Le keskiviikkona 19. heinäkuuta 2023, 0.32.26 EEST Lynne a écrit :
> > User-space access to the cycle counter has been deemed a security threat
> > due to the Cycle Drift attack, and is disabled as of OpenSBI 1.3.
> >
> > If FFmpeg does not support Linux perf, FFmpeg will get _no_ performance
> >
Jul 17, 2023, 20:09 by r...@remlab.net:
> Le maanantaina 17. heinäkuuta 2023, 20.48.40 EEST Lynne a écrit :
>
>> >> > But I still argue that that is, either way, completely negligible
>> >> > compared
>> >> > to the *existing* overhead. Each loop is making 4 system calls, and
>> >> > each
>> >> >
Le maanantaina 17. heinäkuuta 2023, 20.48.40 EEST Lynne a écrit :
> >> > But I still argue that that is, either way, completely negligible
> >> > compared
> >> > to the *existing* overhead. Each loop is making 4 system calls, and
> >> > each
> >> > of those system call requires a direct call (to PL
Jul 17, 2023, 07:18 by r...@remlab.net:
> Le sunnuntaina 16. heinäkuuta 2023, 23.32.21 EEST Lynne a écrit :
>
>> Introducing additional overhead in the form of a dereference is a point
>> where instability can creep in. Can you guarantee that a context will
>> always remain in L1D cache,
>>
>
> L1
Le sunnuntaina 16. heinäkuuta 2023, 23.32.21 EEST Lynne a écrit :
> Introducing additional overhead in the form of a dereference is a point
> where instability can creep in. Can you guarantee that a context will
> always remain in L1D cache,
L1D is not involved here. In version 2, the pointers are
Jul 15, 2023, 22:13 by r...@remlab.net:
> Le lauantaina 15. heinäkuuta 2023, 20.43.26 EEST Lynne a écrit :
>
>> Jul 15, 2023, 10:26 by r...@remlab.net:
>> > Le lauantaina 15. heinäkuuta 2023, 11.05.51 EEST Lynne a écrit :
>> >> Jul 14, 2023, 20:29 by r...@remlab.net:
>> >> > This makes all calls t
Le lauantaina 15. heinäkuuta 2023, 20.43.26 EEST Lynne a écrit :
> Jul 15, 2023, 10:26 by r...@remlab.net:
> > Le lauantaina 15. heinäkuuta 2023, 11.05.51 EEST Lynne a écrit :
> >> Jul 14, 2023, 20:29 by r...@remlab.net:
> >> > This makes all calls to the bench start and stop functions via
> >> > f
Jul 15, 2023, 10:26 by r...@remlab.net:
> Le lauantaina 15. heinäkuuta 2023, 11.05.51 EEST Lynne a écrit :
>
>> Jul 14, 2023, 20:29 by r...@remlab.net:
>> > This makes all calls to the bench start and stop functions via
>> > function pointers. While the primary goal is to support run-time
>> > sel
Le lauantaina 15. heinäkuuta 2023, 11.05.51 EEST Lynne a écrit :
> Jul 14, 2023, 20:29 by r...@remlab.net:
> > This makes all calls to the bench start and stop functions via
> > function pointers. While the primary goal is to support run-time
> > selection of the performance measurement back-end in
Jul 14, 2023, 20:29 by r...@remlab.net:
> This makes all calls to the bench start and stop functions via
> function pointers. While the primary goal is to support run-time
> selection of the performance measurement back-end in later commits,
> this has the side benefit of containing platform depen
10 matches
Mail list logo
