From: Shreyansh Jain <shreyansh.jain@nxp.com>
To: Olivier Matz <olivier.matz@6wind.com>
Cc: <dev@dpdk.org>, <thomas.monjalon@6wind.com>, <hemant.agrawal@nxp.com>
Subject: Re: [dpdk-dev] [PATCH 1/2] drivers/mempool: add stack mempool handler as driver
Date: Mon, 27 Mar 2017 10:24:51 +0530 [thread overview]
Message-ID: <452b48e9-c94e-954e-6692-94692f456dfb@nxp.com> (raw)
In-Reply-To: <20170324172227.5bdfac8c@platinum>
Hello Olivier,
On Friday 24 March 2017 09:52 PM, Olivier Matz wrote:
> Hi Shreyansh,
>
> On Mon, 20 Mar 2017 15:33:09 +0530, Shreyansh Jain <shreyansh.jain@nxp.com> wrote:
>> CONFIG_RTE_DRIVER_MEMPOOL_STACK option added to common_base.
>> Stack mempool handler moved from lib/librte_mempool into drivers/mempool.
>>
>> With this patch, the Stack mempool handler registration is optional and
>> toggled via the configuration file. In case disabled (N), it would imply
>> request for creating of mempool would fail.
>>
>> Signed-off-by: Shreyansh Jain <shreyansh.jain@nxp.com>
>> ---
>> config/common_base | 5 +
>> drivers/Makefile | 1 +
>> drivers/mempool/Makefile | 36 +++++
>> drivers/mempool/stack/Makefile | 55 ++++++++
>> drivers/mempool/stack/rte_mempool_stack.c | 147 +++++++++++++++++++++
>> .../mempool/stack/rte_mempool_stack_version.map | 4 +
>> lib/librte_mempool/Makefile | 1 -
>> lib/librte_mempool/rte_mempool_stack.c | 147 ---------------------
>> 8 files changed, 248 insertions(+), 148 deletions(-)
>> create mode 100644 drivers/mempool/Makefile
>> create mode 100644 drivers/mempool/stack/Makefile
>> create mode 100644 drivers/mempool/stack/rte_mempool_stack.c
>> create mode 100644 drivers/mempool/stack/rte_mempool_stack_version.map
>> delete mode 100644 lib/librte_mempool/rte_mempool_stack.c
>
>
>
> I tried to pass the mempool autotest, and it issues a segfault.
> I think the libraries are missing in rte.app.mk, so no handler is
> registered.
>
> Adding the following code in lib/librte_mempool/rte_mempool_ops.c
> fixes the crash.
>
> ops = rte_mempool_get_ops(mp->ops_index);
> + if (ops == NULL || ops->alloc == NULL)
> + return -ENOTSUP;
> return ops->alloc(mp);
>
> Now that drivers are not linked to the mempool library, it can
> happen that there is no handler. Could you please add this patch in your
> patchset?
Indeed. I will add this code set as first patch. Thanks for suggestion/fix.
>
> I also checked that compilation works in shared lib mode. It looks good,
> but I think there will be a behavior change if CONFIG_RTE_EAL_PMD_PATH
> is empty (default). This option is probably used by distros to indicate
> where dpdk plugins are installed, and when it is set, all drivers of
> this directory are loaded, so in that case it won't change.
>
> But when unset, the drivers have to be loaded manually, and with this
> change, the mempool driver will have to be loaded with the -d eal option.
> Could you please check what occurs in that case? At least see if it
> displays a nice error or if it crashes. I suspect it will crash
Ok. I will try this and if there is any issue, fix it.
>
> Also, the MAINTAINERS file should be updated.
Yes, that is something I thought of updating but left it out before
sending the patch.
One confirmation: I assume that maintainers need to be added with
"drivers/mempool/ring" and "drivers/mempool/stack" folders.
Who should be marked as maintainer - You (because of existing
lib/librte_mempool ownership) or I (because I have moved the code from
lib/librte_mempool)?
I think it would continue to be you but wanted to take your
confirmation before adding the lines.
>
>
> Thanks,
> Olivier
>
>
-
Shreyansh
next prev parent reply other threads:[~2017-03-27 4:48 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-03-20 10:03 Shreyansh Jain
2017-03-20 10:03 ` [dpdk-dev] [PATCH 2/2] drivers/mempool: add ring " Shreyansh Jain
2017-03-20 14:50 ` [dpdk-dev] [PATCH 1/2] drivers/mempool: add stack " Hunt, David
2017-03-21 4:55 ` Shreyansh Jain
2017-03-21 6:02 ` Wiles, Keith
2017-03-21 6:25 ` Shreyansh Jain
2017-03-21 6:28 ` Shreyansh Jain
2017-03-21 14:45 ` Wiles, Keith
2017-03-24 16:22 ` Olivier Matz
2017-03-27 4:54 ` Shreyansh Jain [this message]
2017-03-27 7:22 ` Olivier Matz
2017-03-28 11:42 ` Shreyansh Jain
2017-03-29 8:18 ` Olivier Matz
2017-03-29 12:55 ` Shreyansh Jain
2017-03-30 12:35 ` Olivier Matz
2017-03-31 5:29 ` [dpdk-dev] [PATCH 1/3] mempool: fix segfault for unlinked mempool handler Shreyansh Jain
2017-03-31 5:29 ` [dpdk-dev] [PATCH 2/3] mempool: introduce ring mempool driver Shreyansh Jain
2017-03-31 5:29 ` [dpdk-dev] [PATCH 3/3] mempool: introduce stack " Shreyansh Jain
2017-03-31 5:31 ` [dpdk-dev] [PATCH 1/3] mempool: fix segfault for unlinked mempool handler Shreyansh Jain
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=452b48e9-c94e-954e-6692-94692f456dfb@nxp.com \
--to=shreyansh.jain@nxp.com \
--cc=dev@dpdk.org \
--cc=hemant.agrawal@nxp.com \
--cc=olivier.matz@6wind.com \
--cc=thomas.monjalon@6wind.com \
/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).