eHEA driver issues from net-2.6.24
Subsystems (2)
In testing the new NAPI improvements on ehea, I get the following:
kernel BUG at include/linux/netdevice.h:318!
cpu 0x1: Vector: 700 (Program Check) at [c00000000f613ac0]
pc: d000000000091054: .ehea_poll+0x1e8/0x1334 [ehea]
lr: c0000000003fe394: .net_rx_action+0x1b8/0x254
sp: c00000000f613d40
msr: 8000000000029032
current = 0xc0000005aff9e8b0
paca = 0xc0000000006e1300
pid = 0, comm = swapper
kernel BUG at include/linux/netdevice.h:318!
enter ? for help
[c00000000f613e40] c0000000003fe394 .net_rx_action+0x1b8/0x254 [c00000000f613ef0] c000000000057b70 .__do_softirq+0xa8/0x164 [c00000000f613f90] c000000000024438 .call_do_softirq+0x14/0x24 [c000000b8ffbf9f0] c00000000000bd30 .do_softirq+0x68/0xac [c000000b8ffbfa80] c000000000057cc4 .irq_exit+0x54/0x6c [c000000b8ffbfb00] c00000000000c358 .do_IRQ+0x170/0x1ac [c000000b8ffbfb90] c000000000004780 hardware_interrupt_entry+0x18/0x98 --- Exception: 501 (Hardware Interrupt) at c000000000010bdc .cpu_idle+0x114/0x1e0 [c000000b8ffbfe80] c000000000010bd0 .cpu_idle+0x108/0x1e0 (unreliable) [c000000b8ffbff00] c000000000026db0 .start_secondary+0x160/0x184 [c000000b8ffbff90] c000000000008364 .start_secondary_prolog+0xc/0x10
I'm a little confused if the port_napi_enable() is being called when the device is initialized, but then again, this is all new to me (should it be called in ehea_open?). I see it called on some reset routines, but not on the first initialization.
Also, on this code, in ehea_sense_port_attr()
/* Number of default QPs */
if (use_mcs)
port->num_def_qps = cb0->num_default_qps;
else
port->num_def_qps = 1;When using napi, since we have multi-queue napi support now, wouldn't we want to use all the default qps instead of 1?
Thanks,
-Andrew
David Miller
maintainerFrom: Andrew Theurer <[email protected]>
Date: Wed, 22 Aug 2007 16:55:03 -0500
Thanks for finally getting to test this, I thought nobody would test this until it got merged into 2.6.24 :-/
Quoted 15 lines
kernel BUG at
include/linux/netdevice.h:318!
enter ? for help
[c00000000f613e40]c0000000003fe394.net_rx_action+0x1b8/0x254 [c00000000f613ef0]c000000000057b70.__do_softirq+0xa8/0x164 [c00000000f613f90]c000000000024438.call_do_softirq+0x14/0x24 [c000000b8ffbf9f0]c00000000000bd30.do_softirq+0x68/0xac [c000000b8ffbfa80]c000000000057cc4.irq_exit+0x54/0x6c [c000000b8ffbfb00]c00000000000c358.do_IRQ+0x170/0x1ac [c000000b8ffbfb90]c000000000004780hardware_interrupt_entry+0x18/0x98--- Exception: 501 (Hardware Interrupt) atc000000000010bdc.cpu_idle+0x114/0x1e0 [c000000b8ffbfe80] c000000000010bd0 .cpu_idle+0x108/0x1e0 (unreliable) [c000000b8ffbff00] c000000000026db0 .start_secondary+0x160/0x184 [c000000b8ffbff90] c000000000008364 .start_secondary_prolog+0xc/0x10
Quoted 4 lines
I'm a little confused if the
port_napi_enable()is being called when the device is initialized, but then again, this is all new to me (should it be called inehea_open?). I see it called on some reset routines, but not on the first initialization.
This is similar to the problem that Arnaldo hit a few minutes ago in the VIA Rhine driver.
You can't only make a napi_enable() call when there has been a previous napi_disable().
One way to fix this would be to forcefully napi_disable() on all the per-port NAPI structs at the beginning of ehea_open(), which should set things up to satisfy the pre-condition of the napi_enable() calls.
You'll need to audit the entire driver to make sure this invariant is held properly.
Quoted 11 lines
Also, on this code, in
ehea_sense_port_attr()/* Number of default QPs */
if (use_mcs) port->num_def_qps = cb0->num_default_qps; else port->num_def_qps = 1;When using napi, since we have multi-queue napi support now, wouldn't we want to use all the default qps instead of 1?
I don't know how this hardware works, you tell me :-)
Quoted 28 lines · David Miller
From: Andrew Theurer <[email protected]>
Date: Wed, 22 Aug 2007 16:55:03 -0500Thanks for finally getting to test this, I thought nobody would test this until it got merged into 2.6.24 :-/
Quoted 10 lines
kernel BUG at
include/linux/netdevice.h:318!
enter ? for help
[c00000000f613e40]c0000000003fe394.net_rx_action+0x1b8/0x254 [c00000000f613ef0]c000000000057b70.__do_softirq+0xa8/0x164 [c00000000f613f90]c000000000024438.call_do_softirq+0x14/0x24 [c000000b8ffbf9f0]c00000000000bd30.do_softirq+0x68/0xac [c000000b8ffbfa80]c000000000057cc4.irq_exit+0x54/0x6c [c000000b8ffbfb00]c00000000000c358.do_IRQ+0x170/0x1ac [c000000b8ffbfb90]c000000000004780hardware_interrupt_entry+0x18/0x98--- Exception: 501 (Hardware Interrupt) atc000000000010bdcQuoted 7 lines
.cpu_idle+0x114/0x1e0 [c000000b8ffbfe80] c000000000010bd0 .cpu_idle+0x108/0x1e0 (unreliable) [c000000b8ffbff00] c000000000026db0 .start_secondary+0x160/0x184 [c000000b8ffbff90] c000000000008364 .start_secondary_prolog+0xc/0x10I'm a little confused if the
port_napi_enable()is being called when the device is initialized, but then again, this is all new to me (should it be called inehea_open?). I see it called on some reset routines, but not on the first initialization.This is similar to the problem that Arnaldo hit a few minutes ago in the VIA Rhine driver.
Quoted 8 lines
You can't only make a
napi_enable()call when there has been a previousnapi_disable().One way to fix this would be to forcefully
napi_disable()on all the per-port NAPI structs at the beginning ofehea_open(), which should set things up to satisfy the pre-condition of thenapi_enable()calls.
OK, Ill try this.
Quoted 13 lines
You'll need to audit the entire driver to make sure this invariant is held properly.
Quoted 9 lines
Also, on this code, in
ehea_sense_port_attr()/* Number of default QPs */
if (use_mcs) port->num_def_qps = cb0->num_default_qps; else port->num_def_qps = 1;
Quoted 5 lines
Quoted 2 lines
When using napi, since we have multi-queue napi support now, wouldn't we want to use all the default qps instead of 1?
I don't know how this hardware works, you tell me :-)
Heh, I don't know it well, either. Maybe Jan Bernd can chime in.
Thanks for your help,
-Andrew
Quoted 7 lines · Andrew Theurer
Quoted 6 lines · David Miller
From: Andrew Theurer <[email protected]>
Date: Wed, 22 Aug 2007 16:55:03 -0500Thanks for finally getting to test this, I thought nobody would test this until it got merged into 2.6.24 :-/
Yes, sorry for the delay.
Quoted 28 lines
Quoted 28 lines
Quoted 16 lines
kernel BUG at
include/linux/netdevice.h:318!
enter ? for help
[c00000000f613e40]c0000000003fe394.net_rx_action+0x1b8/0x254 [c00000000f613ef0]c000000000057b70.__do_softirq+0xa8/0x164 [c00000000f613f90]c000000000024438.call_do_softirq+0x14/0x24 [c000000b8ffbf9f0]c00000000000bd30.do_softirq+0x68/0xac [c000000b8ffbfa80]c000000000057cc4.irq_exit+0x54/0x6c [c000000b8ffbfb00]c00000000000c358.do_IRQ+0x170/0x1ac [c000000b8ffbfb90]c000000000004780hardware_interrupt_entry+0x18/0x98--- Exception: 501 (Hardware Interrupt) atc000000000010bdc.cpu_idle+0x114/0x1e0 [c000000b8ffbfe80] c000000000010bd0 .cpu_idle+0x108/0x1e0 (unreliable) [c000000b8ffbff00] c000000000026db0 .start_secondary+0x160/0x184 [c000000b8ffbff90] c000000000008364 .start_secondary_prolog+0xc/0x10Quoted 5 lines
I'm a little confused if the
port_napi_enable()is being called when the device is initialized, but then again, this is all new to me (should it be called inehea_open?). I see it called on some reset routines, but not on the first initialization.This is similar to the problem that Arnaldo hit a few minutes ago in the VIA Rhine driver.
You can't only make a
napi_enable()call when there has been a previousnapi_disable().
Quoted 6 lines
Quoted 5 lines
One way to fix this would be to forcefully
napi_disable()on all the per-port NAPI structs at the beginning ofehea_open(), which should set things up to satisfy the pre-condition of thenapi_enable()calls.OK, Ill try this.
Let me fix this. I'll try to get it done today.
Quoted 13 lines
Quoted 13 lines
You'll need to audit the entire driver to make sure this invariant is held properly.
Quoted 9 lines
Also, on this code, in
ehea_sense_port_attr()/* Number of default QPs */
if (use_mcs) port->num_def_qps = cb0->num_default_qps; else port->num_def_qps = 1;
Quoted 7 lines
Quoted 6 lines
Quoted 3 lines
When using napi, since we have multi-queue napi support now, wouldn't we want to use all the default qps instead of 1?
I don't know how this hardware works, you tell me :-)
Heh, I don't know it well, either. Maybe Jan Bernd can chime in.
We'd like to keep the possibility to switch back to a single queue for now. However, we could activate multi queue support as default now. I'll include this in the patch.
Quoted 6 lines
Thanks for your help,
-Andrew
David Miller
maintainerFrom: Jan-Bernd Themann <[email protected]>
Date: Thu, 23 Aug 2007 08:55:29 +0200
Quoted 2 lines
We'd like to keep the possibility to switch back to a single queue for now.
Please do not do this, we already have way too much configurability out there.
If you have the physical hardware queues enabled, use multiqueue napi support.
If you add a knob to use or not use multi-napi, this makes life more miserable for your users and your driver more complicated and harder to maintain.
Hi David,
Quoted 8 lines · David Miller
From: Jan-Bernd Themann <[email protected]>
Date: Thu, 23 Aug 2007 08:55:29 +0200Quoted 2 lines
We'd like to keep the possibility to switch back to a single queue for now.
Please do not do this, we already have way too much configurability out there.
ok, we decided to remove the switch for kernel 2.6.24
Regards,
Jan-Bernd