-by: Pallantla Poornima
---
Reviewed-by: Anatoly Burakov
--
Thanks,
Anatoly
On 15-Sep-18 6:20 AM, Nikolay Nikolaev wrote:
Use the typedef version of struct VhostUserMsg. Also unify the related
parameter name.
Signed-off-by: Nikolay Nikolaev
---
I'm probably missing some background on this, but according to DPDK
coding style guide, typedef structs are to be avoided [
Currently, command-line switches for legacy mem mode or single-file
segments mode are only stored in internal config. This leads to a
situation where these flags have to always match between primary
and secondary, which is bad for usability.
Fix this by storing these flags in the shared config as
The vdpa sample application creates vhost-user sockets by using the
vDPA backend. vDPA stands for vhost Data Path Acceleration which utilizes
virtio ring compatible devices to serve virtio driver directly to enable
datapath acceleration. As vDPA driver can help to set up vhost datapath,
this applic
Signed-off-by: Xiaolong Ye
---
lib/librte_vhost/rte_vdpa.h| 3 +++
lib/librte_vhost/rte_vhost_version.map | 1 +
lib/librte_vhost/vdpa.c| 6 ++
3 files changed, 10 insertions(+)
diff --git a/lib/librte_vhost/rte_vdpa.h b/lib/librte_vhost/rte_vdpa.h
index 90465ca26
Hi,
This patchset introduces vdpa sample to demonstrate the vDPA use case.
v3 changes:
* list cmd would show queue number and supported features of vdpa devices.
* address Xiao's review comments
v2 changes:
* fix a compilation error reported by Rosen
* improve create cmd in interactive mode and
-Original Message-
> Date: Thu, 20 Sep 2018 15:20:53 +
> From: Honnappa Nagarahalli
> To: Jerin Jacob , "Phil Yang (Arm
> Technology China)"
> CC: "dev@dpdk.org" , nd ,
> "kkokkilaga...@caviumnetworks.com" ,
> "Gavin Hu (Arm Technology China)" ,
> "ferruh.yi...@intel.com"
> Subje
This patchset fixes numerous issues with musl compatibility
in the core EAL libraries. It does not fix anything beyond
core EAL (so, PCI driver is still broken, so are a few other
drivers), but it's a good start.
Tested on container with Alpine Linux. Alpine dependencies:
build-base bsd-compat-he
When built against musl, fcntl.h doesn't silently get included.
Fix by including it explicitly.
Bugzilla ID: 33
Signed-off-by: Anatoly Burakov
---
lib/librte_eal/linuxapp/eal/eal_hugepage_info.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/lib/librte_eal/linuxapp/eal/eal_hugepage_info.c
Musl already has PAGE_SIZE defined, and our define clashed with it.
Rename our define to SYS_PAGE_SIZE.
Bugzilla ID: 36
Signed-off-by: Anatoly Burakov
---
drivers/bus/pci/linux/pci_vfio.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/bus/pci/linux/pci_vfio.c b/drivers/bus/pci/li
We use _GNU_SOURCE all over the place, but often times we miss
defining it, resulting in broken builds on musl. Rather than
fixing every library's and driver's and application's makefile,
fix it by simply defining _GNU_SOURCE by default for all
Linuxapp builds.
Signed-off-by: Anatoly Burakov
---
When built against musl, fcntl.h doesn't silently get included.
Fix by including it explicitly.
Bugzilla ID: 31
Signed-off-by: Anatoly Burakov
---
lib/librte_eal/common/eal_common_memory.c | 1 +
lib/librte_eal/linuxapp/eal/eal_memory.c | 1 +
2 files changed, 2 insertions(+)
diff --git a/lib
When built against musl, fcntl.h doesn't silently get included.
Fix by including it explicitly.
Bugzilla ID: 34
Signed-off-by: Anatoly Burakov
---
lib/librte_eal/common/eal_common_fbarray.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/lib/librte_eal/common/eal_common_fbarray.c
b/lib/lib
Musl wraps various string functions such as strlcpy in order to
harden them. However, the fortify wrappers are included without
including the actual string functions being wrapped, which
throws missing definition compile errors. Fix by including
string.h in string functions header.
Signed-off-by:
Musl complains about pthread id being of wrong size, because on
musl, pthread_t is a struct pointer, not an unsinged int. Fix the
printing code by casting pthread id to unsigned pointer type and
adjusting the format specifier to be of appropriate size.
Signed-off-by: Anatoly Burakov
---
lib/libr
> -Original Message-
> > Date: Wed, 19 Sep 2018 21:42:39 +0800
> > From: Phil Yang
> > To: dev@dpdk.org
> > CC: n...@arm.com, jerin.ja...@caviumnetworks.com,
> > kkokkilaga...@caviumnetworks.com, honnappa.nagaraha...@arm.com,
> > gavin...@arm.com
> > Subject: [PATCH v2 2/3] kni: fix kni fi
>
> Current Intel tx prepare function does not properly handle the
> case where only IP checksum is requested, without requesting
> any L4 checksum or TSO: IP checksum is not properly reset to 0
> and output packet may contain invalid IP checksum.
>
> Fixes: 4fb7e803eb1a ("ethdev: add Tx prepa
Hi Konstantin,
On 9/18/2018 6:12 PM, Ananyev, Konstantin wrote:
I am not saying this should be the ONLY way to do as it does not work
very well with non NPU/FPGA class of SoC.
So how about making the proposed IPSec library as plugin/driver to
rte_security.
As I mentioned above, I don't think t
Adding Bruce in CC
Thanks
M.P.Jananee
>-Original Message-
>From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Jananee
>Parthasarathy
>Sent: Thursday, September 20, 2018 6:54 PM
>To: dev@dpdk.org
>Cc: Pattan, Reshma ; Chaitanya Babu, TalluriX
>
>Subject: [dpdk-dev] [PATCH] test: restruc
From: Ian Dolzhansky
Fixes: a46f8d584eb8 ("net/failsafe: add fail-safe PMD")
Cc: sta...@dpdk.org
Signed-off-by: Ian Dolzhansky
Signed-off-by: Andrew Rybchenko
Acked-by: Gaetan Rivet
---
drivers/net/failsafe/failsafe_ops.c | 10 ++
1 file changed, 10 insertions(+)
diff --git a/driver
From: Ian Dolzhansky
Support Tx queue deferred start.
Signed-off-by: Ian Dolzhansky
Signed-off-by: Andrew Rybchenko
Acked-by: Gaetan Rivet
---
doc/guides/nics/features/failsafe.ini | 2 +-
doc/guides/rel_notes/release_18_11.rst | 4 +-
drivers/net/failsafe/failsafe_ether.c | 44 +
Since the topic is raised in multicast address list patchset, I'd like
to highlight it here as well. Current version uses failsafe ops directly
on sync to synchronize queues state which iterates over all sub-devices.
For already in sync sub-devices it does not go to driver since ethdev
functions ch
From: Ian Dolzhansky
Support Rx queue deferred start.
Signed-off-by: Ian Dolzhansky
Signed-off-by: Andrew Rybchenko
Acked-by: Gaetan Rivet
---
doc/guides/nics/features/failsafe.ini | 1 +
doc/guides/rel_notes/release_18_11.rst | 7 ++
drivers/net/failsafe/failsafe_ether.c | 44 ++
From: Ian Dolzhansky
Signed-off-by: Ian Dolzhansky
Signed-off-by: Andrew Rybchenko
Acked-by: Gaetan Rivet
---
app/test-pmd/cmdline.c | 91 +
doc/guides/rel_notes/release_18_11.rst | 8 ++
doc/guides/testpmd_app_ug/testpmd_funcs.rst | 7 ++
3 fil
On Thu, 20 Sep 2018 14:22:08 +0100
Luca Boccassi wrote:
> From: Bruce Richardson
>
> Signed-off-by: Bruce Richardson
> Signed-off-by: Luca Boccassi
Tested on Fedora 28 (sphinx 1.7.5) and RHEL 7.5 (sphinx 1.1.3)
Tested-by: Timothy Redaelli
From: Bruce Richardson
Signed-off-by: Bruce Richardson
Signed-off-by: Luca Boccassi
---
v2: send on behalf of Bruce on request.
- tell sphinx to create .doctrees working files in the
parent of the target directory so that they don't get
installed
- change the output director
On 9/3/18 7:56 PM, Ferruh Yigit wrote:
On 8/29/2018 8:16 AM, Andrew Rybchenko wrote:
From: Ian Dolzhansky
Signed-off-by: Ian Dolzhansky
Signed-off-by: Andrew Rybchenko
---
app/test-pmd/cmdline.c | 91 ++
doc/guides/rel_notes/release_18_11.rst | 6 +
From: Chaitanya Babu Talluri
Divided main test to smaller logical tests.
Registered with UT framework.
Added cleanup of the resources else ring creation fails
during consecutive test runs.
Freed the allocated mempool, rings and uninitalized the drivers.
Signed-off-by: Chaitanya Babu Talluri
---
On Wed, 19 Sep 2018 14:48:27 +0100
Luca Boccassi wrote:
> From: Bruce Richardson
>
> Signed-off-by: Bruce Richardson
> Signed-off-by: Luca Boccassi
> ---
> v2: send on behalf of Bruce on request.
> - tell sphinx to create .doctrees working files in the
> parent of the target directo
Currently, DPDK will skip mapping some areas (or even an entire BAR)
if MSI-X table happens to be in them but is smaller than page size.
Kernels 4.16+ will allow mapping MSI-X BARs [1], and will report this
as a capability flag. Capability flags themselves are also only
supported since kernel 4.6
> -Original Message-
> From: long...@viettel.com.vn [mailto:long...@viettel.com.vn]
> Sent: Wednesday, September 19, 2018 9:17 AM
> To: Pattan, Reshma
> Cc: dev@dpdk.org; Bao-Long Tran
> Subject: Incorrect latencystats implementation
>
>
> I have submit a patch to implement the trivi
20/09/2018 11:09, Gaëtan Rivet:
> On Thu, Sep 20, 2018 at 05:46:37PM +0900, Hideyuki Yamashita wrote:
> > Hello,
> >
> > From dpdk 18.08 release rte_eth_dev_attach and
> > rte_eth_dev_detach becom deprecated API and
> > it is recommended to replace with rte_eal_hotplug_add
> > and rte_eal_hotplu
20/09/2018 14:26, Anatoly Burakov:
> When compiling on FreeBSD, lots of warnings/errors are thrown for
> unused parameter. Fix these by marking the parameters as unused
> in the code.
>
> Fixes: 1009ba1704f9 ("mem: add internal API to get and set segment fd")
> Fixes: 3a44687139eb ("mem: allow que
From: Chas Williams
Some PMDs, especially ones with vector receives, require a minimum number
of receive buffers in order to receive any packets. If the first slave
read leaves less than this number available, a read from the next slave
may return 0 implying that the slave doesn't have any packe
When NUMA-aware hugepages config option is set, we rely on
libnuma to tell the kernel to allocate hugepages on a specific
NUMA node. However, we allocate node mask before we check if
NUMA is available in the first place, which, according to
the manpage [1], causes undefined behaviour.
Fix by only
On Thu, Sep 20, 2018 at 2:28 AM Matan Azrad wrote:
>
> Hi Chas
> Please see small comments.
>
> > From: Chas Williams
> > Some PMDs, especially ones with vector receives, require a minimum
> > number of receive buffers in order to receive any packets. If the first
> > slave
> > read leaves less
When compiling on FreeBSD, lots of warnings/errors are thrown for
unused parameter. Fix these by marking the parameters as unused
in the code.
Fixes: 1009ba1704f9 ("mem: add internal API to get and set segment fd")
Fixes: 3a44687139eb ("mem: allow querying offset into segment fd")
Signed-off-by:
Hi,
Thanks, I have sent a v2.
Any comment on the problem of dropped mbuf that I described in the cover
letter? In our application the max_latency_ns metric is useless since
after running for a while it would always take on obviously incorrect value
(up to a few minutes). I suspect the impact o
On 9/19/18 6:12 PM, Gaëtan Rivet wrote:
On Wed, Sep 19, 2018 at 04:50:57PM +0200, Gaëtan Rivet wrote:
Hi,
Sorry about the delay on this, overall it looks ok;
I have an issue however, see inline.
On Mon, Sep 03, 2018 at 07:55:22AM +0100, Andrew Rybchenko wrote:
From: Evgeny Im
Signed-off-by:
Add API to allow creating new malloc heaps. They will be created
with socket ID's going above RTE_MAX_NUMA_NODES, to avoid clashing
with internal heaps.
Signed-off-by: Anatoly Burakov
---
lib/librte_eal/common/include/rte_malloc.h | 19
lib/librte_eal/common/malloc_heap.c| 30 ++
Add API to detach from existing chunk of external memory in a
process.
Signed-off-by: Anatoly Burakov
---
lib/librte_eal/common/include/rte_malloc.h | 27 ++
lib/librte_eal/common/rte_malloc.c | 27 ++
lib/librte_eal/rte_eal_version.map |
We will need to refer to external heaps in some way. While we use
heap ID's internally, for external API use it has to be something
more user-friendly. So, we will be using a string to uniquely
identify a heap.
Signed-off-by: Anatoly Burakov
---
lib/librte_eal/common/include/rte_malloc_heap.h |
We will be assigning "invalid" socket ID's to external heap, and
malloc will now be able to verify if a supplied socket ID is in
fact a valid one, rendering parameter checks for sockets
obsolete.
This changes the semantics of what we understand by "socket ID",
so document the change in the release
Add an API to add externally allocated memory to malloc heap. The
memory will be stored in memseg lists like regular DPDK memory.
Multiple segments are allowed within a heap. If IOVA table is
not provided, IOVA addresses are filled in with RTE_BAD_IOVA.
Signed-off-by: Anatoly Burakov
---
lib/lib
Previously, to calculate length of memory area covered by a memseg
list, we would've needed to multiply page size by length of fbarray
backing that memseg list. This is not obvious and unnecessarily
low level, so store length in the memseg list itself.
Signed-off-by: Anatoly Burakov
---
drivers/
Introduce an example application demonstrating the use of
external memory support. This is a simple application based on
skeleton app, but instead of using internal DPDK memory, it is
using externally allocated memory.
The RX/TX and init path is a carbon-copy of skeleton app, with
no modifications
When we will be creating external heaps, they will have their own
"fake" socket ID, so add a function that will map the heap name
to its socket ID.
Signed-off-by: Anatoly Burakov
---
lib/librte_eal/common/include/rte_malloc.h | 14
lib/librte_eal/common/rte_malloc.c | 37 +++
Add a guide for external memory sample application. The application
is identical to Basic Forwarding example in everything except parts
of initialization code, so the bits that are identical will not be
described.
It is also not necessary to describe how external memory is being
allocated due to t
We will be assigning "invalid" socket ID's to external heap, and
malloc will now be able to verify if a supplied socket ID is in
fact a valid one, rendering parameter checks for sockets
obsolete.
Signed-off-by: Anatoly Burakov
---
lib/librte_pipeline/rte_pipeline.c | 3 +--
1 file changed, 1 ins
In order to use external memory in multiple processes, we need to
attach to primary process's memseg lists, so add a new API to do
that. It is the responsibility of the user to ensure that memory
is accessible and that it has been previously added to the malloc
heap by another process.
Signed-off-
Add simple unit tests to test external memory support.
The tests are pretty basic and mostly consist of checking
if invalid API calls are handled correctly, plus a simple
allocation/deallocation test for malloc and memzone.
Signed-off-by: Anatoly Burakov
---
test/test/Makefile| 1 +
When we allocate and use DPDK memory, we need to be able to
differentiate between DPDK hugepage segments and segments that
were made part of DPDK but are externally allocated. Add such
a property to memseg lists.
This breaks the ABI, so bump the EAL library ABI version and
document the change in r
Document the addition of external memory support to DPDK.
Signed-off-by: Anatoly Burakov
---
doc/guides/rel_notes/release_18_11.rst | 5 +
1 file changed, 5 insertions(+)
diff --git a/doc/guides/rel_notes/release_18_11.rst
b/doc/guides/rel_notes/release_18_11.rst
index 63bbb1b51..9a05c9980
Add a short chapter on usage of external memory in DPDK to the
Programmer's Guide.
Signed-off-by: Anatoly Burakov
---
.../prog_guide/env_abstraction_layer.rst | 38 +++
1 file changed, 38 insertions(+)
diff --git a/doc/guides/prog_guide/env_abstraction_layer.rst
b/doc/guid
Add an API to remove memory from specified heaps. This will first
check if all elements within the region are free, and that the
region is the original region that was added to the heap (by
comparing its length to length of memory addressed by the
underlying memseg list).
Signed-off-by: Anatoly Bu
We will be assigning "invalid" socket ID's to external heap, and
malloc will now be able to verify if a supplied socket ID is in
fact a valid one, rendering parameter checks for sockets
obsolete.
Signed-off-by: Anatoly Burakov
---
lib/librte_sched/rte_sched.c | 2 +-
1 file changed, 1 insertion(
Add an API to destroy specified heap.
Signed-off-by: Anatoly Burakov
---
lib/librte_eal/common/include/rte_malloc.h | 23 +
lib/librte_eal/common/malloc_heap.c| 22
lib/librte_eal/common/malloc_heap.h| 3 ++
lib/librte_eal/common/rte_malloc.c | 58 ++
Switch over all parts of EAL to use heap ID instead of NUMA node
ID to identify heaps. Heap ID for DPDK-internal heaps is NUMA
node's index within the detected NUMA node list. Heap ID for
external heaps will be order of their creation.
Signed-off-by: Anatoly Burakov
---
config/common_base
We will be assigning "invalid" socket ID's to external heap, and
malloc will now be able to verify if a supplied socket ID is in
fact a valid one, rendering parameter checks for sockets
obsolete.
Signed-off-by: Anatoly Burakov
---
lib/librte_flow_classify/rte_flow_classify.c | 3 +--
1 file chan
This is a proposal to enable using externally allocated memory
in DPDK.
In a nutshell, here is what is being done here:
- Index internal malloc heaps by NUMA node index, rather than NUMA
node itself (external heaps will have ID's in order of creation)
- Add identifier string to malloc heap, to
1 - 100 of 133 matches
Mail list logo
