[alsa-devel] [PATCH 0/3] Fixes for the new generic parser

newer
[alsa-devel] [PATCH] ASoC: wm8978:...

David Henningsson

16 Jan 2013 16 Jan '13

11:45 a.m.

Hi Takashi,

I've run my hda-emu summary test tool to check for errors in the 110 codecs we have of certified systems (those in codecs/canonical) against the test/hda-migrate branch.

After these fixes and the one I sent yesterday, there are no remaining regressions (as seen by this test tool) compared to the current sound tree.

David Henningsson (3): ALSA: hda - Check array bounds in get_input_path ALSA: hda - fix wrong adc_idx in generic parser ALSA: hda - initialize channel counts correctly

sound/pci/hda/hda_generic.c | 16 ++++++++++++---- 1 file changed, 12 insertions(+), 4 deletions(-)

-- 1.7.9.5

Show replies by date

David Henningsson

16 Jan 16 Jan

11:45 a.m.

New subject: [alsa-devel] [PATCH 1/3] ALSA: hda - Check array bounds in get_input_path

This gives us some additional safety.

Signed-off-by: David Henningsson david.henningsson@canonical.com --- sound/pci/hda/hda_generic.c | 8 ++++++++ 1 file changed, 8 insertions(+)

diff --git a/sound/pci/hda/hda_generic.c b/sound/pci/hda/hda_generic.c index 932e6a1..577acc4 100644 --- a/sound/pci/hda/hda_generic.c +++ b/sound/pci/hda/hda_generic.c @@ -2397,8 +2397,16 @@ static int create_input_ctls(struct hda_codec *codec) static struct nid_path *get_input_path(struct hda_codec *codec, int adc_idx, int imux_idx) { struct hda_gen_spec *spec = codec->spec; + if (imux_idx < 0 || imux_idx >= HDA_MAX_NUM_INPUTS) { + snd_BUG(); + return NULL; + } if (spec->dyn_adc_switch) adc_idx = spec->dyn_adc_idx[imux_idx]; + if (adc_idx < 0 || adc_idx >= AUTO_CFG_MAX_OUTS) { + snd_BUG(); + return NULL; + } return snd_hda_get_path_from_idx(codec, spec->input_paths[imux_idx][adc_idx]); }

-- 1.7.9.5

David Henningsson

11:45 a.m.

New subject: [alsa-devel] [PATCH 2/3] ALSA: hda - fix wrong adc_idx in generic parser

We use knew->index for adc_idx when we create "Capture Volume" and "Capture Switch", so use the same to retrieve adc_idx.

Signed-off-by: David Henningsson david.henningsson@canonical.com --- sound/pci/hda/hda_generic.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/sound/pci/hda/hda_generic.c b/sound/pci/hda/hda_generic.c index 577acc4..aac8eb9 100644 --- a/sound/pci/hda/hda_generic.c +++ b/sound/pci/hda/hda_generic.c @@ -2426,7 +2426,7 @@ static int mux_enum_get(struct snd_kcontrol *kcontrol, { struct hda_codec *codec = snd_kcontrol_chip(kcontrol); struct hda_gen_spec *spec = codec->spec; - unsigned int adc_idx = snd_ctl_get_ioffidx(kcontrol, &ucontrol->id); + unsigned int adc_idx = kcontrol->id.index;

ucontrol->value.enumerated.item[0] = spec->cur_mux[adc_idx]; return 0; @@ -2436,7 +2436,7 @@ static int mux_enum_put(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_value *ucontrol) { struct hda_codec *codec = snd_kcontrol_chip(kcontrol); - unsigned int adc_idx = snd_ctl_get_ioffidx(kcontrol, &ucontrol->id); + unsigned int adc_idx = kcontrol->id.index; return mux_select(codec, adc_idx, ucontrol->value.enumerated.item[0]); } @@ -2468,7 +2468,7 @@ static int cap_put_caller(struct snd_kcontrol *kcontrol, int i, adc_idx, err = 0;

imux = &spec->input_mux; - adc_idx = snd_ctl_get_ioffidx(kcontrol, &ucontrol->id); + adc_idx = kcontrol->id.index; mutex_lock(&codec->control_mutex); /* we use the cache-only update at first since multiple input paths * may shared the same amp; by updating only caches, the redundant

-- 1.7.9.5

David Henningsson

11:45 a.m.

New subject: [alsa-devel] [PATCH 3/3] ALSA: hda - initialize channel counts correctly

Even a single DAC can output two channels, so the channel count is twice the number of DACs.

Signed-off-by: David Henningsson david.henningsson@canonical.com --- sound/pci/hda/hda_generic.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/sound/pci/hda/hda_generic.c b/sound/pci/hda/hda_generic.c index aac8eb9..3ef38da 100644 --- a/sound/pci/hda/hda_generic.c +++ b/sound/pci/hda/hda_generic.c @@ -1399,7 +1399,7 @@ static int fill_and_eval_dacs(struct hda_codec *codec, }

spec->ext_channel_count = spec->min_channel_count = - spec->multiout.num_dacs; + spec->multiout.num_dacs * 2;

if (spec->multi_ios == 2) { for (i = 0; i < 2; i++)

-- 1.7.9.5

Takashi Iwai

12:01 p.m.

At Wed, 16 Jan 2013 11:45:34 +0100, David Henningsson wrote:

...

Hi Takashi,

I've run my hda-emu summary test tool to check for errors in the 110 codecs we have of certified systems (those in codecs/canonical) against the test/hda-migrate branch.

After these fixes and the one I sent yesterday, there are no remaining regressions (as seen by this test tool) compared to the current sound tree.

Thanks!

I applied these to test/hda-gen-parser branch now.

I'll take some test shots today, and if everything looks OK, I'm going to merge the patches to sound git tree for-next branch tomorrow or later.

Takashi

...

David Henningsson (3): ALSA: hda - Check array bounds in get_input_path ALSA: hda - fix wrong adc_idx in generic parser ALSA: hda - initialize channel counts correctly

sound/pci/hda/hda_generic.c | 16 ++++++++++++---- 1 file changed, 12 insertions(+), 4 deletions(-)

-- 1.7.9.5

4359

Age (days ago)

4359

Last active (days ago)

List overview

Download

4 comments

2 participants

tags (0)

participants (2)

David Henningsson
Takashi Iwai