Hi Shahaf,
On Sun, Nov 05, 2017 at 05:44:49PM +0200, Shahaf Shuler wrote:
> While the PMD avoids from creating hash RXQ with no hash fields and
> array of queues after the port was allready started, it lacks such
> protection when re-creating the flows after the port restarts.
>
> This may lead t
Hi Jerin
On 11/3/2017 8:56 PM, Jerin Jacob Wrote:
-Original Message-
[...]
g like that.
Ok, but how to distinguish following 2 options?
No clearly understood this question. For arm64 case, you can add
CONFIG_RTE_RING_USE_C11_MEM_MODEL=y in config/defconfig_arm64-armv8a-*
Sorry for
On Monday 06 November 2017 07:11 AM, Thomas Monjalon wrote:
> The wording changes have been done in the API without breaking
> the ABI. The deprecated fields and symbols can be removed later
> when an another ABI change will be required.
> The deprecation notice can be removed.
>
> The release not
On Monday 06 November 2017 07:11 AM, Thomas Monjalon wrote:
> The function rte_mempool_virt2phy() is renamed to rte_mempool_virt2iova().
> The new function has one less parameter because it is unused.
> The deprecated function is kept as an alias to avoid breaking the API.
>
> Signed-off-by: Thoma
On Monday 06 November 2017 07:11 AM, Thomas Monjalon wrote:
> The struct fields phys_addr_t rte_mempool_objhdr.physaddr and
> rte_mempool_memhdr.phys_addr are renamed to rte_iova_t iova.
> The deprecated names are kept in an anonymous union to avoid breaking
> the API.
>
> Signed-off-by: Thomas M
On Monday 06 November 2017 07:11 AM, Thomas Monjalon wrote:
> The struct rte_memzone field .phys_addr is renamed to .iova.
> The deprecated name is kept in an anonymous union to avoid breaking
> the API.
>
> Signed-off-by: Thomas Monjalon
> ---
Acked-by: Santosh Shukla
On Monday 06 November 2017 07:11 AM, Thomas Monjalon wrote:
> The function rte_malloc_virt2phy() is renamed to rte_malloc_virt2iova().
> The deprecated name is kept as an alias to avoid breaking the API.
>
> Signed-off-by: Thomas Monjalon
> ---
Acked-by: Santosh Shukla
On Monday 06 November 2017 07:11 AM, Thomas Monjalon wrote:
> The function rte_mem_virt2phy() is kept and used in functions which
> works only with physical addresses.
> For all other calls this function is replaced by rte_mem_virt2iova()
> which does a direct mapping (no conversion) in the VA cas
On Monday 06 November 2017 07:11 AM, Thomas Monjalon wrote:
> If the IOVA mode is not using physical addresses,
> no need to log an error about physical address issue.
>
> Signed-off-by: Thomas Monjalon
> ---
Acked-by: Santosh Shukla
On Monday 06 November 2017 07:11 AM, Thomas Monjalon wrote:
> The IO virtual addresses may be used instead of physical addresses.
> As IOVA is more generic, it should be used in most places instead
> of physical address wording.
>
> Signed-off-by: Thomas Monjalon
> ---
> lib/librte_eal/common/in
Hi,
> -Original Message-
> From: Li, Xiaoyun
> Sent: Monday, November 6, 2017 10:42 AM
> To: Lu, Wenzhuo
> Cc: dev@dpdk.org; Li, Xiaoyun ; sta...@dpdk.org
> Subject: [PATCH v2] net/e1000: fix cannot getting Rx interrupts issue
>
> When using VFIO and MSIX interrupt mode, cannot get Rx in
-Original Message-
> Date: Fri, 3 Nov 2017 16:13:51 +0530
> From: Guduri Prathyusha
> To: tomasz.kante...@intel.com
> CC: jianbo@arm.com, guduriprathyu...@gmail.com,
> konstantin.anan...@intel.com, dev@dpdk.org, Guduri Prathyusha
>
> Subject: [dpdk-dev] [PATCH v2] examples/l3fwd: f
Hi, Maxime
> -Original Message-
> From: Maxime Coquelin [mailto:maxime.coque...@redhat.com]
> Sent: Friday, November 3, 2017 11:57 PM
> To: dev@dpdk.org; y...@fridaylinux.org; Yao, Lei A
> Cc: m...@redhat.com
> Subject: Re: [PATCH] vhost: postpone ring addresses translations at kick time
When using VFIO and MSIX interrupt mode, cannot get Rx interrupts. Because
when enabling the interrupt vectors, the offset is computed in a way which
only supports IGB_UIO. But the offset should be different when using VFIO.
This patch fixes this issue.
Fixes: c3cd3de0ab50 ("igb: enable Rx queue i
The function pci_get_sysfs_path was moved from EAL to the PCI driver.
The namespace is now fixed by adding "rte_" prefix.
The map files are fixed by removing the symbol from EAL and adding
it to the PCI driver.
It is an API break but it is probably not used by applications.
Anyway this API is alr
From: Santosh Shukla
Renamed data type from phys_addr_t to rte_iova_t.
Signed-off-by: Santosh Shukla
Reviewed-by: Anatoly Burakov
Signed-off-by: Thomas Monjalon
---
drivers/net/ark/ark_ddm.c | 2 +-
drivers/net/ark/ark_ddm.h | 4 +--
drivers/net/ark/ark_et
The wording changes have been done in the API without breaking
the ABI. The deprecated fields and symbols can be removed later
when an another ABI change will be required.
The deprecation notice can be removed.
The release notes describe the new available API with IOVA wording.
Signed-off-by: Tho
From: Santosh Shukla
Renamed data type from phys_addr_t to rte_iova_t.
Signed-off-by: Santosh Shukla
Reviewed-by: Anatoly Burakov
Signed-off-by: Thomas Monjalon
---
app/test-crypto-perf/cperf_test_common.c | 2 +-
app/test-crypto-perf/cperf_test_vectors.h | 4 ++--
doc/guides/prog_guide/c
The following inline functions and macros have been renamed to be
consistent with the IOVA wording:
rte_mbuf_data_dma_addr -> rte_mbuf_data_iova
rte_mbuf_data_dma_addr_default -> rte_mbuf_data_iova_default
rte_pktmbuf_mtophys-> rte_pktmbuf_iova
rte_pktmbuf_mtophys_offset ->
From: Santosh Shukla
Rename buf_physaddr to buf_iova.
Keep the deprecated name in an anonymous union to avoid breaking
the API.
Signed-off-by: Santosh Shukla
Reviewed-by: Anatoly Burakov
Signed-off-by: Thomas Monjalon
---
app/test-crypto-perf/cperf_test_common.c | 4 ++--
doc/guides/prog_gu
The struct fields phys_addr_t rte_mempool_objhdr.physaddr and
rte_mempool_memhdr.phys_addr are renamed to rte_iova_t iova.
The deprecated names are kept in an anonymous union to avoid breaking
the API.
Signed-off-by: Thomas Monjalon
---
drivers/mempool/dpaa2/dpaa2_hw_mempool.h| 2 +-
dr
The functions rte_mempool_populate_phys() and
rte_mempool_populate_phys_tab() are renamed to
rte_mempool_populate_iova() and rte_mempool_populate_iova_tab().
The deprecated functions are kept as aliases to avoid breaking the API.
Signed-off-by: Thomas Monjalon
---
lib/librte_mempool/rte_mempool.
The function rte_mempool_virt2phy() is renamed to rte_mempool_virt2iova().
The new function has one less parameter because it is unused.
The deprecated function is kept as an alias to avoid breaking the API.
Signed-off-by: Thomas Monjalon
---
app/test-crypto-perf/cperf_test_common.c | 4 ++--
d
The struct rte_memzone field .phys_addr is renamed to .iova.
The deprecated name is kept in an anonymous union to avoid breaking
the API.
Signed-off-by: Thomas Monjalon
---
drivers/crypto/qat/qat_qp.c | 2 +-
drivers/net/bnx2x/bnx2x.c | 2 +-
drivers/net/bnx2x
From: Santosh Shukla
Use RTE_PTR_DIFF macro in rte_mem_virt2iova api.
Suggested-by: Anatoly Burakov
Signed-off-by: Santosh Shukla
---
lib/librte_eal/common/rte_malloc.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/lib/librte_eal/common/rte_malloc.c
b/lib/librte_eal/com
From: Santosh Shukla
Renaming rte_memseg {.phys_addr} to {.iova}
Keep the deprecated name in an anonymous union to avoid breaking
the API.
Use rte_iova_t and RTE_BAD_IOVA where appropriate in
memory segment handling.
Signed-off-by: Santosh Shukla
Reviewed-by: Anatoly Burakov
Signed-off-by: Th
The function rte_malloc_virt2phy() is renamed to rte_malloc_virt2iova().
The deprecated name is kept as an alias to avoid breaking the API.
Signed-off-by: Thomas Monjalon
---
app/test-crypto-perf/cperf_test_vector_parsing.c | 4 ++--
app/test-crypto-perf/cperf_test_vectors.c| 6 +++---
The function rte_mem_virt2phy() is kept and used in functions which
works only with physical addresses.
For all other calls this function is replaced by rte_mem_virt2iova()
which does a direct mapping (no conversion) in the VA case.
Note: the new function rte_mem_virt2iova() function matches the
b
The IO virtual addresses may be used instead of physical addresses.
As IOVA is more generic, it should be used in most places instead
of physical address wording.
Signed-off-by: Thomas Monjalon
---
lib/librte_eal/common/include/rte_memory.h | 10 +-
1 file changed, 9 insertions(+), 1 del
If the IOVA mode is not using physical addresses,
no need to log an error about physical address issue.
Signed-off-by: Thomas Monjalon
---
lib/librte_eal/linuxapp/eal/eal_memory.c | 9 +
1 file changed, 5 insertions(+), 4 deletions(-)
diff --git a/lib/librte_eal/linuxapp/eal/eal_memory.
This v4 is a big rework to complete the patches sent by Santosh.
The core components for memory handling (memseg, malloc, memzone,
mempool, mbuf) are almost fully updated for IOVA awareness.
It will ease maintenance and upcoming reworks; that's why we should push
it in 17.11 LTS.
The most importan
Hi,
05/11/2017 13:05, James Bensley:
> Hi all,
>
> Please see attached and below a patch which allowed Pktgen to run on
> my laptop which has a built in Intel 82579LM 1G NIC:
Thank you for your contribution.
If you have a bit of time, you can try to send a well formatted patch
which will be bett
Hi,
I think we have a compilation error in ixgbe IPsec for more than one week.
Someone to fix it please?
05/11/2017 21:11, sys_...@intel.com:
> Failure #1
> SUSE11SP2_64 / Linux 3.0.13-0 / GCC 4.5.1
> Config: x86_64-native-linuxapp-gcc
> CONFIG_RTE_LIBRTE_PMD_PCAP=y,
> CONFIG_RTE_
This patch enhances the Rx packet type offload to also report the L4
protocol information in the hw ptype filled by the PMD for each received
packet.
Signed-off-by: Moti Haimovsky
---
V4:
* Removed extra blank line from the end of mlx4_ethdev.c
V3:
* Modifications according to review by Adrien Ma
In open_single_tx_pcap there is a call to pcap_open_dead which calls
malloc to create and return a pcap_t. That object is never freed in
this case. Other places it is freed by passing it back similar to the
way the pcap_dumper_t is in this case.
The pcap_t is only used to create the pcap_dumper_t
While the PMD avoids from creating hash RXQ with no hash fields and
array of queues after the port was allready started, it lacks such
protection when re-creating the flows after the port restarts.
This may lead to inconsist behaviour for flows depending if they were
created before or after the po
This patch enhances the Rx packet type offload to also report the L4
protocol information in the hw ptype filled by the PMD for each received
packet.
Signed-off-by: Moti Haimovsky
---
V3:
* Modifications according to review by Adrien Mazarguil
Re: [PATCH v2] net/mlx4: enhance Rx packet type
See inline
> -Original Message-
> From: Adrien Mazarguil [mailto:adrien.mazarg...@6wind.com]
> Sent: Friday, November 3, 2017 4:23 PM
> To: Mordechay Haimovsky
> Cc: dev@dpdk.org
> Subject: Re: [PATCH v2] net/mlx4: enhance Rx packet type offloads
>
> On Thu, Nov 02, 2017 at 02:14:34PM +0
Hi all,
Please see attached and below a patch which allowed Pktgen to run on
my laptop which has a built in Intel 82579LM 1G NIC:
bensley@ubuntu-laptop:~$ lspci -nn | grep Ether
00:19.0 Ethernet controller [0200]: Intel Corporation 82579LM Gigabit
Network Connection [8086:1502] (rev 04)
bensley@
Hello folks,
This patch removes single interface constraint from the libpcap-based PMD.
The problem it solves is to allow PCAP device consists of more than single
device:
# testpmd --vdev net_pcap0,iface=vethA,iface=vethB,iface=vethC and so..
I found the issue when performed RSS emulation based
Hello folks,
This patch removes single interface constraint from the libpcap-based PMD. The
problem it solves is to allow PCAP device consists of more than single device:
# testpmd --vdev net_pcap0,iface=vethA,iface=vethB,iface=vethC and so..
I found the issue when performed RSS emulation base
41 matches
Mail list logo
