On 12/06/2016 03:07 PM, Takashi Iwai wrote:

On Tue, 06 Dec 2016 07:39:41 +0100, Kai-Heng Feng wrote:

...

Commit [64047d7f4912 ALSA: hda - ignore the assoc and seq when comparing pin configurations] may still fail to match pins on some machines, because the bitmask it used only ignore seq but not assoc. Change the bitmask to also ignore assoc.

So you are ignoring *both* assoc and seq numbers? Or did you intend to ignore only assoc number?

The original intention is to ignoring both assoc and seq numbers. But my patch only ignored the seq, so this patch can fix my previous patch.

In anyway, it'd be better to use a macro like

if ((t_pins->val & ~AC_DEFCFGDEF_ASSOC) == (cfg & ~AC_DEFCFG_ASSOC))

Yes, it is better to use the macro. Let us use the macro in the V2 patch.

Thanks, Hui.

thanks,

Takashi

...

Thanks to Hui Wang for the analysis.

Signed-off-by: Kai-Heng Feng kai.heng.feng@canonical.com

sound/pci/hda/hda_auto_parser.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/sound/pci/hda/hda_auto_parser.c b/sound/pci/hda/hda_auto_parser.c index 4ad29f8..fdc30d0 100644 --- a/sound/pci/hda/hda_auto_parser.c +++ b/sound/pci/hda/hda_auto_parser.c @@ -901,7 +901,7 @@ static bool pin_config_match(struct hda_codec *codec, for (; t_pins->nid; t_pins++) { if (t_pins->nid == nid) { found = 1;

• 		if ((t_pins->val & 0xfffffff0) == (cfg & 0xfffffff0))
  

• 		if ((t_pins->val & 0xffffff00) == (cfg & 0xffffff00))
  		break;
  	else if ((cfg & 0xf0000000) == 0x40000000 && (t_pins->val & 0xf0000000) == 0x40000000)
  		break;
  

-- 2.10.2