Hi,
> -Original Message-
> From: dev On Behalf Of Michael Baum
> Sent: Wednesday, May 27, 2020 11:38 AM
> To: dev@dpdk.org
> Cc: Matan Azrad ; Slava Ovsiienko
> ; sta...@dpdk.org
> Subject: [dpdk-dev] [PATCH 1/6] net/mlx5: fix hairpin Tx queue creation
> error flow
>
> The mlx5_txq_obj_h
Added changes in order to build mempool lib on windows.
Depends-on: patch-70497 ("ring: build on Windows")
Fady Bader (5):
mempool: disable tracing for Windows
eal: disable function versioning on Windows
mempool: use generic memory management
eal: export needed functions for mempool
mem
__rte_trace_mem_per_thread_alloc is used by mempool lib but not exported on
Windows.
the function was exported.
Signed-off-by: Fady Bader
---
lib/librte_eal/rte_eal_version.map | 1 +
1 file changed, 1 insertion(+)
diff --git a/lib/librte_eal/rte_eal_version.map
b/lib/librte_eal/rte_eal_versio
Tracing is not supported for Windows.
Disabled all trace calls in for Windows.
Signed-off-by: Fady Bader
---
lib/librte_mempool/mempool_trace_points.c | 4
lib/librte_mempool/rte_mempool.c | 20
lib/librte_mempool/rte_mempool.h | 22 ++
mempool used Unix memory management calls, which are not supported
on Windows.
Used generic memory management instead.
Signed-off-by: Fady Bader
---
lib/librte_mempool/rte_mempool.c | 21 ++---
1 file changed, 10 insertions(+), 11 deletions(-)
diff --git a/lib/librte_mempool/rte
Mempool didn't compile on Windows.
Needed changes were made to fix this.
Signed-off-by: Fady Bader
---
lib/meson.build | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/lib/meson.build b/lib/meson.build
index 30ed9e05e..c53732383 100644
--- a/lib/meson.build
+++ b/lib/meson.b
Function versioning is not needed on Windows, also the function versioning
implementation is not supported by Windows.
Function versioning was disabled on Windows.
Signed-off-by: Fady Bader
---
lib/librte_eal/include/rte_function_versioning.h | 4
1 file changed, 4 insertions(+)
diff --git
On 5/31/20 3:41 PM, Fady Bader wrote:
> Tracing is not supported for Windows.
> Disabled all trace calls in for Windows.
>
> Signed-off-by: Fady Bader
> ---
> lib/librte_mempool/mempool_trace_points.c | 4
> lib/librte_mempool/rte_mempool.c | 20
> lib/librte_
On 5/31/20 3:41 PM, Fady Bader wrote:
> mempool used Unix memory management calls, which are not supported
> on Windows.
> Used generic memory management instead.
>
> Signed-off-by: Fady Bader
LGTM, but I've expected rte_eal_ prefix, not just rte_.
Anyway, these functions are added early and it
On Sun, May 31, 2020 at 6:12 PM Fady Bader wrote:
>
> Tracing is not supported for Windows.
> Disabled all trace calls in for Windows.
What is the current major issue to enable Trace for windows?.
Trace does not need to rte_malloc support, it can work with malloc as
well incase rte_malloc is not
Hi,
> -Original Message-
> From: Suanming Mou
> Sent: Thursday, May 28, 2020 12:22 PM
> To: Slava Ovsiienko ; Matan Azrad
> ; dev@dpdk.org
> Cc: Raslan Darawsheh ; sta...@dpdk.org
> Subject: [PATCH v2] net/mlx5: fix interrupt installation timing
>
> Currently, the DevX counter query work
Hi,
> -Original Message-
> From: Suanming Mou
> Sent: Thursday, May 28, 2020 10:00 AM
> To: Slava Ovsiienko ; Matan Azrad
> ; dev@dpdk.org
> Cc: Raslan Darawsheh ; sta...@dpdk.org
> Subject: [PATCH v2] net/mlx5: fix secondary process resources release
>
> When secondary process starts, i
Using the current implementation of DPDK, an application cannot
match on fragmented/non-fragmented IPv6 packets in a simple way.
In current implementation:
IPv6 header doesn't contain information regarding the packet
fragmentation.
Fragmented IPv6 packets contain a dedicated extension header, as
d
Hi, Patrick
Please, see below.
>From: PATRICK KEROULAS
>Sent: Friday, May 29, 2020 23:56
>To: Slava Ovsiienko
>Cc: dev@dpdk.org; Vivien Didelot ; Shahaf Shuler
>; Raslan Darawsheh ; Matan Azrad
>>
>Subject: Re: [dpdk-dev] mlx5 & pdump: convert HW timestamps to nanoseconds
>
>
>On Tue, May 26
Hi Ferruh Yigit and others
I am learning data structure defined by dpdk framework and I noticed
a data structure definition below:
enum rte_eth_hash_function {
RTE_ETH_HASH_FUNCTION_DEFAULT = 0,
RTE_ETH_HASH_FUNCTION_TOEPLITZ, /**< Toeplitz */
RTE_ETH_HASH_FUNCTION_SIMPLE_XOR, /
On Sun, 31 May 2020 16:02:21 +0300
Andrew Rybchenko wrote:
[snip]
> >
> > +#ifndef RTE_EXEC_ENV_WINDOWS
> > #include
> > +#endif
>
> I really dislike the approach. It adds too many conditional
> compilation statements which make code harder to read.
>
> You can have dummy implementation o
> -Original Message-
> From: Stephen Hemminger [mailto:step...@networkplumber.org]
> Sent: Saturday, May 30, 2020 1:47 AM
> To: wangyunjian
> Cc: dev@dpdk.org; sthem...@microsoft.com; Lilijun (Jerry)
> ; xudingke ;
> sta...@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH] bus/vmbus: fix wrong al
On Sun, 31 May 2020 17:43:29 +0300
Dekel Peled wrote:
> Using the current implementation of DPDK, an application cannot
> match on fragmented/non-fragmented IPv6 packets in a simple way.
>
> In current implementation:
> IPv6 header doesn't contain information regarding the packet
> fragmentation
File mlx5_common.c includes both specific and non-specific Linux APIs.
Move the Linux specific APIS into a new file named linux/mlx5_common_os.c.
Signed-off-by: Ophir Munk
---
drivers/common/mlx5/Makefile | 1 +
drivers/common/mlx5/linux/meson.build | 1 +
drivers/common/m
File mlx5_nl.c is using Netlink APIs which are Linux specifics.
Move it (including file mlx5_nl.h) to common/mlx5/linux directory.
Signed-off-by: Ophir Munk
---
drivers/common/mlx5/Makefile |2 +-
drivers/common/mlx5/linux/meson.build |1 +
drivers/common/mlx5/linux/mlx5_nl.c
The glue file mlx5_glue.c is based on Linux specifics APIs.
Move it (including file mlx5_glue.h) to common/mlx5/linux directory.
Signed-off-by: Ophir Munk
---
drivers/common/mlx5/Makefile | 12 +-
drivers/common/mlx5/linux/meson.build | 199 +
drivers/common/mlx5/linux/mlx5_glue.
v1:
Initial release
v2:
Following review comments.
Ophir Munk (3):
common/mlx5: move glue files under Linux directory
common/mlx5: move netlink files under Linux directory
common/mlx5: split common file under Linux directory
drivers/common/mlx5/Makefile | 15 +-
drivers/com
Currently, when flow destroyed, some memory resources may still be kept
as cached to help next time create flow more efficiently.
Some system may need the resources to be more flexible with flow create
and destroy. After peak time, with millions of flows destroyed, the
system would prefer the res
Currently, when flow destroyed, some memory resources may still be kept
as cached to help next time create flow more efficiently.
Some system may need the resources to be more flexible with flow create
and destroy. After peak time, with millions of flows destroyed, the
system would prefer the res
While flow destroyed, rdma-core may still cache some resources for more
efficiently flow recreate. In case the peak time that millions of flows
created and destroyed, the cached resources will be very huge.
Currently, rdma-core proivdes the new function to configure the flow
resources not to be ca
Thanks, PSB.
> -Original Message-
> From: Stephen Hemminger
> Sent: Monday, June 1, 2020 8:39 AM
> To: Dekel Peled
> Cc: ferruh.yi...@intel.com; arybche...@solarflare.com; Ori Kam
> ; john.mcnam...@intel.com;
> marko.kovace...@intel.com; Asaf Penso ; Matan
> Azrad ; Eli Britstein ;
> dev
26 matches
Mail list logo
