On Wed, Dec 07, 2016 at 10:13:14AM +, Ananyev, Konstantin wrote:
>
> Hi Yliu,
>
> >
> > On Tue, Dec 06, 2016 at 03:53:42PM +, Ferruh Yigit wrote:
> > > > Please, we need a comment for each driver saying
> > > > "it is OK, we do not need any checksum preparation for TSO"
> > > > or
> > >
y setting loopback
> traffic between Virtio and Vhost. For example, modify the macro
> TXONLY_DEF_PACKET_LEN to the requested packet size in testpmd.h,
> rebuild and start testpmd in both host and guest, then "start" on
> one side and "start tx_first 32" on the other.
>
>
> Signed-off-by: Zhihong Wang
Reviewed-by: Yuanhan Liu
--yliu
On Wed, Dec 07, 2016 at 01:23:48PM +, Jan Wickbom wrote:
> > On Thu, Dec 01, 2016 at 04:26:50PM +0100, Jan Wickbom wrote:
> > > static int
> > > -fdset_fill(fd_set *rfset, fd_set *wfset, struct fdset *pfdset)
> > > +fdset_fill(struct pollfd *rwfds, struct fdset *pfdset)
...
> > > + rwf
On Sun, Dec 04, 2016 at 12:18:22AM +, Jianfeng Tan wrote:
> To support those devices that do not provide packet type info when
> receiving packets, add a new option, --parse-ptype, to analyze
> packet type in the Rx callback.
A quick question: does this patch have anything to do with the virti
On Fri, Dec 02, 2016 at 02:31:13PM +, Jianfeng Tan wrote:
> To support vhost kernel as the backend of net_virtio_user in comming
> patches, we abstract a vhost layer to hide all vhost_user functions.
> - Move vhost_user specific structs and macros into vhost_user.c,
> and only keep common
On Fri, Dec 02, 2016 at 02:31:15PM +, Jianfeng Tan wrote:
> Before the commit 86d59b21468a ("net/virtio: support LRO"), features
> in virtio PMD, is decided and properly set at device initialization
> and will not be changed. But afterward, features could be changed in
> virtio_dev_configure(),
On Mon, Dec 12, 2016 at 05:50:34PM +0100, Jan Wickbom wrote:
> Currently select() is used to monitor file descriptors for vhostuser
> ports. This limits the number of ports possible to create since the
> fd number is used as index in the fd_set and we have seen fds > 1023.
> This patch changes sele
On Mon, Dec 12, 2016 at 06:54:00PM +0100, Maxime Coquelin wrote:
> REPLY_ACK features provide a generic way for QEMU to ensure both
> completion and success of a request.
>
> As described in vhost-user spec in QEMU repository, QEMU sets
> VHOST_USER_NEED_REPLY flag (bit 3) when expecting a reply_a
On Tue, Dec 13, 2016 at 10:57:10AM +0100, Maxime Coquelin wrote:
>
>
> On 12/13/2016 10:56 AM, Yuanhan Liu wrote:
> >On Mon, Dec 12, 2016 at 06:54:00PM +0100, Maxime Coquelin wrote:
> >>REPLY_ACK features provide a generic way for QEMU to ensure both
> >>co
e APIs are good:
Acked-by: Yuanhan Liu
> Signed-off-by: Santosh Shukla
> Signed-off-by: Jerin Jacob
> CC: Huawei Xie
> CC: Yuanhan Liu
Not a big deal, but I think we normally put the 'Cc' above the SoB.
--yliu
On Wed, Dec 14, 2016 at 07:25:30AM +0530, Jerin Jacob wrote:
> patchset 14-28: Replace the raw readl/writel in the drivers with
> new rte_read[b/w/l/q], rte_write[b/w/l/q] eal abstraction
Instead of rte_read[b/w/l/q], there is another typical naming style:
rte_read[8/16/32/64]. Any preferences? If
On Wed, Dec 14, 2016 at 07:25:57AM +0530, Jerin Jacob wrote:
> From: Santosh Shukla
>
> Replace the raw I/O device memory read/write access with eal
> abstraction for I/O device memory read/write access to fix
> portability issues across different architectures.
>
> Signed-off-by: Santosh Shukla
On Wed, Dec 14, 2016 at 07:25:56AM +0530, Jerin Jacob wrote:
> * Following macros are derived from linux/pci_regs.h, however,
> * we can't simply include that header here, as there is no such
> @@ -320,37 +322,37 @@ static const struct virtio_pci_ops legacy_ops = {
> static inline uint8_t
> i
On Tue, Dec 13, 2016 at 02:19:47PM +0100, Jan Wickbom wrote:
> +
> + poll(pfdset->rwfds, numfds, 1000 /* millisecs */);
> +
> + for (i = 0; i < numfds; ) {
> pthread_mutex_lock(&pfdset->fd_mutex);
> +
> pfdentry = &pfdset->fd[i];
>
On Thu, Dec 15, 2016 at 02:50:19AM +0800, Yi Zhang wrote:
> Current ixgbevf driver get max_rx_pktlen = 15872, but in fact PF
> supports 15872-byte jumbo frame and VF only supports 9728-byte jumbo
> frame. If VF is running DPDK driver and set frame_size > 9728 ,PF
> running kernel ixgbe driver will
Hi Olivier,
Firstly sorry for late response!
On Thu, Nov 24, 2016 at 09:56:38AM +0100, Olivier Matz wrote:
> With virtio, doing tso requires to modify the network
> packet data:
I thought more about it this time, and I'm wondering why it's needed.
> - the dpdk API requires to set the l4 checksu
On Wed, Dec 14, 2016 at 09:45:34PM -0800, Santosh Shukla wrote:
> On Wed, Dec 14, 2016 at 11:02:23AM +0800, Yuanhan Liu wrote:
> > On Wed, Dec 14, 2016 at 07:25:56AM +0530, Jerin Jacob wrote:
> > > * Following macros are derived from linux/pci_regs.h, however,
> > >
On Fri, Dec 16, 2016 at 10:19:43AM +, Yang, Zhiyong wrote:
> > > I run the same virtio/vhost loopback tests without NIC.
> > > I can see the throughput drop when running choosing functions at run
> > > time compared to original code as following on the same platform(my
> > machine is haswell)
On Wed, Dec 21, 2016 at 10:20:26AM +0800, Qiming Yang wrote:
> Add two new commands "show port capa " and "show
I think 'cap' is a more well-known shortening for capability than
"capa"?
--yliu
ked (pollfd) array for the fds, e.g. as many fds as the size of
the array.
Also see:
http://dpdk.org/ml/archives/dev/2016-April/037024.html
Reported-by: Patrik Andersson
Signed-off-by: Jan Wickbom
Signed-off-by: Yuanhan Liu
---
For some reason, Jan won't be able to make v6, so I mad
On Wed, Dec 21, 2016 at 10:06:36AM -0800, Stephen Hemminger wrote:
> On Wed, 21 Dec 2016 17:45:13 +0800
> Yuanhan Liu wrote:
>
> > From: Jan Wickbom
> >
> > Currently select() is used to monitor file descriptors for vhostuser
> > ports. This limits the number
ply return 1 to let
EAL skip this device. Thus, the crashed could be fixed.
Fixes: c1f86306a026 ("virtio: add new driver")
Cc: sta...@dpdk.org
Signed-off-by: Yuanhan Liu
---
drivers/net/virtio/virtio_ethdev.c | 9 +
1 file changed, 9 insertions(+)
diff --git a/drivers/net/
Here is a temp to fix few multiple process issues: only did some basic
test, not positive sure (though it's likely) it will fix the issue
reported by Yaron.
Yaron, please help testing. Thanks!
---
Yuanhan Liu (3):
net/virtio: fix crash for secondary process
net/virtio: fix multiple pr
: virtqueue_enqueue error: -14
Fix it by choosing the correct Rx/Tx callbacks for the secondary process.
That is, use vector path if it's given.
Fixes: 8d8393fb1861 ("virtio: pick simple Rx/Tx")
Cc: sta...@dpdk.org
Signed-off-by: Yuanhan Liu
---
drivers/net/virtio/virtio_ethdev.c | 6 +-
ss on init stage.
Fixes: d5bbeefca826 ("virtio: introduce PCI implementation structure")
Cc: sta...@dpdk.org
Reported-by: Yaron Illouz
Signed-off-by: Yuanhan Liu
---
drivers/net/virtio/virtio_ethdev.c | 12
drivers/net/virtio/virtio_pci.c | 4 ++--
dri
On Thu, Dec 22, 2016 at 03:18:40PM +0800, Yuanhan Liu wrote:
> Here is a temp to fix few multiple process issues: only did some basic
Sign.., I meant "attempt" :-/
--yliu
On Mon, Dec 19, 2016 at 09:59:33PM -0800, Gopakumar Choorakkot Edakkunni wrote:
> While I was testing virtio with ubuntu 14.04 kvm as host and dpdk16.07
> linux as guest, quite often I have seen that I get into a situation where
> virtio_recv_mergeable_pkts() gets into a forever loop, after sending
On Fri, Dec 23, 2016 at 07:14:21AM +, Jianfeng Tan wrote:
> In driver probe phase, we obtain device information; and then virtio
> driver will initialize device and stores info, like negotiated
> features, in vhost_user layer; finally, vhost_user gets DRIVER_OK
> notification from virtio driver
On Fri, Dec 23, 2016 at 07:14:22AM +, Jianfeng Tan wrote:
> To support vhost kernel as the backend of net_virtio_user in coming
> patches, we move vhost_user specific structs and macros into
> vhost_user.c, and only keep common definitions in vhost.h.
Good.
> Besides, remove VHOST_USER_MQ fea
On Fri, Dec 23, 2016 at 07:14:23AM +, Jianfeng Tan wrote:
> +typedef int (*vhost_setup_t)(struct virtio_user_dev *dev);
> +typedef int (*vhost_send_request_t)(struct virtio_user_dev *dev,
> + enum vhost_user_request req,
> + void *
On Fri, Dec 23, 2016 at 07:14:24AM +, Jianfeng Tan wrote:
> + * Copyright(c) 2010-2016 Intel Corporation. All rights reserved.
^
I think it should be 2016.
> +/* By default, vhost kernel module allows 64 regions, but DPDK allows
> + * 256 segments. As a relief,
On Fri, Dec 23, 2016 at 07:14:25AM +, Jianfeng Tan wrote:
> When used with vhost kernel backend, we can offload at both directions.
> - From vhost kernel to virtio_user, the offload is enabled so that
> DPDK app can trust the flow is checksum-correct; and if DPDK app
> sends it throug
On Mon, Dec 26, 2016 at 06:58:58AM +, Tan, Jianfeng wrote:
>
>
> > -Original Message-
> > From: Yuanhan Liu [mailto:yuanhan@linux.intel.com]
> > Sent: Monday, December 26, 2016 2:28 PM
> > To: Tan, Jianfeng
> > Cc: dev@dpdk.org; Yigit, Fe
On Mon, Dec 26, 2016 at 06:55:16AM +, Tan, Jianfeng wrote:
>
>
> > -Original Message-
> > From: Yuanhan Liu [mailto:yuanhan@linux.intel.com]
> > Sent: Monday, December 26, 2016 2:27 PM
> > To: Tan, Jianfeng
> > Cc: dev@dpdk.org; Yigit, Fe
On Tue, Dec 27, 2016 at 01:33:14PM +0800, wang.yon...@zte.com.cn wrote:
> There are some descriptions in
> dpdk-16.11\doc\guides\sample_app_ug\vhost.rst,
> "The sample application performs simple packet switching between virtual
> machines based on Media Access Control (MAC) address or Virtual
r (from its own process space). Thus,
everything would work as expected.
---
Yuanhan Liu (6):
ethdev: fix port data mismatched in multiple process model
net/virtio: fix wrong Rx/Tx method for secondary process
net/virtio: store PCI operators pointer locally
net/virtio: store IO port info local
: virtqueue_enqueue error: -14
Fix it by choosing the correct Rx/Tx callbacks for the secondary process.
That is, use vector path if it's given.
Fixes: 8d8393fb1861 ("virtio: pick simple Rx/Tx")
Cc: sta...@dpdk.org
Signed-off-by: Yuanhan Liu
---
v2: fix a checkpatch warning: use {} consistent
PCI device, keeping
the chance of referencing to wrong data minimal.
Fixes: af75078fece3 ("first public release")
Cc: sta...@dpdk.org
Cc: Thomas Monjalon
Cc: Bruce Richardson
Cc: Ferruh Yigit
Signed-off-by: Yuanhan Liu
---
lib/librte_ether/rte_ethdev.c | 58 +++
be done in coming patches. Later patches would also
set them correctly for secondary process, so that the multiple process
model could work.
Signed-off-by: Yuanhan Liu
---
drivers/net/virtio/virtio_ethdev.c | 9 ++---
drivers/net/virtio/virtio_pci.c | 26
Like vtpci_ops, the rte_pci_ioport has to store in local memory. This
is basically for the rte_pci_device field is allocated from process
local memory, but not from shared memory.
Signed-off-by: Yuanhan Liu
---
drivers/net/virtio/virtio_pci.c | 49
Actually, virtio_hw->dev is not used since the beginning when it's
introduced. Remove it.
Signed-off-by: Yuanhan Liu
---
drivers/net/virtio/virtio_pci.c | 2 --
drivers/net/virtio/virtio_pci.h | 1 -
2 files changed, 3 deletions(-)
diff --git a/drivers/net/virtio/virtio_pci.c b/dri
0ad71 ("virtio: use PCI ioport API")
Fixes: d5bbeefca826 ("virtio: introduce PCI implementation structure")
Fixes: 6ba1f63b5ab0 ("virtio: support specification 1.0")
Cc: sta...@kernel.org
Cc: Juho Snellman
Cc: Yaron Illouz
Reported-by: Juho Snellman
Reported-by: Yaron
On Wed, Dec 28, 2016 at 07:02:41PM +0800, Yuanhan Liu wrote:
...
> Cc: sta...@kernel.org
I knew I would make a mistake like this some day :/
Not my first time typing wrong, but it's the first time sending it out.
Sorry for that. Fixed in this reply.
--yliu
On Wed, Dec 28, 2016 at 04:10:51PM -0500, Charles (Chas) Williams wrote:
> From: Wen Chiu
>
> Only increment and decrement nb_started_ports on the first and last
> device start and stop. Otherwise, nb_started_ports can become negative
> if a device is stopped multiple times.
How could you be ab
On Wed, Dec 28, 2016 at 04:10:52PM -0500, Charles (Chas) Williams wrote:
> Start a vhost server once during devinit instead of during device start
> and stop. Some vhost clients, QEMU, don't re-attaching to sockets when
> the vhost server is stopped and later started. Preserve existing behavior
>
On Thu, Dec 29, 2016 at 10:58:11AM -0500, Charles (Chas) Williams wrote:
> On 12/29/2016 03:52 AM, Yuanhan Liu wrote:
> >On Wed, Dec 28, 2016 at 04:10:52PM -0500, Charles (Chas) Williams wrote:
> >>Start a vhost server once during devinit instead of during device start
> &g
On Thu, Dec 29, 2016 at 07:42:51AM +, Tan, Jianfeng wrote:
> Forget to mention a key part, when we bind virtio to vfio-pci driver in the
> VM, we need to insert vfio kernel module like this:
>
> $ modprobe vfio enable_unsafe_noiommu_mode=1
> $ modprobe vfio-pci
Such info should be documated
On Thu, Dec 29, 2016 at 07:30:37AM +, Jianfeng Tan wrote:
> Add handler in virtio_pci_ops to set queue/irq bind.
>
> Signed-off-by: Jianfeng Tan
> ---
> drivers/net/virtio/virtio_pci.c | 29 +
> drivers/net/virtio/virtio_pci.h | 5 +
> 2 files changed, 34 ins
On Thu, Dec 29, 2016 at 07:30:38AM +, Jianfeng Tan wrote:
> /*
> * Two types of mbuf to be cleaned:
> * 1) mbuf that has been consumed by backend but not used by virtio.
> diff --git a/drivers/net/virtio/virtqueue.h b/drivers/net/virtio/virtqueue.h
> index f0bb089..62be136 100644
> --- a/d
On Thu, Dec 29, 2016 at 07:30:39AM +, Jianfeng Tan wrote:
> This patch mainly allocates structure to store queue/irq mapping,
> and configure queue/irq mapping down through PCI ops. It also creates
> eventfds for each Rx queue and tell the kernel about the eventfd/intr
> binding.
>
> Mostly im
On Thu, Dec 29, 2016 at 07:30:41AM +, Jianfeng Tan wrote:
> When closing virtio devices, close eventfds, free the struct to
> store queue/irq mapping.
>
> Signed-off-by: Jianfeng Tan
> ---
> drivers/net/virtio/virtio_ethdev.c | 27 ++-
> 1 file changed, 26 insertions(
On Thu, Dec 29, 2016 at 07:30:42AM +, Jianfeng Tan wrote:
> To support those devices that do not provide packet type info when
> receiving packets, add a new option, --parse-ptype, to analyze
> packet type in the Rx callback.
I think this would be needed for all PMD drivers don't have the PTYP
On Thu, Dec 29, 2016 at 07:30:43AM +, Jianfeng Tan wrote:
> As it gets killed, in SIGINT signal handler, device is not stopped
> and closed. In virtio's case, vector assignment in the KVM is not
> deassigned.
What wrong could happen then?
> This patch will invoke dev_stop() and dev_close() in
On Thu, Dec 29, 2016 at 07:30:34AM +, Jianfeng Tan wrote:
> v2:
> - Add PCI queue/irq config ops.
> - Move rxq interrupt settings before sending DRIVER OK.
>
> Historically, virtio PMD can only be binded to igb_uio or
> uio_pci_generic, and not for vfio-pci. Besides, quote from
> http://dp
On Fri, Dec 30, 2016 at 07:30:19AM +, Tan, Jianfeng wrote:
> Hi,
>
> > -Original Message-
> > From: Yuanhan Liu [mailto:yuanhan@linux.intel.com]
> > Sent: Friday, December 30, 2016 2:40 PM
> > To: Tan, Jianfeng
> > Cc: dev@dpdk.org; step..
On Fri, Dec 30, 2016 at 04:26:27PM -0500, Charles (Chas) Williams wrote:
>
>
> On 12/29/2016 10:15 PM, Yuanhan Liu wrote:
> >On Thu, Dec 29, 2016 at 10:58:11AM -0500, Charles (Chas) Williams wrote:
> >>On 12/29/2016 03:52 AM, Yuanhan Liu wrote:
> >>>On W
On Sun, Jan 01, 2017 at 02:01:56PM -0500, Charles (Chas) Williams wrote:
> If you create a vhost server device, it doesn't create the actual datagram
> socket until you call .dev_start(). If you call .dev_stop() is also
> deletes those sockets. For QEMU clients, this is a problem since QEMU
> doe
On Sun, Jan 01, 2017 at 02:01:57PM -0500, Charles (Chas) Williams wrote:
> .dev_start()/.dev_stop() roughly corresponds to the local device's
> port being up or down. This is different from the remote client being
> connected which is roughtly link up or down. Emulate the behavior by
> separately
On Tue, Jan 03, 2017 at 10:57:55PM -0500, Yong Wang wrote:
> In function vhost_new_device(), current code dose not free 'dev'
> in "i == MAX_VHOST_DEVICE" condition statements. It will lead to a
> memory leak.
Nice catch!
Here are few minor stuff you might need pay attention to for future
contrib
On Wed, Jan 04, 2017 at 03:59:21AM +, Jianfeng Tan wrote:
> virtio_user is not properly reset when users call vtpci_reset(),
> as it ignores VIRTIO_CONFIG_STATUS_RESET status in
> virtio_user_set_status().
>
> This might lead to initialization failure as it starts to re-init
> the device befor
On Wed, Jan 04, 2017 at 03:59:22AM +, Jianfeng Tan wrote:
> To support vhost kernel as the backend of net_virtio_user in coming
> patches, we move vhost_user specific structs and macros into
> vhost_user.c, and only keep common definitions in vhost.h.
>
> Besides, remove VHOST_USER_MQ feature
On Wed, Jan 04, 2017 at 03:59:23AM +, Jianfeng Tan wrote:
> +struct virtio_user_backend_ops ops_user;
Better to qualify it with "extern const" ...
--yliu
On Wed, Jan 04, 2017 at 03:59:24AM +, Jianfeng Tan wrote:
> +static int
> +vhost_kernel_ioctl(struct virtio_user_dev *dev,
> +enum vhost_user_request req,
> +void *arg)
> +{
> + int i, ret = -1;
> + uint64_t req_kernel;
> + struct vhost_memory_kernel
On Wed, Jan 04, 2017 at 06:46:34AM +, Tan, Jianfeng wrote:
> Hi Yuanhan,
>
> > -Original Message-
> > From: Yuanhan Liu [mailto:yuanhan@linux.intel.com]
> > Sent: Wednesday, January 4, 2017 2:03 PM
> > To: Tan, Jianfeng
> > Cc: dev@dpd
On Wed, Jan 04, 2017 at 02:56:50PM +0800, Tan, Jianfeng wrote:
>
>
> On 12/30/2016 2:27 PM, Yuanhan Liu wrote:
> >On Thu, Dec 29, 2016 at 07:30:39AM +, Jianfeng Tan wrote:
> >>This patch mainly allocates structure to store queue/irq mapping,
> >>and configu
On Wed, Jan 04, 2017 at 03:22:08PM +0800, Tan, Jianfeng wrote:
>
> Sorry, I forget to reply this comment.
>
> On 12/26/2016 3:44 PM, Yuanhan Liu wrote:
> >[...]
> >>+
> >>+ /* Does not work when VIRTIO_F_IOMMU_PLATFORM now, why? */
> >Because this fea
On Tue, Jan 03, 2017 at 11:22:42AM -0500, Charles (Chas) Williams wrote:
> If you create a vhost server device, it doesn't create the actual datagram
> socket until you call .dev_start(). If you call .dev_stop() is also
> deletes those sockets. For QEMU clients, this is a problem since QEMU
> doe
On Wed, Jan 04, 2017 at 06:34:40PM +0100, Thomas Monjalon wrote:
> +Cc Sergio (maintainer of the secondary process thing)
>
> 2016-12-28 19:02, Yuanhan Liu:
> > --- a/lib/librte_ether/rte_ethdev.c
> > +++ b/lib/librte_ether/rte_ethdev.c
> > @@ -201,9 +201,6 @@ rte_
Tx buffer
Yuanhan Liu (2):
vhost: fix crash on NUMA
version: 17.05.1
On Wed, Jun 14, 2017 at 10:19:53AM +0200, Jens Freimann wrote:
> On Mon, Jun 12, 2017 at 02:29:04PM -0700, Daniel Verkamp wrote:
> > vsocket->conn_mutex was allocated with pthread_mutex_init() but never
> > freed with pthread_mutex_destroy(). This is a potential memory leak,
> > depending on how p
P csum.
> >
> >Fixes: 859b480d5afd ("vhost: add guest offload setting")
> >Cc: sta...@dpdk.org
> >
> >Cc: Yuanhan Liu
> >Cc: Maxime Coquelin
> >Cc: Jiayu Hu
> >Signed-off-by: Jianfeng Tan
>
> Reviewed-by: Maxime Coquelin
Series applied to dpdk-next-virtio.
Thanks!
--yliu
On Fri, Jun 30, 2017 at 02:43:47PM +, Mcnamara, John wrote:
>
>
> > -Original Message-
> > From: Yang, Zhiyong
> > Sent: Thursday, May 11, 2017 3:17 AM
> > To: dev@dpdk.org
> > Cc: Mcnamara, John ; yuanhan@linux.intel.com;
> > maxime.coque...@redhat.com; Yang, Zhiyong
> > Subject
On Thu, Jun 29, 2017 at 08:01:45AM +, Tan, Jianfeng wrote:
>
>
> > -Original Message-
> > From: Maxime Coquelin [mailto:maxime.coque...@redhat.com]
> > Sent: Thursday, June 29, 2017 3:58 PM
> > To: dev@dpdk.org; y...@fridaylinux.org
> > Cc: i.dyu...@samsung.com; Tan, Jianfeng; sta...@
On Wed, Jun 28, 2017 at 03:40:31PM +0300, Ivan Dyukov wrote:
> To compare enabled features in current device we must use bit
> mask instead of bit position.
>
> CC: sta...@dpdk.org
> Fixes: c843af3aa13e ("vhost: access header only")
>
> Signed-off-by: Ivan Dyukov
Applied to dpdk-next-virtio.
T
Hi all,
The next meeting of the techboard will happen on IRC #dpdk-board,
at 3pm UTC, this Wednesday 5th of July.
The agenda is updated here:
https://annuel.framapad.org/p/r.0c3cc4d1e011214183872a98f6b5c7db
For your convenience, I copied it here.
0. Release blocking issues?
On Mon, Jul 03, 2017 at 05:56:20AM +, Hu, Jiayu wrote:
> > > +/**
> > > + * GRO table, which is used to merge packets. It keeps many reassembly
> > > + * tables of desired GRO types. Applications need to create GRO tables
> > > + * before using rte_gro_reassemble to perform GRO.
> > > + */
> >
Haven't looked at the details yet, and below are some quick comments
after a glimpse.
On Sat, Jul 01, 2017 at 07:08:41PM +0800, Jiayu Hu wrote:
...
> +void *rte_gro_tbl_create(const
> + const struct rte_gro_param *param)
The DPDK style is:
void *
rte_gro_tbl_destroy(...)
Also you sh
Again, just some quick comments after a glimpse.
On Sat, Jul 01, 2017 at 07:08:42PM +0800, Jiayu Hu wrote:
> + for (i = 0; i < nb_pkts; i++) {
> + if (RTE_ETH_IS_IPV4_HDR(pkts[i]->packet_type) &&
> + (pkts[i]->packet_type & RTE_PTYPE_L4_TCP)) {
> +
On Tue, Jul 04, 2017 at 10:50:40AM +0200, Jens Freimann wrote:
> This series is a bunch of small cleanups regarding checking return values etc.
Series applied to dpdk-next-virtio. You probably need update your script
(if you are using one) about my email address.
Thanks.
--yliu
On Tue, Jul 04, 2017 at 10:50:42AM +0200, Jens Freimann wrote:
> +out_mutex:
> + if (pthread_mutex_destroy(&vsocket->conn_mutex))
> + RTE_LOG(ERR, VHOST_CONFIG,
> + "error: failed to destroy connection mutex\n");
One minor comment though: {} should be used for o
On Fri, Jun 23, 2017 at 08:41:37AM -0400, Allain Legacy wrote:
> The rte_eth_dev.data pointer is set to a reference to a static table.
> Attempting to rte_free() it leads to a panic. For example, the
> following commands result in a panic if run in testpmd
>
> testpmd> port attach virtio_user0,
On Fri, Jun 23, 2017 at 12:14:45AM -0400, Yong Wang wrote:
> Since "rte_eal_dev_init()" has been removed, the comment referred to
> it should be modified simultaneously.
>
> Signed-off-by: Yong Wang
Applied to dpdk-next-virtio. Thanks.
--yliu
On Thu, Jul 06, 2017 at 04:30:43PM +0300, Ilya Maximets wrote:
> syscall always returns '-1' on failure and there is no point
> in printing that value. 'errno' is much more informative.
>
> Fixes: 586e39001317 ("vhost: export numa node")
>
> Signed-off-by: Ilya Maximets
Applied to dpdk-next-vir
On Thu, Jun 29, 2017 at 05:28:32PM +0800, Changpeng Liu wrote:
> +.. note::
> +You must check whether you Qemu can support "vhost-user-scsi" or not,
> +the latest Qemu code is recommended.
You should specify the least QEMU version that supports vhost-user scsi here.
> +
> +Common Issues
>
On Fri, Jul 07, 2017 at 02:00:38AM +, Liu, Changpeng wrote:
>
>
> > -Original Message-
> > From: Yuanhan Liu [mailto:y...@fridaylinux.org]
> > Sent: Friday, July 7, 2017 8:51 AM
> > To: Liu, Changpeng
> > Cc: dev@dpdk.org
> > Subject: R
On Sat, Jul 08, 2017 at 12:14:51PM +0800, Changpeng Liu wrote:
> +.. note::
> +You must check whether your Qemu can support "vhost-user-scsi" or not,
> +Qemu v2.9.50 or newer version is required.
QEMU v2.9.50 looks like a stable version, and I don't think they will
backport new features on
On Sat, Jul 08, 2017 at 01:21:37PM +0800, Changpeng Liu wrote:
> vhost-user protocol is common to many virtio devices, such as
> virtio_net/virtio_scsi/virtio_blk. Since DPDK vhost library
> removed the NET specific data structures, the vhost library
> is common to other virtio devices, such as vir
Currently, testpmd just allows to query the RETA info only when the
required size equals to configured RETA size.
This patch allows to query any RETA size <= the configured size. This
helps when the RETA size is big (say 512) and when I just want to peak
few RETA entries.
Signed-off-by: Yuan
Print the valid RTE size range so that user knows what goes wrong.
Signed-off-by: Yuanhan Liu
---
app/test-pmd/cmdline.c | 8 +---
1 file changed, 5 insertions(+), 3 deletions(-)
diff --git a/app/test-pmd/cmdline.c b/app/test-pmd/cmdline.c
index c8faef9..d32a1df 100644
--- a/app/test-pmd
On Fri, Jul 07, 2017 at 02:10:57PM +0530, Hemant Agrawal wrote:
> On 7/7/2017 10:37 AM, Yuanhan Liu wrote:
> >On Sat, Jul 08, 2017 at 01:21:37PM +0800, Changpeng Liu wrote:
> >>vhost-user protocol is common to many virtio devices, such as
> >>virtio_net/virtio_scsi/v
On Fri, Jul 07, 2017 at 02:21:28PM +0200, Thomas Monjalon wrote:
> 08/07/2017 07:21, Changpeng Liu:
> > MAINTAINERS | 2 +
> > doc/guides/sample_app_ug/vhost_scsi.rst | 110 +++
> > examples/Makefile | 2 +-
> > examples/vhost_scsi/Makefile
On Fri, Jul 07, 2017 at 12:52:48PM -0700, Stephen Hemminger wrote:
> While doing code for Hyper-V, noticed that the virtio driver was
> confused about receive versus transmit offloads. The virtio
> checksum offload is L4 (TCP/UDP) only, not IPv4. Also, TSO
> and LRO are not the same.
>
> This may
On Tue, Jul 04, 2017 at 05:13:37PM +0100, Ferruh Yigit wrote:
> @@ -157,8 +164,12 @@ rte_eth_dev_pci_generic_probe(struct rte_pci_device
> *pci_dev,
>
> RTE_FUNC_PTR_OR_ERR_RET(*dev_init, -EINVAL);
> ret = dev_init(eth_dev);
> - if (ret)
> + if (ret) {
> rte_eth
On Mon, Jul 10, 2017 at 01:19:06AM +, Wu, Jingjing wrote:
> > -Original Message-
> > From: Yuanhan Liu [mailto:y...@fridaylinux.org]
> > Subject: [PATCH 2/2] testpmd: give more hint on invalid RETA size
> >
> > Print the valid RTE size range so t
Hi all,
Here is the meeting notes for the last DPDK technical board meeting
held on 2017-07-05.
Member attendees:
- Hemant Agrawal
- Jan Blunck
- Jerin Jacob
- Olivier Matz
- Stephen Hemminger
- Thomas Monjalon
- Yuanhan Liu
0. Release blocking issues
No major issues, but as always, we lack
On Mon, Jul 10, 2017 at 11:48:27AM +0200, Jens Freimann wrote:
> >@@ -668,7 +668,7 @@ rte_vhost_driver_register(const char *path, uint64_t
> >flags)
> > }
> >
> > vhost_user.vsockets[vhost_user.vsocket_cnt++] = vsocket;
> >-
> >+goto out;
> >out_mutex:
> > if (pthread_mutex_destroy
On Tue, Jul 18, 2017 at 01:29:48PM +0100, Ferruh Yigit wrote:
> On 7/18/2017 12:04 PM, Ferruh Yigit wrote:
> > On 7/17/2017 7:35 AM, Tiwei Bie wrote:
> >> Make the code more readable. No functional change.
> >>
> >> Signed-off-by: Tiwei Bie
> >
> > Applied to dpdk-next-net/master, thanks.
>
> An
On Tue, Jul 18, 2017 at 08:43:42AM -0500, Matt Laswell wrote:
> Hi Qiming,
>
> That's fantastic news. Thank you very much for taking the time to figure
> the issue out.
>
> Would it be possible to backport the fix to the 16.11 LTS release? This
> kind of problem seems tailor-made for LTS.
Agr
On Tue, Jul 18, 2017 at 10:26:39AM +, De Lara Guarch, Pablo wrote:
> Hi Akhil,
>
> > -Original Message-
> > From: Akhil Goyal [mailto:akhil.go...@nxp.com]
> > Sent: Tuesday, July 18, 2017 8:34 AM
> > To: dev@dpdk.org; sta...@dpdk.org
> > Cc: hemant.agra...@nxp.com; De Lara Guarch, Pabl
eproduce the problem on 16.11 LTS release? I think maybe it has
> been fixed already.
>
>
> Best Regards,
> Xiaoyun Li
>
>
>
> -----Original Message-
> From: Yuanhan Liu [mailto:y...@fridaylinux.org]
> Sent: Tuesday, July 18, 2017 22:01
> To: Matt Laswell
101 - 200 of 1959 matches
Mail list logo