> -Original Message-
> From: outlook_739db8e1c4bc6fae at outlook.com
> [mailto:outlook_739db8e1c4bc6fae at outlook.com] On Behalf Of Dong.Wang
> Sent: Wednesday, April 15, 2015 2:46 PM
> To: Ananyev, Konstantin; dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH] ixgbe:Add write memory bar
To initialize a structure with zeros, one field was explicitly set
to avoid "missing initializer" bug with old GCC (e.g. 4.4).
This warning is now disabled (commit ) for old versions of GCC,
so the workarounds may be removed.
These initializers should not be needed for static variables but they
ar
With GCC 4.4.7 from CentOS 6.5, the following errors arise:
lib/librte_pmd_ixgbe/ixgbe_rxtx.c: In function ?ixgbe_dev_rx_queue_setup?:
lib/librte_pmd_ixgbe/ixgbe_rxtx.c:2509: error: missing initializer
lib/librte_pmd_ixgbe/ixgbe_rxtx.c:2509: error: (near initialization for
?dev_info.driver_name?)
No warning is needed, it just works.
On Wed, Apr 15, 2015 at 8:39 PM, Ouyang, Changchun <
changchun.ouyang at intel.com> wrote:
>
>
> > -Original Message-
> > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Stephen
> > Hemminger
> > Sent: Wednesday, April 15, 2015 11:20 PM
> > To:
> Hi,
>
>> -Original Message-
>> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of WangDong
>> Sent: Saturday, April 11, 2015 4:34 PM
>> To: dev at dpdk.org
>> Subject: [dpdk-dev] [PATCH] ixgbe:Add write memory barrier for recv pkts.
>>
>> Like transmit packets, before update receiv
Hi,
I have two questions regarding mempools:
- the first is trivial: how do you delete them? Can you? I can't see a
function to do that, and none of the examples are doing such thing. When
exactly it get deleted?
- during creation, cache_size have one requirement: it has to be smaller
than RTE
Hi all,
I have a DPDK app that requires tight time synchronization between end
hosts.
In my setup, the testbed is connected to a Cisco switch that supports the
PTP protocol. I know that the ixgbe driver for Intel 82599 nics support the
PTP time protocol, but I'm not sure if the DPDK framework supp
On Wed, 15 Apr 2015 17:43:02 -0700
Kamraan Nasim wrote:
> >
> > This had me stumped for a while as well. In my case, PostGres9.4 was also
> > running on my system which also used huge pages and came up before my DPDK
> > application causing rte_mempool_create() to ENOMEM.
>
>
> Check which othe
>
> This had me stumped for a while as well. In my case, PostGres9.4 was also
> running on my system which also used huge pages and came up before my DPDK
> application causing rte_mempool_create() to ENOMEM.
Check which other applications are using huge pages:
> lsof | grep huge
And see if you
On Wed, Apr 15, 2015 at 03:09:39PM +, Kavanagh, Mark B wrote:
> Hi,
>
> The recent reimplementation of rte_memcpy in DPDK v2.0.0 seems to have a
> placed an implicit floor on the microarchitecture/Instruction set supported
> by DPDK.
>
> For example, I can't compile head of OVS against DPDK
>-Original Message-
>From: Richardson, Bruce
>Sent: Wednesday, April 15, 2015 5:05 PM
>To: Kavanagh, Mark B
>Cc: dev at dpdk.org
>Subject: Re: [dpdk-dev] Minimum Supported x86 microarchitecture
>
>On Wed, Apr 15, 2015 at 03:09:39PM +, Kavanagh, Mark B wrote:
>> Hi,
>>
>> The recent re
On 15/04/2015 10:30, Sergio Gonzalez Monroy wrote:
> Use flag --as-needed when linking against shared libs, exluding
> eal, malloc, mempool and ring as explained in:
> 7cee5f6b4a9b ("lib: set LDLIBS for each library")
Self-NACK. It is not correct to put the commit id here (it does not
exist after
Hi,
The recent reimplementation of rte_memcpy in DPDK v2.0.0 seems to have a placed
an implicit floor on the microarchitecture/Instruction set supported by DPDK.
For example, I can't compile head of OVS against DPDK 2.0 with gcc without
passing the 'msse3' flag; this points to an implicit minim
Hi,
> From: yangguangjerry at hotmail.com [mailto:yangguangjerry at hotmail.com]
> Sent: Tuesday, April 07, 2015 10:46 AM
> To: De Lara Guarch, Pablo
> Cc: "dev at dpdk.org"
> Subject: Re:[dpdk-dev] [RFC] Cuckoo hash for DPDK 2.1
>
>
> hi?Pablo,
> ??? rte_hash uses Jenkins hash (http://burtlebur
Hi,
These rules parse failure for missing blank space between the start and end of
port. Looks like user guide shown us the wrong format:)
L3fwd-acl sample will request at least one route rule. You can just add one
simple rule as below which forward all packets to port0.
-bash-4.3# cat rule_ipv4
On Tue, Apr 14, 2015 at 7:53 PM, Stephen Hemminger <
stephen at networkplumber.org> wrote:
> On Tue, 14 Apr 2015 14:31:53 -0700
> Ravi Kerur wrote:
>
> > +
> > + for (i = 0; i < 2; i++)
> > + rte_mov32(dst + i * 32, src + i * 32);
> > }
> Unless you force compiler to unroll the l
On Tue, Apr 14, 2015 at 11:32 PM, Pawel Wodkowski <
pawelx.wodkowski at intel.com> wrote:
> On 2015-04-14 23:31, Ravi Kerur wrote:
>
>> +
>> + for (i = 0; i < 8; i++) {
>> + ymm = _mm256_loadu_si256((const __m256i *)(src +
>> i * 32));
>> +
I am trying to build memnic-1.3/pmd on a qemu-kvm running
4.0.0-rc7+ with gcc 4.8.2, and it fails with a few errors, the first of
which is
/path/to/dpdk-2.0.0/build/include/rte_memcpy.h:625:2: error: implicit
declaration of function ?_mm_alignr_epi8?
[-Werror=implicit-function-declaration]
MOVEUN
On Wed, 15 Apr 2015 20:15:18 +0100
Zoltan Kiss wrote:
> Hi,
>
> I have two questions regarding mempools:
>
> - the first is trivial: how do you delete them? Can you? I can't see a
> function to do that, and none of the examples are doing such thing. When
> exactly it get deleted?
You can't d
Hi,
I can`t find the rule format files (rule_ipv4.db and rule_ipv6.db) so I
can`t run the access control example.
I wrote my own rule file based on rule file example in the sample app
document, but the example stopped saying that "Cause: ./ipv4.txt Line 1:
parse rules error".
My rule file likes be
On Thu, Apr 09, 2015 at 02:29:42PM -0700, Stephen Hemminger wrote:
> The Linux style is not to put extra useless paren's around
> the expression passed to return statement.
>
> Signed-off-by: Stephen Hemminger
Acked-by: Bruce Richardson
> ---
> lib/librte_ether/rte_ethdev.c | 298
> ++
On Thu, Apr 09, 2015 at 02:29:41PM -0700, Stephen Hemminger wrote:
> The statistics tables and null mac address should be immutable.
> Fix up get_addr routines to accept const args.
>
> Signed-off-by: Stephen Hemminger
Acked-by: Bruce Richardson
> ---
> lib/librte_ether/rte_ethdev.c | 44
> +
On Thu, Apr 09, 2015 at 02:29:40PM -0700, Stephen Hemminger wrote:
> Fix space after keywords, and other missing whitespace.
>
> Signed-off-by: Stephen Hemminger
Partial NAK for now. Mostly this looks ok, but I disagree with the changes to
the indentation of continued lines. Until we get agreeme
On Thu, Apr 09, 2015 at 02:29:39PM -0700, Stephen Hemminger wrote:
> There is no reason to inline functions that are not in the critical
> path.
>
> Signed-off-by: Stephen Hemminger
Acked-by: Bruce Richardson
> ---
> lib/librte_ether/rte_ethdev.c | 12 ++--
> 1 file changed, 6 insertio
Hi Changchun,
I confirm that blacklisting the first two network interfaces solve the
problem.
Just to be clear, is it the case that it is not possible to have a DPDK
application acquiring selectively virtio network devices without using
the blacklisting (-b) option?
It seems a bit clumsy to ha
On Wed, Apr 08, 2015 at 10:37:50AM -0700, Stephen Hemminger wrote:
> From: Stephen Hemminger
>
> The function documentation was obviously copied and not updated.
>
> Signed-off-by: Stephen Hemminger
Acked-by: Bruce Richardson
> ---
> tools/dpdk_nic_bind.py | 2 +-
> 1 file changed, 1 inserti
Use flag --as-needed when linking against shared libs, exluding
eal, malloc, mempool and ring as explained in:
7cee5f6b4a9b ("lib: set LDLIBS for each library")
Signed-off-by: Sergio Gonzalez Monroy
---
mk/rte.app.mk | 36
1 file changed, 20 insertions(+), 16
Set proper DT_NEEDED entries for shared libraries by explicitly
linking against its dependent libraries (LDLIBS).
Signed-off-by: Sergio Gonzalez Monroy
---
mk/rte.lib.mk | 6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/mk/rte.lib.mk b/mk/rte.lib.mk
index 0d7482d..5d0988f
This patch introduces a new LDLIBS variable to be set per library.
Its purpose is to especify the library's dependent libraries to
be explicitly linked against.
Given the circular dependencies between eal, malloc, mempool and ring,
we work around it by not linking eal against its dependent DPDK li
This is a split of a previous series:
http://dpdk.org/ml/archives/dev/2015-April/016257.html
The series adds proper DT_NEEDED entries to each library when building
shared libraries. The only exception is eal, which is not linked against
its dependent dpdk libraries to avoid circular dependencies.
The integration of using uio_pci_generic broke use of UIO
by igb_uio and other drivers. Go back to using uio IRQ control
through interrupt fd.
This patch assumes that if uio_pci_generic is being used
that the kernel has been fixed to support this
by integrating patch to support IRQ control.
Signe
The driver already supported INTX interrupts but had no in kernel
function to enable and disable them.
It is possible for userspace to do this by accessing PCI config
directly, but this racy and better handled by same mechanism
that already exists in kernel.
Signed-off-by: Stephen Hemminger
---
Hi
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of ???
> Sent: Wednesday, April 15, 2015 3:40 AM
> To: dev at dpdk.org
> Subject: [dpdk-dev] Where is rule_ipv4.db and rule_ipv6.db for l3pwd-acl
> example?
>
> Hi,
> I can`t find the rule format files (rule_i
Here is the correct way to fix it (even if it maybe hard to accept).
I think the most technically correct way to handle this is to fix
UIO driver in kernel to do the IRQ management according to the
API that was already described in the kernel documentation.
Then have the DPDK EAL use the IRQ manage
On Wed, 15 Apr 2015 21:46:27 +0800
Dong.Wang wrote:
> Yes, current implementation works well with IA, and the transmit packets
> function's rte_wmb() is also unneccessary.
>
> But there are two reasons for adding rte_wmb() in recv pkts function:
> 1) The memory barrier in recv pkts function and
On Wed, 15 Apr 2015 07:19:39 +
"Zhou, Danny" wrote:
> Could you please send out the steps for us to reproduce it? I guess you have
> applied
> v6 interrupt patches to perform interrupt related tests, right?
>
> We cannot reproduce it now.
>
> The support to in_kernel uio_pci_generic avoids
On 2015-04-14 23:31, Ravi Kerur wrote:
> +
> + for (i = 0; i < 8; i++) {
> + ymm = _mm256_loadu_si256((const __m256i *)(src + i *
> 32));
> + _mm256_storeu_si256((__m256i *)(dst + i * 32), ymm);
> + }
> +
> n -= 256;
> -
Change the features from bit mask to bit number. This allows the
DPDK driver to use the definitions from Linux[1]. This makes doing
enhancements to DPDK driver easier when new feature bits are added.
More importantly, get rid of the confusing feature bit initialization
code. Remove the double nega
This fixes another of the issues with running virtio on non-KVM
envirionments. For example, Google Compute Engine reports a
ring size of 16K.
If guest virtio requests more slots than available then
the queue should just be initialized to the smaller value.
Conversely, if the number of descriptors
Don't attempt to set the MAC address table unless the host allows it.
Also, don't return a value from mac_table_set since all callers
ignore the return value.
Signed-off-by: Stephen Hemminger
---
lib/librte_pmd_virtio/virtio_ethdev.c | 9 ++---
1 file changed, 6 insertions(+), 3 deletions(-)
Don't try to set features related to receiving unless the
appropriate feature bit has ben negotiated with the host.
This solves some of the issues when using virtio on non-KVM/QEMU
hypervisors.
Signed-off-by: Stephen Hemminger
---
lib/librte_pmd_virtio/virtio_ethdev.c | 20
There are several places in virtio with extra newlines between
calling a function and checking the result.
Remove them to improve readability.
Signed-off-by: Stephen Hemminger
---
lib/librte_pmd_virtio/virtio_ethdev.c | 6 --
1 file changed, 6 deletions(-)
diff --git a/lib/librte_pmd_virtio
These are cleanups and enhancements to virtio driver.
The important fixes are those that allow virtio to work in non-KVM
environments. The driver was not negotiating properly with the host.
Stephen Hemminger (5):
virtio: remove useless new lines
virtio: don't enable/disable rx modes unless sup
Could you please send out the steps for us to reproduce it? I guess you have
applied
v6 interrupt patches to perform interrupt related tests, right?
We cannot reproduce it now.
The support to in_kernel uio_pci_generic avoids using igb_uio this DPDK
specific kernel module,
so it will be much ea
Hi Oliver
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Olivier Deme
> Sent: Wednesday, April 15, 2015 1:16 AM
> To: dev at dpdk.org
> Subject: [dpdk-dev] freeze with dpdk-2.0.0
>
> Hi,
>
> I am trying to use DPDK-2.0.0 on Fedora 20 running in a qemu virt
45 matches
Mail list logo
