| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Date: Wed, 16 Nov 2011 17:42:57 -0800
From: David Decotigny <david.decotigny@...gle.com>
To: Jeff Kirsher <jeffrey.t.kirsher@...el.com>,
Jesse Brandeburg <jesse.brandeburg@...el.com>,
Bruce Allan <bruce.w.allan@...el.com>,
Carolyn Wyborny <carolyn.wyborny@...el.com>,
Don Skidmore <donald.c.skidmore@...el.com>,
Greg Rose <gregory.v.rose@...el.com>,
Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@...el.com>,
Alex Duyck <alexander.h.duyck@...el.com>,
John Ronciak <john.ronciak@...el.com>,
e1000-devel@...ts.sourceforge.net, netdev@...r.kernel.org,
linux-kernel@...r.kernel.org
Cc: Paul Gortmaker <paul.gortmaker@...driver.com>,
David Decotigny <david.decotigny@...gle.com>,
Ying Cai <ycai@...gle.com>
Subject: Re: [PATCH FIX net v1] net-e1000e: Fix default interrupt throttle
rate not set in NIC HW
Sorry, I realized that the same fix could be applied to e1000. Sending
an updated v2 patch shortly...
--
David Decotigny
On Wed, Nov 16, 2011 at 5:18 PM, David Decotigny
<david.decotigny@...gle.com> wrote:
> From: Ying Cai <ycai@...gle.com>
>
> This change ensures that the itr/itr_setting adjustment logic is used,
> even for the default/compiled-in value.
>
> Context:
> When we changed the default InterruptThrottleRate value from default
> (3 = dynamic mode) to 8000 for example, only adapter->itr_setting
> (which controls interrupt coalescing mode) was set to 8000, but
> adapter->itr (which controls the value set in NIC register) was not
> updated accordingly. So from ethtool, it seemed the interrupt
> throttling is enabled at 8000 intr/s, but the NIC actually was
> running in dynamic mode which has lower CPU efficiency especially
> when throughput is not high.
>
>
>
> Signed-off-by: David Decotigny <david.decotigny@...gle.com>
> ---
> drivers/net/ethernet/intel/e1000e/param.c | 98 +++++++++++++++-------------
> 1 files changed, 52 insertions(+), 46 deletions(-)
>
> diff --git a/drivers/net/ethernet/intel/e1000e/param.c b/drivers/net/ethernet/intel/e1000e/param.c
> index 20e93b0..41937e5 100644
> --- a/drivers/net/ethernet/intel/e1000e/param.c
> +++ b/drivers/net/ethernet/intel/e1000e/param.c
> @@ -106,7 +106,7 @@ E1000_PARAM(RxAbsIntDelay, "Receive Absolute Interrupt Delay");
> /*
> * Interrupt Throttle Rate (interrupts/sec)
> *
> - * Valid Range: 100-100000 (0=off, 1=dynamic, 3=dynamic conservative)
> + * Valid Range: 100-100000 or one of: 0=off, 1=dynamic, 3=dynamic conservative
> */
> E1000_PARAM(InterruptThrottleRate, "Interrupt Throttling Rate");
> #define DEFAULT_ITR 3
> @@ -335,53 +335,59 @@ void __devinit e1000e_check_options(struct e1000_adapter *adapter)
>
> if (num_InterruptThrottleRate > bd) {
> adapter->itr = InterruptThrottleRate[bd];
> - switch (adapter->itr) {
> - case 0:
> - e_info("%s turned off\n", opt.name);
> - break;
> - case 1:
> - e_info("%s set to dynamic mode\n", opt.name);
> - adapter->itr_setting = adapter->itr;
> - adapter->itr = 20000;
> - break;
> - case 3:
> - e_info("%s set to dynamic conservative mode\n",
> - opt.name);
> - adapter->itr_setting = adapter->itr;
> - adapter->itr = 20000;
> - break;
> - case 4:
> - e_info("%s set to simplified (2000-8000 ints) "
> - "mode\n", opt.name);
> - adapter->itr_setting = 4;
> - break;
> - default:
> - /*
> - * Save the setting, because the dynamic bits
> - * change itr.
> - */
> - if (e1000_validate_option(&adapter->itr, &opt,
> - adapter) &&
> - (adapter->itr == 3)) {
> - /*
> - * In case of invalid user value,
> - * default to conservative mode.
> - */
> - adapter->itr_setting = adapter->itr;
> - adapter->itr = 20000;
> - } else {
> - /*
> - * Clear the lower two bits because
> - * they are used as control.
> - */
> - adapter->itr_setting =
> - adapter->itr & ~3;
> - }
> - break;
> - }
> +
> + /* Make sure a message is printed for
> + * non-special values. And in case of an
> + * invalid option, display warning, use
> + * default and go through itr/itr_setting
> + * adjustment logic below */
> + if ((adapter->itr < 0 || adapter->itr > 4)
> + && e1000_validate_option(&adapter->itr, &opt,
> + adapter))
> + adapter->itr = opt.def;
> } else {
> - adapter->itr_setting = opt.def;
> + /* if no option specified, use default value
> + and go through the logic below to adjust
> + itr/itr_setting */
> + adapter->itr = opt.def;
> +
> + /* Make sure a message is printed for
> + * non-special default values */
> + if (adapter->itr < 0 || adapter->itr > 4)
> + e_info("%s set to default %d\n",
> + opt.name, adapter->itr);
> + }
> +
> +
> +
> + adapter->itr_setting = adapter->itr;
> + switch (adapter->itr) {
> + case 0:
> + e_info("%s turned off\n", opt.name);
> + break;
> + case 1:
> + e_info("%s set to dynamic mode\n", opt.name);
> + adapter->itr = 20000;
> + break;
> + case 3:
> + e_info("%s set to dynamic conservative mode\n",
> + opt.name);
> adapter->itr = 20000;
> + break;
> + case 4:
> + e_info("%s set to simplified (2000-8000 ints) "
> + "mode\n", opt.name);
> + break;
> + default:
> + /*
> + * Save the setting, because the dynamic bits
> + * change itr.
> + *
> + * Clear the lower two bits because
> + * they are used as control.
> + */
> + adapter->itr_setting &= ~3;
> + break;
> }
> }
> { /* Interrupt Mode */
> --
> 1.7.3.1
>
>
Powered by blists - more mailing lists