* Re: [9front] memfillpoly
@ 2018-04-05 7:25 cinap_lenrek
2018-04-05 8:15 ` Robert Ransom
0 siblings, 1 reply; 4+ messages in thread
From: cinap_lenrek @ 2018-04-05 7:25 UTC (permalink / raw)
To: 9front
memfillpoly is run in the kernel by devdraw. using fp there is
only supported on amd64 kernel right now.
--
cinap
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [9front] memfillpoly
2018-04-05 7:25 [9front] memfillpoly cinap_lenrek
@ 2018-04-05 8:15 ` Robert Ransom
2018-04-05 21:37 ` hiro
0 siblings, 1 reply; 4+ messages in thread
From: Robert Ransom @ 2018-04-05 8:15 UTC (permalink / raw)
To: 9front
On 4/5/18, cinap_lenrek@felloff.net <cinap_lenrek@felloff.net> wrote:
> memfillpoly is run in the kernel by devdraw. using fp there is
> only supported on amd64 kernel right now.
Oh. Thanks.
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [9front] memfillpoly
2018-04-05 8:15 ` Robert Ransom
@ 2018-04-05 21:37 ` hiro
2018-04-07 3:26 ` Robert Ransom
0 siblings, 1 reply; 4+ messages in thread
From: hiro @ 2018-04-05 21:37 UTC (permalink / raw)
To: 9front
have you done any performance tests, or why are you contemplating
these kinds of optimizations?
On 4/5/18, Robert Ransom <rransom.8774@gmail.com> wrote:
> On 4/5/18, cinap_lenrek@felloff.net <cinap_lenrek@felloff.net> wrote:
>> memfillpoly is run in the kernel by devdraw. using fp there is
>> only supported on amd64 kernel right now.
>
> Oh. Thanks.
>
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [9front] memfillpoly
2018-04-05 21:37 ` hiro
@ 2018-04-07 3:26 ` Robert Ransom
0 siblings, 0 replies; 4+ messages in thread
From: Robert Ransom @ 2018-04-07 3:26 UTC (permalink / raw)
To: 9front
On 4/5/18, hiro <23hiro@gmail.com> wrote:
> have you done any performance tests, or why are you contemplating
> these kinds of optimizations?
The current implementation is broken in at least one way, and the
floating-point implementation would be much simpler.
Turns out I may have been wrong about it being faster; the current
implementation does perform the div/mod instructions only a few times
per segment.
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2018-04-07 3:26 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2018-04-05 7:25 [9front] memfillpoly cinap_lenrek
2018-04-05 8:15 ` Robert Ransom
2018-04-05 21:37 ` hiro
2018-04-07 3:26 ` Robert Ransom
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).