| Message ID | 20191210154536.29819-1-tiwai@suse.de (mailing list archive) |
|---|---|
| Headers | show |
| Series | drm/bridge: dw-hdmi: PCM API updates | expand |
On Tue, 10 Dec 2019 16:45:34 +0100, Takashi Iwai wrote: > > Hi, > > this is a patch set for updating ALSA PCM API usages in dw-hdmi > driver. I already tried to "fix" the driver some time ago but it was > utterly wrong. So this is a combination of the revised patch and > another cleanup patch. > > The first one is to change the buffer allocation mechanism in the > driver to the manual allocation of the h/w buffer and the automatic > allocation of PCM stream buffers via the new standard API. The > significant change is that size of the h/w buffer isn't no longer > controlled via ALSA preallocation proc file but rather via the new > module option (if any). > > The second one is a oneliner patch just to remove the superfluous PCM > ops. > > Both need the ALSA PCM core changes in 5.5-rc1, so please apply them > on top of 5.5-rc1 or later. Or, just let me know if I should apply > them through sound git tree. > > > thanks, > > Takashi > > === > > Takashi Iwai (2): > drm/bridge: dw-hdmi: Follow the standard ALSA memalloc way > drm/bridge: dw-hdmi: Drop superfluous ioctl PCM ops Any chance for reviewing these patches? Since this driver is the only one who is still using the old ALSA vmalloc API, I'd like to change it and drop the old API in 5.6. thanks, Takashi
Hi, On 09/01/2020 10:10, Takashi Iwai wrote: > On Tue, 10 Dec 2019 16:45:34 +0100, > Takashi Iwai wrote: >> >> Hi, >> >> this is a patch set for updating ALSA PCM API usages in dw-hdmi >> driver. I already tried to "fix" the driver some time ago but it was >> utterly wrong. So this is a combination of the revised patch and >> another cleanup patch. >> >> The first one is to change the buffer allocation mechanism in the >> driver to the manual allocation of the h/w buffer and the automatic >> allocation of PCM stream buffers via the new standard API. The >> significant change is that size of the h/w buffer isn't no longer >> controlled via ALSA preallocation proc file but rather via the new >> module option (if any). >> >> The second one is a oneliner patch just to remove the superfluous PCM >> ops. >> >> Both need the ALSA PCM core changes in 5.5-rc1, so please apply them >> on top of 5.5-rc1 or later. Or, just let me know if I should apply >> them through sound git tree. >> >> >> thanks, >> >> Takashi >> >> === >> >> Takashi Iwai (2): >> drm/bridge: dw-hdmi: Follow the standard ALSA memalloc way >> drm/bridge: dw-hdmi: Drop superfluous ioctl PCM ops > > Any chance for reviewing these patches? > > Since this driver is the only one who is still using the old ALSA > vmalloc API, I'd like to change it and drop the old API in 5.6. Ping Russel since you reviewed the previous attempt. Neil > > > thanks, > > Takashi >
On Thu, Jan 09, 2020 at 10:10:09AM +0100, Takashi Iwai wrote: > On Tue, 10 Dec 2019 16:45:34 +0100, > Takashi Iwai wrote: > > > > Hi, > > > > this is a patch set for updating ALSA PCM API usages in dw-hdmi > > driver. I already tried to "fix" the driver some time ago but it was > > utterly wrong. So this is a combination of the revised patch and > > another cleanup patch. > > > > The first one is to change the buffer allocation mechanism in the > > driver to the manual allocation of the h/w buffer and the automatic > > allocation of PCM stream buffers via the new standard API. The > > significant change is that size of the h/w buffer isn't no longer > > controlled via ALSA preallocation proc file but rather via the new > > module option (if any). > > > > The second one is a oneliner patch just to remove the superfluous PCM > > ops. > > > > Both need the ALSA PCM core changes in 5.5-rc1, so please apply them > > on top of 5.5-rc1 or later. Or, just let me know if I should apply > > them through sound git tree. > > > > > > thanks, > > > > Takashi > > > > === > > > > Takashi Iwai (2): > > drm/bridge: dw-hdmi: Follow the standard ALSA memalloc way > > drm/bridge: dw-hdmi: Drop superfluous ioctl PCM ops > > Any chance for reviewing these patches? > > Since this driver is the only one who is still using the old ALSA > vmalloc API, I'd like to change it and drop the old API in 5.6. It isn't something I can even test at the moment; I have the platforms but no TV to connect them to.