Re: [PATCH v6 6/6] drm/drm-file: Show finer-grained BO sizes in drm_show_memory_stats

[Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>]

On 22/09/2023 12:03, Adrián Larumbe wrote:
> On 21.09.2023 11:14, Tvrtko Ursulin wrote:
>>
>> On 20/09/2023 16:32, Tvrtko Ursulin wrote:
>>>
>>> On 20/09/2023 00:34, Adrián Larumbe wrote:
>>>> The current implementation will try to pick the highest available size
>>>> display unit as soon as the BO size exceeds that of the previous
>>>> multiplier. That can lead to loss of precision in contexts of low memory
>>>> usage.
>>>>
>>>> The new selection criteria try to preserve precision, whilst also
>>>> increasing the display unit selection threshold to render more accurate
>>>> values.
>>>>
>>>> Signed-off-by: Adrián Larumbe <adrian.larumbe@collabora.com>
>>>> Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
>>>> Reviewed-by: Steven Price <steven.price@arm.com>
>>>> ---
>>>>    drivers/gpu/drm/drm_file.c | 5 ++++-
>>>>    1 file changed, 4 insertions(+), 1 deletion(-)
>>>>
>>>> diff --git a/drivers/gpu/drm/drm_file.c b/drivers/gpu/drm/drm_file.c
>>>> index 762965e3d503..34cfa128ffe5 100644
>>>> --- a/drivers/gpu/drm/drm_file.c
>>>> +++ b/drivers/gpu/drm/drm_file.c
>>>> @@ -872,6 +872,8 @@ void drm_send_event(struct drm_device *dev, struct
>>>> drm_pending_event *e)
>>>>    }
>>>>    EXPORT_SYMBOL(drm_send_event);
>>>> +#define UPPER_UNIT_THRESHOLD 100
>>>> +
>>>>    static void print_size(struct drm_printer *p, const char *stat,
>>>>                   const char *region, u64 sz)
>>>>    {
>>>> @@ -879,7 +881,8 @@ static void print_size(struct drm_printer *p,
>>>> const char *stat,
>>>>        unsigned u;
>>>>        for (u = 0; u < ARRAY_SIZE(units) - 1; u++) {
>>>> -        if (sz < SZ_1K)
>>>> +        if ((sz & (SZ_1K - 1)) &&
>>>
>>> IS_ALIGNED worth it at all?
>>>
>>>> +            sz < UPPER_UNIT_THRESHOLD * SZ_1K)
>>>>                break;
>>>
>>> Excuse me for a late comment (I was away). I did not get what what is
>>> special about a ~10% threshold? Sounds to me just going with the lower
>>> unit, when size is not aligned to the higher one, would be better than
>>> sometimes precision-sometimes-not.
>>
>> FWIW both current and the threshold option make testing the feature very
>> annoying.
> 
> How so?

I have to build in the knowledge of implementation details of 
print_size() into my IGT in order to use the right size BOs, so test is 
able to verify stats move as expected. It just feels wrong.

>> So I'd really propose we simply use smaller unit when unaligned.
> 
> Like I said in the previous reply, for drm files whose overall BO size sum is enormous
> but not a multiple of a MiB, this would render huge number representations in KiB.
> I don't find this particularly comfortable to read, and then this extra precision
> would mean nothing to nvtop or gputop, which would have to scale the size to their
> available screen dimensions when plotting them.

I don't think numbers in KiB are so huge.

And I don't think people will end up reading them manually a lot anyway, 
since you have to hunt the pid, and fd, etc.. It is much more realistic 
that some tool like gputop will be used.

And I don't think consistency of units across drivers or whatever 
matters. Even better to keep userspace parser on their toes and make 
then follow drm-usage-stats.rst and not any implementations, at some 
point in time.

Regards,

Tvrtko