[alsa-devel] [PATCH] ad1848 and cs4231 busy loop replacement
rene.herman at gmail.com
Mon Sep 10 15:13:24 CEST 2007
On 09/10/2007 02:42 PM, Krzysztof Helt wrote:
> On 9/10/07, Rene Herman <rene.herman at gmail.com> wrote:
>> Well, no, sorry, but I consider that to be completely breaking the logic of
>> the code.
> No I changed the logic of this code not to wait for specifically for
> callibration "start" but into calibration "under way".
No, waiting for calibration the be under way is what my 0/1 ms does. You are
waiting for it to be nearly done, which is complete nonsense. One line below
we are waiting for 250 ms (generally with _one_ pass through the loop -- we
only wake up through signals) anyway!
The no delay at all from cs4231 is the logic -- when we've dropped MCE, ACI
comes up (when auto-calibrating) and we only wait for it to finish. For
ad1848, ACI up may take 5 cycles from MCE down so we delay 1 ms so we know
we're testing correctly.
>> Your: wait unconditionally until calibration _nearly_ done, then go wait for
>> it for 250 ms to be really done.
>> Mine: wait unconditionally until calibration has started, then go wait for
>> it for 250 ms to finish.
[ ... ]
> So the only difference is 6 (or 1) ms and this time will be spent in
> the loop anyway. Are we arguing 1ms (for CS4231) in 250ms waiting
No, we are arguing maintaining code. Do not obscure the code flow for no
reason. Fix your logic or (for what it's worth) I am going to NAK the change.
> I don't understand "keep them in sync".
In sync source-code wise. While the no delay from cs4231 may be the rule,
ad1848 needs a small delay so if you'd wanted to keep them looking the same
I wouldn't care about a 1 ms delay for cs4231 as well. If you don't, fine as
More information about the Alsa-devel