All of lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH] e1000e: changed some expensive calls of udelay to usleep_range
@ 2017-09-05 20:27 ` Kardonik Michael
  0 siblings, 0 replies; 8+ messages in thread
From: Kardonik Michael @ 2017-09-05 20:27 UTC (permalink / raw)
  To: leoyang.li, michael.kardonik, Jeff Kirsher, Jesse Brandeburg,
	Shannon Nelson, Carolyn Wyborny, Don Skidmore, Matthew Vick,
	John Ronciak, Mitch Williams, intel-wired-lan, netdev,
	linux-kernel
  Cc: Matthew Tan

Calls to udelay are not preemtable by userspace so userspace
applications experience a large (~200us) latency when running on core0. 
Instead usleep_range can be used to be more friendly to userspace
since it is preemtable. This is due to udelay using busy-wait loops
while usleep_rang uses hrtimers instead. It is recommended to use
udelay when the delay is <10us since at that precision overhead of
usleep_range hrtimer setup causes issues. However, the replaced calls
are for 50us and 100us so this should not be not an issue.
There is no open bug that this patch is fixing, but we see a good
boost in zero loss performance of specific user space application 
(dpdk l3fwd) when this patch is applied: we get from 32% of 10Gb line 
to 49%.

Signed-off-by: Matthew Tan <matthew.tan_1@nxp.com>
Signed-off-by: Michael Kardonik <michael.kardonik@nxp.com>

---
 drivers/net/ethernet/intel/e1000e/phy.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/intel/e1000e/phy.c b/drivers/net/ethernet/intel/e1000e/phy.c
index de13aea..e318fdc 100644
--- a/drivers/net/ethernet/intel/e1000e/phy.c
+++ b/drivers/net/ethernet/intel/e1000e/phy.c
@@ -158,7 +158,7 @@ s32 e1000e_read_phy_reg_mdic(struct e1000_hw *hw, u32 offset, u16 *data)
 	 * the lower time out
 	 */
 	for (i = 0; i < (E1000_GEN_POLL_TIMEOUT * 3); i++) {
-		udelay(50);
+		usleep_range(50, 60);
 		mdic = er32(MDIC);
 		if (mdic & E1000_MDIC_READY)
 			break;
@@ -183,7 +183,7 @@ s32 e1000e_read_phy_reg_mdic(struct e1000_hw *hw, u32 offset, u16 *data)
 	 * reading duplicate data in the next MDIC transaction.
 	 */
 	if (hw->mac.type == e1000_pch2lan)
-		udelay(100);
+		usleep_range(100, 110);
 
 	return 0;
 }
@@ -222,7 +222,7 @@ s32 e1000e_write_phy_reg_mdic(struct e1000_hw *hw, u32 offset, u16 data)
 	 * the lower time out
 	 */
 	for (i = 0; i < (E1000_GEN_POLL_TIMEOUT * 3); i++) {
-		udelay(50);
+		usleep_range(50, 60);
 		mdic = er32(MDIC);
 		if (mdic & E1000_MDIC_READY)
 			break;
@@ -246,7 +246,7 @@ s32 e1000e_write_phy_reg_mdic(struct e1000_hw *hw, u32 offset, u16 data)
 	 * reading duplicate data in the next MDIC transaction.
 	 */
 	if (hw->mac.type == e1000_pch2lan)
-		udelay(100);
+		usleep_range(100, 110);
 
 	return 0;
 }
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 8+ messages in thread
* [PATCH] e1000e: changed some expensive calls of udelay to usleep_range
@ 2017-08-23 15:59 Matthew Tan
  2017-09-04 16:25 ` Pavel Machek
  0 siblings, 1 reply; 8+ messages in thread
From: Matthew Tan @ 2017-08-23 15:59 UTC (permalink / raw)
  To: jeffrey.t.kirsher
  Cc: michael.kardonik, carolyn.wyborny, donald.c.skidmore,
	bruce.w.allan, john.ronciak, mitch.a.williams, intel-wired-lan,
	netdev, linux-kernel, Matthew Tan

    Calls to udelay are not preemtable by userspace so userspace
    applications experience a large (~200us) latency when running on core
    0. Instead usleep_range can be used to be more friendly to userspace
    since it is preemtable. This is due to udelay using busy-wait loops
    while usleep_rang uses hrtimers instead. It is recommended to use
    udelay when the delay is <10us since at that precision overhead of
    usleep_range hrtimer setup causes issues. However, the replaced calls
    are for 50us and 100us so this should not be not an issue.

Signed-off-by: Matthew Tan <matthew.tan_1@nxp.com>
---
 drivers/net/ethernet/intel/e1000e/phy.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/intel/e1000e/phy.c b/drivers/net/ethernet/intel/e1000e/phy.c
index de13aea..e318fdc 100644
--- a/drivers/net/ethernet/intel/e1000e/phy.c
+++ b/drivers/net/ethernet/intel/e1000e/phy.c
@@ -158,7 +158,7 @@ s32 e1000e_read_phy_reg_mdic(struct e1000_hw *hw, u32 offset, u16 *data)
 	 * the lower time out
 	 */
 	for (i = 0; i < (E1000_GEN_POLL_TIMEOUT * 3); i++) {
-		udelay(50);
+		usleep_range(40, 60);
 		mdic = er32(MDIC);
 		if (mdic & E1000_MDIC_READY)
 			break;
@@ -183,7 +183,7 @@ s32 e1000e_read_phy_reg_mdic(struct e1000_hw *hw, u32 offset, u16 *data)
 	 * reading duplicate data in the next MDIC transaction.
 	 */
 	if (hw->mac.type == e1000_pch2lan)
-		udelay(100);
+		usleep_range(90, 100);
 
 	return 0;
 }
@@ -222,7 +222,7 @@ s32 e1000e_write_phy_reg_mdic(struct e1000_hw *hw, u32 offset, u16 data)
 	 * the lower time out
 	 */
 	for (i = 0; i < (E1000_GEN_POLL_TIMEOUT * 3); i++) {
-		udelay(50);
+		usleep_range(40, 60);
 		mdic = er32(MDIC);
 		if (mdic & E1000_MDIC_READY)
 			break;
@@ -246,7 +246,7 @@ s32 e1000e_write_phy_reg_mdic(struct e1000_hw *hw, u32 offset, u16 data)
 	 * reading duplicate data in the next MDIC transaction.
 	 */
 	if (hw->mac.type == e1000_pch2lan)
-		udelay(100);
+		usleep_range(90, 110);
 
 	return 0;
 }
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 8+ messages in thread
end of thread, other threads:[~2017-09-12  1:34 UTC | newest]

Thread overview: 8+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2017-09-05 20:27 [PATCH] e1000e: changed some expensive calls of udelay to usleep_range Kardonik Michael
2017-09-05 20:27 ` [Intel-wired-lan] " Kardonik Michael
2017-09-10  6:56 ` Neftin, Sasha
2017-09-12  1:34 ` Brown, Aaron F
2017-09-12  1:34   ` [Intel-wired-lan] " Brown, Aaron F
2017-09-12  1:34   ` Brown, Aaron F
  -- strict thread matches above, loose matches on Subject: below --
2017-08-23 15:59 Matthew Tan
2017-09-04 16:25 ` Pavel Machek

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.