DPDK patches and discussions
 help / color / mirror / Atom feed
From: Ferruh Yigit <ferruh.yigit@intel.com>
To: Ophir Munk <ophirmu@mellanox.com>,
	Adrien Mazarguil <adrien.mazarguil@6wind.com>
Cc: dev@dpdk.org, Thomas Monjalon <thomas@monjalon.net>,
	Olga Shern <olgas@mellanox.com>, Matan Azrad <matan@mellanox.com>
Subject: Re: [dpdk-dev] [PATCH v4 0/7] new mlx4 datapath bypassing ibverbs
Date: Thu, 5 Oct 2017 19:54:46 +0100	[thread overview]
Message-ID: <c3bd8f90-2028-d50a-8a29-8c756d740b8b@intel.com> (raw)
In-Reply-To: <a15d5daa-cdcf-e563-7984-feaddd74740f@intel.com>

On 10/5/2017 7:48 PM, Ferruh Yigit wrote:
> On 10/5/2017 10:33 AM, Ophir Munk wrote:
>> v4 (Ophir):
>> - Split "net/mlx4: restore Rx scatter support" commit from "net/mlx4: 
>>   restore full Rx support bypassing Verbs" commit
>>
>> v3 (Adrien):
>> - Drop a few unrelated or unnecessary changes such as the removal of
>>   MLX4_PMD_TX_MP_CACHE.
>> - Move device checksum support detection code to its previous location.
>> - Fix include guard in mlx4_prm.h.
>> - Reorder #includes alphabetically.
>> - Replace MLX4_TRANSPOSE() macro with documented inline function.
>> - Remove extra spaces and blank lines.
>> - Use uint8_t * instead of char * for buffers.
>> - Replace mlx4_get_cqe() macro with a documented inline function.
>> - Replace several unsigned int with uint32_t.
>> - Add consistency to field names (sge_n => sges_n).
>> - Make mbuf size checks in RX queue setup function similar to mlx5.
>> - Update various comments.
>> - Fix indentation.
>> - Replace run-time endian conversion with static ones where possible.
>> - Reorder fields in struct rxq and struct txq for consistency, remove
>>   one level of unnecessary inner structures.
>> - Fix memory leak on Tx bounce buffer.
>> - Update commit logs.
>> - Fix remaining checkpatch warnings.
>>
>> v2 (Matan):
>> Rearange patches.
>> Semantics.
>> Enhancements.
>> Fix compilation issues.
>>
>> Moti Haimovsky (6):
>>   net/mlx4: add simple Tx bypassing Verbs
>>   net/mlx4: restore full Rx support bypassing Verbs
>>   net/mlx4: restore Tx gather support
>>   net/mlx4: restore Tx checksum offloads
>>   net/mlx4: restore Rx offloads
>>   net/mlx4: add loopback Tx from VF
>>
>> Ophir Munk (1):
>>   net/mlx4: restore Rx scatter support
> 
> Hi Ophir,
> 
> I am a little confused, can you please help me?
> 
> Currently both mlx4 and mlx5 should support both rdma-core and MLX-OFED,
> is this correct?
> 
> When I try to compile these patches with rdma-core, it is giving warning
> for shared library [1].

Ahh, this is because missing library in Makefile, can you please send a
new version to fix this?

   diff --git a/drivers/net/mlx4/Makefile b/drivers/net/mlx4/Makefile
   index 0515cd7ef..3b3a02047 100644
   --- a/drivers/net/mlx4/Makefile
   +++ b/drivers/net/mlx4/Makefile
   @@ -54,7 +54,7 @@ CFLAGS += -D_BSD_SOURCE
    CFLAGS += -D_DEFAULT_SOURCE
    CFLAGS += -D_XOPEN_SOURCE=600
    CFLAGS += $(WERROR_FLAGS)
   -LDLIBS += -libverbs
   +LDLIBS += -libverbs -lmlx4

    # A few warnings cannot be avoided in external headers.
    CFLAGS += -Wno-error=cast-qual

> 
> If I try to compile with mlx-ofed, getting missing header error [2].

Can you please clarify this?

> 
> What is the dependency for mlx4 driver now?
> 
> 
> [1]
> mlx4_rxq.o: In function `mlx4_rxq_setup':
> .../dpdk/drivers/net/mlx4/mlx4_rxq.c:393: undefined reference to
> `mlx4dv_init_obj'
> mlx4_txq.o: In function `mlx4_txq_setup':
> .../dpdk/drivers/net/mlx4/mlx4_txq.c:429: undefined reference to
> `mlx4dv_init_obj'
> 
> 
> [2]
> In file included from .../drivers/net/mlx4/mlx4_flow.c:66:
> .../drivers/net/mlx4/mlx4_rxtx.h:43:10: fatal error:
> 'infiniband/mlx4dv.h' file not found
> #include <infiniband/mlx4dv.h>
>          ^~~~~~~~~~~~~~~~~~~~~
> 

  reply	other threads:[~2017-10-05 18:54 UTC|newest]

Thread overview: 61+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-08-24 15:54 [dpdk-dev] [PATCH 0/5] new mlx4 Tx " Moti Haimovsky
2017-08-24 15:54 ` [dpdk-dev] [PATCH 1/5] net/mlx4: add simple Tx " Moti Haimovsky
2017-08-24 15:54 ` [dpdk-dev] [PATCH 2/5] net/mlx4: support multi-segments Tx Moti Haimovsky
2017-08-24 15:54 ` [dpdk-dev] [PATCH 3/5] net/mlx4: refine setting Tx completion flag Moti Haimovsky
2017-08-24 15:54 ` [dpdk-dev] [PATCH 4/5] net/mlx4: add Tx checksum offloads Moti Haimovsky
2017-08-24 15:54 ` [dpdk-dev] [PATCH 5/5] net/mlx4: add loopback Tx from VF Moti Haimovsky
2017-10-03 10:48 ` [dpdk-dev] [PATCH v2 0/6] new mlx4 datapath bypassing ibverbs Matan Azrad
2017-10-03 10:48   ` [dpdk-dev] [PATCH v2 1/6] net/mlx4: add simple Tx " Matan Azrad
2017-10-03 10:48   ` [dpdk-dev] [PATCH v2 2/6] net/mlx4: get back Rx flow functionality Matan Azrad
2017-10-03 10:48   ` [dpdk-dev] [PATCH v2 3/6] net/mlx4: support multi-segments Tx Matan Azrad
2017-10-03 10:48   ` [dpdk-dev] [PATCH v2 4/6] net/mlx4: get back Tx checksum offloads Matan Azrad
2017-10-03 10:48   ` [dpdk-dev] [PATCH v2 5/6] net/mlx4: get back Rx " Matan Azrad
2017-10-03 22:26     ` Ferruh Yigit
2017-10-03 10:48   ` [dpdk-dev] [PATCH v2 6/6] net/mlx4: add loopback Tx from VF Matan Azrad
2017-10-03 22:27   ` [dpdk-dev] [PATCH v2 0/6] new mlx4 datapath bypassing ibverbs Ferruh Yigit
2017-10-04 18:48   ` [dpdk-dev] [PATCH v3 " Adrien Mazarguil
2017-10-04 18:48     ` [dpdk-dev] [PATCH v3 1/6] net/mlx4: add simple Tx bypassing Verbs Adrien Mazarguil
2017-10-04 18:48     ` [dpdk-dev] [PATCH v3 2/6] net/mlx4: restore full Rx support " Adrien Mazarguil
2017-10-04 18:48     ` [dpdk-dev] [PATCH v3 3/6] net/mlx4: restore Tx gather support Adrien Mazarguil
2017-10-04 18:48     ` [dpdk-dev] [PATCH v3 4/6] net/mlx4: restore Tx checksum offloads Adrien Mazarguil
2017-10-04 18:48     ` [dpdk-dev] [PATCH v3 5/6] net/mlx4: restore Rx offloads Adrien Mazarguil
2017-10-04 18:48     ` [dpdk-dev] [PATCH v3 6/6] net/mlx4: add loopback Tx from VF Adrien Mazarguil
2017-10-05  9:33     ` [dpdk-dev] [PATCH v4 0/7] new mlx4 datapath bypassing ibverbs Ophir Munk
2017-10-05  9:33       ` [dpdk-dev] [PATCH v4 1/7] net/mlx4: add simple Tx bypassing Verbs Ophir Munk
2017-10-05  9:33       ` [dpdk-dev] [PATCH v4 2/7] net/mlx4: restore full Rx support " Ophir Munk
2017-10-05  9:33       ` [dpdk-dev] [PATCH v4 3/7] net/mlx4: restore Rx scatter support Ophir Munk
2017-10-05  9:33       ` [dpdk-dev] [PATCH v4 4/7] net/mlx4: restore Tx gather support Ophir Munk
2017-10-05  9:33       ` [dpdk-dev] [PATCH v4 5/7] net/mlx4: restore Tx checksum offloads Ophir Munk
2017-10-05  9:33       ` [dpdk-dev] [PATCH v4 6/7] net/mlx4: restore Rx offloads Ophir Munk
2017-10-05  9:33       ` [dpdk-dev] [PATCH v4 7/7] net/mlx4: add loopback Tx from VF Ophir Munk
2017-10-05 11:40       ` [dpdk-dev] [PATCH v4 0/7] new mlx4 datapath bypassing ibverbs Adrien Mazarguil
2017-10-05 18:48       ` Ferruh Yigit
2017-10-05 18:54         ` Ferruh Yigit [this message]
2017-10-11 18:31       ` [dpdk-dev] [PATCH v5 0/5] " Adrien Mazarguil
2017-10-11 18:31         ` [dpdk-dev] [PATCH v5 1/5] net/mlx4: add Tx bypassing Verbs Adrien Mazarguil
2017-10-11 18:31         ` [dpdk-dev] [PATCH v5 2/5] net/mlx4: add Rx " Adrien Mazarguil
2017-10-11 18:32         ` [dpdk-dev] [PATCH v5 3/5] net/mlx4: restore Tx checksum offloads Adrien Mazarguil
2017-10-11 18:32         ` [dpdk-dev] [PATCH v5 4/5] net/mlx4: restore Rx offloads Adrien Mazarguil
2017-10-11 18:32         ` [dpdk-dev] [PATCH v5 5/5] net/mlx4: add loopback Tx from VF Adrien Mazarguil
2017-10-12 12:29         ` [dpdk-dev] [PATCH v6 0/5] new mlx4 datapath bypassing ibverbs Adrien Mazarguil
2017-10-12 12:29           ` [dpdk-dev] [PATCH v6 1/5] net/mlx4: add Tx bypassing Verbs Adrien Mazarguil
2017-10-12 12:29           ` [dpdk-dev] [PATCH v6 2/5] net/mlx4: add Rx " Adrien Mazarguil
2017-10-12 12:29           ` [dpdk-dev] [PATCH v6 3/5] net/mlx4: restore Tx checksum offloads Adrien Mazarguil
2017-10-12 12:29           ` [dpdk-dev] [PATCH v6 4/5] net/mlx4: restore Rx offloads Adrien Mazarguil
2017-10-12 12:30           ` [dpdk-dev] [PATCH v6 5/5] net/mlx4: add loopback Tx from VF Adrien Mazarguil
2017-10-24  6:29           ` [dpdk-dev] [PATCH v6 0/5] new mlx4 datapath bypassing ibverbs gowrishankar muthukrishnan
2017-10-24  8:49             ` gowrishankar muthukrishnan
2017-10-24  9:55               ` Nélio Laranjeiro
2017-10-24 10:01                 ` Adrien Mazarguil
2017-10-24 16:59           ` Ferruh Yigit
2017-10-04 21:48   ` [dpdk-dev] [PATCH v3 0/7] " Ophir Munk
2017-10-04 21:49     ` [dpdk-dev] [PATCH v3 1/7] net/mlx4: add simple Tx " Ophir Munk
2017-10-04 21:49     ` [dpdk-dev] [PATCH v3 2/7] net/mlx4: get back Rx flow functionality Ophir Munk
2017-10-04 21:49     ` [dpdk-dev] [PATCH v3 3/7] net/mlx4: support multi-segments Rx Ophir Munk
2017-10-04 21:49     ` [dpdk-dev] [PATCH v3 4/7] net/mlx4: support multi-segments Tx Ophir Munk
2017-10-04 21:49     ` [dpdk-dev] [PATCH v3 5/7] net/mlx4: get back Tx checksum offloads Ophir Munk
2017-10-04 21:49     ` [dpdk-dev] [PATCH v3 6/7] net/mlx4: get back Rx " Ophir Munk
2017-10-04 21:49     ` [dpdk-dev] [PATCH v3 7/7] net/mlx4: add loopback Tx from VF Ophir Munk
2017-10-04 22:37     ` [dpdk-dev] [PATCH v3 0/7] new mlx4 datapath bypassing ibverbs Ferruh Yigit
2017-10-04 22:46       ` Thomas Monjalon
2017-10-24 11:56 ` [dpdk-dev] [PATCH 0/5] new mlx4 Tx " Nélio Laranjeiro

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=c3bd8f90-2028-d50a-8a29-8c756d740b8b@intel.com \
    --to=ferruh.yigit@intel.com \
    --cc=adrien.mazarguil@6wind.com \
    --cc=dev@dpdk.org \
    --cc=matan@mellanox.com \
    --cc=olgas@mellanox.com \
    --cc=ophirmu@mellanox.com \
    --cc=thomas@monjalon.net \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).