From: Nicolas Chautru <nicolas.chautru@intel.com>
To: thomas@monjalon.net, akhil.goyal@nxp.com, dev@dpdk.org
Cc: ferruh.yigit@intel.com, Nic Chautru <nicolas.chautru@intel.com>
Subject: [dpdk-dev] [PATCH v3 04/14] baseband/turbo_sw: support large size code block
Date: Wed, 4 Mar 2020 10:54:52 -0800 [thread overview]
Message-ID: <1583348102-13253-5-git-send-email-nicolas.chautru@intel.com> (raw)
In-Reply-To: <1583348102-13253-1-git-send-email-nicolas.chautru@intel.com>
From: Nic Chautru <nicolas.chautru@intel.com>
This is to support cases when the input data for
decoding a code block is larger than 64kB and would
not fit as a contiguous block of data into one
mbuf. In that case the length from the opearation
supersedes the mbug default structure.
Signed-off-by: Nic Chautru <nicolas.chautru@intel.com>
---
app/test-bbdev/test_bbdev_perf.c | 40 +++++++++++++++++++-----
doc/guides/rel_notes/release_20_05.rst | 6 ++++
drivers/baseband/turbo_sw/bbdev_turbo_software.c | 11 ++++---
3 files changed, 45 insertions(+), 12 deletions(-)
diff --git a/app/test-bbdev/test_bbdev_perf.c b/app/test-bbdev/test_bbdev_perf.c
index d8db58e..d46966d 100644
--- a/app/test-bbdev/test_bbdev_perf.c
+++ b/app/test-bbdev/test_bbdev_perf.c
@@ -764,6 +764,7 @@ typedef int (test_case_function)(struct active_device *ad,
{
int ret;
unsigned int i, j;
+ bool large_input = false;
for (i = 0; i < n; ++i) {
char *data;
@@ -774,24 +775,47 @@ typedef int (test_case_function)(struct active_device *ad,
op_type, n * ref_entries->nb_segments,
mbuf_pool->size);
- TEST_ASSERT_SUCCESS(((seg->length + RTE_PKTMBUF_HEADROOM) >
- (uint32_t)UINT16_MAX),
- "Given data is bigger than allowed mbuf segment size");
-
+ if (seg->length > 64000) {
+ /*
+ * Special case when DPDK mbuf cannot handle
+ * the required input size
+ */
+ printf("Warning: Larger input size than DPDK mbuf %d\n",
+ seg->length);
+ large_input = true;
+ } else {
+ TEST_ASSERT_SUCCESS(
+ ((seg->length + RTE_PKTMBUF_HEADROOM)
+ > (uint32_t)UINT16_MAX),
+ "Given data is bigger than allowed mbuf segment size"
+ );
+ }
bufs[i].data = m_head;
bufs[i].offset = 0;
bufs[i].length = 0;
if ((op_type == DATA_INPUT) || (op_type == DATA_HARQ_INPUT)) {
- data = rte_pktmbuf_append(m_head, seg->length);
- TEST_ASSERT_NOT_NULL(data,
+ if ((op_type == DATA_INPUT) && large_input) {
+ /* Allocate a fake overused mbuf */
+ data = rte_malloc(NULL, 128 * 1024, 0);
+ memcpy(data, seg->addr, seg->length);
+ m_head->buf_addr = data;
+ m_head->buf_iova = rte_mem_virt2phy(data);
+ m_head->data_off = 0;
+ m_head->data_len = seg->length;
+ } else {
+ data = rte_pktmbuf_append(m_head, seg->length);
+ TEST_ASSERT_NOT_NULL(data,
"Couldn't append %u bytes to mbuf from %d data type mbuf pool",
seg->length, op_type);
- TEST_ASSERT(data == RTE_PTR_ALIGN(data, min_alignment),
+ TEST_ASSERT(data == RTE_PTR_ALIGN(
+ data, min_alignment),
"Data addr in mbuf (%p) is not aligned to device min alignment (%u)",
data, min_alignment);
- rte_memcpy(data, seg->addr, seg->length);
+ rte_memcpy(data, seg->addr, seg->length);
+ }
+
bufs[i].length += seg->length;
for (j = 1; j < ref_entries->nb_segments; ++j) {
diff --git a/doc/guides/rel_notes/release_20_05.rst b/doc/guides/rel_notes/release_20_05.rst
index 2190eaf..d6c3dfb 100644
--- a/doc/guides/rel_notes/release_20_05.rst
+++ b/doc/guides/rel_notes/release_20_05.rst
@@ -56,6 +56,12 @@ New Features
Also, make sure to start the actual text at the margin.
=========================================================
+* **Updated the TURBO_SW bbdev PMD.**
+
+ Updated the ``turbo_sw`` bbdev driver with changes including:
+
+ * Support for large size code block not fitting in one mbuf segment.
+ * Exposes the accurate LLR decimal assumption.
Removed Items
-------------
diff --git a/drivers/baseband/turbo_sw/bbdev_turbo_software.c b/drivers/baseband/turbo_sw/bbdev_turbo_software.c
index 5ca8ca1..ea3fecb 100644
--- a/drivers/baseband/turbo_sw/bbdev_turbo_software.c
+++ b/drivers/baseband/turbo_sw/bbdev_turbo_software.c
@@ -1335,7 +1335,7 @@ struct turbo_sw_queue {
static inline void
process_ldpc_dec_cb(struct turbo_sw_queue *q, struct rte_bbdev_dec_op *op,
- uint8_t c, uint16_t out_length, uint16_t e,
+ uint8_t c, uint16_t out_length, uint32_t e,
struct rte_mbuf *m_in,
struct rte_mbuf *m_out_head, struct rte_mbuf *m_out,
struct rte_mbuf *m_harq_in,
@@ -1617,8 +1617,8 @@ struct turbo_sw_queue {
struct rte_bbdev_stats *queue_stats)
{
uint8_t c, r = 0;
- uint16_t e, out_length;
- uint16_t crc24_overlap = 0;
+ uint32_t e;
+ uint16_t out_length, crc24_overlap = 0;
struct rte_bbdev_op_ldpc_dec *dec = &op->ldpc_dec;
struct rte_mbuf *m_in = dec->input.data;
struct rte_mbuf *m_harq_in = dec->harq_combined_input.data;
@@ -1661,7 +1661,10 @@ struct turbo_sw_queue {
e = (r < dec->tb_params.cab) ?
dec->tb_params.ea : dec->tb_params.eb;
- seg_total_left = rte_pktmbuf_data_len(m_in) - in_offset;
+ if (e < 64000) /* Special case handling when overusing mbuf */
+ seg_total_left = rte_pktmbuf_data_len(m_in) - in_offset;
+ else
+ seg_total_left = e;
process_ldpc_dec_cb(q, op, c, out_length, e,
m_in, m_out_head, m_out,
--
1.8.3.1
next prev parent reply other threads:[~2020-03-04 18:55 UTC|newest]
Thread overview: 115+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-02-27 4:38 [dpdk-dev] [PATCH v1 00/14] bbdev new features Nicolas Chautru
2020-02-27 4:38 ` [dpdk-dev] [PATCH v1 01/14] bbdev: add capability flag for filler bits inclusion in HARQ Nicolas Chautru
2020-02-27 4:38 ` [dpdk-dev] [PATCH v1 02/14] bbdev: expose device HARQ buffer size at device level Nicolas Chautru
2020-02-27 4:38 ` [dpdk-dev] [PATCH v1 03/14] baseband/turbo_sw: fix the exposed LLR decimals assumption Nicolas Chautru
2020-02-27 4:38 ` [dpdk-dev] [PATCH v1 04/14] baseband/turbo_sw: support large size code block Nicolas Chautru
2020-02-27 4:38 ` [dpdk-dev] [PATCH v1 05/14] test-bbdev: rename FPGA LTE macros to be more explicit Nicolas Chautru
2020-02-27 4:39 ` [dpdk-dev] [PATCH v1 06/14] test-bbdev: support HARQ validation Nicolas Chautru
2020-02-27 4:39 ` [dpdk-dev] [PATCH v1 07/14] test-bbdev: support for performance tests Nicolas Chautru
2020-02-27 4:39 ` [dpdk-dev] [PATCH v1 08/14] test-bbdev: support for LDPC interrupt test Nicolas Chautru
2020-02-27 4:39 ` [dpdk-dev] [PATCH v1 09/14] test-bbdev: support for offload test for LDPC Nicolas Chautru
2020-02-27 4:39 ` [dpdk-dev] [PATCH v1 10/14] test-bbdev: vectors update Nicolas Chautru
2020-02-27 4:39 ` [dpdk-dev] [PATCH v1 11/14] doc: update of testbbdev documentation Nicolas Chautru
2020-02-27 4:39 ` [dpdk-dev] [PATCH v1 12/14] drivers/baseband: add PMD for FPGA 5GNR FEC Nicolas Chautru
2020-02-27 4:39 ` [dpdk-dev] [PATCH v1 13/14] test-bbdev: add support for FPGA driver initialization Nicolas Chautru
2020-02-27 4:39 ` [dpdk-dev] [PATCH v1 14/14] doc: add feature matrix table for bbdev devices Nicolas Chautru
2020-03-04 18:22 ` [dpdk-dev] [PATCH v2 00/15] bbdev new features Nicolas Chautru
2020-03-04 18:22 ` [dpdk-dev] [PATCH v2 01/15] bbdev: add capability flag for filler bits inclusion in HARQ Nicolas Chautru
2020-03-04 18:22 ` [dpdk-dev] [PATCH v2 02/15] bbdev: expose device HARQ buffer size at device level Nicolas Chautru
2020-03-04 18:22 ` [dpdk-dev] [PATCH v2 03/15] baseband/turbo_sw: fix the exposed LLR decimals assumption Nicolas Chautru
2020-03-04 18:22 ` [dpdk-dev] [PATCH v2 04/15] baseband/turbo_sw: support large size code block Nicolas Chautru
2020-03-04 18:22 ` [dpdk-dev] [PATCH v2 05/15] test-bbdev: rename FPGA LTE macros to be more explicit Nicolas Chautru
2020-03-04 18:22 ` [dpdk-dev] [PATCH v2 06/15] test-bbdev: support HARQ validation Nicolas Chautru
2020-03-04 18:22 ` [dpdk-dev] [PATCH v2 07/15] test-bbdev: support for performance tests Nicolas Chautru
2020-03-04 18:22 ` [dpdk-dev] [PATCH v2 08/15] test-bbdev: support for LDPC interrupt test Nicolas Chautru
2020-03-04 18:22 ` [dpdk-dev] [PATCH v2 09/15] test-bbdev: support for offload test for LDPC Nicolas Chautru
2020-03-04 18:22 ` [dpdk-dev] [PATCH v2 10/15] test-bbdev: vectors update Nicolas Chautru
2020-03-04 18:22 ` [dpdk-dev] [PATCH v2 11/15] doc: update of testbbdev documentation Nicolas Chautru
2020-03-04 18:22 ` [dpdk-dev] [PATCH v2 12/15] drivers/baseband: add PMD for FPGA 5GNR FEC Nicolas Chautru
2020-03-04 18:22 ` [dpdk-dev] [PATCH v2 13/15] test-bbdev: add support for FPGA driver initialization Nicolas Chautru
2020-03-04 18:22 ` [dpdk-dev] [PATCH v2 14/15] doc: add feature matrix table for bbdev devices Nicolas Chautru
2020-03-04 18:22 ` [dpdk-dev] [PATCH v2 15/15] doc: add release note for bbdev PMD update Nicolas Chautru
2020-03-04 18:54 ` [dpdk-dev] [PATCH v3 00/14] bbdev new features Nicolas Chautru
2020-03-04 18:54 ` [dpdk-dev] [PATCH v3 01/14] bbdev: add capability flag for filler bits inclusion in HARQ Nicolas Chautru
2020-03-13 10:42 ` Dave Burley
2020-03-25 9:53 ` Akhil Goyal
2020-03-26 2:47 ` Chautru, Nicolas
2020-03-04 18:54 ` [dpdk-dev] [PATCH v3 02/14] bbdev: expose device HARQ buffer size at device level Nicolas Chautru
2020-03-13 10:47 ` Dave Burley
2020-03-04 18:54 ` [dpdk-dev] [PATCH v3 03/14] baseband/turbo_sw: fix the exposed LLR decimals assumption Nicolas Chautru
2020-03-04 18:54 ` Nicolas Chautru [this message]
2020-03-13 16:10 ` [dpdk-dev] [PATCH v3 04/14] baseband/turbo_sw: support large size code block Dave Burley
2020-03-25 10:18 ` Akhil Goyal
2020-03-26 2:50 ` Chautru, Nicolas
2020-03-04 18:54 ` [dpdk-dev] [PATCH v3 05/14] test-bbdev: rename FPGA LTE macros to be more explicit Nicolas Chautru
2020-03-13 10:55 ` Dave Burley
2020-03-04 18:54 ` [dpdk-dev] [PATCH v3 06/14] test-bbdev: support HARQ validation Nicolas Chautru
2020-03-25 13:23 ` Akhil Goyal
2020-03-26 2:53 ` Chautru, Nicolas
2020-03-04 18:54 ` [dpdk-dev] [PATCH v3 07/14] test-bbdev: support for performance tests Nicolas Chautru
2020-03-13 10:58 ` Dave Burley
2020-03-25 13:47 ` Akhil Goyal
2020-03-26 2:56 ` Chautru, Nicolas
2020-03-04 18:54 ` [dpdk-dev] [PATCH v3 08/14] test-bbdev: support for LDPC interrupt test Nicolas Chautru
2020-03-13 10:59 ` Dave Burley
2020-03-04 18:54 ` [dpdk-dev] [PATCH v3 09/14] test-bbdev: support for offload test for LDPC Nicolas Chautru
2020-03-25 13:54 ` Akhil Goyal
2020-03-26 2:57 ` Chautru, Nicolas
2020-03-04 18:54 ` [dpdk-dev] [PATCH v3 10/14] test-bbdev: vectors update Nicolas Chautru
2020-03-13 11:02 ` Dave Burley
2020-03-25 13:58 ` Akhil Goyal
2020-03-26 2:59 ` Chautru, Nicolas
2020-03-04 18:54 ` [dpdk-dev] [PATCH v3 11/14] doc: update of testbbdev documentation Nicolas Chautru
2020-03-13 11:03 ` Dave Burley
2020-03-25 14:48 ` Akhil Goyal
2020-03-26 3:02 ` Chautru, Nicolas
2020-03-04 18:55 ` [dpdk-dev] [PATCH v3 12/14] drivers/baseband: add PMD for FPGA 5GNR FEC Nicolas Chautru
2020-03-13 11:03 ` Dave Burley
2020-03-25 14:59 ` Akhil Goyal
2020-03-26 3:16 ` Chautru, Nicolas
2020-03-26 5:27 ` Akhil Goyal
2020-03-30 0:09 ` Chautru, Nicolas
2020-03-04 18:55 ` [dpdk-dev] [PATCH v3 13/14] test-bbdev: add support for FPGA driver initialization Nicolas Chautru
2020-03-13 11:05 ` Dave Burley
2020-03-04 18:55 ` [dpdk-dev] [PATCH v3 14/14] doc: add feature matrix table for bbdev devices Nicolas Chautru
2020-03-13 11:05 ` Dave Burley
2020-03-12 20:10 ` [dpdk-dev] [PATCH v3 00/14] bbdev new features Chautru, Nicolas
2020-03-13 7:15 ` Akhil Goyal
2020-03-24 23:47 ` Chautru, Nicolas
2020-03-26 2:45 ` [dpdk-dev] [PATCH v4 00/10] " Nicolas Chautru
2020-03-26 2:45 ` [dpdk-dev] [PATCH v4 01/10] bbdev: add capability flag for filler bits inclusion in HARQ Nicolas Chautru
2020-03-26 2:45 ` [dpdk-dev] [PATCH v4 02/10] bbdev: expose device HARQ buffer size at device level Nicolas Chautru
2020-03-26 2:45 ` [dpdk-dev] [PATCH v4 03/10] baseband/turbo_sw: fix the exposed LLR decimals assumption Nicolas Chautru
2020-03-26 2:45 ` [dpdk-dev] [PATCH v4 04/10] baseband/turbo_sw: support large size code block Nicolas Chautru
2020-03-26 2:45 ` [dpdk-dev] [PATCH v4 05/10] test-bbdev: rename FPGA LTE macros to be more explicit Nicolas Chautru
2020-03-26 2:45 ` [dpdk-dev] [PATCH v4 06/10] test-bbdev: support HARQ validation Nicolas Chautru
2020-03-26 2:45 ` [dpdk-dev] [PATCH v4 07/10] test-bbdev: support for performance tests Nicolas Chautru
2020-03-26 2:45 ` [dpdk-dev] [PATCH v4 08/10] test-bbdev: support for LDPC interrupt test Nicolas Chautru
2020-03-26 2:45 ` [dpdk-dev] [PATCH v4 09/10] test-bbdev: support for offload test for LDPC Nicolas Chautru
2020-03-26 2:45 ` [dpdk-dev] [PATCH v4 10/10] test-bbdev: vectors update Nicolas Chautru
2020-03-26 3:27 ` [dpdk-dev] [PATCH v5 00/10] bbdev new features Nicolas Chautru
2020-03-26 3:27 ` [dpdk-dev] [PATCH v5 01/10] bbdev: add capability flag for filler bits inclusion in HARQ Nicolas Chautru
2020-03-27 11:44 ` Dave Burley
2020-03-26 3:27 ` [dpdk-dev] [PATCH v5 02/10] bbdev: expose device HARQ buffer size at device level Nicolas Chautru
2020-03-27 11:45 ` Dave Burley
2020-03-26 3:27 ` [dpdk-dev] [PATCH v5 03/10] baseband/turbo_sw: fix the exposed LLR decimals assumption Nicolas Chautru
2020-03-27 11:45 ` Dave Burley
2020-03-26 3:27 ` [dpdk-dev] [PATCH v5 04/10] baseband/turbo_sw: support large size code block Nicolas Chautru
2020-03-27 11:45 ` Dave Burley
2020-03-26 3:27 ` [dpdk-dev] [PATCH v5 05/10] test-bbdev: rename FPGA LTE macros to be more explicit Nicolas Chautru
2020-03-27 11:45 ` Dave Burley
2020-03-26 3:27 ` [dpdk-dev] [PATCH v5 06/10] test-bbdev: support HARQ validation Nicolas Chautru
2020-03-27 11:47 ` Dave Burley
2020-03-26 3:27 ` [dpdk-dev] [PATCH v5 07/10] test-bbdev: support for performance tests Nicolas Chautru
2020-03-27 11:47 ` Dave Burley
2020-03-26 3:27 ` [dpdk-dev] [PATCH v5 08/10] test-bbdev: support for LDPC interrupt test Nicolas Chautru
2020-03-27 11:47 ` Dave Burley
2020-03-26 3:27 ` [dpdk-dev] [PATCH v5 09/10] test-bbdev: support for offload test for LDPC Nicolas Chautru
2020-03-27 11:47 ` Dave Burley
2020-03-26 3:27 ` [dpdk-dev] [PATCH v5 10/10] test-bbdev: vectors update Nicolas Chautru
2020-03-27 11:47 ` Dave Burley
2020-03-27 11:43 ` [dpdk-dev] [PATCH v5 00/10] bbdev new features Dave Burley
2020-04-05 16:41 ` Akhil Goyal
2020-03-03 18:08 ` [dpdk-dev] [PATCH v1 00/14] " Chautru, Nicolas
2020-03-04 8:23 ` Thomas Monjalon
2020-03-04 18:28 ` Chautru, Nicolas
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=1583348102-13253-5-git-send-email-nicolas.chautru@intel.com \
--to=nicolas.chautru@intel.com \
--cc=akhil.goyal@nxp.com \
--cc=dev@dpdk.org \
--cc=ferruh.yigit@intel.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).