DPDK patches and discussions
 help / color / mirror / Atom feed
From: Wang Xiao W <xiao.w.wang@intel.com>
To: jing.d.chen@intel.com
Cc: dev@dpdk.org
Subject: [dpdk-dev] [PATCH v2 04/16] fm10k/base: use bitshift for itr_scale
Date: Wed, 27 Jan 2016 11:50:35 +0800	[thread overview]
Message-ID: <1453866647-16215-5-git-send-email-xiao.w.wang@intel.com> (raw)
In-Reply-To: <1453866647-16215-1-git-send-email-xiao.w.wang@intel.com>

Upstream community wishes us to use bitshift instead of a divisor,
because this is faster, and prevents any need for a '0' check. In our
case, this even works out because default Gen3 will be 0.

Because of this, we are also able to remove the check for non-zero value
in the vf code path since that will already be the default Gen3 case.

Signed-off-by: Wang Xiao W <xiao.w.wang@intel.com>
---
 drivers/net/fm10k/base/fm10k_type.h | 6 +++---
 drivers/net/fm10k/base/fm10k_vf.c   | 4 ----
 2 files changed, 3 insertions(+), 7 deletions(-)

diff --git a/drivers/net/fm10k/base/fm10k_type.h b/drivers/net/fm10k/base/fm10k_type.h
index 62fa73f..44187b1 100644
--- a/drivers/net/fm10k/base/fm10k_type.h
+++ b/drivers/net/fm10k/base/fm10k_type.h
@@ -352,9 +352,9 @@ struct fm10k_hw;
 #define FM10K_TDLEN(_n)		((0x40 * (_n)) + 0x8002)
 #define FM10K_TDLEN_ITR_SCALE_SHIFT		9
 #define FM10K_TDLEN_ITR_SCALE_MASK		0x00000E00
-#define FM10K_TDLEN_ITR_SCALE_GEN1		4
-#define FM10K_TDLEN_ITR_SCALE_GEN2		2
-#define FM10K_TDLEN_ITR_SCALE_GEN3		1
+#define FM10K_TDLEN_ITR_SCALE_GEN1		2
+#define FM10K_TDLEN_ITR_SCALE_GEN2		1
+#define FM10K_TDLEN_ITR_SCALE_GEN3		0
 #define FM10K_TPH_TXCTRL(_n)	((0x40 * (_n)) + 0x8003)
 #define FM10K_TPH_TXCTRL_DESC_TPHEN		0x00000020
 #define FM10K_TPH_TXCTRL_DESC_RROEN		0x00000200
diff --git a/drivers/net/fm10k/base/fm10k_vf.c b/drivers/net/fm10k/base/fm10k_vf.c
index 7822ab6..39bc927 100644
--- a/drivers/net/fm10k/base/fm10k_vf.c
+++ b/drivers/net/fm10k/base/fm10k_vf.c
@@ -159,10 +159,6 @@ STATIC s32 fm10k_init_hw_vf(struct fm10k_hw *hw)
 			     FM10K_TDLEN_ITR_SCALE_MASK) >>
 			    FM10K_TDLEN_ITR_SCALE_SHIFT;
 
-	/* ensure a non-zero itr scale */
-	if (!hw->mac.itr_scale)
-		hw->mac.itr_scale = FM10K_TDLEN_ITR_SCALE_GEN3;
-
 	return FM10K_SUCCESS;
 }
 
-- 
1.9.3

  parent reply	other threads:[~2016-01-27  3:51 UTC|newest]

Thread overview: 69+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-01-21 10:35 [dpdk-dev] [PATCH 00/16] fm10k: update shared code from ND team Wang Xiao W
2016-01-21 10:35 ` [dpdk-dev] [PATCH 01/16] fm10k/base: cleanup namespace pollution and correct typecast Wang Xiao W
2016-01-21 10:35 ` [dpdk-dev] [PATCH 02/16] fm10k/base: use bitshift for itr_scale Wang Xiao W
2016-01-21 10:35 ` [dpdk-dev] [PATCH 03/16] fm10k/base: reset max_queues on init_hw_vf failure Wang Xiao W
2016-01-21 10:35 ` [dpdk-dev] [PATCH 04/16] fm10k/base: document ITR scale workaround in VF TDLEN register Wang Xiao W
2016-01-21 10:35 ` [dpdk-dev] [PATCH 05/16] fm10k/base: fix checkpatch warning Wang Xiao W
2016-01-21 10:35 ` [dpdk-dev] [PATCH 06/16] fm10k/base: use BIT macro instead of open-coded bit-shifting of 1 Wang Xiao W
2016-01-21 10:35 ` [dpdk-dev] [PATCH 07/16] fm10k/base: do not use CamelCase Wang Xiao W
2016-01-21 10:35 ` [dpdk-dev] [PATCH 08/16] fm10k/base: use memcpy for mac addr copy Wang Xiao W
2016-01-21 10:35 ` [dpdk-dev] [PATCH 09/16] fm10k/base: allow removal of is_slot_appropriate function Wang Xiao W
2016-01-21 10:35 ` [dpdk-dev] [PATCH 10/16] fm10k/base: consistently use VLAN ID when referencing vid variables Wang Xiao W
2016-01-21 10:35 ` [dpdk-dev] [PATCH 11/16] fm10k/base: fix comment per upstream review changes Wang Xiao W
2016-01-21 10:35 ` [dpdk-dev] [PATCH 12/16] fm10k/base: TLV structures must be 4byte aligned, not 1byte aligned Wang Xiao W
2016-01-21 10:35 ` [dpdk-dev] [PATCH 13/16] fm10k/base: move constants to the right of binary operators Wang Xiao W
2016-01-21 10:35 ` [dpdk-dev] [PATCH 14/16] fm10k/base: minor cleanups Wang Xiao W
2016-01-21 10:36 ` [dpdk-dev] [PATCH 15/16] fm10k: use default mailbox message handler for pf Wang Xiao W
2016-01-22 21:31   ` Bruce Richardson
2016-01-25  2:31     ` Wang, Xiao W
2016-01-26 20:16       ` Bruce Richardson
2016-01-27  1:57         ` Wang, Xiao W
2016-01-27  3:50   ` [dpdk-dev] [PATCH v2 00/16] fm10k: update shared code Wang Xiao W
2016-01-27  3:50     ` [dpdk-dev] [PATCH v2 01/16] fm10k: use default mailbox message handler for pf Wang Xiao W
2016-02-16 10:50       ` Bruce Richardson
2016-02-18 10:25         ` Wang, Xiao W
2016-02-19 11:06       ` [dpdk-dev] [PATCH v3 00/18] fm10k: update shared code Wang Xiao W
2016-02-19 11:06         ` [dpdk-dev] [PATCH v3 01/18] fm10k: use default mailbox message handler for PF Wang Xiao W
2016-02-19 11:06         ` [dpdk-dev] [PATCH v3 02/18] fm10k/base: correct typecast in fm10k_update_xc_addr_pf Wang Xiao W
2016-02-19 11:06         ` [dpdk-dev] [PATCH v3 03/18] fm10k/base: cleanup namespace pollution Wang Xiao W
2016-02-19 11:06         ` [dpdk-dev] [PATCH v3 04/18] fm10k/base: use bitshift for itr_scale Wang Xiao W
2016-02-19 11:06         ` [dpdk-dev] [PATCH v3 05/18] fm10k/base: reset max_queues on init_hw_vf failure Wang Xiao W
2016-02-19 11:06         ` [dpdk-dev] [PATCH v3 06/18] fm10k/base: document ITR scale workaround in VF TDLEN register Wang Xiao W
2016-02-19 11:06         ` [dpdk-dev] [PATCH v3 07/18] fm10k/base: cleanup lines over 80 characters Wang Xiao W
2016-02-19 11:06         ` [dpdk-dev] [PATCH v3 08/18] fm10k/base: cleanup useless else Wang Xiao W
2016-02-19 11:06         ` [dpdk-dev] [PATCH v3 09/18] fm10k/base: use BIT macro instead of open-coded bit-shifting of 1 Wang Xiao W
2016-02-19 11:06         ` [dpdk-dev] [PATCH v3 10/18] fm10k/base: do not use CamelCase Wang Xiao W
2016-02-19 11:06         ` [dpdk-dev] [PATCH v3 11/18] fm10k/base: use memcpy for mac addr copy Wang Xiao W
2016-02-19 11:06         ` [dpdk-dev] [PATCH v3 12/18] fm10k/base: allow removal of is_slot_appropriate function Wang Xiao W
2016-02-19 11:07         ` [dpdk-dev] [PATCH v3 13/18] fm10k/base: consistently use VLAN ID when referencing vid variables Wang Xiao W
2016-02-19 11:07         ` [dpdk-dev] [PATCH v3 14/18] fm10k/base: imporve comment per upstream review changes Wang Xiao W
2016-02-19 11:07         ` [dpdk-dev] [PATCH v3 15/18] fm10k/base: fix TLV structures alignment Wang Xiao W
2016-02-19 11:07         ` [dpdk-dev] [PATCH v3 16/18] fm10k/base: move constants to the right of binary operators Wang Xiao W
2016-02-19 11:07         ` [dpdk-dev] [PATCH v3 17/18] fm10k/base: minor cleanups Wang Xiao W
2016-02-19 11:07         ` [dpdk-dev] [PATCH v3 18/18] fm10k/base: remove unused struct element Wang Xiao W
2016-02-29  2:30         ` [dpdk-dev] [PATCH v3 00/18] fm10k: update shared code Ding, HengX
2016-03-08 16:27           ` Bruce Richardson
2016-03-08 13:24         ` Bruce Richardson
2016-03-08 14:15           ` Wang, Xiao W
2016-03-08 14:25             ` Chen, Jing D
2016-03-08 14:32               ` Wang, Xiao W
2016-03-08 14:53               ` Bruce Richardson
2016-01-27  3:50     ` [dpdk-dev] [PATCH v2 02/16] fm10k/base: add macro definitions that are needed Wang Xiao W
2016-02-16 10:52       ` Bruce Richardson
2016-01-27  3:50     ` [dpdk-dev] [PATCH v2 03/16] fm10k/base: cleanup namespace pollution and correct typecast Wang Xiao W
2016-02-16 10:55       ` Bruce Richardson
2016-01-27  3:50     ` Wang Xiao W [this message]
2016-01-27  3:50     ` [dpdk-dev] [PATCH v2 05/16] fm10k/base: reset max_queues on init_hw_vf failure Wang Xiao W
2016-01-27  3:50     ` [dpdk-dev] [PATCH v2 06/16] fm10k/base: document ITR scale workaround in VF TDLEN register Wang Xiao W
2016-01-27  3:50     ` [dpdk-dev] [PATCH v2 07/16] fm10k/base: fix checkpatch warning Wang Xiao W
2016-01-27  3:50     ` [dpdk-dev] [PATCH v2 08/16] fm10k/base: use BIT macro instead of open-coded bit-shifting of 1 Wang Xiao W
2016-01-27  3:50     ` [dpdk-dev] [PATCH v2 09/16] fm10k/base: do not use CamelCase Wang Xiao W
2016-01-27  3:50     ` [dpdk-dev] [PATCH v2 10/16] fm10k/base: use memcpy for mac addr copy Wang Xiao W
2016-01-27  3:50     ` [dpdk-dev] [PATCH v2 11/16] fm10k/base: allow removal of is_slot_appropriate function Wang Xiao W
2016-01-27  3:50     ` [dpdk-dev] [PATCH v2 12/16] fm10k/base: consistently use VLAN ID when referencing vid variables Wang Xiao W
2016-01-27  3:50     ` [dpdk-dev] [PATCH v2 13/16] fm10k/base: fix comment per upstream review changes Wang Xiao W
2016-01-27  3:50     ` [dpdk-dev] [PATCH v2 14/16] fm10k/base: TLV structures must be 4byte aligned, not 1byte aligned Wang Xiao W
2016-01-27  3:50     ` [dpdk-dev] [PATCH v2 15/16] fm10k/base: move constants to the right of binary operators Wang Xiao W
2016-01-27  3:50     ` [dpdk-dev] [PATCH v2 16/16] fm10k/base: minor cleanups Wang Xiao W
2016-02-16  8:11     ` [dpdk-dev] [PATCH v2 00/16] fm10k: update shared code Chen, Jing D
2016-01-21 10:36 ` [dpdk-dev] [PATCH 16/16] fm10k/base: add macro definitions that are needed Wang Xiao W

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=1453866647-16215-5-git-send-email-xiao.w.wang@intel.com \
    --to=xiao.w.wang@intel.com \
    --cc=dev@dpdk.org \
    --cc=jing.d.chen@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).