16/09/2013 14:42, Patrick Mahan :
> On Sep 16, 2013, at 1:29 PM, Thomas Monjalon
wrote:
> > There is no need to check for bars mapping, especially BAR0 is not
> > required. If bars mapping failed, then pci_uio_map_resource will fail and
> > we won't reach this check. So get rid of BAR0 check.
> >
The new release can be downloaded here:
http://dpdk.org/browse/dpdk/tag/?id=v1.3.1r3
Changelog
- fixes for:
* KNI build with Linux 3.10
* tests build with GCC 4.8
* vmxnet3-usermap initialization
* rte_malloc dependenc
> There is no need to check for bars mapping, especially BAR0 is not required.
> If bars mapping failed, then pci_uio_map_resource will fail and we won't
> reach this check. So get rid of BAR0 check.
> Besides, pci_uio_map_resource should only be called for Intel devices.
> The flag RTE_PCI_DRV_NEE
Hi Sujith,
NetBSD/Rump is the only open source TCP/IP stack for DPDK. Some people
may have tried to port LwIP too. As far as we know, only 6WIND has a
robust and fully compliant stack which provides socket APIs to the
applications on top of the DPDK. I'd be pleased to get a list of options
for
From: David Marchand
There is no need to check for bars mapping, especially BAR0 is not required.
If bars mapping failed, then pci_uio_map_resource will fail and we won't reach
this check. So get rid of BAR0 check.
Besides, pci_uio_map_resource should only be called for Intel devices.
The flag RT
hi all,
In "testpmd", the default nb_pkt_per_burst value is 16, that is, receive 16
packets one time and then forward.
Now, I set nb_pkt_per_burst=1, and after the packet received, delay 50us,
because I have to add some packet processing steps between receive and
transmit. However, in
I agree with Thomas, the check is not needed.
After this all settles out, I have a patch to add support for I/O
resource tracking in base DPDK. It makes it easier to do the right
thing in drivers that need to use I/O instead of memory.
16/09/2013 13:22, Nicolas Dichtel :
> Le 13/09/2013 16:14, Thomas Monjalon a ?crit :
> > GCC 4.8 was producing this error:
> > argument to ?sizeof? in ?strncmp? call is the same expression as the
> > second source; did you mean to provide an explicit length?
> > [-Werror=sizeof-pointer-
12/09/2013 18:09, Nicolas Dichtel :
> Le 29/07/2013 14:44, Thomas Monjalon a ?crit :
> > - The flags NETIF_F_HW_VLAN_* have been renamed to
> > NETIF_F_HW_VLAN_CTAG_*. See Linux commit
> > f646968f8f7c624587de729115d802372b9063dd.
> >
> > - The VLAN protocol must be specified.
> > See Linux commit
12/09/2013 17:34, Nicolas Dichtel :
> Le 29/07/2013 14:44, Thomas Monjalon a ?crit :
> > hlist API has changes.
> > See Linux commit b67bfe0d42cac56c512dd5da4b1b347a23f4b70a.
> >
> > Signed-off-by: Thomas Monjalon
>
> Acked-by: Nicolas Dichtel
pushed
--
Thomas
16/09/2013 13:21, Nicolas Dichtel :
> Le 13/09/2013 16:14, Thomas Monjalon a ?crit :
> > - timecompare (used for hardware timestamping) has been removed.
> > (see Linux commit 65f8f9a1c1db831e5159e3e3e50912d1f214cd0c)
> > Simply disable HW_TIME_STAMP feature because it is not used by KNI.
> > For t
I totally disagree with this statement. I am currently working on a non-intel
device that does need UIO support (I copied the igb UIO to create a new UIO
driver). This device not only needs bar0 but also bar1. I've modified the eal
PCI support code to support this behavior and change this beh
Le 13/09/2013 16:14, Thomas Monjalon a ?crit :
> GCC 4.8 was producing this error:
> argument to ?sizeof? in ?strncmp? call is the same expression as the
> second source; did you mean to provide an explicit length?
> [-Werror=sizeof-pointer-memaccess]
>
> Signed-off-by: Thomas Mon
Le 13/09/2013 16:14, Thomas Monjalon a ?crit :
> - timecompare (used for hardware timestamping) has been removed.
> (see Linux commit 65f8f9a1c1db831e5159e3e3e50912d1f214cd0c)
> Simply disable HW_TIME_STAMP feature because it is not used by KNI.
> For this purpose, kcompat.h must be included before
16/09/2013 11:43, Ivan Boule :
> On 09/13/2013 03:38 PM, Thomas Monjalon wrote:
> > Some Poll-Mode Drivers (PMD) are not reconfigurable and,
> > thus, do not implement (rx|tx)_queue_release functions.
> > For these drivers, the functions rte_eth_dev_(rx|tx)_queue_config
> > must return an ENOTSUP e
On 09/13/2013 03:38 PM, Thomas Monjalon wrote:
> Some Poll-Mode Drivers (PMD) are not reconfigurable and,
> thus, do not implement (rx|tx)_queue_release functions.
> For these drivers, the functions rte_eth_dev_(rx|tx)_queue_config
> must return an ENOTSUP error only when reconfiguring,
> but not a
13/09/2013 16:23, Ivan Boule :
> Add into the `rte_eth_stats` data structure 4 (64-bit) counters
> of XOFF/XON pause frames received and sent on a given port.
>
> Update em, igb, and ixgbe drivers to return the value of the 4 XOFF/XON
> counters through the `rte_eth_stats_get` function exported by
Yes.
-Original Message-
From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Sambath Kumar
Balasubramanian
Sent: Saturday, September 14, 2013 9:38 PM
To: St Leger, Jim
Cc: dev at dpdk.org
Subject: Re: [dpdk-dev] Merge with 1.4.1??
Hi All,
Just wanted to confirm that 1.4.1 is the latest
18 matches
Mail list logo
