Now that the queue spinlocks have been added to the rx and
tx burst functions the memcpy of the slave data is no
longer necessary, so it has been removed.
Signed-off-by: Bernard Iremonger
Acked-by: Konstantin Ananyev
---
drivers/net/bonding/rte_eth_bond_pmd.c | 71
Use rte_spinlock_trylock() in the rx/tx burst functions to
take the queue spinlock.
Signed-off-by: Bernard Iremonger
Acked-by: Konstantin Ananyev
---
drivers/net/bonding/rte_eth_bond_pmd.c | 116 -
1 file changed, 84 insertions(+), 32 deletions(-)
diff --git
When adding or removing a slave device from the bonding device
the rx and tx queue spinlocks should be held.
Signed-off-by: Bernard Iremonger
Acked-by: Konstantin Ananyev
---
drivers/net/bonding/rte_eth_bond_api.c | 52 --
1 file changed, 49 insertions(+), 3
At present it is possible to add and remove slave devices from the
bonding device while traffic is running. This can result in
segmentation faults occurring in the rx and tx burst functions.
To resolve this issue spinlocks have been added to the rx and tx
queues.
Now when a slave is added or
Add spinlock to bonding rx and tx queues.
Take spinlock in rx and tx burst functions.
Take all spinlocks in slave add and remove functions.
With spinlocks in place remove memcpy of slaves.
Changes in v3:
Rebase to latest master.
Drop patches 4 and 5 from v2 patchset.
Update commit messages on
Hi,
please see inline
2016-06-08 10:53 GMT+02:00 Thomas Monjalon :
> Hi Zyta,
>
> 2016-06-01 09:56, zr at semihalf.com:
> > rte_eth_dev_get_reg_length and rte_eth_dev_get_reg callbacks
> > do not provide register size to the app in any way. It is
> > needed to allocate proper number of bytes
Fixes: b6ea6408fbc7 ("ethdev: store numa_node per device")
Signed-off-by: Bernard Iremonger
---
app/test-pmd/testpmd.c | 20 +---
1 file changed, 13 insertions(+), 7 deletions(-)
diff --git a/app/test-pmd/testpmd.c b/app/test-pmd/testpmd.c
index 991457d..1e13c36 100644
---
Fixes: edab33b1c01d ("app/testpmd: support port hotplug")
Signed-off-by: Bernard Iremonger
---
app/test-pmd/testpmd.c | 7 ++-
1 file changed, 6 insertions(+), 1 deletion(-)
diff --git a/app/test-pmd/testpmd.c b/app/test-pmd/testpmd.c
index c7ab8a0..991457d 100644
---
Move call to init_fwd_streams from start_port function
to start_packet_forwarding function.
Signed-off-by: Bernard Iremonger
---
app/test-pmd/testpmd.c | 11 ++-
1 file changed, 6 insertions(+), 5 deletions(-)
diff --git a/app/test-pmd/testpmd.c b/app/test-pmd/testpmd.c
index
Use this function in stop_port and close_port functions.
Signed-off-by: Bernard Iremonger
---
app/test-pmd/testpmd.c | 18 ++
app/test-pmd/testpmd.h | 2 ++
2 files changed, 20 insertions(+)
diff --git a/app/test-pmd/testpmd.c b/app/test-pmd/testpmd.c
index f22d1b6..6f68a18
Remove fwd_config_setup from fwd_config_display and check that
forwarding has been setup before displaying forwarding configuration.
Add call to fwd_config_setup for corelist, coremask, nbcore setup.
Add call to fwd_config_setup for portlist, portmask, nbport setup.
Signed-off-by: Bernard
Add calls to port_is_forwarding function.
Signed-off-by: Bernard Iremonger
---
app/test-pmd/testpmd.c | 24 ++--
1 file changed, 10 insertions(+), 14 deletions(-)
diff --git a/app/test-pmd/testpmd.c b/app/test-pmd/testpmd.c
index 7a1e470..f22d1b6 100644
---
Remove checks on test_done variable.
Remove code to update forwarding configuration.
Fixes: edab33b1c01d ("app/testpmd: support port hotplug")
Signed-off-by: Bernard Iremonger
---
app/test-pmd/testpmd.c | 28 +---
1 file changed, 1 insertion(+), 27 deletions(-)
diff
Add function port_is_forwarding to check whether
a port is forwarding or not.
Signed-off-by: Bernard Iremonger
---
app/test-pmd/config.c | 18 +-
app/test-pmd/testpmd.h | 3 ++-
2 files changed, 19 insertions(+), 2 deletions(-)
diff --git a/app/test-pmd/config.c
Modify testpmd to allow stop, close, detach and attach
of a port without stopping forwarding.
Changes in v3:
rebase to latest master.
added patch 9
app/testpmd: stop forwarding on exit
reworked and renamed patch 4
app/testpmd: remove fwd_config_setup from fwd_config_display
Changes in v2:
Hi Thiago,
> Any clue?
Sure. This is the reason: in dpdk-16.04/lib/librte_eal/linuxapp/igb_uio we have:
#ifndef PCI_MSIX_ENTRY_SIZE
#define PCI_MSIX_ENTRY_SIZE 16
#define PCI_MSIX_ENTRY_LOWER_ADDR 0
#define PCI_MSIX_ENTRY_UPPER_ADDR 4
#define PCI_MSIX_ENTRY_DATA
Compile DPDK with clang, below line in virtio_rxtx.c could be
optimized with four "VMOVAPS ymm, m256".
memset(>fake_mbuf, 0, sizeof(rxvq->fake_mbuf));
This instruction requires memory address is 32-byte aligned.
Or, it leads to segfault. Although only tested with Clang 3.6.0,
it can be
In ip_pipeline app, the structure app_thread_data needs to be aligned to
the cache line boundary as threads on different cpu cores are accessing
fields of the app->thread_data and having this structure not aligned on
cacheline boundary leads to false cacheline sharing.
Fixes: 7f64b9c004aa
This patches fixes problem of reusing index of outmost loop in nested
loops. This bug will lead to failure when starting a multi queue
virtio device: rx queues (except from the first one) cannot be started,
expecially their vq_ring cannot be initialized, so that when invoking
rx func on these
Hi Stephen,
> -Original Message-
> From: Stephen Hemminger [mailto:stephen at networkplumber.org]
> Sent: Saturday, June 11, 2016 2:12 AM
> To: Lu, Wenzhuo
> Cc: dev at dpdk.org; Tao, Zhe
> Subject: Re: [dpdk-dev] [PATCH 2/8] lib/librte_ether: defind RX/TX lock mode
>
> On Wed, 8 Jun
Hi Olivier,
> -Original Message-
> From: Olivier Matz [mailto:olivier.matz at 6wind.com]
> Sent: Thursday, June 9, 2016 3:51 PM
> To: Lu, Wenzhuo; Stephen Hemminger
> Cc: dev at dpdk.org; Tao, Zhe
> Subject: Re: [dpdk-dev] [PATCH 2/8] lib/librte_ether: defind RX/TX lock mode
>
> Hi,
>
>
Hi Konstantin,
> -Original Message-
> From: Ananyev, Konstantin
> Sent: Wednesday, June 8, 2016 5:20 PM
> To: Lu, Wenzhuo; Tao, Zhe; dev at dpdk.org
> Cc: Richardson, Bruce; Chen, Jing D; Liang, Cunming; Wu, Jingjing; Zhang,
> Helin
> Subject: RE: [PATCH v4 2/8] lib/librte_ether: defind
Hi Konstantin,
> -Original Message-
> From: Ananyev, Konstantin
> Sent: Wednesday, June 8, 2016 4:42 PM
> To: Lu, Wenzhuo; Tao, Zhe; dev at dpdk.org
> Cc: Richardson, Bruce; Chen, Jing D; Liang, Cunming; Wu, Jingjing; Zhang,
> Helin
> Subject: RE: [PATCH v4 4/8] ixgbe: implement device
Hi Konstantin, Thomas,
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Wednesday, June 8, 2016 5:00 PM
> To: Ananyev, Konstantin; Lu, Wenzhuo
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH] examples: add a new example for link reset
>
>
Add a new virtual device named vhost-user, which can be used just like
eth_ring, eth_null, etc. To reuse the code of original virtio, we do
some adjustment in virtio_ethdev.c, such as remove key _static_ of
eth_virtio_dev_init() so that it can be reused in virtual device; and
we add some check to
This patch implements another new instance of struct virtio_pci_ops to
drive the virtio-user virtual device. Instead of rd/wr ioport or PCI
configuration space, this virtual pci driver will rd/wr the virtual
device struct virtio_user_hw, and when necessary, invokes APIs provided
by device
Two device emulation layer APIs are added for virtio driver to call:
- virtio_user_start_device()
- virtio_user_stop_device()
- virtio_user_dev_init()
- virtio_user_dev_uninit()
These APIs will get called by virtio driver, and they call vhost adapter
layer APIs to implement the
This patch is to provide vhost adapter layer implementations. Instead
of relying on a hypervisor to translate between device emulation and
vhost backend, here we directly talk with vhost backend through the
vhost file. Depending on the type of vhost file,
- vhost-user is used if the given path
This patch is related to how to calculate relative address for vhost
backend.
The principle is that: based on one or multiple shared memory regions,
vhost maintains a reference system with the frontend start address,
backend start address, and length for each segment, so that each
frontend
This patch is to move phys addr check from virtio_dev_queue_setup
to pci ops. To makt that happen, make sure virtio_ops.setup_queue
return the result if we pass through the check.
Signed-off-by: Jianfeng Tan
Signed-off-by: Huawei Xie
Acked-by: Yuanhan Liu
---
v7:
- CONFIG_RTE_VIRTIO_VDEV -> CONFIG_RTE_VIRTIO_USER; and corresondingly,
RTE_VIRTIO_VDEV -> RTE_VIRTIO_USER.
- uint64_t -> uintptr_t, so that it can be compiled on 32-bit platform.
- Rebase on latest dpdk-next-virtio branch.
- Abandon abstracting related code into vring_hdr_desc_init(),
31 matches
Mail list logo
