Thread (21 messages) 21 messages, 5 authors, 2021-05-20

Re: [net-next 6/6] can: mcp251xfd: mcp251xfd_regmap_crc_read(): work around broken CRC on TBC register

From: Patrick Menschel <hidden>
Date: 2021-05-09 07:46:35
Also in: linux-can

Am 08.05.21 um 20:36 schrieb Patrick Menschel:
Am 07.05.21 um 10:25 schrieb Marc Kleine-Budde:
quoted
On 07.05.2021 08:21:57, Patrick Menschel wrote:
quoted
quoted
quoted
quoted
Would it be possible for you to pull these patches into a v5.10 branch
in your linux-rpi repo [1]?
Here you are:

https://github.com/marckleinebudde/linux-rpi/tree/v5.10-rpi/backport-performance-improvements

I've included the UINC performance enhancements, too. The branch is compiled
tested only, though. I'll send a pull request to the rpi kernel after I've
testing feedback from you.
Drew, Patrick, have you tested this branch? If so I'll send a pull
request to the raspi kernel.
quoted
not yet. Thanks for reminding me. I'll start a native build on a pi0w asap.

Is there any test application or stress test that I should run?
No, not any particular, do your normal (stress) testing.
Following up on this.

...

Do I have to change my test?
Hi Marc,

I changed my test to 1 hour and removed the sleep statement.
Still no measurable difference for performance and no CRC Errors with
both kernels.

Apparently the test is hard on the CPU, I have two pytest processes
listed in htop one with 80%CPU and one with 60% CPU, approx 30% ram
usage of 512MB. I have no clue how it reaches the CPU values, there
should be only one CPU on the pi0w.


### 5.10.17+ on pi0w ###

2021-05-09 08:02:56 [    INFO] 725649 frames in 1:00:00
(test_socketcan.py:890)


### 5.10.31-performance-backports+ on pi0w ###

2021-05-09 09:13:32 [    INFO] 715936 frames in 1:00:00
(test_socketcan.py:890)


I'll switch boards to a pi3b and test again with these settings.

Best Regards,
Patrick
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help