On Thu, Feb 04, 2021 at 06:52:01PM -0800, John Hubbard wrote: > On 2/4/21 5:44 PM, Minchan Kim wrote: > > On Thu, Feb 04, 2021 at 04:24:20PM -0800, John Hubbard wrote: > > > On 2/4/21 4:12 PM, Minchan Kim wrote: > > > ... > > > > > > Then, how to know how often CMA API failed? > > > > > > > > > > Why would you even need to know that, *in addition* to knowing specific > > > > > page allocation numbers that failed? Again, there is no real-world motivation > > > > > cited yet, just "this is good data". Need more stories and support here. > > > > > > > > Let me give an example. > > > > > > > > Let' assume we use memory buffer allocation via CMA for bluetooth > > > > enable of device. > > > > If user clicks the bluetooth button in the phone but fail to allocate > > > > the memory from CMA, user will still see bluetooth button gray. > > > > User would think his touch was not enough powerful so he try clicking > > > > again and fortunately CMA allocation was successful this time and > > > > they will see bluetooh button enabled and could listen the music. > > > > > > > > Here, product team needs to monitor how often CMA alloc failed so > > > > if the failure ratio is steadily increased than the bar, > > > > it means engineers need to go investigation. > > > > > > > > Make sense? > > > > > > > > > > Yes, except that it raises more questions: > > > > > > 1) Isn't this just standard allocation failure? Don't you already have a way > > > to track that? > > > > > > Presumably, having the source code, you can easily deduce that a bluetooth > > > allocation failure goes directly to a CMA allocation failure, right? > > Still wondering about this... It would work if we have full source code and stack are not complicated for every usecases. Having said, having a good central place automatically popped up is also beneficial for not to add similar statistics for each call sites. Why do we have too many item in slab sysfs instead of creating each call site inventing on each own? > > > > > > > Anyway, even though the above is still a little murky, I expect you're right > > > that it's good to have *some* indication, somewhere about CMA behavior... > > > > > > Thinking about this some more, I wonder if this is really /proc/vmstat sort > > > of data that we're talking about. It seems to fit right in there, yes? > > > > Thing is CMA instance are multiple, cma-A, cma-B, cma-C and each of CMA > > heap has own specific scenario. /proc/vmstat could be bloated a lot > > while CMA instance will be increased. > > > > Yes, that would not fit in /proc/vmstat...assuming that you really require > knowing--at this point--which CMA heap is involved. And that's worth poking > at. If you get an overall indication in vmstat that CMA is having trouble, > then maybe that's all you need to start digging further. I agree it could save to decide whether I should go digging further but anyway, I need to go though each of instance once it happens. In that, what I need is per-CMA statistics, not global. I am happy to implement it but I'd like to say it's not my case. > > It's actually easier to monitor one or two simpler items than it is to monitor > a larger number of complicated items. And I get the impression that this is > sort of a top-level, production software indicator. Let me clarify one more time. What I'd like to get ultimately is per-CMA statistics instead of global vmstat for the usecase at this moment. Global vmstat could help the decision whether I should go deeper but it ends up needing per-CMA statistics. And I'd like to keep them in sysfs, not debugfs since it should be stable as a telemetric. What points do you disagree in this view?
