Re: [PATCH] drm/amd/display: Remove check for 0 kHz clock values

[Paul Menzel <pmenzel+amd-gfx@molgen.mpg.de>]

[-- Attachment #1: Type: text/plain, Size: 2795 bytes --]

Dear Nicholas,


On 7/15/19 2:57 PM, Kazlauskas, Nicholas wrote:
> On 7/15/19 6:34 AM, Paul Menzel wrote:
>> From 09c1952466752033722b02d9c7e5532e1982f6d9 Mon Sep 17 00:00:00 2001
>> From: Paul Menzel <pmenzel@molgen.mpg.de>
>> Date: Sat, 13 Jul 2019 20:33:49 +0200
>>
>> This basically reverts commit 00893681a0ff4 (drm/amd/display: Reject
>> PPLib clock values if they are invalid).
>>
>> 0 kHz values are a thing on at least the boards below.
>>
>> 1.  MSI MS-7A37/B350M MORTAR (MS-7A37), BIOS 1.G1 05/17/2018
>> 2.  MSI B450M Mortar, 2400G on 4.19.8
>> 3.  Gigabyte Technology Co., Ltd. X470 AORUS ULTRA GAMING/X470 AORUS
>>      ULTRA GAMING-CF, BIOS F30 04/16/2019
>>
>> Asserting instead of giving a useful error message to the user, so they
>> can understand what is going on and how to possible fix things, might be
>> good for development, but is a bad user experience, so should not be on
>> production systems. So, remove the check for now.
>>
>> Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=107296
>> Tested: MSI MS-7A37/B350M MORTAR (MS-7A37)
>> Signed-off-by: Paul Menzel <pmenzel@molgen.mpg.de>
> 
> The two assertions should probably just be replaced with 
> DC_LOG_DEBUG(...) instead - this will drop the callstack on boot for 
> production systems.
> 
> Dropping the whole validation also means that we're going to be taking 
> the table as-is and overriding the defaults - which isn't something we'd 
> actually want to do.
> 
> I do think it's fine to just reduce this to a debug message since you'd 
> see this on any 2400G/AM4 (as far as I'm aware), and only for the fCLK 
> table (the tables always come from PPLIB/SMU).

Where can I find more information on this for 2400G/AM4? What about
2200(?) and so on?

If it’s expected, that there is 0 kHz values in there, the code should
deal with that, and filter that out. I do not understand, that the whole
table is invalidated and default values are used instead. That sounds
buggy.


Kind regards,

Paul


>> ---
>>   drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c | 5 -----
>>   1 file changed, 5 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c b/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c
>> index 1b4b51657f5e..edaaae5754fe 100644
>> --- a/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c
>> +++ b/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c
>> @@ -1362,11 +1362,6 @@ static bool verify_clock_values(struct dm_pp_clock_levels_with_voltage *clks)
>>   	if (clks->num_levels == 0)
>>   		return false;
>>   
>> -	for (i = 0; i < clks->num_levels; i++)
>> -		/* Ensure that the result is sane */
>> -		if (clks->data[i].clocks_in_khz == 0)
>> -			return false;
>> -
>>   	return true;
>>   }


[-- Attachment #2: S/MIME Cryptographic Signature --]
[-- Type: application/pkcs7-signature, Size: 5174 bytes --]