From mboxrd@z Thu Jan 1 00:00:00 1970
Return-Path: <arybchenko@solarflare.com>
Received: from dispatch1-us1.ppe-hosted.com (dispatch1-us1.ppe-hosted.com
[148.163.129.52]) by dpdk.org (Postfix) with ESMTP id 98A2B5F14
for <dev@dpdk.org>; Mon, 26 Mar 2018 18:10:10 +0200 (CEST)
X-Virus-Scanned: Proofpoint Essentials engine
Received: from webmail.solarflare.com (webmail.solarflare.com [12.187.104.26])
(using TLSv1 with cipher ECDHE-RSA-AES256-SHA (256/256 bits))
(No client certificate requested)
by mx1-us4.ppe-hosted.com (Proofpoint Essentials ESMTP Server) with ESMTPS id
21A9FBC006E; Mon, 26 Mar 2018 16:10:09 +0000 (UTC)
Received: from ocex03.SolarFlarecom.com (10.20.40.36) by
ocex03.SolarFlarecom.com (10.20.40.36) with Microsoft SMTP Server (TLS) id
15.0.1044.25; Mon, 26 Mar 2018 09:10:06 -0700
Received: from opal.uk.solarflarecom.com (10.17.10.1) by
ocex03.SolarFlarecom.com (10.20.40.36) with Microsoft SMTP Server (TLS) id
15.0.1044.25 via Frontend Transport; Mon, 26 Mar 2018 09:10:05 -0700
Received: from uklogin.uk.solarflarecom.com (uklogin.uk.solarflarecom.com
[10.17.10.10])
by opal.uk.solarflarecom.com (8.13.8/8.13.8) with ESMTP id w2QGA4p5003630;
Mon, 26 Mar 2018 17:10:04 +0100
Received: from uklogin.uk.solarflarecom.com (localhost.localdomain [127.0.0.1])
by uklogin.uk.solarflarecom.com (8.13.8/8.13.8) with ESMTP id w2QGA4Z0024909;
Mon, 26 Mar 2018 17:10:04 +0100
From: Andrew Rybchenko <arybchenko@solarflare.com>
To: <dev@dpdk.org>
CC: Olivier MATZ <olivier.matz@6wind.com>, Thomas Monjalon
<thomas@monjalon.net>, Anatoly Burakov <anatoly.burakov@intel.com>, Santosh
Shukla <santosh.shukla@caviumnetworks.com>, Jerin Jacob
<jerin.jacob@caviumnetworks.com>, Hemant Agrawal <hemant.agrawal@nxp.com>,
Shreyansh Jain <shreyansh.jain@nxp.com>
Date: Mon, 26 Mar 2018 17:09:40 +0100
Message-ID: <1522080591-24705-1-git-send-email-arybchenko@solarflare.com>
X-Mailer: git-send-email 1.8.2.3
In-Reply-To: <1516713372-10572-1-git-send-email-arybchenko@solarflare.com>
References: <1516713372-10572-1-git-send-email-arybchenko@solarflare.com>
MIME-Version: 1.0
Content-Type: text/plain
X-MDID: 1522080609-9pofIu6DFqg0
Subject: [dpdk-dev] [PATCH v3 00/11] mempool: prepare to add bucket driver
X-BeenThere: dev@dpdk.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: DPDK patches and discussions <dev.dpdk.org>
List-Unsubscribe: <https://dpdk.org/ml/options/dev>,
<mailto:dev-request@dpdk.org?subject=unsubscribe>
List-Archive: <http://dpdk.org/ml/archives/dev/>
List-Post: <mailto:dev@dpdk.org>
List-Help: <mailto:dev-request@dpdk.org?subject=help>
List-Subscribe: <https://dpdk.org/ml/listinfo/dev>,
<mailto:dev-request@dpdk.org?subject=subscribe>
X-List-Received-Date: Mon, 26 Mar 2018 16:10:11 -0000
The patch series should be applied on top of [7].
The initial patch series [1] is split into two to simplify processing.
The second series relies on this one and will add bucket mempool driver
and related ops.
The patch series has generic enhancements suggested by Olivier.
Basically it adds driver callbacks to calculate required memory size and
to populate objects using provided memory area. It allows to remove
so-called capability flags used before to tell generic code how to
allocate and slice allocated memory into mempool objects.
Clean up which removes get_capabilities and register_memory_area is
not strictly required, but I think right thing to do.
Existing mempool drivers are updated.
rte_mempool_populate_iova_tab() is also deprecated in v2 as agreed in [2].
Unfortunately it requires addition of -Wno-deprecated-declarations flag
to librte_mempool since the function is used by deprecated earlier
rte_mempool_populate_phys_tab(). If the later may be removed in the
release, we can avoid addition of the flag to allow usage of deprecated
functions.
One open question remains from previous review [3].
The patch series interfere with memory hotplug for DPDK [4] ([5] to be
precise). So, rebase may be required.
A new patch is added to the series to rename MEMPOOL_F_NO_PHYS_CONTIG
as MEMPOOL_F_NO_IOVA_CONTIG as agreed in [6].
MEMPOOL_F_CAPA_PHYS_CONTIG is not renamed since it removed in this
patchset.
It breaks ABI since changes rte_mempool_ops. Also it removes
rte_mempool_ops_register_memory_area() and
rte_mempool_ops_get_capabilities() since corresponding callbacks are
removed.
Internal global functions are not listed in map file since it is not
a part of external API.
[1] https://dpdk.org/ml/archives/dev/2018-January/088698.html
[2] https://dpdk.org/ml/archives/dev/2018-March/093186.html
[3] https://dpdk.org/ml/archives/dev/2018-March/093329.html
[4] https://dpdk.org/ml/archives/dev/2018-March/092070.html
[5] https://dpdk.org/ml/archives/dev/2018-March/092088.html
[6] https://dpdk.org/ml/archives/dev/2018-March/093345.html
[7] https://dpdk.org/ml/archives/dev/2018-March/093196.html
v2 -> v3:
- fix build error in mempool/dpaa: prepare to remove register memory area op
v1 -> v2:
- deprecate rte_mempool_populate_iova_tab()
- add patch to fix memory leak if no objects are populated
- add patch to rename MEMPOOL_F_NO_PHYS_CONTIG
- minor fixes (typos, blank line at the end of file)
- highlight meaning of min_chunk_size (when it is virtual or
physical contiguous)
- make sure that mempool is initialized in rte_mempool_populate_anon()
- move patch to ensure that mempool is initialized earlier in the series
RFCv2 -> v1:
- split the series in two
- squash octeontx patches which implement calc_mem_size and populate
callbacks into the patch which removes get_capabilities since it is
the easiest way to untangle the tangle of tightly related library
functions and flags advertised by the driver
- consistently name default callbacks
- move default callbacks to dedicated file
- see detailed description in patches
RFCv1 -> RFCv2:
- add driver ops to calculate required memory size and populate
mempool objects, remove extra flags which were required before
to control it
- transition of octeontx and dpaa drivers to the new callbacks
- change info API to get information from driver required to
API user to know contiguous block size
- remove get_capabilities (not required any more and may be
substituted with more in info get API)
- remove register_memory_area since it is substituted with
populate callback which can do more
- use SPDX tags
- avoid all objects affinity to single lcore
- fix bucket get_count
- deprecate XMEM API
- avoid introduction of a new function to flush cache
- fix NO_CACHE_ALIGN case in bucket mempool
Andrew Rybchenko (9):
mempool: fix memhdr leak when no objects are populated
mempool: rename flag to control IOVA-contiguous objects
mempool: add op to calculate memory size to be allocated
mempool: add op to populate objects using provided memory
mempool: remove callback to get capabilities
mempool: deprecate xmem functions
mempool/octeontx: prepare to remove register memory area op
mempool/dpaa: prepare to remove register memory area op
mempool: remove callback to register memory area
Artem V. Andreev (2):
mempool: ensure the mempool is initialized before populating
mempool: support flushing the default cache of the mempool
doc/guides/rel_notes/deprecation.rst | 12 +-
doc/guides/rel_notes/release_18_05.rst | 33 ++-
drivers/mempool/dpaa/dpaa_mempool.c | 13 +-
drivers/mempool/octeontx/rte_mempool_octeontx.c | 64 ++++--
drivers/net/thunderx/nicvf_ethdev.c | 2 +-
lib/librte_mempool/Makefile | 6 +-
lib/librte_mempool/meson.build | 17 +-
lib/librte_mempool/rte_mempool.c | 179 ++++++++-------
lib/librte_mempool/rte_mempool.h | 280 +++++++++++++++++-------
lib/librte_mempool/rte_mempool_ops.c | 37 ++--
lib/librte_mempool/rte_mempool_ops_default.c | 51 +++++
lib/librte_mempool/rte_mempool_version.map | 10 +-
test/test/test_mempool.c | 31 ---
13 files changed, 485 insertions(+), 250 deletions(-)
create mode 100644 lib/librte_mempool/rte_mempool_ops_default.c
--
2.7.4