From: Srikanth Yalavarthi <syalavarthi@marvell.com>
Cc: <dev@dpdk.org>, <syalavarthi@marvell.com>,
<sshankarnara@marvell.com>, <aprabhu@marvell.com>,
<ptakkar@marvell.com>
Subject: [PATCH v2 0/3] Spec changes to support multi I/O models
Date: Wed, 20 Sep 2023 00:19:07 -0700 [thread overview]
Message-ID: <20230920071910.10428-1-syalavarthi@marvell.com> (raw)
In-Reply-To: <20230830155303.30380-1-syalavarthi@marvell.com>
This series implements changes to mldev spec to extend support
for ML models with multiple inputs and outputs. Changes include
introduction of I/O layout to support packed and split buffers
for model input and output. Extended the rte_ml_model_info
structure to support multiple inputs and outputs.
Updated rte_ml_op and quantize / dequantize APIs to support an
array of input and output ML buffer segments.
Support for batches option is dropped from test application.
v2:
- Minor fixes
- Cleanup of application help
v1:
- Initial changes
Srikanth Yalavarthi (3):
mldev: add support for arbitrary shape dimensions
mldev: introduce support for IO layout
mldev: drop input and output size get APIs
app/test-mldev/ml_options.c | 16 -
app/test-mldev/ml_options.h | 2 -
app/test-mldev/test_inference_common.c | 420 +++++++++++++++++--------
app/test-mldev/test_inference_common.h | 6 +
app/test-mldev/test_model_common.c | 6 -
app/test-mldev/test_model_common.h | 1 -
doc/guides/tools/testmldev.rst | 6 -
drivers/ml/cnxk/cn10k_ml_dev.h | 3 +
drivers/ml/cnxk/cn10k_ml_model.c | 84 +++--
drivers/ml/cnxk/cn10k_ml_model.h | 12 +
drivers/ml/cnxk/cn10k_ml_ops.c | 135 +++-----
lib/mldev/meson.build | 2 +-
lib/mldev/mldev_utils.c | 30 --
lib/mldev/mldev_utils.h | 16 -
lib/mldev/rte_mldev.c | 50 +--
lib/mldev/rte_mldev.h | 201 +++++-------
lib/mldev/rte_mldev_core.h | 68 +---
lib/mldev/version.map | 3 -
18 files changed, 506 insertions(+), 555 deletions(-)
--
2.41.0
next prev parent reply other threads:[~2023-09-20 7:19 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-08-30 15:52 [PATCH v1 " Srikanth Yalavarthi
2023-08-30 15:53 ` [PATCH v1 1/3] mldev: add support for arbitrary shape dimensions Srikanth Yalavarthi
2023-08-30 15:53 ` [PATCH v1 2/3] mldev: introduce support for IO layout Srikanth Yalavarthi
2023-08-30 15:53 ` [PATCH v1 3/3] mldev: drop input and output size get APIs Srikanth Yalavarthi
2023-09-20 7:19 ` Srikanth Yalavarthi [this message]
2023-09-20 7:19 ` [PATCH v2 1/3] mldev: add support for arbitrary shape dimensions Srikanth Yalavarthi
2023-09-20 7:19 ` [PATCH v2 2/3] mldev: introduce support for IO layout Srikanth Yalavarthi
2023-09-20 7:19 ` [PATCH v2 3/3] mldev: drop input and output size get APIs Srikanth Yalavarthi
2023-10-03 6:10 ` Anup Prabhu
2023-09-27 18:11 ` [PATCH v3 0/4] Spec changes to support multi I/O models Srikanth Yalavarthi
2023-09-27 18:11 ` [PATCH v3 1/4] mldev: add support for arbitrary shape dimensions Srikanth Yalavarthi
2023-09-27 18:11 ` [PATCH v3 2/4] mldev: introduce support for IO layout Srikanth Yalavarthi
2023-09-27 18:11 ` [PATCH v3 3/4] mldev: drop input and output size get APIs Srikanth Yalavarthi
2023-09-27 18:11 ` [PATCH v3 4/4] mldev: update release notes for 23.11 Srikanth Yalavarthi
2023-09-29 3:39 ` Jerin Jacob
2023-10-02 9:59 ` [EXT] " Srikanth Yalavarthi
2023-10-02 9:58 ` [PATCH v4 0/3] Spec changes to support multi I/O models Srikanth Yalavarthi
2023-10-02 9:58 ` [PATCH v4 1/3] mldev: add support for arbitrary shape dimensions Srikanth Yalavarthi
2023-10-04 14:42 ` Anup Prabhu
2023-10-05 9:12 ` Shivah Shankar Shankar Narayan Rao
2023-10-02 9:58 ` [PATCH v4 2/3] mldev: introduce support for IO layout Srikanth Yalavarthi
2023-10-05 9:10 ` Shivah Shankar Shankar Narayan Rao
2023-10-02 9:58 ` [PATCH v4 3/3] mldev: drop input and output size get APIs Srikanth Yalavarthi
2023-10-03 6:12 ` Anup Prabhu
2023-10-05 9:06 ` Shivah Shankar Shankar Narayan Rao
2023-10-11 14:45 ` [PATCH v4 0/3] Spec changes to support multi I/O models Thomas Monjalon
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=20230920071910.10428-1-syalavarthi@marvell.com \
--to=syalavarthi@marvell.com \
--cc=aprabhu@marvell.com \
--cc=dev@dpdk.org \
--cc=ptakkar@marvell.com \
--cc=sshankarnara@marvell.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).