On Wed, Dec 09, 2015 at 02:19:58PM +0100, Kamil Rytarowski wrote:
> Currently rte_eal_check_module() detects Linux kernel modules via reading
> /proc/modules. Built-in ones aren't listed there and therefore they are not
> being found by the script.
>
> Add support for checking built-in modules wit
Hi Kamil,
First of all, sorry for no one has reviewed your patches for over one
month! You may want to ping more often (say, per week) next time if it
still happenes :)
Another thing is that there is no maintainer for tools code.
On Wed, Dec 09, 2015 at 02:19:57PM +0100, Kamil Rytarowski wrote:
Jan,
I was waiting for some others feedbacks before going into the code.
Glad to see you already tried this.
On Mon, Jan 18, 2016 at 3:58 PM, Jan Viktorin
wrote:
> On Thu, 14 Jan 2016 11:38:16 +0100
> David Marchand wrote:
>> - no need for a rte_pci_driver reference in rte_pci_device, since w
On Mon, Jan 18, 2016 at 06:13:09PM +0900, Tetsuya Mukawa wrote:
> +struct virtio_pci_access_ops {
> + uint8_t (*legacy_read8)(struct virtio_hw *hw, uint8_t *addr);
> + uint16_t (*legacy_read16)(struct virtio_hw *hw, uint16_t *addr);
> + uint32_t (*legacy_read32)(struct virtio_hw *hw, ui
Hi Tetsuya,
On 1/18/2016 5:13 PM, Tetsuya Mukawa wrote:
> The patches abstract pci access method of virtio-net PMD.
> The patch should be on Yuanhan's below patch series.
> - [PATCH v4 0/8] virtio 1.0 enabling for virtio pmd driver
>
>
> Tetsuya Mukawa (3):
>virtio: Change the parameter orde
This is useful when sections have duplicate names.
Signed-off-by: Rich Lane
---
v1->v2:
- Added new symbol to version script.
lib/librte_cfgfile/rte_cfgfile.c | 16
lib/librte_cfgfile/rte_cfgfile.h | 23 +++
lib/librte_cfgfile/rte_cfgfile
On Mon, Jan 18, 2016 at 12:47 PM, Yuanhan Liu
wrote:
> On Mon, Jan 18, 2016 at 12:15:40PM +0530, Santosh Shukla wrote:
>> I am testing for virtio 1.0 and 0.95 for arm including your patch,
>> soon we;ll post the patch series that is rebased on / dependent on
>> below patchset:
>> - virtio 1.0
>> -
VLAN filtering was always performed, even if hw_vlan_filter was
disabled. During device initialization, default filter
RTE_MACVLAN_PERFECT_MATCH was applied. In this situation, all incoming
VLAN frames were dropped by the card (increase of the register RUPP - Rx
Unsupported Protocol).
In order to
This patch adds below function pointers to abstract pci access method.
- legacy_read8/16/32
- legacy_write8/16/32
- modern_read8/16/32
- modern_write8/16/32
- map_pci_cfg
- unmap_pci_cfg
- get_cfg_addr
- read_pci_cfg
This layer will be used when virtio-net PMD supports container extension.
To abstract pci access method, the patch moves below function
to "virtio_pci.c".
- rte_eal_pci_unmap_device()
Signed-off-by: Tetsuya Mukawa
---
drivers/net/virtio/virtio_ethdev.c | 2 +-
drivers/net/virtio/virtio_pci.c| 11 +++
drivers/net/virtio/virtio_pci.h| 1 +
3 files cha
The patch change the parameter order of below functions.
- io_write8()
- io_write16()
- io_write32()
This changig are needed to add a new layer to abstract accessing
method.
Signed-off-by: Tetsuya Mukawa
---
drivers/net/virtio/virtio_pci.c | 66 -
1 fil
The patches abstract pci access method of virtio-net PMD.
The patch should be on Yuanhan's below patch series.
- [PATCH v4 0/8] virtio 1.0 enabling for virtio pmd driver
Tetsuya Mukawa (3):
virtio: Change the parameter order of io_write8/16/32()
virtio: move rte_eal_pci_unmap_device() to vir
On 2016/01/14 15:41, Tetsuya Mukawa wrote:
>
>>> 2. Abstraction of read/write accesses.
>>>
>>> It may be difficult to cleanly rebase my patches on this patches,
>>> because virtio_read_caps() is not abstracted.
>>> Let me describe it more.
>>> So far, we need to handle below 3 virtio-net devices..
On Mon, Jan 18, 2016 at 5:12 PM, Stephen Hemminger <
stephen at networkplumber.org> wrote:
> On Fri, 15 Jan 2016 16:18:01 +
> Ferruh Yigit wrote:
>
> > This work is to make DPDK ports more visible and to enable using common
> > Linux tools to configure DPDK ports.
> >
> > Patch is based on KN
On Mon, 18 Jan 2016 17:48:51 -0600
Jay Rolette wrote:
> On Mon, Jan 18, 2016 at 5:12 PM, Stephen Hemminger <
> stephen at networkplumber.org> wrote:
>
> > On Fri, 15 Jan 2016 16:18:01 +
> > Ferruh Yigit wrote:
> >
> > > This work is to make DPDK ports more visible and to enable using common
On 1/15/2016 4:10 PM, Pavel Fedin wrote:
> Hello!
>
>> If this is the case, i am wondering whether we should include
>> "malfunctioning clients" in commit message. It triggers me to think if
>> there are existing buggy implementations.
> Well... Can you suggest how to rephrase it? May be "if a cl
On 1/15/2016 12:34 PM, Yuanhan Liu wrote:
> +void
> +vtpci_write_dev_config(struct virtio_hw *hw, uint64_t offset,
> + void *src, int length)
> +{
> + hw->vtpci_ops->write_dev_cfg(hw, offset, src, length);
missed changing src to const.
> +}
.On 1/15/2016 12:34 PM, Yuanhan Liu wrote:
> Modern (v1.0) virtio pci device defines several pci capabilities.
> Each cap has a configure structure corresponding to it, and the
> cap.bar and cap.offset fields tell us where to find it.
>
[snip]
> +
> +static inline void
> +io_write64_twopart(uint64_
On 2016/01/15 13:36, Yuanhan Liu wrote:
> v4: - mark "src" arg as const for write_dev_cfg operation
>
> - remove unnessary inline, and likely/unlikely
>
> v3: - export pci_unmap_device as well; and invoke it at virtio
> uninit stage.
>
> - fixed same data corruption bug reported by Qi
We define to_pci_driver helper macro here.
Suggested-by: David Marchand
Signed-off-by: Jan Viktorin
---
app/test/virtual_pmd.c | 2 +-
drivers/net/virtio/virtio_ethdev.c | 25 +
lib/librte_cryptodev/rte_cryptodev.c| 3 ++-
lib/librte_eal/commo
Suggested-by: David Marchand
Signed-off-by: Jan Viktorin
---
drivers/net/e1000/em_ethdev.c | 14 ++---
drivers/net/e1000/igb_ethdev.c | 24 +++---
drivers/net/enic/enic_main.c | 4 ++--
drivers/net/fm10k/fm10k_ethdev.c | 1
Signed-off-by: Jan Viktorin
---
app/test/virtual_pmd.c | 2 +-
lib/librte_eal/bsdapp/eal/eal_pci.c | 2 +-
lib/librte_eal/common/eal_common_pci.c | 4 ++--
lib/librte_eal/common/include/rte_dev.h | 1 +
lib/librte_eal/common/include/rte_pci.h | 1 -
lib/librte_eal/linuxapp/e
Suggested-by: David Marchand
Signed-off-by: Jan Viktorin
---
lib/librte_eal/common/eal_common_pci.c | 6 +++---
lib/librte_eal/common/include/rte_dev.h | 3 +++
lib/librte_eal/common/include/rte_pci.h | 1 -
3 files changed, 6 insertions(+), 4 deletions(-)
diff --git a/lib/librte_eal/common/ea
Signed-off-by: Jan Viktorin
---
lib/librte_eal/common/include/rte_common.h | 16
1 file changed, 16 insertions(+)
diff --git a/lib/librte_eal/common/include/rte_common.h
b/lib/librte_eal/common/include/rte_common.h
index b58a384..bf6560d 100644
--- a/lib/librte_eal/common/inclu
Define a general represenation of a device and driver in DPDK. The goal is to
get rid of the deep dependency on the PCI. Higher level code should not
reference the bus-specific structures as it does not need to.
PCI infrastructure embeds those structures. Other infrastructures will do the
same.
S
Hello,
based on the recent discussions [1], I have prepared a little preview of
patches reflecting the basic ideas. That is:
* generalization of device and driver structures
* embedding a generic device and driver structure in the bus-specific ones
* moving some members from PCI-specific structur
On 1/15/2016 12:34 PM, Yuanhan Liu wrote:
> -static void
> +static int
> virtio_negotiate_features(struct virtio_hw *hw)
> {
> uint64_t host_features;
> @@ -949,6 +949,22 @@ virtio_negotiate_features(struct virtio_hw *hw)
> hw->guest_features = vtpci_negotiate_features(hw, host_featur
2016-01-16 16:53, David Marchand:
> On Thu, Jan 14, 2016 at 12:46 PM, Jan Viktorin
> wrote:
> > On Thu, 14 Jan 2016 11:38:16 +0100
> > David Marchand wrote:
> >> Here is a proposal of a big cleanup in ethdev (cryptodev would have to
> >> follow) and eal structures.
[...]
> >> ABI is most likely
Hello Declan,
On Mon, Jan 18, 2016 at 3:54 PM, Declan Doherty
wrote:
> In your proposal above, having an bus type enumeration in the rte_device
> which specifies the bus type might be simpler than having to parse a
> specific name formatting.
This is not simpler.
This is useless afaics.
"upper"
On 1/15/2016 12:34 PM, Yuanhan Liu wrote:
> Modern (v1.0) virtio pci device defines several pci capabilities.
> Each cap has a configure structure corresponding to it, and the
> cap.bar and cap.offset fields tell us where to find it.
>
[snip]
>
> +static void *
> +get_cfg_addr(struct rte_pci_de
> -Original Message-
> From: David Marchand [mailto:david.marchand at 6wind.com]
> Sent: Saturday, January 16, 2016 3:11 PM
> To: Mcnamara, John
> Cc: dev at dpdk.org; Thomas Monjalon
> Subject: Re: [dpdk-dev] [PATCH 09/11] doc: refresh headers list
>
> Hello John,
>
> On Tue, Jan 12, 2
I've lost some To/CC in the e-mail, so forwarding to dpdk-dev...
Begin forwarded message:
Date: Mon, 18 Jan 2016 15:58:34 +0100
From: Jan Viktorin
To: David Marchand
Subject: Re: Proposal for a big eal / ethdev cleanup
Hello David,
I am playing around a little bit with the code according to
On Sun, Jan 17, 2016 at 11:57:18AM -0800, Rich Lane wrote:
> The common vhost code only supported a single mmap per device. vhost-user
> worked around this by saving the address/length/fd of each mmap after the end
> of the rte_virtio_memory struct. This only works if the vhost-user code frees
> de
Hi,
Since there was neither positive nor negative feedback so far I wanted to
ask if this patch is ok?
Especially since we had a discussion on the approach I chose in the first
approach to this issue I'd be happy about a ack/nak.
Kind Regards,
Christian
Christian Ehrhardt
Software Engineer, Ubu
Hi,
On 01/15/2016 10:00 AM, Panu Matilainen wrote:
diff --git a/lib/librte_cmdline/cmdline_rdline.h
b/lib/librte_cmdline/cmdline_rdline.h
index b9aad9b..72e2dad 100644
--- a/lib/librte_cmdline/cmdline_rdline.h
+++ b/lib/librte_cmdline/cmdline_rdline.h
@@ -93,7 +93,7 @
Hi Kamil,
2015-12-09 14:19, Kamil Rytarowski:
> Currently dpdk_nic_bind.py detects Linux kernel modules via reading
> /proc/modules. Built-in ones aren't listed there and therefore they are not
> being found by the script.
>
> Add support for checking built-in modules with parsing the sysfs files
Hi Declan,
On 01/12/2016 07:11 PM, Declan Doherty wrote:
> In this rfc I'm looking to get some feedback on a proposal to change the
> cryptodev burst API from the current implementation of accepting burst
> of rte_mbuf's to a burst API based on rte_crypto_op's.
>
> -static inline uint16_t
> -rte_
On Mon, Jan 18, 2016 at 12:15:40PM +0530, Santosh Shukla wrote:
> I am testing for virtio 1.0 and 0.95 for arm including your patch,
> soon we;ll post the patch series that is rebased on / dependent on
> below patchset:
> - virtio 1.0
> - vfio-noiommu
> - KDRV check by huawei
>
> IMO, we should st
On Fri, 15 Jan 2016 16:18:01 +
Ferruh Yigit wrote:
> This work is to make DPDK ports more visible and to enable using common
> Linux tools to configure DPDK ports.
>
> Patch is based on KNI but contains only control functionality of it,
> also this patch does not include any Linux kernel net
On 01/14/2016 09:28 PM, Santosh Shukla wrote:
> So far virtio handle rw access for uio / ioport interface, This patch to
> extend
> the support for vfio interface. For that introducing private struct
> virtio_vfio_dev{
> - is_vfio
> - pci_dev
> };
> Signed-off-by: Santosh Shukl
On 01/17/2016 05:58 AM, Rich Lane wrote:
> This is useful when sections have duplicate names.
>
> Signed-off-by: Rich Lane
> ---
> lib/librte_cfgfile/rte_cfgfile.c | 16
> lib/librte_cfgfile/rte_cfgfile.h | 23 +++
> 2 files changed, 39 insertions(+)
>
Thi
On 14/01/16 10:38, David Marchand wrote:
> Hello all,
>
> Here is a proposal of a big cleanup in ethdev (cryptodev would have to
> follow) and eal structures.
> This is something I wanted to do for quite some time and the arrival of
> a new bus makes me think we need it.
>
> This is an alternativ
Wrong list, sorry
> -Original Message-
> From: De Lara Guarch, Pablo
> Sent: Monday, January 18, 2016 1:48 PM
> To: bombermag at gmail.com
> Cc: dev at dpdk.org; De Lara Guarch, Pablo
> Subject: RE: [dpdk-users] Broken hash table functions with CFLAGS -O3
>
> > -Original Message-
On Fri, Jan 15, 2016 at 07:12:04PM +0530, Santosh Shukla wrote:
> On Fri, Jan 15, 2016 at 6:13 PM, Santosh Shukla wrote:
> > On Fri, Jan 15, 2016 at 11:57 AM, Yuanhan Liu
> > wrote:
> >> On Thu, Jan 14, 2016 at 06:58:31PM +0530, Santosh Shukla wrote:
> >>> So far virtio handle rw access for uio /
> -Original Message-
> From: users [mailto:users-bounces at dpdk.org] On Behalf Of Dmitriy Yakovlev
> Sent: Friday, January 15, 2016 1:20 AM
> To: users at dpdk.org
> Subject: [dpdk-users] Broken hash table functions with CFLAGS -O3
>
> Hello.
> I trying to use hash tables for adding struc
> From: Ravi Kerur [mailto:rkerur at gmail.com]
> Sent: Monday, December 21, 2015 11:13 PM
> To: dev at dpdk.org
> Cc: Richardson, Bruce; Doherty, Declan; Ananyev, Konstantin; Ravi Kerur
> Subject: [PATCH v1] Modify and modularize l3fwd code
>
> v1:
> > Rebase to latest code base for DPDK
We could do something ? la modinfo, but let's keep it simple for now.
With this, you can extract the devices that need to be bound to uio / vfio
with tools like objdump :
$ objdump -j rte_pci_id_uio -s build/lib/librte_pmd_fm10k.so
Contents of section rte_pci_id_uio:
15760 8680a415 868
Now that all pci device ids are in their respective drivers, we can remove
this header.
Signed-off-by: David Marchand
---
doc/api/doxy-api-index.md | 1 -
lib/librte_eal/common/Makefile | 2 +-
lib/librte_eal/common/include/rte_pci_dev_ids.h | 93
Reused defines from the driver and moved broadcom vendor id macro.
Signed-off-by: David Marchand
---
Changes since v1:
- indent fix
drivers/net/bnx2x/bnx2x.c | 3 +-
drivers/net/bnx2x/bnx2x_ethdev.c| 21 +++--
lib/librte_eal/common/include/rte_pci_dev
Moved cisco vendor id since the driver had no such information.
Signed-off-by: David Marchand
---
Changes since v1:
- indent fix
drivers/net/enic/enic_ethdev.c | 12
lib/librte_eal/common/include/rte_pci_dev_ids.h | 17 -
2 files changed, 4 inserti
Moved vmware device ids macro since the driver had no such information.
Signed-off-by: David Marchand
---
Changes since v1:
- indent fix
drivers/net/vmxnet3/vmxnet3_ethdev.c| 9 -
lib/librte_eal/common/include/rte_pci_dev_ids.h | 16
2 files changed, 4 ins
Reused defines from virtio_pci.h.
Signed-off-by: David Marchand
---
Changes since v1:
- indent fix
drivers/net/virtio/virtio_ethdev.c | 7 ++-
lib/librte_eal/common/include/rte_pci_dev_ids.h | 17 -
2 files changed, 2 insertions(+), 22 deletions(-)
diff --git
Since the base driver already defines all pci device ids, no need to
redefine them, let's just drop the previous RTE_PCI_DEV_ID_DECL* stuff.
Signed-off-by: David Marchand
---
Changes since v1:
- indent fix
drivers/net/fm10k/fm10k_ethdev.c| 6 ++---
lib/librte_eal/common/includ
Since the base driver already defines all pci device ids, no need to
redefine them, let's just drop the previous RTE_PCI_DEV_ID_DECL* stuff.
Signed-off-by: David Marchand
---
Changes since v1:
- indent fix
drivers/net/i40e/i40e_ethdev.c | 20 +++--
drivers/net/i40e/i40e_et
test application and kni still want to know ixgbe pci devices.
So let's create a header in the driver that will be used by them.
Signed-off-by: David Marchand
---
app/test-pmd/Makefile | 2 +
app/test-pmd/cmdline.c | 2 +-
app/test/Makefile
test application and kni still want to know e1000 pci devices.
So let's create headers in the driver that will be used by them.
Signed-off-by: David Marchand
---
app/test/Makefile | 3 +
app/test/test_pci.c | 3 +-
drivers/net/e1000/e
This patchset moves all pci device ids from eal to the pmds that need them
(patches 1 to 8).
Global pci device id list is then removed (patch 9).
In last patch, all those device ids are put in a dedicated section for
retrieval by external tools.
Changes since v1:
- indent fixes in i40e, fm10k, vi
On Mon, Jan 18, 2016 at 12:29 PM, Jason Wang wrote:
>
>
> On 01/14/2016 09:28 PM, Santosh Shukla wrote:
>> So far virtio handle rw access for uio / ioport interface, This patch to
>> extend
>> the support for vfio interface. For that introducing private struct
>> virtio_vfio_dev{
>> - is_vf
On Mon, Jan 18, 2016 at 11:41 AM, Yuanhan Liu
wrote:
> On Fri, Jan 15, 2016 at 07:12:04PM +0530, Santosh Shukla wrote:
>> On Fri, Jan 15, 2016 at 6:13 PM, Santosh Shukla
>> wrote:
>> > On Fri, Jan 15, 2016 at 11:57 AM, Yuanhan Liu
>> > wrote:
>> >> On Thu, Jan 14, 2016 at 06:58:31PM +0530, Sant
On Sun, 17 Jan 2016 22:05:09 -0500
Zhihong Wang wrote:
> This patch set optimizes DPDK memcpy for AVX512 platforms, to make full
> utilization of hardware resources and deliver high performance.
>
> In current DPDK, memcpy holds a large proportion of execution time in
> libs like Vhost, especial
Ferruh Yigit writes:
> This work is to make DPDK ports more visible and to enable using common
> Linux tools to configure DPDK ports.
This is a good goal. Only question - why use an additional kernel module
to do this? Is it _JUST_ for ethtool support? I think the other stuff
can be accomplished
2016-01-16 16:10, David Marchand:
> On Tue, Jan 12, 2016 at 3:06 PM, Mcnamara, John
> wrote:
> > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of David Marchand
> >> Since we are going to remove a header in next commit, let's first refresh
> >> documentation.
> >
> > I don't like these par
ping?
W dniu 16.12.2015 o 15:14, Kamil Rytarowski pisze:
> ping?
>
> W dniu 09.12.2015 o 14:19, Kamil Rytarowski pisze:
>> Currently dpdk_nic_bind.py detects Linux kernel modules via reading
>> /proc/modules. Built-in ones aren't listed there and therefore they
>> are not
>> being found by the sc
63 matches
Mail list logo
