[dpdk-dev] [PATCH 0/2 v4] Fix two compile issues with i686 platform

2014-12-11 15:28, Qiu, Michael: > On 2014/12/11 21:26, Neil Horman wrote: > > On Thu, Dec 11, 2014 at 01:56:06AM +0100, Thomas Monjalon wrote: > >>> These two issues are both introuduced by commit b77b5639: > >>> mem: add huge page sizes for IBM Power > >>> > >>> Michael Qiu (2): > >>>

[dpdk-dev] [PATCH RFC v2 08/12] lib/librte_vhost: vhost-user support

> -Original Message- > From: Xie, Huawei > Sent: Thursday, December 11, 2014 10:13 AM > To: 'Linhaifeng'; dev at dpdk.org > Cc: haifeng.lin at intel.com > Subject: RE: [dpdk-dev] [PATCH RFC v2 08/12] lib/librte_vhost: vhost-user > support > > > > > Only support one vhost-user port ? >

[dpdk-dev] [PATCH] vmxnet3: set txq_flags in default TX conf

> -Original Message- > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Pablo de Lara > Sent: Thursday, December 11, 2014 5:08 PM > To: dev at dpdk.org > Cc: root > Subject: [dpdk-dev] [PATCH] vmxnet3: set txq_flags in default TX conf > > Since commit fbde27f19ab8f "get default

[dpdk-dev] i40e 3rdparty QSFP module support

Hi, I have an i40e card with a Finisar QSFP (40G) module. The dpdk driver does not seem to detect the correct mac type for this QSFP module. Inside: STATIC enum i40e_media_type i40e_get_media_type(struct i40e_hw *hw) hw->phy.link_info.phy_type has value 0, which does not match up

[dpdk-dev] Build sucess DPDK 1.7.1 in UBunutu, testpmd failed with librte_distributor.so

Hi! I am trying these commands. based on suggestion http://dpdk.org/ml/archives/dev/2014-October/006247.html $RTE_TARGET=/x86_64-ivshmem-linuxapp-gcc $make CONFIG_RTE_BUILD_COMBINE_LIBS=y CONFIG_RTE_BUILD_SHARED_LIB=y install T="$RTE_TARGET" EXTRA_LDFLAGS=--no-as-needed Build sucessfully.

[dpdk-dev] [PATCH RFC v2 08/12] lib/librte_vhost: vhost-user support

> > Only support one vhost-user port ? Do you mean vhost server by "port"? If that is the case, yes, now only one vhost server is supported for multiple virtio devices. As stated in the cover letter, we have requirement and plan for multiple server support, though I am not sure if it is

[dpdk-dev] [PATCH] vmxnet3: set txq_flags in default TX conf

Since commit fbde27f19ab8f "get default Rx/Tx configuration from dev info", a default RX/TX configuration can be used for all PMDs. In case of vmxnet3, the whole structure was zeroed and not filled out. The PMD does not support multi segments or offload functions, so txq_flags should have those

[dpdk-dev] [PATCH v2] ixgbe: fix multi-process support

On Thu, Dec 11, 2014 at 04:58:42PM +, Pattan, Reshma wrote: > > > > -Original Message- > > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Bruce Richardson > > Sent: Friday, December 5, 2014 1:46 PM > > To: dev at dpdk.org > > Subject: [dpdk-dev] [PATCH v2] ixgbe: fix

[dpdk-dev] [PATCH v3 25/28] eal/pci: Remove pci_probe/close_all_drivers()

(2014/12/11 15:02), Qiu, Michael wrote: > Why you do not call pci_invoke_all_drivers() directly in former patches? > > What you do is first modify then remove I have received comments for former patches internally. And you know, this patch series are long. For those reviewers, I didn't want

[dpdk-dev] [PATCH v3 23/28] eal/pci: Add rte_eal_pci_probe_one and rte_eal_pci_close_one

(2014/12/11 14:54), Qiu, Michael wrote: > On 12/9/2014 2:34 PM, Tetsuya Mukawa wrote: >> The functions are used for probe and close a device. >> First the function tries to find a device that has the specfied PCI address. >> Then, probe or close the device. >> >> Signed-off-by: Tetsuya Mukawa >>

[dpdk-dev] [PATCH 0/2 v4] Fix two compile issues with i686 platform

On 2014/12/11 21:26, Neil Horman wrote: > On Thu, Dec 11, 2014 at 01:56:06AM +0100, Thomas Monjalon wrote: >>> These two issues are both introuduced by commit b77b5639: >>> mem: add huge page sizes for IBM Power >>> >>> Michael Qiu (2): >>> Fix compile issue with hugepage_sz in 32-bit

[dpdk-dev] rte_config.h missing in DPDK 1.7.1

Hi Marc! Thank for your response. But I tried double check.Propelly set those values. Actually I found the file inside x86_64-native-linuxapp-gcc/include/rte_config.h. But when I compile DPDK OVS, it does not find the file. I got same error/ lib/dpdk-link.c:25:24: fatal error: rte_config.h: No

[dpdk-dev] [PATCH v3 18/28] eal/pci: Prevent double registrations for pci_device_list

(2014/12/11 12:24), Qiu, Michael wrote: > If you modify one function, make sure all places, where call the > function, have been modified accordingly in same patch. > > Please do not split it. Because we must make sure every patch could > make dpdk work( I think without this patch, it will have

[dpdk-dev] [PATCH RFC v2 08/12] lib/librte_vhost: vhost-user support

On 2014/12/11 5:37, Huawei Xie wrote: > vhost-user support > > > Signed-off-by: Huawei Xie > --- > lib/librte_vhost/Makefile | 5 +- > lib/librte_vhost/vhost-net.h | 4 + > lib/librte_vhost/vhost_cuse/virtio-net-cdev.c | 9 + >

[dpdk-dev] [PATCH v3 21/28] eal/pci: Fix pci_probe_all_drivers to share code with closing function

(2014/12/11 13:46), Qiu, Michael wrote: > On 12/11/2014 11:52 AM, Qiu, Michael wrote: >> On 12/9/2014 2:34 PM, Tetsuya Mukawa wrote: >>> pci_close_all_drivers() will be implemented after the patch. >>> To share a part of code between thses 2 functions, The patch fixes >>> pci_probe_all_drivers()

[dpdk-dev] [PATCH v3 13/28] eal/pci: Prevent double registration for devargs_list

Hi Michael, (2014/12/09 23:55), Qiu, Michael wrote: > On 2014/12/9 14:33, Tetsuya Mukawa wrote: >> The patch fixes rte_eal_devargs_add() not to register same device twice. >> >> Signed-off-by: Tetsuya Mukawa >> --- >> lib/librte_eal/common/eal_common_devargs.c | 35 >>

[dpdk-dev] [PATCH v3 07/28] ethdev: Add functions to know which port is attached or detached

(2014/12/11 12:35), Qiu, Michael wrote: > On 12/11/2014 11:12 AM, Tetsuya Mukawa wrote: >> Hi Michael, >> >> (2014/12/09 23:39), Qiu, Michael wrote: >>> On 2014/12/9 14:32, Tetsuya Mukawa wrote: The patch adds rte_eth_dev_save() and rte_eth_dev_get_changed_port(). rte_eth_dev_save() is

[dpdk-dev] [PATCH RFC v2 08/12] lib/librte_vhost: vhost-user support

On 2014/12/11 5:37, Huawei Xie wrote: > vhost-user support > > > Signed-off-by: Huawei Xie > --- > lib/librte_vhost/Makefile | 5 +- > lib/librte_vhost/vhost-net.h | 4 + > lib/librte_vhost/vhost_cuse/virtio-net-cdev.c | 9 + >

[dpdk-dev] [PATCH] librte_net: fix TX checksum offload problem with IPv6 packet.

Hi Konstantin, On 12/11/2014 12:49 PM, Konstantin Ananyev wrote: > For rte_ipv6_phdr_cksum() gcc 4.8.* with "-O3" not always generates > correct code. > Sometimes it 'forgets' to put len and proto fields of psd_header on the stack. > To overcome that problem and speedup things a bit, refactored

[dpdk-dev] [PATCH] kni: fix build on RHEL6.5

RHEL6.5 kernel is based on 2.6.32. But there are two changing from 2.6.35: 1. socket struct is changed It wrappered previous wait_queue_head_t of socket to struct socket_wq. So for the kernel older than 2.6.35, we should directly use socket->wait instead. 2. new function sk_sleep() This function

[dpdk-dev] [PATCH 14/15] app/test: turn off cpu flag checks for tile architecture

>-Original Message- >From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Neil Horman >Sent: Tuesday, December 09, 2014 11:03 PM >To: Zhigang Lu >Cc: dev at dpdk.org >Subject: Re: [dpdk-dev] [PATCH 14/15] app/test: turn off cpu flag checks for tile >architecture > >On Mon, Dec 08, 2014

[dpdk-dev] rte_config.h missing in DPDK 1.7.1

Hi! I am seeking help on installing DPDK OVS with DPDK 1.7.1 on ubuntu 14.04. If any body knows solutin, please let me. I compile in a typical way for DPDK 1.7.1 with x86_64_native-linuxaap-gcc. command: make config T=x86_64-native-linuxapp-gcc make It compiled and worked. I am able to run

[dpdk-dev] [PATCH v4] mbuf: fix of enabling all newly added RX error flags

Hi Helin, On 12/10/2014 11:29 PM, Zhang, Helin wrote: > --- a/lib/librte_mbuf/rte_mbuf.h > +++ b/lib/librte_mbuf/rte_mbuf.h > @@ -83,12 +83,7 @@ extern "C" { > #define PKT_RX_RSS_HASH (1ULL << 1) /**< RX packet with >> RSS hash result. */ > #define PKT_RX_FDIR

[dpdk-dev] [PATCH v3 15/28] eal/pci: Add probe and close function for virtual drivers

Hi Michael, (2014/12/10 0:51), Qiu, Michael wrote: > On 2014/12/9 14:33, Tetsuya Mukawa wrote: >> The patch adds rte_eal_dev_init_one() and rte_eal_dev_close_one(). >> These are used for attaching and detaching virtual devices. >> >> Signed-off-by: Tetsuya Mukawa >> --- >>

[dpdk-dev] [PATCH v3 07/28] ethdev: Add functions to know which port is attached or detached

Hi Michael, (2014/12/09 23:39), Qiu, Michael wrote: > On 2014/12/9 14:32, Tetsuya Mukawa wrote: >> The patch adds rte_eth_dev_save() and rte_eth_dev_get_changed_port(). >> rte_eth_dev_save() is used for saving current rte_eth_dev structures. >> rte_eth_dev_get_changed_port() receives the

[dpdk-dev] [PATCH v3 03/28] eal/pci: Replace pci address comparison code by eal_compare_pci_addr

Hi Michael, (2014/12/09 23:22), Qiu, Michael wrote: > On 2014/12/9 14:32, Tetsuya Mukawa wrote: >> This patch replaces pci_addr_comparison() and memcmp() of pci addresses by >> eal_compare_pci_addr(). >> >> Signed-off-by: Tetsuya Mukawa >> --- >> lib/librte_eal/bsdapp/eal/eal_pci.c | 16

[dpdk-dev] [PATCH v3 3/3] app/testpmd:change tx_checksum command and csum forwarding engine

Hi Jijiang, Some more comments, in addition to the one I've made in the cover letter. Reference link for patchwork readers: http://dpdk.org/ml/archives/dev/2014-December/009886.html On 12/10/2014 02:03 AM, Jijiang Liu wrote: > --- a/app/test-pmd/cmdline.c > +++ b/app/test-pmd/cmdline.c > @@

[dpdk-dev] [PATCH] librte_net: fix TX checksum offload problem with IPv6 packet.

For rte_ipv6_phdr_cksum() gcc 4.8.* with "-O3" not always generates correct code. Sometimes it 'forgets' to put len and proto fields of psd_header on the stack. To overcome that problem and speedup things a bit, refactored rte_raw_cksum() by splitting ipv6 pseudo-header csum calculation into 3

[dpdk-dev] In DPDK 1.7.1, the link status of the interface using virtio driver is always down.

Hi Changchun, I found you had done follow change to allow the virtio interface startup when the link is down. Is there any scenario causing link down for virtio interface? diff --git a/lib/librte_pmd_virtio/virtio_ethdev.c b/lib/librte_pmd_virtio/virtio_ethdev.c index 78018f9..4bff0fe 100644

[dpdk-dev] [PATCH v3 2/3] i40e:support outer IPv4 checksum capability

On 12/10/2014 02:03 AM, Jijiang Liu wrote: > The DEV_TX_OFFLOAD_OUTER_IPV4_CKSUM flag is added in i40e capability set, > which means the i40e supports TX checksum offload of tunneling packet. > > Signed-off-by: Jijiang Liu > --- > lib/librte_pmd_i40e/i40e_ethdev.c |3 ++- > 1 files

[dpdk-dev] [PATCH v3 1/3] librte_ether:add outer IP offload capability flag

On 12/10/2014 02:03 AM, Jijiang Liu wrote: > If the flag is set in a PMD, which means the NIC(s) support TX checksum > offload of tunneling packet. > > Signed-off-by: Jijiang Liu > --- > lib/librte_ether/rte_ethdev.h |1 + > 1 files changed, 1 insertions(+), 0 deletions(-) > > diff --git

[dpdk-dev] [PATCH RFC v2 00/12] lib/librte_vhost: vhost-user support

Hi Xie, (2014/12/11 7:04), Xie, Huawei wrote: > This patch set is based on latest vhost. > It fixes vhost-user memory map/unmap alignment issue. > It uses VHOST_USER_GET_VRING_BASE as the message for vhost device stop > in vhost-user. > It uses VHOST_SET_VRING_KICK as the

[dpdk-dev] [PATCH v3 0/3] enhance TX checksum command and csum forwarding engine

Hi Jijiang, Sorry for the late review, I was very busy these last days. Please find my comments below. On 12/10/2014 02:03 AM, Jijiang Liu wrote: > In the current codes, the "tx_checksum set (ip|udp|tcp|sctp|vxlan) (hw|sw) > (port-id)" command is not easy to understand and extend, so the patch

[dpdk-dev] [PATCH v4 0/4] compatibility fallback and replacement of kernel function invoking

On 12/11/2014 09:00 AM, Thomas Monjalon wrote: >> The related kernel function is: >> - pci_num_vf, it is introduced from upstream linux-2.6.34. For RHEL-based >> kernel, it is defined from RHEL5.9. >> >> - kstrtoul, this function is united kernel API to replace strict_strtoul in >> the furture.

[dpdk-dev] [PATCH v3 14/28] eal/pci: Add rte_eal_devargs_remove

Hi Michael, (2014/12/10 0:36), Qiu, Michael wrote: > I don't know if other reviewers ask you to split so many patches. But I > would like merge some of them, because some are doing same affairs(just > using different args), others should be add/remove affairs. Those could > be merge to one patch,

[dpdk-dev] [PATCH v3 3/3] doc: add VM power mgmt app

> -Original Message- > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Pablo de Lara > Sent: Tuesday, December 9, 2014 12:02 PM > To: dev at dpdk.org > Subject: [dpdk-dev] [PATCH v3 3/3] doc: add VM power mgmt app > > Added new section in sample app UG for > the new VM power

[dpdk-dev] [PATCH v3 2/3] doc: add vm power mgmt request sequence svg

> -Original Message- > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Pablo de Lara > Sent: Tuesday, December 9, 2014 12:02 PM > To: dev at dpdk.org > Subject: [dpdk-dev] [PATCH v3 2/3] doc: add vm power mgmt request sequence svg > > Added second of the two figures in the VM

[dpdk-dev] A question about hugepage initialization time

On Wed, Dec 10, 2014 at 07:16:59PM +, L?szl? Vadkerti wrote: > na ez :) > > On Wed, 10 Dec 2014, Bruce Richardson wrote: > > > On Wed, Dec 10, 2014 at 09:29:26AM -0500, Neil Horman wrote: > >> On Wed, Dec 10, 2014 at 10:32:25AM +, Bruce Richardson wrote: > >>> On Tue, Dec 09, 2014 at

[dpdk-dev] [PATCH v3 1/3] doc: add vm power mgmt overview svg

> -Original Message- > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Pablo de Lara > Sent: Tuesday, December 9, 2014 12:02 PM > To: dev at dpdk.org > Subject: [dpdk-dev] [PATCH v3 1/3] doc: add vm power mgmt overview svg > > Added first of the two figures in the VM power

[dpdk-dev] [PATCH] doc: fix typos in ring lib

> -Original Message- > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Pablo de Lara > Sent: Tuesday, December 9, 2014 12:28 PM > To: dev at dpdk.org > Subject: [dpdk-dev] [PATCH] doc: fix typos in ring lib > > Ring library section in PG had a couple of typos, in the text and in

[dpdk-dev] [RFC PATCH 6/7] sample: add new sample for multi-pthread

Signed-off-by: Cunming Liang --- examples/multi-pthread/Makefile | 57 ++ examples/multi-pthread/main.c | 232 examples/multi-pthread/main.h | 46 3 files changed, 335 insertions(+) create mode 100644

[dpdk-dev] [RFC PATCH 5/7] testpmd: support multi-pthread mode

Signed-off-by: Cunming Liang --- app/test-pmd/cmdline.c | 41 app/test-pmd/testpmd.c | 84 +- app/test-pmd/testpmd.h | 1 + 3 files changed, 125 insertions(+), 1 deletion(-) diff --git a/app/test-pmd/cmdline.c

[dpdk-dev] [RFC PATCH 4/7] eal: add simple API for multi-pthread

Signed-off-by: Cunming Liang --- config/common_linuxapp | 1 + lib/librte_eal/common/include/rte_eal.h | 10 +++ lib/librte_eal/linuxapp/eal/eal_thread.c | 105 ++- 3 files changed, 115 insertions(+), 1 deletion(-) diff --git

[dpdk-dev] [RFC PATCH 3/7] ring: use linear-tid as ring debug stats index

Signed-off-by: Cunming Liang --- lib/librte_ring/rte_ring.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/lib/librte_ring/rte_ring.h b/lib/librte_ring/rte_ring.h index 3920830..c038a4f 100644 --- a/lib/librte_ring/rte_ring.h +++ b/lib/librte_ring/rte_ring.h @@ -189,9

[dpdk-dev] [RFC PATCH 2/7] mempool: use linear-tid as mempool cache index

Signed-off-by: Cunming Liang --- lib/librte_mempool/rte_mempool.h | 22 +++--- 1 file changed, 11 insertions(+), 11 deletions(-) diff --git a/lib/librte_mempool/rte_mempool.h b/lib/librte_mempool/rte_mempool.h index 3314651..bf4117b 100644 --- a/lib/librte_mempool/rte_mempool.h

[dpdk-dev] [RFC PATCH 1/7] eal: add linear thread id as pthread-local variable

Signed-off-by: Cunming Liang --- lib/librte_eal/common/include/rte_eal.h | 5 ++ lib/librte_eal/common/include/rte_lcore.h | 12 lib/librte_eal/linuxapp/eal/eal_thread.c | 115 -- 3 files changed, 126 insertions(+), 6 deletions(-) diff --git

[dpdk-dev] [RFC PATCH 0/7] support multi-phtread per lcore

Scope & Usage Scenario DPDK usually pin pthread per core to avoid task switch overhead. It gains performance a lot, but it's not efficient in all cases. In some cases, it may too expensive to use the whole core for a lightweight workload. It's a reasonable demand to

[dpdk-dev] librte_distributor.so: cannot open shared object file: No such file or directory

Hi! I compiled DPDK v1.7.1 on ubunutu 14.04. I used following commands to compille. cd DPDKexport RTE_SDK=$(pwd)export RTE_TARGET="x86_64-ivshmem-linuxapp-gcc" make CONFIG_RTE_BUILD_COMBINE_LIBS=y CONFIG_RTE_BUILD_SHARED_LIB=y install T="$RTE_TARGET" I mounted hugepages and

[dpdk-dev] Pipeline and RTE_CORE_MAX=128

Hello, After last change for RTE_CORE_MAX=128 the rte_pipeline still supports only 64 ports as it is optimized to work on u64 variables. Are there any plans to extend pipeline design to use 128, 256 or even 512 ports using AVX/SSE extensions available? Regards, Mirek

[dpdk-dev] [RFC PATCH 0/7] support multi-phtread per lcore

Thank you Cunming for explanation. What about DPDK timers? They also depend on rte_lcore_id() to avoid spinlocks. Mirek > -Original Message- > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Cunming Liang > Sent: Thursday, December 11, 2014 3:05 AM > To: dev at dpdk.org >

[dpdk-dev] [PATCH v3 20/28] eal/pci: Add rte_eal_pci_close_one_driver

On Thu, Dec 11, 2014 at 03:41:06AM +, Qiu, Michael wrote: > On 12/9/2014 2:33 PM, Tetsuya Mukawa wrote: > > The function is used for closing the specified driver and device. > > > > Signed-off-by: Tetsuya Mukawa > > --- > > lib/librte_eal/common/eal_private.h | 15 + > >

[dpdk-dev] [PATCH v3] test-pmd: Fix pointer aliasing error

Thomas, Michael, Wouldn't it cause unaligned memory access (new changes as well as the previous code)? Wondering if get_unaligned/put_unaligned macros similar to the ones used in kernel be ported to user-space? Thanks, Ravi On Wed, Dec 10, 2014 at 4:54 PM, Thomas Monjalon wrote: > > > >

[dpdk-dev] lib: include rte_memory.h for __rte_cache_aligned

On Thu, Dec 11, 2014 at 01:36:54AM +0100, Thomas Monjalon wrote: > Hi Neil, > > 2014-12-10 19:28, Neil Horman: > > On Wed, Dec 10, 2014 at 07:09:03PM +, Jia Yu wrote: > > > Hi Neil, > > > > > > Moving __rte_cache_aligned right after struct keyword will help. On the > > > other hand,

[dpdk-dev] In DPDK 1.7.1, the link status of the interface using virtio driver is always down.

Hi, The result is still the same. [root at EIPU-0(KVMCluster) /root] # ./testpmd -c 3 -n 4 -- --burst=64 -i --txq=1 --rxq=1 --txqflags=0x EAL: Cannot read numa node link for lcore 0 - using physical package id instead EAL: Detected lcore 0 as core 0 on socket 0 EAL: Cannot read numa node

[dpdk-dev] [PATCH 14/15] app/test: turn off cpu flag checks for tile architecture

On Thu, Dec 11, 2014 at 12:43:36PM +0800, Tony Lu wrote: > >-Original Message- > >From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Neil Horman > >Sent: Tuesday, December 09, 2014 11:03 PM > >To: Zhigang Lu > >Cc: dev at dpdk.org > >Subject: Re: [dpdk-dev] [PATCH 14/15] app/test:

[dpdk-dev] In DPDK 1.7.1, the link status of the interface using virtio driver is always down.

Hi, > -Original Message- > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Fu, Weiyi (NSN - > CN/Hangzhou) > Sent: Thursday, December 11, 2014 3:57 PM > To: dev at dpdk.org > Subject: [dpdk-dev] In DPDK 1.7.1, the link status of the interface using > virtio > driver is always

[dpdk-dev] In DPDK 1.7.1, the link status of the interface using virtio driver is always down.

Hi, We are using the l2fwd based on DPDK 1.7.1 and found out that the link status of the interface using virtio driver is always down. Is there any precondition to let the link up? # ./pci_unbind.py --status Network devices using IGB_UIO driver :00:04.0

[dpdk-dev] [PATCH v3 0/6] i40e VMDQ support

Tested-by: Min Cao Patch name: i40e VMDQ support Brief description: Test Flag: Tested-by Tester name:min.cao at intel.com Result summary: total 1 cases, 1 passed, 0 failed Test Case 1: Name:

[dpdk-dev] [PATCH v3 25/28] eal/pci: Remove pci_probe/close_all_drivers()

Why you do not call pci_invoke_all_drivers() directly in former patches? What you do is first modify then remove Thanks, Michael On 12/9/2014 2:34 PM, Tetsuya Mukawa wrote: > These functions are actually wrappers of pci_invoke_all_drivers(). > Just call it directly. > > Signed-off-by:

[dpdk-dev] [PATCH v3 23/28] eal/pci: Add rte_eal_pci_probe_one and rte_eal_pci_close_one

On 12/9/2014 2:34 PM, Tetsuya Mukawa wrote: > The functions are used for probe and close a device. > First the function tries to find a device that has the specfied PCI address. > Then, probe or close the device. > > Signed-off-by: Tetsuya Mukawa > --- > lib/librte_eal/common/eal_common_pci.c |

[dpdk-dev] In DPDK 1.7.1, the link status of the interface using virtio driver is always down.

Hi, I have seen this issue happen on older kernels like 2.6.32-220.el6.x86_64 while it works with no issues on a recent kernel like 3.10.x. Further, I found this issue was happening due to /sys/bus/pci/devices//msi_irqs dir not being enumerated in older kernels resulting in hw->use_msix=0. This

[dpdk-dev] [PATCH RFC v2 12/12] lib/librte_vhost: cleanup when vhost user socket connection is closed

close the memory region file descriptor close the kick/callfd vSwitch needs to run endlessly. resource leak is deadly issue. Signed-off-by: Huawei Xie --- lib/librte_vhost/vhost_user/vhost-net-user.c | 7 ++-- lib/librte_vhost/vhost_user/virtio-net-user.c | 59 +++

[dpdk-dev] [PATCH RFC v2 10/12] lib/librte_vhost: vhost-user memory region map

deals with vhost user memory map/unmap alignment Signed-off-by: Huawei Xie --- lib/librte_vhost/rte_virtio_net.h | 2 + lib/librte_vhost/vhost-net.h | 2 - lib/librte_vhost/vhost_user/vhost-net-user.h | 3 +- lib/librte_vhost/vhost_user/virtio-net-user.c |

[dpdk-dev] [PATCH RFC v2 09/12] lib/librte_vhost: minor fix

Signed-off-by: Huawei Xie --- lib/librte_vhost/vhost_user/vhost-net-user.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/lib/librte_vhost/vhost_user/vhost-net-user.c b/lib/librte_vhost/vhost_user/vhost-net-user.c index 841d7e6..6b9ebd7 100644 ---

[dpdk-dev] [PATCH RFC v2 08/12] lib/librte_vhost: vhost-user support

vhost-user support Signed-off-by: Huawei Xie --- lib/librte_vhost/Makefile | 5 +- lib/librte_vhost/vhost-net.h | 4 + lib/librte_vhost/vhost_cuse/virtio-net-cdev.c | 9 + lib/librte_vhost/vhost_user/vhost-net-user.c | 422 ++

[dpdk-dev] [PATCH RFC v2 07/12] lib/librte_vhost: async event and callback

Signed-off-by: Huawei Xie --- lib/librte_vhost/vhost_user/fd_man.c | 205 +++ lib/librte_vhost/vhost_user/fd_man.h | 64 +++ 2 files changed, 269 insertions(+) create mode 100644 lib/librte_vhost/vhost_user/fd_man.c create mode 100644

[dpdk-dev] [PATCH RFC v2 06/12] lib/librte_vhost: cuse_set_memory_table

cuse_set_memory_table Signed-off-by: Huawei Xie --- lib/librte_vhost/Makefile | 2 +- lib/librte_vhost/vhost-net.h | 4 +- lib/librte_vhost/vhost_cuse/vhost-net-cdev.c | 7 +- lib/librte_vhost/vhost_cuse/virtio-net-cdev.c | 85 +++

[dpdk-dev] [PATCH RFC v2 05/12] lib/librte_vhost: host_memory_map refine

host_memory_map only maps partial memory of target process into current process through shared backed file. Signed-off-by: Huawei Xie --- lib/librte_vhost/vhost_cuse/virtio-net-cdev.c | 43 +-- 1 file changed, 20 insertions(+), 23 deletions(-) diff --git

[dpdk-dev] [PATCH RFC v2 04/12] lib/librte_vhost: copy of host_memory_map from virtio-net.c to new file virtio-net-cdev.c

create virtio-net-cdev.c and copy host_memory_map into it. Signed-off-by: Huawei Xie --- lib/librte_vhost/vhost_cuse/virtio-net-cdev.c | 258 ++ 1 file changed, 258 insertions(+) create mode 100644 lib/librte_vhost/vhost_cuse/virtio-net-cdev.c diff --git

[dpdk-dev] [PATCH RFC v2 02/12] lib/librte_vhost: rename vhost-net-cdev.h as vhost-net.h

vhost-net.h is the shared header file for both vhost-cuse and vhost-user Signed-off-by: Huawei Xie --- lib/librte_vhost/vhost-net-cdev.h| 113 --- lib/librte_vhost/vhost-net.h | 113 +++

[dpdk-dev] [PATCH RFC v2 01/12] lib/librte_vhost: mov vhost-cuse implementation to vhost_cuse directory

create vhost_cuse directory and move vhost-net-cdev.c to vhost_cuse directory Signed-off-by: Huawei Xie --- lib/librte_vhost/Makefile| 4 +- lib/librte_vhost/vhost-net-cdev.c| 389 --- lib/librte_vhost/vhost_cuse/vhost-net-cdev.c | 389

[dpdk-dev] [PATCH RFC v2 00/12] lib/librte_vhost: vhost-user support

This patchset refines vhost library to support both vhost-cuse and vhost-user. Huawei Xie (12): create vhost_cuse directory and move vhost-net-cdev.c to vhost_cuse directory rename vhost-net-cdev.h as vhost-net.h move eventfd_copy logic out from virtio-net.c to vhost-net-cdev.c exact

[dpdk-dev] [PATCH v3 21/28] eal/pci: Fix pci_probe_all_drivers to share code with closing function

On 12/11/2014 11:52 AM, Qiu, Michael wrote: > On 12/9/2014 2:34 PM, Tetsuya Mukawa wrote: >> pci_close_all_drivers() will be implemented after the patch. >> To share a part of code between thses 2 functions, The patch fixes >> pci_probe_all_drivers() first. >> >> Signed-off-by: Tetsuya Mukawa >>

[dpdk-dev] [PATCH v3 07/28] ethdev: Add functions to know which port is attached or detached

On 12/11/2014 11:12 AM, Tetsuya Mukawa wrote: > Hi Michael, > > (2014/12/09 23:39), Qiu, Michael wrote: >> On 2014/12/9 14:32, Tetsuya Mukawa wrote: >>> The patch adds rte_eth_dev_save() and rte_eth_dev_get_changed_port(). >>> rte_eth_dev_save() is used for saving current rte_eth_dev structures.

[dpdk-dev] [PATCH v3 18/28] eal/pci: Prevent double registrations for pci_device_list

If you modify one function, make sure all places, where call the function, have been modified accordingly in same patch. Please do not split it. Because we must make sure every patch could make dpdk work( I think without this patch, it will have some issue after you applied "Replace pci address

[dpdk-dev] [PATCH v2] add one option memory-only for secondary processes

Hi, sorry for the delay. > Subject: RE: [dpdk-dev] [PATCH v2] add one option memory-only for secondary > processes > > Hi, Hiroshi, > Yes, you are right, in order to avoid such problem, while create the mempool, > which shall be shared between the primary > process and those secondary

[dpdk-dev] [RFC PATCH 0/7] support multi-phtread per lcore

Steve, Great write up. Nice explanation of 1) per-lcore numbering and 2) Multi-producer/consumer enqueue -dequeue. Thanks, -Original Message- From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Cunming Liang Sent: Wednesday, December 10, 2014 6:05 PM To: dev at dpdk.org Subject:

[dpdk-dev] [PATCH] mk: fix link to combined library

From: Hiroshi Shimamoto The application should be linked to the single combined library in the condition that both of CONFIG_RTE_BUILD_COMBINE_LIB and CONFIG_RTE_BUILD_SHARED_LIB are enabled. The current makefile generates an application that links to each library.

[dpdk-dev] [PATCH v2] app/test: fix memory needs after RTE_MAX_LCORE was increased to 128

2014-12-10 15:40, Thomas Monjalon: > 2014-12-09 10:11, Pablo de Lara: > > Since commit b91c67e5a693211862aa7dc3b78630b4e856c2af, > > maximum number of cores is 128, which has increase > > the total memory necessary for a rte_mempool structure, > > as the per-lcore local cache has been doubled in

[dpdk-dev] [PATCH] examples/vhost: increase MAX_QUEUE number

> increase MAX_QUEUES from 256 to 512 > In vhost example, MAX_QUEUES macro should be the maximum possible queue > number of the port. > Theoretically we should only set up the queues that are used, i.e., first rx > queue of each pool, or > at most queues from 0 to MAX_QUEUES. Before we revise

[dpdk-dev] [PATCH v2] i40e: bug fix of querying reta

> There is a bug in querying reta, of storing the data to the correct entry. > Code changes is straightforward for this bug. > > Signed-off-by: Helin Zhang > --- > lib/librte_pmd_i40e/i40e_ethdev.c| 2 +- > lib/librte_pmd_i40e/i40e_ethdev_vf.c | 2 +- > 2 files changed, 2 insertions(+), 2

[dpdk-dev] [PATCH] enic: fix of compile error

> Compile warnings/errors was found on gcc 4.7.2 as follows. Variables > was reported of being used but uninitialized. Assigning an initial > value to it is needed. > > lib/librte_pmd_enic/vnic/vnic_dev.c: In function vnic_dev_get_mac_addr: > lib/librte_pmd_enic/vnic/vnic_dev.c:393:16: error: a1

[dpdk-dev] [PATCH v2] bond: fix for mac assignment to slaves device

> > -V2: > > Tidies up the slave_remove function as per Pawel's comments. > > > > Adding call to mac_address_slaves_update from the lsc handler when the > > first slave become active to propagate any mac changes made while > > devices are inactive > > > > Changed removing slave logic to use

[dpdk-dev] [PATCH v4 0/4] compatibility fallback and replacement of kernel function invoking

> The related kernel function is: > - pci_num_vf, it is introduced from upstream linux-2.6.34. For RHEL-based > kernel, it is defined from RHEL5.9. > > - kstrtoul, this function is united kernel API to replace strict_strtoul in > the furture. It is introduced from linux-2.6.39. For RHEL6, it is

[dpdk-dev] [PATCH] rte_eth_ctrl: rename rte_eth_fdir_flow.ip6_flow

> > The name of the rte_eth_fdir_flow's rte_eth_ipv6_flow attribute, > > 'ip6_flow', clashes with a macro defined in > > /usr/include/netinet/ip6.h, such that when DPDK is linked with an > > application that uses the afforementioned header, the macro is > > expanded within the DPDK struct, causing

[dpdk-dev] [PATCH 0/2 v4] Fix two compile issues with i686 platform

> These two issues are both introuduced by commit b77b5639: > mem: add huge page sizes for IBM Power > > Michael Qiu (2): > Fix compile issue with hugepage_sz in 32-bit system > Fix compile issue of eal with icc compile Acked-by: Thomas Monjalon Applied Thanks -- Thomas

[dpdk-dev] lib: include rte_memory.h for __rte_cache_aligned

2014-12-09 10:22, Neil Horman: > On Tue, Dec 09, 2014 at 09:53:18AM +0100, Olivier MATZ wrote: > > On 12/08/2014 04:04 PM, Neil Horman wrote: > > >On Fri, Nov 07, 2014 at 09:28:09AM -0800, Jia Yu wrote: > > >>Include rte_memory.h for lib files that use __rte_cache_aligned > > >>attribute. > > >> >

[dpdk-dev] mk: fix build 32bits shared libs on 64bits system

> > Incompatible libraries error when building shared libraries for 32bits on > > a 64bits system. > > Fix issue by passing CPU_CFLAGS to CC when LINK_USING_CC is enabled. > > > > Signed-off-by: Sergio Gonzalez Monroy > > Acked-by: Pablo de Lara > > Acked-by: Neil Horman Applied Thanks --

[dpdk-dev] mk: fix LDFLAGS for shared lib

2014-12-08 23:44, Hiroshi Shimamoto: > > On Wed, Dec 03, 2014 at 11:33:54PM +, Hiroshi Shimamoto wrote: > > > From: Hiroshi Shimamoto > > > > > > Only CPU_LDFLAGS is used in mk/rte.sharelib.mk. > > > It should be LDFLAGS to build the library with correct linkage options. > > > > > >

[dpdk-dev] [PATCH] Avoid possible memory cpoy when sort hugepages

On 12/11/2014 5:37 AM, Jay Rolette wrote: > On Wed, Dec 10, 2014 at 12:39 PM, Ananyev, Konstantin < > konstantin.ananyev at intel.com> wrote: > >>> I just got through replacing that entire function in my repo with a call >> to qsort() from the standard library last night myself. Faster >>>

[dpdk-dev] lib: include rte_memory.h for __rte_cache_aligned

Hi Neil, 2014-12-10 19:28, Neil Horman: > On Wed, Dec 10, 2014 at 07:09:03PM +, Jia Yu wrote: > > Hi Neil, > > > > Moving __rte_cache_aligned right after struct keyword will help. On the > > other hand, enforcing this rule for existing (100+) and future definitions > > will be difficult.