From: "Avijit Pandey" <Avijit@voereir.com>
To: "nicolas.chautru@intel.com" <nicolas.chautru@intel.com>,
"dev@dpdk.org" <dev@dpdk.org>
Subject: Re: Validation error in LDPC decoding
Date: Mon, 1 Jul 2024 05:28:40 +0000 [thread overview]
Message-ID: <AS8PR02MB683708002C4FA18EB3995974CCD32@AS8PR02MB6837.eurprd02.prod.outlook.com> (raw)
In-Reply-To: <AS8PR02MB68376A490051FE73A1B0740BCCD62@AS8PR02MB6837.eurprd02.prod.outlook.com>
[-- Attachment #1.1: Type: text/plain, Size: 3487 bytes --]
Hello Nicolas,
Tagging you, as you helped in solving previous queries related to BBdev.
Any insights or guidance to help resolve this issue would be highly appreciated. If you need any more details, please feel free to ask.
Thank you for your time and assistance!
Best Regards,
Avijit Pandey
Cloud SME | VoerEir AB
+91 9598570190<tel:+917506391870>
[Title: LinkedIn - Description: image of LinkedIn icon]<https://www.linkedin.com/in/avijit-pandey-603206154/>
[Image]
From: Avijit Pandey <Avijit@voereir.com>
Date: Wednesday, 26 June 2024 at 16:52
To: dev@dpdk.org <dev@dpdk.org>
Subject: Validation error in LDPC decoding
Hello Devs,
I hope this email finds you well.
I am reaching out to seek assistance regarding an issue I am facing in DPDK-BBdev utility, (v22.11).
I am performing some tests over a 5G Accelerator, ACC200 using DPDK-BBdev. I am creating custom test-vector files to generate custom load over the ACC200 device. Currently, I am performing the LDPC decoding test through code block mode. I am creating the test-vector files with help of an open-source library, sionna (https://nvlabs.github.io/sionna/). Following are the steps to create input0 and output0 used in BBdev vector file. I have pulled out this code form the following reference: https://github.com/NVlabs/sionna/discussions/54.
k, n, q_m = 32, 64, 2
encoder = sn.fec.ldpc.LDPC5GEncoder(k, n, num_bits_per_symbol=q_m)
c = encoder(u)
e = c.numpy()
for i in range(e.shape[0]):
for j in flipped_positions[i]:
e[i][j] = 1.0 if e[i][j] == 0.0 else 0.0
e = tf.convert_to_tensor(e, np.float32)
decoder = sn.fec.ldpc.LDPC5GDecoder(encoder, hard_out=True, return_infobits=True, num_iter=20, cn_type="minsum")
w = decoder(2*e - 1)
When I use w as input0 and u as output0 in BBdev vector file I face following validation error:
Enqueue 32 reqs (phys 0x17ff80000) to reg 0x1180000000
Debug : MMIO Enqueue
TestCase check_dec_status_and_ordering() line 1936 failed: op_status (8) != expected_status (0)
TestCase validate_ldpc_dec_op() line 2328 failed (err -1): Checking status and ordering for decoder failed
TestCase latency_test_ldpc_dec() line 4645 failed (err -1): Validation failed!
727:get_stats_from_queues() Got stats on 0
762:rte_bbdev_stats_get() Retrieved stats of device 0
602:rte_bbdev_stop() Stopped device 0
TestCase [ 0] : validation_tc failed
647:rte_bbdev_close() Closed device 0
Please find below the BBdev vector file created for the same
op_type =
RTE_BBDEV_OP_LDPC_DEC
input0=
0xff01ff01,0x0101ff01,0x0101ff01,0x01010101,0x0101ff01,0xff010101,0x01010101,0x01010101,0xff01ff01,0x0101ff01,0xff010101,0xffffff01,0xff0101ff,0x0101ff01,0xffff0101,0xff0101ff
output0=
0xffffffff
basegraph=
2
z_c=
8
q_m=
2
n_filler=
28
n_cb=
300
e=
64
rv_index =
0
code_block_mode =
1
iter_max =
20
expected_iter_count =
20
op_flags =
RTE_BBDEV_LDPC_ITERATION_STOP_ENABLE
expected_status =
OK
I would like to understand what incorrect configuration am I using?
Any insights or guidance to help resolve this issue would be highly appreciated. If you need any more details, please feel free to ask.
Thank you for your time and assistance!
Best Regards,
Avijit Pandey
Cloud SME | VoerEir AB
+91 9598570190<tel:+917506391870>
[Title: LinkedIn - Description: image of LinkedIn icon]<https://www.linkedin.com/in/avijit-pandey-603206154/>
[Image]
[-- Attachment #1.2: Type: text/html, Size: 26147 bytes --]
[-- Attachment #2: image003.png --]
[-- Type: image/png, Size: 1438 bytes --]
[-- Attachment #3: image004.png --]
[-- Type: image/png, Size: 8370 bytes --]
prev parent reply other threads:[~2024-07-01 5:28 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-06-26 11:22 Avijit Pandey
2024-07-01 5:28 ` Avijit Pandey [this message]
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=AS8PR02MB683708002C4FA18EB3995974CCD32@AS8PR02MB6837.eurprd02.prod.outlook.com \
--to=avijit@voereir.com \
--cc=dev@dpdk.org \
--cc=nicolas.chautru@intel.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).