2016-04-08 11:22, Thomas Monjalon:
> These stats will be compiled when adding +debug
> to test-build.sh targets:
> CONFIG_RTE_LIBRTE_IP_FRAG_TBL_STAT
> CONFIG_RTE_SCHED_COLLECT_STATS
> CONFIG_RTE_PORT_STATS_COLLECT
> CONFIG_RTE_TABLE_STATS_COLLECT
> CONFIG_RTE_PIPELINE
2016-04-12 00:13, Yuanhan Liu:
> On Mon, Apr 11, 2016 at 05:39:05PM +0200, Thomas Monjalon wrote:
> > The git messages have three parts:
> > 1/ the headline
> > 2/ the explanations
> > 3/ the footer tags
> >
> > The headline helps to quickly browse an history or catch instantly the
> > purpose of
2016-04-11 23:54, Thomas Monjalon:
> Replace some hard-coded section numbers by dynamic links.
>
> Signed-off-by: Thomas Monjalon
Applied
Replace some hard-coded section numbers by dynamic links.
Signed-off-by: Thomas Monjalon
---
doc/guides/nics/virtio.rst | 2 +-
doc/guides/prog_guide/ip_fragment_reassembly_lib.rst | 2 +-
doc/guides/prog_guide/kernel_nic_interface.rst | 2 ++
doc/guides/prog_guid
Changed symbol on NIC overview table from X to Y to help
clarify the indicated features are supported. The X caused
confusion for some readers.
Also, added * character to indicate partially supported
features. This can be used in the future to direct the reader
to more specific details in the indi
2016-04-11 19:30, Ferruh Yigit:
> fix vhost-kni compile errors because of Linux kernel API changes
>
> - SOCK_ASYNC_WAITDATA renamed to SOCKWQ_ASYNC_WAITDATA
> Linux commit id: 9cd3e072
> Updated in Linux kernel 4.4
>
> - sk_alloc() gets new parameter
> Linux commit id: 11aa9c28b
> Update
2016-04-11 18:55, Jasvinder Singh:
> Fixes the buffer overflow that occurs due to following;
>
> 1. When the input packet burst does not meet the conditions: (a) being
> contiguous (first n bits set in pkts_mask, all the other bits cleared)
> and (b) containing a full burst, i.e. at least tx_burst
2016-04-11 18:54, Jasvinder Singh:
> Fixes the variable bsz_mask type form unit32_t to unit64_t
>
> Fixes: 4d97e8b565cc ("port: ethdev")
> Fixes: 304c8091e90a ("port: add ethdev writer nodrop")
> Fixes: 8dceb6aa6ecf ("port: hierarchical scheduler")
> Fixes: 3e5966837a09 ("port: new Tx burst implem
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Monday, April 11, 2016 10:55 PM
> To: Mcnamara, John
> Cc: dev at dpdk.org
> Subject: [PATCH] doc: fix references in guides
>
> Replace some hard-coded section numbers by dynamic links.
>
> Signe
From: Harry van Haaren
Add roadmap items for 16.07, remove 16.04 from schedule.
Signed-off-by: Harry van Haaren
Signed-off-by: Thomas Monjalon
---
dev/roadmap.html | 67 +---
1 file changed, 35 insertions(+), 32 deletions(-)
v2:
- add more
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Sridhar.V.Iyer
> Sent: Thursday, April 7, 2016 8:24 PM
> To: dev at dpdk.org
> Subject: [dpdk-dev] Packet drops at lower tc transmit-rates.
>
> Hi all,
>
> We are using DPDK 1.7 in our application.
> We are r
2016-04-11 11:24, John Guzik:
> --- a/doc/guides/prog_guide/ip_fragment_reassembly_lib.rst
> +++ b/doc/guides/prog_guide/ip_fragment_reassembly_lib.rst
> -For more information about direct and indirect mbufs, refer to the *DPDK
> Programmers guide 7.7 Direct and Indirect Buffers.*
> +For more info
2016-04-11 11:15, Stephen Hemminger:
> On Mon, 11 Apr 2016 12:46:16 +0200
> Simon K?gstr?m wrote:
> > In file included from [...]lib/librte_table/rte_table_lpm.c:43:0:
> > [...]/dpdk.build/include/rte_lpm.h:484:25: fatal error: rte_lpm_sse.h:
> > No such file or directory
> > #include "rte_lpm_ss
Hi Cristian,
Yes, I mostly agree with your suggestions:
1. We should fix the two obvious bugs (1a and 1b) right away. Jasvinder's
patches look fine.
2. We should take no immediate action on the issue I raised about PMDs
(vector IXGBE) not enqueuing more than 32 packets. We can discuss and
debate;
fix vhost-kni compile errors because of Linux kernel API changes
- SOCK_ASYNC_WAITDATA renamed to SOCKWQ_ASYNC_WAITDATA
Linux commit id: 9cd3e072
Updated in Linux kernel 4.4
- sk_alloc() gets new parameter
Linux commit id: 11aa9c28b
Updated in Linux kernel 4.2
New parameter is: "@kern:
Hi Robert,
I am doing a quick summary below on the changes proposed by these patches:
1. [PRIORITY 1] Bug fixing:
a) Fix buffer overflow issue in rte_port_ring.c (writer, writer_nodrop): double
the tx_buf buffer size (applicable for current code approach)
b) Fix issue with handling burst sizes b
2016-04-11 09:52, Stephen Hemminger:
> I wonder if this matrix would be better if auto-generated some how.
> Either statically from source scan, or dynamically via simple app?
It would be a good exercise.
But I'm afraid the AI would be too complex to judge if the support is
enough to be valid (e.g
2016-04-11 17:54, Ferruh Yigit:
> +#if LINUX_VERSION_CODE >= KERNEL_VERSION(4,4,0)
> + (!test_and_set_bit(SOCKWQ_ASYNC_NOSPACE, &q->sock->flags) &&
> +#else
> (!test_and_set_bit(SOCK_ASYNC_NOSPACE, &q->sock->flags) &&
> +#endif
You could avoid some issues with backported feature
On 4/11/2016 6:04 PM, Thomas Monjalon wrote:
> 2016-04-11 17:54, Ferruh Yigit:
>> +#if LINUX_VERSION_CODE >= KERNEL_VERSION(4,4,0)
>> +(!test_and_set_bit(SOCKWQ_ASYNC_NOSPACE, &q->sock->flags) &&
>> +#else
>> (!test_and_set_bit(SOCK_ASYNC_NOSPACE, &q->sock->flags) &&
>> +#endif
>
Fixes the buffer overflow that occurs due to following;
1. When the input packet burst does not meet the conditions: (a) being
contiguous (first n bits set in pkts_mask, all the other bits cleared)
and (b) containing a full burst, i.e. at least tx_burst_sz packets
(n >= tx_burst_size). This is the
Fixes the variable bsz_mask type form unit32_t to unit64_t
Fixes: 4d97e8b565cc ("port: ethdev")
Fixes: 304c8091e90a ("port: add ethdev writer nodrop")
Fixes: 8dceb6aa6ecf ("port: hierarchical scheduler")
Fixes: 3e5966837a09 ("port: new Tx burst implementation of ring writer")
Fixes: 5f4cd47309d6 (
2016-04-11 16:10, Don Provan:
> I can't believe you guys are seriously considering changing the prefix from
> rte_. That's a nightmare at the practical level, but it really doesn't make
> as much sense as some of you seem to think. I've always been really impressed
> that the names were prefixed
vm_power_manager utilize libvirt API virDomainGetVcpuPinInfo for
retrieve domU vcpu information. This API implemented from version 0.9.3.
Suse11 SP3 32bit default libvirt version is 0.8.8, so there'll be build
error. Add judgement in sample Makefile to alarm unsupport environment.
examples/vm_powe
CID 13307 (#1 of 1): Copy into fixed size buffer (STRING_OVERFLOW)
fixed_size_dest: You might overrun the 128 byte fixed-size string fwd_modes
by copying fwd_eng->fwd_mode_name without checking the length.
Fixes: 769ce6b17835 ("app/testpmd: list forwarding engines")
Signed-off-by: Tomasz Kulasek
fix vhost-kni compile errors because of Linux kernel API changes
- SOCK_ASYNC_WAITDATA renamed to SOCKWQ_ASYNC_WAITDATA
Linux commit id: 9cd3e072
Updated in Linux kernel 4.4
- sk_alloc() gets new parameter
Linux commit id: 11aa9c28b
Updated in Linux kernel 4.2
New parameter is: "@kern:
The git messages have three parts:
1/ the headline
2/ the explanations
3/ the footer tags
The headline helps to quickly browse an history or catch instantly the
purpose of a commit. Making it short with some consistent wording
allows to easily parse it or match some patterns.
The explanations mus
Hi all,
I tried compiling DPDK from source using setup.sh script in tools
directory, but I get this error
[32] Remove KNI module
[33] Remove hugepage mappings
[34] Exit Script
Option: 14
/bin/sh: line 1: cc: command not found
cp: cannot stat
?/root/workspace/dpdk/x86_64-native-linuxapp-gcc/.co
vm_power_manager utilize libvirt API virDomainGetVcpuPinInfo for
retrieve domU vcpu information. This API implemented from version 0.9.3.
Suse11 SP3 32bit default libvirt version is 0.8.8, so there'll be build
error. Add judgement in sample Makefile to skip unsupport environment.
examples/vm_power
2016-03-17 15:47, Christian Ehrhardt:
> With LIBRTE_PMD_XENVIRT enabled testpmd is built in a way to ONLY work
> in XEN environments.
> It will surface as:
>PMD: gntalloc: ioctl error
>EAL: Error - exiting with code: 1
> Cause: Creation of mbuf pool for socket 0 failed
>
> With LIBRTE
An issue is found that DCB cannot be configged on ixgbe
NICs. It's said the TX queue number is not right.
On ixgbe the max TX queue number is not fixed, it depends
on the multi-queue mode. The API rte_eth_dev_configure
should be used to config this mode. But the input of this
API includes TX queue
I can't believe you guys are seriously considering changing the prefix from
rte_. That's a nightmare at the practical level, but it really doesn't make as
much sense as some of you seem to think. I've always been really impressed that
the names were prefixed with rte_ instead of dpdk_. While the
> > Add a new section on tested platforms and nics to the release notes.
> >
> > Signed-off-by: Qian Xu
> > Signed-off-by: John McNamara
>
> Acked-by: Pablo de Lara
Applied, thanks
thanks ferruh yigit!!
cheers
On Mon, Apr 11, 2016 at 2:57 PM, chintu hetam wrote:
> I tried vhost-user mode, and during init of ./vhost-switch it fails
> with
> PMD: eth_ixgbe_dev_init(): port 0 vendorID=0x8086 deviceID=0x1528
> pf queue num: 0, configured vmdq pool num: 64, each vmdq pool has
I tried vhost-user mode, and during init of ./vhost-switch it fails
with
PMD: eth_ixgbe_dev_init(): port 0 vendorID=0x8086 deviceID=0x1528
pf queue num: 0, configured vmdq pool num: 64, each vmdq pool has 2 queues
EAL: Error - exiting with code: 1
Cause: Cannot initialize network ports
it's fail
> vm_power_manager utilize libvirt API virDomainGetVcpuPinInfo to
> retrieve domU vcpu information. This API is implemented from version 0.9.3.
> Suse11 SP3 32bit default libvirt version is 0.8.8.
>
> examples/vm_power_manager/channel_manager.c:
> channel_manager.c:117:3: error: implicit declarati
>Hi all,
>
>I tried compiling DPDK from source using setup.sh script in tools
>directory, but I get this error
>
>[32] Remove KNI module
>[33] Remove hugepage mappings
>
>[34] Exit Script
The way I compile DPDK is this way:
# cd
# export RTE_SDK=`pwd`
# export RTE_TARGET=x86_64-native-linuxapp-g
Yes, that is correct. Closing the socket on failure needs to be added.
Regards,
Patrik
On 04/11/2016 11:34 AM, Christian Ehrhardt wrote:
> I like the approach as well to go for the fix for robustness first.
>
> I was accidentally able to find another testcase to hit the same root
> cause.
> A
Hi!
I'm upgrading from DPDK 2.1 to 16.04-rc4, and have a new build issue
which I didn't see before. It's in the librte_table and happens from
time to time (unfrequently) in my out-of-tree build. It looks like a
race between comilation and SYMLINK-FILE:
[...]
== Build lib/librte_table
CC rte_tab
From: Marvin Liu
vm_power_manager utilize libvirt API virDomainGetVcpuPinInfo to
retrieve domU vcpu information. This API is implemented from version 0.9.3.
Suse11 SP3 32bit default libvirt version is 0.8.8.
examples/vm_power_manager/channel_manager.c:
channel_manager.c:117:3: error: implicit de
From: Marvin Liu
vm_power_manager utilize libvirt API virDomainGetVcpuPinInfo to
retrieve domU vcpu information. This API is implemented from version 0.9.3.
Suse11 SP3 32bit default libvirt version is 0.8.8.
examples/vm_power_manager/channel_manager.c:
channel_manager.c:117:3: error: implicit de
On 04/11/2016 12:12 PM, Thomas Monjalon wrote:
> 2016-04-11 11:33, Olivier Matz:
>> --- a/app/test/test_func_reentrancy.c
>> +++ b/app/test/test_func_reentrancy.c
>> @@ -78,7 +78,7 @@ typedef void (*case_clean_t)(unsigned lcore_id);
>> #define MAX_ITER_TIMES (16)
>> #define
2016-04-11 11:33, Olivier Matz:
> --- a/app/test/test_func_reentrancy.c
> +++ b/app/test/test_func_reentrancy.c
> @@ -78,7 +78,7 @@ typedef void (*case_clean_t)(unsigned lcore_id);
> #define MAX_ITER_TIMES (16)
> #define MAX_LPM_ITER_TIMES (8)
>
> -#define
2016-03-30 15:27, Bruce Richardson:
> On Wed, Mar 30, 2016 at 09:46:34AM +0800, Yuanhan Liu wrote:
> > On Tue, Mar 29, 2016 at 11:29:46PM +0200, Thomas Monjalon wrote:
> > > The git messages have three parts:
> > > 1/ the headline
> > > 2/ the explanations
> > > 3/ the footer tags
> > >
> > > The
2016-04-11 16:24, Wenzhuo Lu:
> An issue is found that DCB cannot be configged on ixgbe
> NICs. It's said the TX queue number is not right.
> On ixgbe the max TX queue number is not fixed, it depends
> on the multi-queue mode. The API rte_eth_dev_configure
> should be used to config this mode. But
CID 13289 (#1-2 of 2): Resource leak (RESOURCE_LEAK):
The system resource will not be reclaimed and reused, reducing the future
availability of the resource.
In pci_vfio_get_group_fd: Leak of memory or pointers to system resources
Fixes: ff0b67d1c868 ("vfio: DMA mapping")
Signed-off-by: Daniel M
vm_power_manager utilize libvirt API virDomainGetVcpuPinInfo for
retrieve domU vcpu information. This API implemented from version 0.9.3.
Suse11 SP3 32bit default libvirt version is 0.8.8, so there'll be build
error. Add judgement in sample Makefile to skip unsupport environment.
examples/vm_power
On Mon, Apr 11, 2016 at 12:32:52PM +0200, Thomas Monjalon wrote:
> From: Marvin Liu
>
> vm_power_manager utilize libvirt API virDomainGetVcpuPinInfo to
> retrieve domU vcpu information. This API is implemented from version 0.9.3.
> Suse11 SP3 32bit default libvirt version is 0.8.8.
>
> examples/
I like the approach as well to go for the fix for robustness first.
I was accidentally able to find another testcase to hit the same root cause.
Adding guests with 15 vhost_user based NICs each while having rxq for
openvswitch-dpdk set to 4 and multiqueue for the guest devices at 4 already
breaks
The mempool element size is set to 0, but 4 bytes are written in
my_obj_init():
uint32_t *objnum = obj;
memset(obj, 0, mp->elt_size);
*objnum = i;
Change the MEMPOOL_ELT_SIZE constant to sizeof(uint32_t). This fixes
memory corruptions since we were writting outside of the object
boundaries.
2016-04-11 11:26, Thomas Monjalon:
> 2016-04-11 16:50, Marvin Liu:
> > Fixes: 2e099bc5d104 ("fix split of compiler and linker options")
>
> As commented earlier, I don't think it is the origin of the issue.
>
> > +$(info "vm_power_manager required libvirt version >= 0.9.3, please update
> > libv
2016-04-11 16:50, Marvin Liu:
> Fixes: 2e099bc5d104 ("fix split of compiler and linker options")
As commented earlier, I don't think it is the origin of the issue.
> +$(info "vm_power_manager required libvirt version >= 0.9.3, please update
> libvirt-devel first")
"required" should be "requires
---
doc/guides/prog_guide/ip_fragment_reassembly_lib.rst | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/doc/guides/prog_guide/ip_fragment_reassembly_lib.rst
b/doc/guides/prog_guide/ip_fragment_reassembly_lib.rst
index 1d3d4ac..196d93c 100644
--- a/doc/guides/prog_guide/ip_fra
On Mon, 11 Apr 2016 12:46:16 +0200
Simon K?gstr?m wrote:
> Hi!
>
> I'm upgrading from DPDK 2.1 to 16.04-rc4, and have a new build issue
> which I didn't see before. It's in the librte_table and happens from
> time to time (unfrequently) in my out-of-tree build. It looks like a
> race between com
> Subject: [PATCH] eal: fix resource leak
>
> CID 13289 (#1-2 of 2): Resource leak (RESOURCE_LEAK):
> The system resource will not be reclaimed and reused, reducing the future
> availability of the resource.
> In pci_vfio_get_group_fd: Leak of memory or pointers to system resources
>
> Fixes: ff0
Thanks Xie.
I am trying to test FreeBSD-9.3-virtio as guest. Somewhere in the forum i
found virtio-kni combination reaching around 2.7 Gbps performance, which is
enough for my test, though i dint find equivalent driver performance
characterization for qemu-vhost user space combination.
Also, as pe
I wonder if this matrix would be better if auto-generated some how.
Either statically from source scan, or dynamically via simple app?
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Monday, April 11, 2016 5:27 PM
> To: Liu, Yong
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v2] examples/vm_power_manager: fix build
> with libvirt version < 0.9.3
>
> 2016-04-11 16:50, M
2016-04-11 11:45, Marvin Liu:
> vm_power_manager utilize libvirt API virDomainGetVcpuPinInfo for
> retrieve domU vcpu information. This API implemented from version 0.9.3.
> Suse11 SP3 32bit default libvirt version is 0.8.8, so there'll be build
> error. Add judgement in sample Makefile to skip uns
Thanks Thomas, I'll send v2 for it.
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Monday, April 11, 2016 3:24 PM
> To: Liu, Yong
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH] examples/vm_power_manager: fix build with
> libvirt version
On 4/10/2016 7:26 AM, chintu hetam wrote:
> I am compiling DPDK 2.2 on Fedora 23 and i am seeing following build error
> /home/vcr/devel/dpdk/dpdk-2.2.0/build/build/lib/librte_eal/linuxapp/kni/kni_vhost.c:
> In function ?kni_sock_poll?:
> /home/vcr/devel/dpdk/dpdk-2.2.0/build/build/lib/librte_eal/l
I fully agree with this course of action.
Thank you,
Patrik
On 04/08/2016 08:47 AM, Xie, Huawei wrote:
> On 4/7/2016 10:52 PM, Christian Ehrhardt wrote:
>> I totally agree to that there is no deterministic rule what to expect.
>> The only rule is that #fd certainly always is > #vhost_user devic
On 4/11/2016 1:29 AM, Zhe Tao wrote:
>
> +/* Check the share memory in case the QEMU doesn't set the share option
> + * as on for the memory-backend-file object in the QEMU command line.
> + */
> +
> +int
> +vhost_check_mem_shared(struct vhost_device_ctx ctx)
> +{
> + struct virtio_net *dev;
> > CID 13289 (#1-2 of 2): Resource leak (RESOURCE_LEAK):
> > The system resource will not be reclaimed and reused, reducing the future
> > availability of the resource.
> > In pci_vfio_get_group_fd: Leak of memory or pointers to system resources
> >
> > Fixes: ff0b67d1c868 ("vfio: DMA mapping")
>
Thomas
Could you help check the updated doc to see if you can merge? Thx. You asked me
to do it quickly so John has submitted the patch for me in a short time.
Thanks
Qian
-Original Message-
From: Mcnamara, John
Sent: Friday, April 08, 2016 11:22 PM
To: dev at dpdk.org
Cc: thomas.monja
The reason cause this problem is that in QEMU, when assign the
memory-backend-file without share option, will cause QEMU mmap the mem file
without using the MAP_SHARED flag, so the page cache for that file will not
shared between other processes, all the upated to the mapping area in the VM
will no
65 matches
Mail list logo
