DPDK patches and discussions
 help / color / mirror / Atom feed
From: Bruce Richardson <bruce.richardson@intel.com>
To: helin.zhang@intel.com
Cc: dev@dpdk.org, ferruh.yigit@intel.com,
	Bruce Richardson <bruce.richardson@intel.com>
Subject: [dpdk-dev] [PATCH v2 2/2] net/i40e: remove option to disable offload flags
Date: Thu,  6 Apr 2017 12:32:17 +0100	[thread overview]
Message-ID: <20170406113217.28236-3-bruce.richardson@intel.com> (raw)
In-Reply-To: <20170406113217.28236-1-bruce.richardson@intel.com>

Having packets received without any offload flags given in the mbuf is not
very useful, and performance tests with testpmd indicates little to no
benefit is got with the current code by turning off the flags. This makes
the build-time option pointless, so we can remove it.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
---
V2: include changes to alitvec as well as neon and sse files
---
 config/common_base                       |  1 -
 doc/guides/nics/i40e.rst                 |  5 -----
 drivers/net/i40e/i40e_rxtx_vec_altivec.c | 11 -----------
 drivers/net/i40e/i40e_rxtx_vec_common.h  |  8 --------
 drivers/net/i40e/i40e_rxtx_vec_neon.c    | 11 -----------
 drivers/net/i40e/i40e_rxtx_vec_sse.c     | 11 -----------
 6 files changed, 47 deletions(-)

diff --git a/config/common_base b/config/common_base
index 7d7290c..434d406 100644
--- a/config/common_base
+++ b/config/common_base
@@ -184,7 +184,6 @@ CONFIG_RTE_LIBRTE_I40E_DEBUG_TX=n
 CONFIG_RTE_LIBRTE_I40E_DEBUG_TX_FREE=n
 CONFIG_RTE_LIBRTE_I40E_RX_ALLOW_BULK_ALLOC=y
 CONFIG_RTE_LIBRTE_I40E_INC_VECTOR=y
-CONFIG_RTE_LIBRTE_I40E_RX_OLFLAGS_ENABLE=y
 CONFIG_RTE_LIBRTE_I40E_16BYTE_RX_DESC=n
 CONFIG_RTE_LIBRTE_I40E_QUEUE_NUM_PER_PF=64
 CONFIG_RTE_LIBRTE_I40E_QUEUE_NUM_PER_VF=4
diff --git a/doc/guides/nics/i40e.rst b/doc/guides/nics/i40e.rst
index 6ae1d46..d5efa3f 100644
--- a/doc/guides/nics/i40e.rst
+++ b/doc/guides/nics/i40e.rst
@@ -105,11 +105,6 @@ Please note that enabling debugging options may affect system performance.
   Toggle the use of Vector PMD instead of normal RX/TX path.
   To enable vPMD for RX, bulk allocation for Rx must be allowed.
 
-- ``CONFIG_RTE_LIBRTE_I40E_RX_OLFLAGS_ENABLE`` (default ``y``)
-
-  Toggle to enable RX ``olflags``.
-  This is only meaningful when Vector PMD is used.
-
 - ``CONFIG_RTE_LIBRTE_I40E_16BYTE_RX_DESC`` (default ``n``)
 
   Toggle to use a 16-byte RX descriptor, by default the RX descriptor is 32 byte.
diff --git a/drivers/net/i40e/i40e_rxtx_vec_altivec.c b/drivers/net/i40e/i40e_rxtx_vec_altivec.c
index 2f6f70a..9831438 100644
--- a/drivers/net/i40e/i40e_rxtx_vec_altivec.c
+++ b/drivers/net/i40e/i40e_rxtx_vec_altivec.c
@@ -130,14 +130,6 @@ i40e_rxq_rearm(struct i40e_rx_queue *rxq)
 	I40E_PCI_REG_WRITE(rxq->qrx_tail, rx_id);
 }
 
-/* Handling the offload flags (olflags) field takes computation
- * time when receiving packets. Therefore we provide a flag to disable
- * the processing of the olflags field when they are not needed. This
- * gives improved performance, at the cost of losing the offload info
- * in the received packet
- */
-#ifdef RTE_LIBRTE_I40E_RX_OLFLAGS_ENABLE
-
 static inline void
 desc_to_olflags_v(vector unsigned long descs[4], struct rte_mbuf **rx_pkts)
 {
@@ -202,9 +194,6 @@ desc_to_olflags_v(vector unsigned long descs[4], struct rte_mbuf **rx_pkts)
 	rx_pkts[2]->ol_flags = (uint64_t)vlan0[0];
 	rx_pkts[3]->ol_flags = (uint64_t)vlan0[1];
 }
-#else
-#define desc_to_olflags_v(desc, rx_pkts) do {} while (0)
-#endif
 
 #define PKTLEN_SHIFT     10
 
diff --git a/drivers/net/i40e/i40e_rxtx_vec_common.h b/drivers/net/i40e/i40e_rxtx_vec_common.h
index 76031fe..952fd4b 100644
--- a/drivers/net/i40e/i40e_rxtx_vec_common.h
+++ b/drivers/net/i40e/i40e_rxtx_vec_common.h
@@ -224,14 +224,6 @@ i40e_rx_vec_dev_conf_condition_check_default(struct rte_eth_dev *dev)
 	struct rte_eth_rxmode *rxmode = &dev->data->dev_conf.rxmode;
 	struct rte_fdir_conf *fconf = &dev->data->dev_conf.fdir_conf;
 
-#ifndef RTE_LIBRTE_I40E_RX_OLFLAGS_ENABLE
-	/* whithout rx ol_flags, no VP flag report */
-	if (rxmode->hw_vlan_strip != 0 ||
-	    rxmode->hw_vlan_extend != 0 ||
-	    rxmode->hw_ip_checksum != 0)
-		return -1;
-#endif
-
 	/* no fdir support */
 	if (fconf->mode != RTE_FDIR_MODE_NONE)
 		return -1;
diff --git a/drivers/net/i40e/i40e_rxtx_vec_neon.c b/drivers/net/i40e/i40e_rxtx_vec_neon.c
index bd7239b..902fb1f 100644
--- a/drivers/net/i40e/i40e_rxtx_vec_neon.c
+++ b/drivers/net/i40e/i40e_rxtx_vec_neon.c
@@ -116,14 +116,6 @@ i40e_rxq_rearm(struct i40e_rx_queue *rxq)
 	I40E_PCI_REG_WRITE(rxq->qrx_tail, rx_id);
 }
 
-/* Handling the offload flags (olflags) field takes computation
- * time when receiving packets. Therefore we provide a flag to disable
- * the processing of the olflags field when they are not needed. This
- * gives improved performance, at the cost of losing the offload info
- * in the received packet
- */
-#ifdef RTE_LIBRTE_I40E_RX_OLFLAGS_ENABLE
-
 static inline void
 desc_to_olflags_v(uint64x2_t descs[4], struct rte_mbuf **rx_pkts)
 {
@@ -187,9 +179,6 @@ desc_to_olflags_v(uint64x2_t descs[4], struct rte_mbuf **rx_pkts)
 	rx_pkts[2]->ol_flags = vgetq_lane_u32(vlan0, 2);
 	rx_pkts[3]->ol_flags = vgetq_lane_u32(vlan0, 3);
 }
-#else
-#define desc_to_olflags_v(descs, rx_pkts) do {} while (0)
-#endif
 
 #define PKTLEN_SHIFT     10
 
diff --git a/drivers/net/i40e/i40e_rxtx_vec_sse.c b/drivers/net/i40e/i40e_rxtx_vec_sse.c
index c43d1c3..f91e32d 100644
--- a/drivers/net/i40e/i40e_rxtx_vec_sse.c
+++ b/drivers/net/i40e/i40e_rxtx_vec_sse.c
@@ -123,14 +123,6 @@ desc_to_olflags_v(struct i40e_rx_queue *rxq, __m128i descs[4] __rte_unused,
 	const __m128i mbuf_init = _mm_set_epi64x(0, rxq->mbuf_initializer);
 	__m128i rearm0, rearm1, rearm2, rearm3;
 
-/* Handling the offload flags (olflags) field takes computation
- * time when receiving packets. Therefore we provide a flag to disable
- * the processing of the olflags field when they are not needed. This
- * gives improved performance, at the cost of losing the offload info
- * in the received packet
- */
-#ifdef RTE_LIBRTE_I40E_RX_OLFLAGS_ENABLE
-
 	__m128i vlan0, vlan1, rss, l3_l4e;
 
 	/* mask everything except RSS, flow director and VLAN flags
@@ -211,9 +203,6 @@ desc_to_olflags_v(struct i40e_rx_queue *rxq, __m128i descs[4] __rte_unused,
 	rearm1 = _mm_blend_epi16(mbuf_init, _mm_slli_si128(vlan0, 4), 0x10);
 	rearm2 = _mm_blend_epi16(mbuf_init, vlan0, 0x10);
 	rearm3 = _mm_blend_epi16(mbuf_init, _mm_srli_si128(vlan0, 4), 0x10);
-#else
-	rearm0 = rearm1 = rearm2 = rearm3 = mbuf_init;
-#endif
 	_mm_store_si128((__m128i *)&rx_pkts[0]->rearm_data, rearm0);
 	_mm_store_si128((__m128i *)&rx_pkts[1]->rearm_data, rearm1);
 	_mm_store_si128((__m128i *)&rx_pkts[2]->rearm_data, rearm2);
-- 
2.9.3

  parent reply	other threads:[~2017-04-06 11:32 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-04-03 14:39 [dpdk-dev] [PATCH 0/2] reduce writes in i40e driver Bruce Richardson
2017-04-03 14:39 ` [dpdk-dev] [PATCH 1/2] net/i40e: eliminate mbuf write on rearm Bruce Richardson
2017-04-03 14:39 ` [dpdk-dev] [PATCH 2/2] net/i40e: remove option to disable offload flags Bruce Richardson
2017-04-03 14:45   ` Ferruh Yigit
2017-04-03 14:51     ` Bruce Richardson
2017-04-06  5:53 ` [dpdk-dev] [PATCH 0/2] reduce writes in i40e driver Jianbo Liu
2017-04-06  5:56 ` Jianbo Liu
2017-04-06 11:32 ` [dpdk-dev] [PATCH v2 " Bruce Richardson
2017-04-06 11:32   ` [dpdk-dev] [PATCH v2 1/2] net/i40e: eliminate mbuf write on rearm Bruce Richardson
2017-04-06 16:24     ` Zhang, Qi Z
2017-04-07  8:26       ` Bruce Richardson
2017-04-06 11:32   ` Bruce Richardson [this message]
2017-04-10 13:13   ` [dpdk-dev] [PATCH v2 0/2] reduce writes in i40e driver Ferruh Yigit

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=20170406113217.28236-3-bruce.richardson@intel.com \
    --to=bruce.richardson@intel.com \
    --cc=dev@dpdk.org \
    --cc=ferruh.yigit@intel.com \
    --cc=helin.zhang@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).