Message ID | 20200814053926.10789-1-hui.wang@canonical.com |
---|---|
Headers | show |
Series | alsa/sof: support 1 and 3 dmics | expand |
On 14.08.20 07:39, Hui Wang wrote: > BugLink: https://bugs.launchpad.net/bugs/1891585 > > [Impact] > On a Dell tigerlake laptop, the sof driver fails to detect the dmic, > it alwasys report 0 dmics on the machine. So users doesn't have > internal mic to use. This does not warrant a "high" importance for SRU. I lowered it to "medium". Also I saw on the bug report that there are tasks for linux-firmware. Is there any hard dependency or ordering restriction between the kernel and the linux-firmware change? -Stefan > > [Fix] > Intel write a patch to fix this issue, if NHLT table returns 1 or > 3 dmics, the sof driver will not fail like before, it can support > 1 or 3 dmics after applying this patch. > > [Test Case] > Boot the kernel with this patch, and put sof-hda-generic-1ch.tplg in > the linux-firmware, the sof driver could detect the dmic and users > have a workable internal mic after booting up. > > [Regression Risk] > This patch adds new support for 1 and 3 dmics, it doesn't change > anything on existing functions, so the regression posibility is very > low. > > > Jaska Uimonen (1): > UBUNTU: SAUCE: ASoC: SOF: intel: hda: support also devices with 1 and > 3 dmics > > sound/soc/sof/intel/hda.c | 8 +++++++- > 1 file changed, 7 insertions(+), 1 deletion(-) >
On 2020/8/14 下午3:44, Stefan Bader wrote: > On 14.08.20 07:39, Hui Wang wrote: >> BugLink: https://bugs.launchpad.net/bugs/1891585 >> >> [Impact] >> On a Dell tigerlake laptop, the sof driver fails to detect the dmic, >> it alwasys report 0 dmics on the machine. So users doesn't have >> internal mic to use. > This does not warrant a "high" importance for SRU. I lowered it to "medium". > Also I saw on the bug report that there are tasks for linux-firmware. Is there > any hard dependency or ordering restriction between the kernel and the > linux-firmware change? OK, it is fine to set it to medium. There is no hard dependency or ordering restriction between kernel and firmware. If users install the firmware first, it will not bring any impact to old kernel since the old kernel doesn't load these 2 new files. If users install the new kernel on the machines with 0 dmics, 2 dmics or 4 dmics first, it will not bring any impact too since the kernel will not load those 2 new files. If users install the new kernel on the machines with 1 dmic or 3 dmics first, the kernel will try to load the -1ch.tplg or -3ch.tplg, but there is no these 2 files since users has not updated linux-firmware, then the sof driver will fail. But this is not a regression since the audio never work with old kernel . Thanks, Hui. > > -Stefan > >> [Fix] >> Intel write a patch to fix this issue, if NHLT table returns 1 or >> 3 dmics, the sof driver will not fail like before, it can support >> 1 or 3 dmics after applying this patch. >> >> [Test Case] >> Boot the kernel with this patch, and put sof-hda-generic-1ch.tplg in >> the linux-firmware, the sof driver could detect the dmic and users >> have a workable internal mic after booting up. >> >> [Regression Risk] >> This patch adds new support for 1 and 3 dmics, it doesn't change >> anything on existing functions, so the regression posibility is very >> low. >> >> >> Jaska Uimonen (1): >> UBUNTU: SAUCE: ASoC: SOF: intel: hda: support also devices with 1 and >> 3 dmics >> >> sound/soc/sof/intel/hda.c | 8 +++++++- >> 1 file changed, 7 insertions(+), 1 deletion(-) >> >
On 14.8.2020 8.39, Hui Wang wrote: > BugLink: https://bugs.launchpad.net/bugs/1891585 > > [Impact] > On a Dell tigerlake laptop, the sof driver fails to detect the dmic, > it alwasys report 0 dmics on the machine. So users doesn't have > internal mic to use. > > [Fix] > Intel write a patch to fix this issue, if NHLT table returns 1 or > 3 dmics, the sof driver will not fail like before, it can support > 1 or 3 dmics after applying this patch. > > [Test Case] > Boot the kernel with this patch, and put sof-hda-generic-1ch.tplg in > the linux-firmware, the sof driver could detect the dmic and users > have a workable internal mic after booting up. > > [Regression Risk] > This patch adds new support for 1 and 3 dmics, it doesn't change > anything on existing functions, so the regression posibility is very > low. > > > Jaska Uimonen (1): > UBUNTU: SAUCE: ASoC: SOF: intel: hda: support also devices with 1 and > 3 dmics > > sound/soc/sof/intel/hda.c | 8 +++++++- > 1 file changed, 7 insertions(+), 1 deletion(-) > applied to oem-5.6, thanks
Changed "Chery picked from" to "cherry picked from commit" Applied to Focal/master-next. Thanks! Ian On 2020-08-14 13:39:25 , Hui Wang wrote: > BugLink: https://bugs.launchpad.net/bugs/1891585 > > [Impact] > On a Dell tigerlake laptop, the sof driver fails to detect the dmic, > it alwasys report 0 dmics on the machine. So users doesn't have > internal mic to use. > > [Fix] > Intel write a patch to fix this issue, if NHLT table returns 1 or > 3 dmics, the sof driver will not fail like before, it can support > 1 or 3 dmics after applying this patch. > > [Test Case] > Boot the kernel with this patch, and put sof-hda-generic-1ch.tplg in > the linux-firmware, the sof driver could detect the dmic and users > have a workable internal mic after booting up. > > [Regression Risk] > This patch adds new support for 1 and 3 dmics, it doesn't change > anything on existing functions, so the regression posibility is very > low. > > > Jaska Uimonen (1): > UBUNTU: SAUCE: ASoC: SOF: intel: hda: support also devices with 1 and > 3 dmics > > sound/soc/sof/intel/hda.c | 8 +++++++- > 1 file changed, 7 insertions(+), 1 deletion(-) > > -- > 2.17.1 > > > -- > kernel-team mailing list > kernel-team@lists.ubuntu.com > https://lists.ubuntu.com/mailman/listinfo/kernel-team
On 2020/8/21 上午6:00, Ian May wrote: > Changed "Chery picked from" to "cherry picked from commit" > Applied to Focal/master-next. Thanks! > > Ian Thanks. > > On 2020-08-14 13:39:25 , Hui Wang wrote: >> BugLink: https://bugs.launchpad.net/bugs/1891585 >> >> [Impact] >> On a Dell tigerlake laptop, the sof driver fails to detect the dmic, >> it alwasys report 0 dmics on the machine. So users doesn't have >> internal mic to use. >> >> [Fix] >> Intel write a patch to fix this issue, if NHLT table returns 1 or >> 3 dmics, the sof driver will not fail like before, it can support >> 1 or 3 dmics after applying this patch. >> >> [Test Case] >> Boot the kernel with this patch, and put sof-hda-generic-1ch.tplg in >> the linux-firmware, the sof driver could detect the dmic and users >> have a workable internal mic after booting up. >> >> [Regression Risk] >> This patch adds new support for 1 and 3 dmics, it doesn't change >> anything on existing functions, so the regression posibility is very >> low. >> >> >> Jaska Uimonen (1): >> UBUNTU: SAUCE: ASoC: SOF: intel: hda: support also devices with 1 and >> 3 dmics >> >> sound/soc/sof/intel/hda.c | 8 +++++++- >> 1 file changed, 7 insertions(+), 1 deletion(-) >> >> -- >> 2.17.1 >> >> >> -- >> kernel-team mailing list >> kernel-team@lists.ubuntu.com >> https://lists.ubuntu.com/mailman/listinfo/kernel-team
Hi Ian, BTW, please help take a look at this patchset https://lists.ubuntu.com/archives/kernel-team/2020-August/112593.html, our oem project is waiting for this patchset to be merged to focal kernel. Thanks. On 2020/8/21 上午8:03, Hui Wang wrote: > > On 2020/8/21 上午6:00, Ian May wrote: >> Changed "Chery picked from" to "cherry picked from commit" >> Applied to Focal/master-next. Thanks! >> >> Ian > > Thanks. > >> >> On 2020-08-14 13:39:25 , Hui Wang wrote: >>> BugLink: https://bugs.launchpad.net/bugs/1891585 >>> >>> [Impact] >>> On a Dell tigerlake laptop, the sof driver fails to detect the dmic, >>> it alwasys report 0 dmics on the machine. So users doesn't have >>> internal mic to use. >>> >>> [Fix] >>> Intel write a patch to fix this issue, if NHLT table returns 1 or >>> 3 dmics, the sof driver will not fail like before, it can support >>> 1 or 3 dmics after applying this patch. >>> >>> [Test Case] >>> Boot the kernel with this patch, and put sof-hda-generic-1ch.tplg in >>> the linux-firmware, the sof driver could detect the dmic and users >>> have a workable internal mic after booting up. >>> >>> [Regression Risk] >>> This patch adds new support for 1 and 3 dmics, it doesn't change >>> anything on existing functions, so the regression posibility is very >>> low. >>> >>> >>> Jaska Uimonen (1): >>> UBUNTU: SAUCE: ASoC: SOF: intel: hda: support also devices with 1 >>> and >>> 3 dmics >>> >>> sound/soc/sof/intel/hda.c | 8 +++++++- >>> 1 file changed, 7 insertions(+), 1 deletion(-) >>> >>> -- >>> 2.17.1 >>> >>> >>> -- >>> kernel-team mailing list >>> kernel-team@lists.ubuntu.com >>> https://lists.ubuntu.com/mailman/listinfo/kernel-team