I just quickly looked into the code and instead of releasing memory or simply
set it to NULL (patch:
http://thread.gmane.org/gmane.comp.networking.dpdk.devel/4683), you can zero
it out and it should work perfectly, you can give it a quick try.
//rte_free(ring->buf_info);
memset(ring->buf_info,
> Signed-off-by: Pablo de Lara
Acked-by: Thomas Monjalon
Applied
Thanks
--
Thomas
2014-10-04 20:19, Nicola?s Pernas Maradei:
> Print an error message to the user when trying to start/stop a rx/tx queue and
> this function is not supported by the PMD driver. The patch does not check if
> the return value is -EINVAL because testpmd is already validating the port and
> queue id.
>
> > New in Patch v2:
> >
> > - Fixes an issue in eth_dev_start/stop where a single interface was always
> > opened/closed even though pcap files had been selected for rx/tx streams.
> > - The link_status was not being properly updated in case of using a single
> > interface for rx/tx streams.
> >
> > Fix the descriptor initialization loop, so that it initializes the
> > i40e_tx_desc::cmd_type_offset_bsz for the correct index into the tx_ring
> > array.
> >
> > Previously it would use the index once to initialize the txd local
> > variable, then
> > again when setting cmd_type_offset_bsz.
2014-10-01 23:27, Pablo de Lara:
> RTE_DRV_LOG macro has changed, so now it requires at least 2 arguments,
> which have been added in ixgbe_bypass, specifying log type, to fix
> compilation error
>
> Signed-off-by: Pablo de Lara
Acked-by: Thomas Monjalon
Applied with this commit log:
ixgbe:
2014-10-05 01:16, Keith Wiles:
> When enabling RTE_LIBRTE_MEMPOOL_DEBUG and compiling with clang
> compiler an error occurs, because ifdefed code now includes GCC pragmas.
>
> GCC 4.4 is when push_options and pop_options pragma show up.
>
> Rework to include Thomas?s suggestion to drop push/pop p
> > > Following "options parsing" patchset (commit d7cb626f and 489a9d6c), core
> > > detection is not working correctly on bsd.
> > >
> > > ./x86_64-native-bsdapp-gcc/app/test -c f -n 4 -- -i
> > > [...]
> > > EAL: lcore 0 unavailable
> > > EAL: invalid coremask
> > >
> > > Align bsd to linux:
> >
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Declan Doherty
> Sent: Tuesday, September 30, 2014 10:58 AM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH v4 0/8] link bonding
>
> v4:
> - Rebased to account for changes in master.
> - Fix for rte_eth_bon
Hi Keith,
2014-10-03 21:27, Wiles, Roger Keith:
> Cleanup the code some to remove n_orig variable that was not required.
>
> Update the comments to __mempool_get_bulk to state the correct return value.
>
Hi Thomas:
Comments to this way of split?
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Huawei Xie
> Sent: Wednesday, October 08, 2014 11:55 AM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH v6 00/25] user space vhost library
>
> This set of patches tr
Hi:
I am reading document about distributor library which is added in DPDK 1.7.
The document mentioned the packets are dynamically load balanced between a set
of worker cores.
So I am wondering is this the only reason we need distributor library? What
else could it give us?
Do we have any perfor
Hi,
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Jingjing Wu
> Sent: Thursday, September 25, 2014 7:59 AM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH 1/4] lib/librte_ether: new filter APIs definition
>
> Define new APIs to support configure multi-k
This patch disables compilation complain from lower GCC version (less than 4.6).
Signed-off-by: Changchun Ouyang
---
lib/librte_pmd_ixgbe/Makefile | 5 +
1 file changed, 5 insertions(+)
diff --git a/lib/librte_pmd_ixgbe/Makefile b/lib/librte_pmd_ixgbe/Makefile
index 0b647bd..3588047 100644
Hi,
I have seen that in the Patchwork tool, one of the possible status for a patch
is "Accepted". Who is responsible to change the status to that, the developer
or the maintainer? And when it is acked or applied?
Thanks,
Pablo
Remove n_orig variable as it is not required.
Signed-off-by: Keith Wiles
---
lib/librte_mempool/rte_mempool.h | 9 +++--
1 file changed, 3 insertions(+), 6 deletions(-)
diff --git a/lib/librte_mempool/rte_mempool.h b/lib/librte_mempool/rte_mempool.h
index 95f19f9..c602cbc 100644
--- a/lib/l
> -Original Message-
> From: Neil Horman [mailto:nhorman at tuxdriver.com]
> Sent: Thursday, October 9, 2014 12:37 PM
> To: Pattan, Reshma
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH RFC] librte_reorder: new reorder library
>
> On Thu, Oct 09, 2014 at 10:27:55AM +, Pattan,
Hi Sergio,
2014-10-09 10:37, Sergio Gonzalez Monroy:
> On Wed, Oct 08, 2014 at 07:05:32PM +0200, Thomas Monjalon wrote:
> > Hi Sergio,
> >
> > 2014-10-06 17:09, Sergio Gonzalez Monroy:
> > > --- a/mk/rte.module.mk
> > > +++ b/mk/rte.module.mk
> > > @@ -78,7 +78,7 @@ build: _postbuild
> > > $(MOD
Include external shared libs dependencies when building shared
libraries.
Signed-off-by: Sergio Gonzalez Monroy
---
mk/rte.app.mk | 4 ++--
mk/rte.lib.mk | 8 +++-
mk/rte.sharelib.mk | 7 ++-
3 files changed, 15 insertions(+), 4 deletions(-)
diff --git a/mk/rte.app.mk b/mk/rte
Rewrite rules to avoid duplicated code.
Signed-off-by: Sergio Gonzalez Monroy
---
mk/rte.sharelib.mk | 37 +++--
1 file changed, 15 insertions(+), 22 deletions(-)
diff --git a/mk/rte.sharelib.mk b/mk/rte.sharelib.mk
index b32bd35..8fc6548 100644
--- a/mk/rte.shar
Remove all code related to building a separated static libs, and by default
only build a single/combined library when building static libs.
Build both separated and combined libraries when building shared libs.
Signed-off-by: Sergio Gonzalez Monroy
---
mk/rte.lib.mk | 39 +++-
Remove the configuration variable CONFIG_RTE_BUILD_COMBINE_LIBS and any
code related to it.
Signed-off-by: Sergio Gonzalez Monroy
---
config/common_bsdapp | 3 +--
config/common_linuxapp | 3 +--
mk/rte.lib.mk | 34 --
mk/rte.sdkbuild.mk | 2 +-
Signed-off-by: Sergio Gonzalez Monroy
---
mk/rte.app.mk | 160 +-
1 file changed, 2 insertions(+), 158 deletions(-)
diff --git a/mk/rte.app.mk b/mk/rte.app.mk
index 34dff2a..5e00e67 100644
--- a/mk/rte.app.mk
+++ b/mk/rte.app.mk
@@ -60,164
This patch just enables to link libs with CC as we currently do for apps.
Signed-off-by: Sergio Gonzalez Monroy
---
mk/rte.sharelib.mk | 12 +++-
1 file changed, 11 insertions(+), 1 deletion(-)
diff --git a/mk/rte.sharelib.mk b/mk/rte.sharelib.mk
index c0a811a..73ce709 100644
--- a/mk/r
As per the proposal, this patch set does:
- Remove CONFIG_RTE_BUILD_COMBINE_LIBS as a configuration option.
- For static library, build a single/combined library.
- For shared libraries, build both individual/separated and single/combined
libraries.
- Link apps only against single/combined l
Hi Reshma,
A few comments and questions about your design...
1) How do you envision the reorder library to be used? Based on the
description, it seems like the expectation is that packet order would be
maintained at either the interface/port level or maybe at the RX queue
level. Is that right or
On Thu, Oct 09, 2014 at 10:21:35AM -0700, Matthew Hall wrote:
> On Thu, Oct 09, 2014 at 12:09:42PM -0400, Neil Horman wrote:
> > From what you've said above, sequence assignment needs to occur prior to any
> > order breaking event. That means you either need to do it in individual
> > PMD's
> > o
On Thu, Oct 09, 2014 at 08:16:04AM -0700, Stephen Hemminger wrote:
> On Thu, 9 Oct 2014 11:48:38 +0100
> Daniel Mrzyglod wrote:
>
> > script was expecting /lib/modules/$(uname -r)/kernel/drivers/uio/uio.ko but
> > in fedora 21
> > there are Compressed kernel modules - xz (LZMA)
> >
> > Signed-
Hi Helin,
inline and snipped. Let me know after reading this mail if you believe I
can already submit the v2 with the changes you suggested.
On 09/10/14 10:57, Zhang, Helin wrote:
> [snip]
I don't think the approach of pre-allocating on the first
rte_kni_alloc() would work (I already d
On Thu, Oct 09, 2014 at 02:36:06PM +, Pattan, Reshma wrote:
>
>
> > -Original Message-
> > From: Neil Horman [mailto:nhorman at tuxdriver.com]
> > Sent: Thursday, October 9, 2014 12:37 PM
> > To: Pattan, Reshma
> > Cc: dev at dpdk.org
> > Subject: Re: [dpdk-dev] [PATCH RFC] librte_reo
script was expecting /lib/modules/$(uname -r)/kernel/drivers/uio/uio.ko but in
fedora 21
there are Compressed kernel modules - xz (LZMA)
Signed-off-by: Daniel Mrzyglod
---
tools/setup.sh | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/tools/setup.sh b/tools/setup.sh
index
FYI, on Monday, there will be a presentation of DPDK applications
running with Openstack:
http://sched.co/XnatsK
Best regards,
Vincent
At least on kernels 3.15 or newer, wrong compiler flags are set when building
kernel modules.
v2:
Remove unnecessary reset of CC when being included from the kernel.
Signed-off-by: Sergio Gonzalez Monroy
---
mk/rte.module.mk | 2 +-
mk/toolchain/clang/rte.vars.mk | 5 +
mk/too
On Thu, Oct 09, 2014 at 09:10:04AM +0100, Bruce Richardson wrote:
> On Wed, Oct 08, 2014 at 07:15:04PM +0200, Thomas Monjalon wrote:
> > 2014-10-06 16:57, Sergio Gonzalez Monroy:
> > > At least on kernels 3.15 or newer, DPDK build is broken for CLANG target.
> > > The issue is that the kernel build
Hi Helin,
Inline and snipped. Thanks for the additional comments.
On 09/10/14 10:33, Zhang, Helin wrote:
> [snip]
>>> [snip]
>> It adds a new API call, rte_kni_init(max_kni_ifaces) that shall be
>> called before any call to rte_kni_alloc() if KNI is used.
>>> To avoid the additional inter
On Wed, Oct 08, 2014 at 03:36:55PM -0700, Matthew Hall wrote:
> On Tue, Oct 07, 2014 at 10:55:11AM +0100, Sergio Gonzalez Monroy wrote:
> > I don't see how this particular patch would force people to change their
> > code,
> > though in all likelihood they will have to as a result of ABI changes i
I had tried with Stephen's patch but after stop is done and when we
call start it crashed at vmxnet3_dev_start()->
vmxnet3_dev_rxtx_init()->vmxnet3_post_rx_bufs() as buf_info is freed
and is not allocated again. buf_info is allocated in
vmxnet3_dev_rx_queue_setup() which would be called once at the
On Wed, Oct 08, 2014 at 07:05:32PM +0200, Thomas Monjalon wrote:
> Hi Sergio,
>
> 2014-10-06 17:09, Sergio Gonzalez Monroy:
> > --- a/mk/rte.module.mk
> > +++ b/mk/rte.module.mk
> > @@ -78,7 +78,7 @@ build: _postbuild
> > $(MODULE).ko: $(SRCS_LINKS)
> > @if [ ! -f $(notdir Makefile) ]; then l
> -Original Message-
> From: Neil Horman [mailto:nhorman at tuxdriver.com]
> Sent: Wednesday, October 8, 2014 8:16 PM
> To: Pattan, Reshma
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH RFC] librte_reorder: new reorder library
>
> On Wed, Oct 08, 2014 at 02:11:34PM +, Pattan,
On Wed, Oct 08, 2014 at 05:38:52PM +0200, Thomas Monjalon wrote:
> Please could you explain why patch is needed?
>
Hi Thomas,
Basically this patch just adds missing dependencies to DPDK DSOs. so the
DL knows about them.
As an example, if we were to build an application against DPDK without
addi
On Thu, Oct 09, 2014 at 12:09:42PM -0400, Neil Horman wrote:
> From what you've said above, sequence assignment needs to occur prior to any
> order breaking event. That means you either need to do it in individual PMD's
> on RX, or in the rte_eth library if you want to make it common. On the TX
On Wed, Oct 08, 2014 at 04:07:28PM -0700, Matthew Hall wrote:
> On Wed, Oct 08, 2014 at 06:55:41PM -0400, Neil Horman wrote:
> > I think because there is a possibility that multiple workers may be used
> > for a
> > single tx queue.
> >
> > Neil
>
> OK, so, in my application packets are RX'ed t
On Thu, Oct 09, 2014 at 10:14:21AM +0100, Bruce Richardson wrote:
> Hi Matthew,
>
> What you are doing will indeed work, and it's the way the vast majority of
> the sample apps are written. However, this will not always work for everyone
> else, sadly.
>
> First off, with RSS, there are a numbe
Hi Helin,
Thanks for the comments. Inline and snipped
On 09/10/14 09:32, Zhang, Helin wrote:
> Hi Marc
>
> Good explanation! Thank you very much! I add more comments for your code
> changes.
> One common comment for annotation, in DPDK, we do not use "//" to start
> annotation.
OK, "//" => "/
On Wed, Oct 08, 2014 at 07:15:04PM +0200, Thomas Monjalon wrote:
> 2014-10-06 16:57, Sergio Gonzalez Monroy:
> > At least on kernels 3.15 or newer, DPDK build is broken for CLANG target.
> > The issue is that the kernel build system sets the flags before including
> > DPDK makefile and therefore as
Hi Helin,
On 09/10/14 08:01, Zhang, Helin wrote:
> Hi Marc
>
> Thanks for the idea on KNI! I have comments/questions as follows. Please
> correct me if I am wrong!
>
>> -Original Message-
>> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Marc Sune
>> Sent: Monday, September 29, 2
Hi Marc
Ohh, I made a stupid proposal of adding a field per port. Sorry! I still have
more comments inlined.
> -Original Message-
> From: Marc Sune [mailto:marc.sune at bisdn.de]
> Sent: Thursday, October 9, 2014 4:45 PM
> To: Zhang, Helin
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev]
Hi Marc
Thanks for your clarification! I added more comments.
> -Original Message-
> From: Marc Sune [mailto:marc.sune at bisdn.de]
> Sent: Thursday, October 9, 2014 3:53 PM
> To: Zhang, Helin
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH] KNI: use a memzone pool for KNI alloc/re
On Thu, 9 Oct 2014 11:48:38 +0100
Daniel Mrzyglod wrote:
> script was expecting /lib/modules/$(uname -r)/kernel/drivers/uio/uio.ko but
> in fedora 21
> there are Compressed kernel modules - xz (LZMA)
>
> Signed-off-by: Daniel Mrzyglod
> ---
> tools/setup.sh | 3 ++-
> 1 file changed, 2 inser
Hi Iain
It seems that all you wanted to use is about flow director or ether type filter
which were all done by Jingjing
Yes, after we sent out patches to dpdk.org, there a lot of code changes have
been merged into its master branch. Patching directly to the latest branch
might fail. I have two
On Thu, Oct 09, 2014 at 10:27:55AM +, Pattan, Reshma wrote:
>
>
> > -Original Message-
> > From: Neil Horman [mailto:nhorman at tuxdriver.com]
> > Sent: Wednesday, October 8, 2014 8:16 PM
> > To: Pattan, Reshma
> > Cc: dev at dpdk.org
> > Subject: Re: [dpdk-dev] [PATCH RFC] librte_reo
Hi Marc
Good explanation! Thank you very much! I add more comments for your code
changes.
One common comment for annotation, in DPDK, we do not use "//" to start
annotation.
> -Original Message-
> From: Marc Sune [mailto:marc.sune at bisdn.de]
> Sent: Thursday, October 9, 2014 3:05 PM
>
On Thu, Oct 09, 2014 at 05:20:31AM +, Wu, Jingjing wrote:
> Hi, Neil
>
> To have rte_ip.h include netinet/in.h directly is also a choice.
>
> But netinet/in.h contains a lot of extra stuff, and these may be useless some
> DPDK applications, such as classification.
> rte_ip.h provides a more
Hi Marc
Thanks for the idea on KNI! I have comments/questions as follows. Please
correct me if I am wrong!
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Marc Sune
> Sent: Monday, September 29, 2014 6:16 PM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATC
Hi, Neil
To have rte_ip.h include netinet/in.h directly is also a choice.
But netinet/in.h contains a lot of extra stuff, and these may be useless some
DPDK applications, such as classification.
rte_ip.h provides a more simplify way for the IP protocol layer.
-Original Message-
From: Ne
Hi, Thomas
The reason why I didn't remover all the macros in rte_ip.h is:
netinet/in.h contains a lot of extra references to the sockets definitions,
some external functions declarations, etc. These are useless some DPDK
applications, such as classification.
rte_ip.h provides a more simplify wa
Acked-by: Helin Zhang
Thanks to James for the good catch!
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Jim Harris
> Sent: Thursday, October 2, 2014 6:00 AM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH] i40e: fix i40e_reset_tx_queue
> cmd_type_offse
Hi Iain
1.7.1 might not include newly developed/fixed filter features. See my comments
in specific filters below.
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Iain Rankin
> Sent: Tuesday, October 7, 2014 10:37 PM
> To: dev at dpdk.org
> Subject: [dpdk-de
vhost lib is turned off by default.
vhost lib is based on cuse, which requires fuse development package to be
installed.
Signed-off-by: Huawei Xie
---
config/common_linuxapp| 8
lib/Makefile | 1 +
lib/librte_vhost/Makefile | 48 ++
1) FIXME:concurrent calls to vhost set mem table from different guests could
cause mem_temp to be overrided.
2) TODO:cmpset cost quite some cpu cyles. allow app to disable this feature if
there is no contention in real workload.
3) FIXME:fix scatter gather mbuf copy to vhost vring chained buffers
fixed serious coding style issues reported by checkpatch
Signed-off-by: Huawei Xie
---
lib/librte_vhost/rte_virtio_net.h | 85 +++--
lib/librte_vhost/vhost-net-cdev.c | 252 +++---
lib/librte_vhost/vhost-net-cdev.h | 33 +++--
lib/librte_vhost/vhost_rxtx
priv field could be used to store application specific context.
Signed-off-by: Huawei Xie
---
lib/librte_vhost/rte_virtio_net.h | 1 +
1 file changed, 1 insertion(+)
diff --git a/lib/librte_vhost/rte_virtio_net.h
b/lib/librte_vhost/rte_virtio_net.h
index 13d75fe..adf35e4 100644
--- a/lib/librt
add "static" for some variable definitions
Signed-off-by: Huawei Xie
---
lib/librte_vhost/vhost-net-cdev.c | 8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/lib/librte_vhost/vhost-net-cdev.c
b/lib/librte_vhost/vhost-net-cdev.c
index 4dbd295..ce9f4ae 100644
--- a/lib/li
header file cleanups
Signed-off-by: Huawei Xie
---
lib/librte_vhost/rte_virtio_net.h | 9 +
lib/librte_vhost/vhost-net-cdev.c | 2 +-
lib/librte_vhost/vhost-net-cdev.h | 11 ++-
lib/librte_vhost/vhost_rxtx.c | 26 ++
lib/librte_vhost/virtio-net.c
VHOST_SUPPORTED_FEATURES is the feature mask that vhost lib supports.
VHOST_FEATURES is the feature mask vhost currently supports after some features
are turned on/off.
Signed-off-by: Huawei Xie
---
lib/librte_vhost/virtio-net.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff -
define PRINT_PACKET and LOG_DEBUG macro
Signed-off-by: Huawei Xie
---
lib/librte_vhost/vhost-net-cdev.h | 31 +++
1 file changed, 31 insertions(+)
diff --git a/lib/librte_vhost/vhost-net-cdev.h
b/lib/librte_vhost/vhost-net-cdev.h
index fa67caf..6885bbf 100644
--- a/
rte_vring_available_entries returns the available number of vring entries.
gpa_to_vva translates guest physical address to vhost virtual address.
rte_vhost_feature_disable/enable disables or enables VHOST features.
rte_vhost_feature_get returns currently supported VHOST features.
rte_vhost_enqueue_
rename init_virtio_net as rte_vhost_callback_register API.
rte_vhost_callback_register register the callbacks called when a
vhost device is created and ready to be added to data processing core
or is de-actived by guest.
Signed-off-by: Huawei Xie
---
lib/librte_vhost/rte_virtio_net.h | 4 ++--
rename register_cuse_device as rte_vhost_driver_register API.
rename start_session_loop as rte_vhost_driver_session_start API.
Signed-off-by: Huawei Xie
---
lib/librte_vhost/rte_virtio_net.h | 5 +
lib/librte_vhost/vhost-net-cdev.c | 4 ++--
lib/librte_vhost/vhost-net-cdev.h | 2 --
3 files
vhost_net_device_ops is internal implementation in vhost lib.
register_cuse_device will be vhost driver register API. There is no need for it
to know
the internal vhost ops. Instead, that ops is retrieved in register_cuse_device
through get_virtio_net_callbacks.
Signed-off-by: Huawei Xie
---
l
Signed-off-by: Huawei Xie
---
lib/librte_vhost/vhost-net-cdev.c | 16 +---
1 file changed, 5 insertions(+), 11 deletions(-)
diff --git a/lib/librte_vhost/vhost-net-cdev.c
b/lib/librte_vhost/vhost-net-cdev.c
index 8cf806a..957f0c8 100644
--- a/lib/librte_vhost/vhost-net-cdev.c
+++ b/
This structure is moved to virtio_net.c. It is related to internal virtio
device management,
so it should not be exposed to other files.
Signed-off-by: Huawei Xie
---
lib/librte_vhost/rte_virtio_net.h | 9 -
lib/librte_vhost/virtio-net.c | 8
2 files changed, 8 insertions(
rte_vhost_enqueue_burst will call virtio_dev_rx and virtio_dev_merge_rx
respectively depending on
whether merge-able feature is negotiated or not in the vhost device.
virtio_dev_merge_tx is renamed to rte_vhost_dequeue_burst.
Signed-off-by: Huawei Xie
---
lib/librte_vhost/vhost_rxtx.c | 16
define PACKET_BURST
Signed-off-by: Huawei Xie
---
lib/librte_vhost/vhost_rxtx.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/lib/librte_vhost/vhost_rxtx.c b/lib/librte_vhost/vhost_rxtx.c
index ab76512..0690e23 100644
--- a/lib/librte_vhost/vhost_rxtx.c
+++ b/lib/librte_vhost/vhost_rxtx.
queue_id parameter is added to vhost rx/tx functions for multiple queue support
in future.
Signed-off-by: Huawei Xie
---
lib/librte_vhost/vhost_rxtx.c | 20 +---
1 file changed, 17 insertions(+), 3 deletions(-)
diff --git a/lib/librte_vhost/vhost_rxtx.c b/lib/librte_vhost/vhost
As a lib, we have no idea the app defined mbuf size. This patch will calculate
mbuf size dynamically.
Signed-off-by: Huawei Xie
---
lib/librte_vhost/vhost_rxtx.c | 17 +++--
1 file changed, 7 insertions(+), 10 deletions(-)
diff --git a/lib/librte_vhost/vhost_rxtx.c b/lib/librte_vhos
This patch makes virtio_dev_merge_tx return the received packets to app layer.
Previously virtio_tx_route is called to route these packets and then free them.
Signed-off-by: Huawei Xie
---
lib/librte_vhost/vhost_rxtx.c | 12 +++-
1 file changed, 7 insertions(+), 5 deletions(-)
diff --g
retry is used to wait some time and retry when there are not enough descriptors.
App could implement this policy easily if it needs.
Signed-off-by: Huawei Xie
---
lib/librte_vhost/vhost_rxtx.c | 34 --
1 file changed, 34 deletions(-)
diff --git a/lib/librte_vhost
Currently zero copy feature isn't generic as it couples closely with nic. It
isn't put in
the vhost lib in this version.
gpa(guest physical address) to hpa(host physical address) mapping region logic
is removed.
Signed-off-by: Huawei Xie
---
lib/librte_vhost/rte_virtio_net.h | 19
lib/li
The following logics will be moved to vhost example.
1. mac learning, which is used to learn the mac address from the first
transmitted packet of guest and bind
the vhost device to a queue in a pool of VMDQ.
2. VMDQ mac/vlan filter: Each pool the vhost device is bind to is assigned a
mac/vlan f
Remove all other codes and only keep virtio_dev_rx, copy_from_mbuf_to_vring,
virtio_dev_merge_rx, virtio_dev_merge_tx.
Previous vhost merge-able feature introduces another version of tx function,
virtio_dev_merge_tx. Actually it is not
related to merge-able feature but is the fix for memcpy betw
vhost_rxtx.c will provide vhost vring enqueue/dequeue API.
rte_virtio_net.h is used as the API header file.
Signed-off-by: Huawei Xie
---
lib/librte_vhost/main.c | 3727 -
lib/librte_vhost/rte_virtio_net.h | 161 ++
lib/librte_vhost/vhost_rxtx.c
Those files will be refactored in subsequent patches to form user space vhost
library.
Makefile and main.h are removed.
Signed-off-by: Huawei Xie
---
examples/vhost/Makefile | 60 -
examples/vhost/eventfd_link/Makefile | 39 -
examples/vhost/eventfd_link/eventfd
This set of patches transforms and refactors vhost example to a user
space vhost library.
This library implements a user space vhost cuse driver, and provides
generic APIs for user space ethernet vSwitch to integrate us-vhost for
fast packet switching with guest virtio.
vhost lib consists of five
84 matches
Mail list logo
