* [PATCH v2] ALSA: emu10k1: minor optimizations
@ 2023-04-22 12:15 Oswald Buddenhagen
2023-04-22 15:33 ` Takashi Iwai
0 siblings, 1 reply; 2+ messages in thread
From: Oswald Buddenhagen @ 2023-04-22 12:15 UTC (permalink / raw)
To: alsa-devel; +Cc: Takashi Iwai
- In snd_emu10k1_look_for_ctl(), evaluate the cheap condition first
- In _snd_emu10k1_{audigy_,}init_efx(), don't use expensive bit setting
loops to fill arrays
- In snd_emu_proc_ptr_reg_read(), remove useless condition - iobase can
be only 0 or 0x20
Signed-off-by: Oswald Buddenhagen <oswald.buddenhagen@gmx.de>
---
v2:
- fix fill sizes for sblive by using the previous hard-coded numbers.
it might be a good idea to introduce constants for that, but that's
for another patch.
---
sound/pci/emu10k1/emufx.c | 18 +++++++-----------
sound/pci/emu10k1/emuproc.c | 5 +----
2 files changed, 8 insertions(+), 15 deletions(-)
diff --git a/sound/pci/emu10k1/emufx.c b/sound/pci/emu10k1/emufx.c
index 6cf7c8b1de47..66f450377888 100644
--- a/sound/pci/emu10k1/emufx.c
+++ b/sound/pci/emu10k1/emufx.c
@@ -641,8 +641,8 @@ snd_emu10k1_look_for_ctl(struct snd_emu10k1 *emu,
list_for_each_entry(ctl, &emu->fx8010.gpr_ctl, list) {
kcontrol = ctl->kcontrol;
if (kcontrol->id.iface == id->iface &&
- !strcmp(kcontrol->id.name, id->name) &&
- kcontrol->id.index == id->index)
+ kcontrol->id.index == id->index &&
+ !strcmp(kcontrol->id.name, id->name))
return ctl;
}
return NULL;
@@ -1213,7 +1213,7 @@ static int snd_emu10k1_audigy_dsp_convert_32_to_2x16(
static int _snd_emu10k1_audigy_init_efx(struct snd_emu10k1 *emu)
{
- int err, i, z, gpr, nctl;
+ int err, z, gpr, nctl;
int bit_shifter16;
const int playback = 10;
const int capture = playback + (SND_EMU10K1_PLAYBACK_CHANNELS * 2); /* we reserve 10 voices */
@@ -1245,12 +1245,10 @@ static int _snd_emu10k1_audigy_init_efx(struct snd_emu10k1 *emu)
icode->code = icode->tram_addr_map + 256;
/* clear free GPRs */
- for (i = 0; i < 512; i++)
- set_bit(i, icode->gpr_valid);
+ memset(icode->gpr_valid, 0xff, 512 / 8);
/* clear TRAM data & address lines */
- for (i = 0; i < 256; i++)
- set_bit(i, icode->tram_valid);
+ memset(icode->tram_valid, 0xff, 256 / 8);
strcpy(icode->name, "Audigy DSP code for ALSA");
ptr = 0;
@@ -1886,12 +1884,10 @@ static int _snd_emu10k1_init_efx(struct snd_emu10k1 *emu)
icode->code = icode->tram_addr_map + 160;
/* clear free GPRs */
- for (i = 0; i < 256; i++)
- set_bit(i, icode->gpr_valid);
+ memset(icode->gpr_valid, 0xff, 256 / 8);
/* clear TRAM data & address lines */
- for (i = 0; i < 160; i++)
- set_bit(i, icode->tram_valid);
+ memset(icode->tram_valid, 0xff, 160 / 8);
strcpy(icode->name, "SB Live! FX8010 code for ALSA v1.2 by Jaroslav Kysela");
ptr = 0; i = 0;
diff --git a/sound/pci/emu10k1/emuproc.c b/sound/pci/emu10k1/emuproc.c
index 6e20cca9c98f..bec72dc60a41 100644
--- a/sound/pci/emu10k1/emuproc.c
+++ b/sound/pci/emu10k1/emuproc.c
@@ -477,10 +477,7 @@ static void snd_emu_proc_ptr_reg_read(struct snd_info_entry *entry,
for(i = offset; i < offset+length; i++) {
snd_iprintf(buffer, "%02X: ",i);
for (j = 0; j < voices; j++) {
- if(iobase == 0)
- value = snd_ptr_read(emu, 0, i, j);
- else
- value = snd_ptr_read(emu, 0x20, i, j);
+ value = snd_ptr_read(emu, iobase, i, j);
snd_iprintf(buffer, "%08lX ", value);
}
snd_iprintf(buffer, "\n");
--
2.40.0.152.g15d061e6df
^ permalink raw reply related [flat|nested] 2+ messages in thread* Re: [PATCH v2] ALSA: emu10k1: minor optimizations
2023-04-22 12:15 [PATCH v2] ALSA: emu10k1: minor optimizations Oswald Buddenhagen
@ 2023-04-22 15:33 ` Takashi Iwai
0 siblings, 0 replies; 2+ messages in thread
From: Takashi Iwai @ 2023-04-22 15:33 UTC (permalink / raw)
To: Oswald Buddenhagen; +Cc: alsa-devel
On Sat, 22 Apr 2023 14:15:19 +0200,
Oswald Buddenhagen wrote:
>
> - In snd_emu10k1_look_for_ctl(), evaluate the cheap condition first
> - In _snd_emu10k1_{audigy_,}init_efx(), don't use expensive bit setting
> loops to fill arrays
> - In snd_emu_proc_ptr_reg_read(), remove useless condition - iobase can
> be only 0 or 0x20
>
> Signed-off-by: Oswald Buddenhagen <oswald.buddenhagen@gmx.de>
> ---
> v2:
> - fix fill sizes for sblive by using the previous hard-coded numbers.
> it might be a good idea to introduce constants for that, but that's
> for another patch.
Thanks, applied now.
Takashi
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2023-04-22 15:34 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2023-04-22 12:15 [PATCH v2] ALSA: emu10k1: minor optimizations Oswald Buddenhagen
2023-04-22 15:33 ` Takashi Iwai
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox