- May 07, 2019
-
-
Pierre-Louis Bossart authored
Add PCI IDs for LP and H skews. Signed-off-by:
Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com> Signed-off-by:
Takashi Iwai <tiwai@suse.de>
-
- Apr 30, 2019
-
-
Takashi Iwai authored
Currently the IRQ handler in HD-audio controller driver is registered before the chip initialization. That is, we have some window opened between the azx_acquire_irq() call and the CORB/RIRB setup. If an interrupt is triggered in this small window, the IRQ handler may access to the uninitialized RIRB buffer, which leads to a NULL dereference Oops. This is usually no big problem since most of Intel chips do register the IRQ via MSI, and we've already fixed the order of the IRQ enablement and the CORB/RIRB setup in the former commit b61749a8 ("sound: enable interrupt after dma buffer initialization"), hence the IRQ won't be triggered in that room. However, some platforms use a shared IRQ, and this may allow the IRQ trigger by another source. Another possibility is the kdump environment: a stale interrupt might be present in there, the IRQ handler can be falsely triggered as well. For covering this small race, let's move the azx_acquire_irq() call after hda_intel_init_chip() call. Although this is a bit radical change, it can cover more widely than checking the CORB/RIRB setup locally in the callee side. Reported-by:
Liwei Song <liwei.song@windriver.com> Signed-off-by:
-
- Apr 08, 2019
-
-
Hui Wang authored
Recently we set CONFIG_SND_HDA_POWER_SAVE_DEFAULT to 1 when configuring the kernel, then two machines were reported to have noise after installing the new kernel. Put them in the blacklist, the noise disappears. https://bugs.launchpad.net/bugs/1821663 Cc: <stable@vger.kernel.org> Signed-off-by:
Hui Wang <hui.wang@canonical.com> Signed-off-by:
-
- Mar 19, 2019
-
-
Hui Wang authored
The commit 3baffc4a (ALSA: hda/intel: Refactoring PM code) changed the behaviour of azx_resume(), it triggers the jackpoll_work after applying this commit. This change introduced a new issue, all codecs are runtime active after S3, and will not call runtime_suspend() automatically. The root cause is the jackpoll_work calls snd_hda_power_up/down_pm, and it calls up_pm before snd_hdac_enter_pm is called, while calls the down_pm in the middle of enter_pm and leave_pm is called. This makes the dev->power.usage_count unbalanced after S3. To fix it, let azx_resume() don't trigger jackpoll_work as before it did. Fixes: 3baffc4a ("ALSA: hda/intel: Refactoring PM code") Signed-off-by:
-
- Mar 18, 2019
-
-
Jaroslav Kysela authored
Another machine which does not like the power saving (noise): https://bugzilla.redhat.com/show_bug.cgi?id=1689623 Also, reorder the Lenovo C50 entry to keep the table sorted. Reported-by:
<hs.guimaraes@outlook.com> Signed-off-by:
Jaroslav Kysela <perex@perex.cz> Cc: <stable@vger.kernel.org> Signed-off-by:
-
- Feb 01, 2019
-
-
Takashi Iwai authored
In the current code, the codec registration may happen both at the codec bind time and the end of the controller probe time. In a rare occasion, they race with each other, leading to Oops due to the still uninitialized card device. This patch introduces a simple flag to prevent the codec registration at the codec bind time as long as the controller probe is going on. The controller probe invokes snd_card_register() that does the whole registration task, and we don't need to register each piece beforehand. Cc: <stable@vger.kernel.org> Signed-off-by:
-
- Jan 01, 2019
-
-
Takashi Iwai authored
This essentially reverts the commits c337104b ("ALSA: HD-Audio: SKL+: abort probe if DSP is present and Skylake driver selected") and d82b51c8 ("ALSA: HD-Audio: SKL+: force HDaudio legacy or SKL+ driver selection") for the path of legacy HD-audio controller (snd-hda-intel). The automatic DSP detection and skip of binding with the legacy driver caused regressions on several machines like Dell XPS13. They give the PCI class 0x40380 indicating the availability of DSP while they don't work with ASoC SKL driver (yet). As the support of ASoC driver for such devices isn't available, it's better to revert the whole DSP-detection-and-skip behavior of the legacy driver, so that we can get the old good driver working on such devices. The pci_binding option for ASoC SKL driver is still kept so that it can work without blacklisting. Fixes: c337104b ("ALSA: HD-Audio: SKL+: abort probe if DSP is present and Skylake driver selected") Reported-by:
Linus Torvalds <torvalds@linux-foundation.org> Reported-by:
Hans de Goede <hdegoede@redhat.com> Reported-by:
Azat Khuzhin <dohardgopro@gmail.com> Cc: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com> Signed-off-by:
-
- Dec 19, 2018
-
-
Pierre-Louis Bossart authored
For HDaudio and Skylake drivers, add module parameter "pci_binding" When pci_binding == 0 (AUTO), the PCI class/subclass info is used to select drivers based on the presence of the DSP. pci_binding == 1 (LEGACY) forces the use of the HDAudio legacy driver, even if the DSP is present. pci_binding == 2 (ASOC) forces the use of the ASOC driver. The information on the DSP presence is bypassed. The value for the module parameter needs to be identical for both drivers. This parameter is intended as a back-up solution if the automatic detection fails or when the DSP usage fails. Such cases should be reported on the alsa-devel mailing list for analysis. Signed-off-by:
-
Pierre-Louis Bossart authored
Now that the SST/Skylake driver supports per platform selectors, we can add logic to automatically select the right driver. If the Skylake driver is selected for a specific platform, and the DSP is detected at run-time based on the PCI class/subclass/prog-if information, the legacy HDaudio driver aborts the probe. This will result in a single driver probing and remove the need for modprobe blacklists. Follow-up patches will add a module parameter to bypass the logic if this automatic detection fails, or if the Skylake driver is unable to actually support the platform (firmware authentication, missing topology file, hardware issue, etc). The same mechanism will be used to conflicts generated by the same PCI ID being registered by both legacy HDAuudio and SOF drivers for Intel platforms. In other words SOF will not require changes to the HDaudio legacy. Signed-off-by:
-
- Dec 11, 2018
-
-
Takashi Iwai authored
After the recent refactoring, snd_hdac_display_power() doesn't return any error, hence it can be defined to return void. This makes many error checks redundant and allows us to reduce them gracefully. Signed-off-by: -
Takashi Iwai authored
When an error occurs in azx_probe_continue(), we should release the display power. However, the current code ignores it and releases the display power only for HSW/BDW cases. Fix it. Signed-off-by: -
Takashi Iwai authored
snd_hdac_display_power() can be called even for a HDA controller without DRM binding. The same is true for other helpers, snd_hdac_i915_set_bclk() and snd_hdac_set_codec_wakeup(). So all superfluous AZX_DCAPS_I915_POWERWELL checks in hda_intel.c can be dropped, and the definition of AZX_DCAPS_I915_POWERWELL itself can be removed as well. This simplifies the code a lot. Signed-off-by: -
Takashi Iwai authored
The current HD-audio code manages the DRM audio power via too complex redirections, and this seems even still unbalanced in a corner case as Intel DRM CI has been intermittently reporting. This patch is a big surgery for addressing the complexity and the possible unbalance. Basically the patch changes the display PM in the following ways: - Both HD-audio controller and codec drivers call a single helper, snd_hdac_display_power(). (Formerly, the display power control from a codec was done indirectly via link_power bus ops.) - snd_hdac_display_power() receives the codec address index. For turning on/off from the controller, pass HDA_CODEC_IDX_CONTROLLER. - snd_hdac_display_power() doesn't manage refcounts any longer, but keeps the power status in bitmap. If any of controller or codecs is turned on, the function updates the DRM power state via get_power() or put_power(). Also this refactor allows us more cleanup: - The link_power bus ops is dropped, so there is no longer indirect management, as mentioned in the above. - hdac_device link_power_control flag is moved to hda_codec display_power_control flag, as it's only for HDA legacy. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=106525 Signed-off-by:
-
- Dec 09, 2018
-
-
Takashi Iwai authored
Make unified suspend / resume helpers and call them from both the runtime- and the system-PM callbacks for simplifying code. There are slight changes of call orders, but there shouldn't be any functional difference after refactoring. Signed-off-by:
-
- Nov 29, 2018
-
-
Kai-Heng Feng authored
It's similar to other AMD audio devices, it also supports D3, which can save some power drain. Signed-off-by:
Kai-Heng Feng <kai.heng.feng@canonical.com> Cc: <stable@vger.kernel.org> Signed-off-by:
-
- Nov 23, 2018
-
-
Hans de Goede authored
Power-saving is causing plops on audio start/stop on the built-in audio of the nForce 430 based ASRock N68C-S UCC motherboard, add this model to the power_save blacklist. BugLink: https://bugzilla.redhat.com/show_bug.cgi?id=1525104 Cc: <stable@vger.kernel.org> Signed-off-by:
-
- Oct 16, 2018
-
-
Hans de Goede authored
Power-saving is causing plops on audio start/stop on Dell Precision T3600 laptops and Intel DZ77BH boards, add these to the power_save blacklist. BugLink: https://bugzilla.redhat.com/show_bug.cgi?id=1525104 Signed-off-by:
-
- Sep 13, 2018
-
-
Takashi Iwai authored
The recent change of vga_switcheroo allowed the runtime PM for HD-audio on AMD GPUs, but this also resulted in a regression. When the HD-audio controller driver gets runtime-suspended, HD-audio link is turned off, and the hotplug notification is ignored. This leads to the inconsistent audio state (the connection isn't notified and ELD is ignored). The best fix would be to implement the proper ELD notification via the audio component, but it's still not ready. As a quick workaround, this patch adds the check of runtime_idle and allows the runtime suspend only when the vga_switcheroo is bound with discrete GPU. That is, a system with a single GPU and APU would be again without runtime PM to keep the HD-audio link for the hotplug notification and ELD read out. Also, the codec->auto_runtime_pm flag is set only for the discrete GPU at the time GPU gets bound via vga_switcheroo (i.e. only dGPU is forcibly runtime-PM enabled), so that APU can still get the ELD notification. For identifying which GPU is bound, a new vga_switcheroo client callback, gpu_bound, is implemented. The vga_switcheroo simply calls this when GPU is bound, and tells whether it's dGPU or APU. Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=200945 Fixes: 07f4f97d ("vga_switcheroo: Use device link for HDA controller") Reported-by:
Jian-Hong Pan <jian-hong@endlessm.com> Tested-by:
Lukas Wunner <lukas@wunner.de> Signed-off-by:
-
- Aug 30, 2018
-
-
Pierre-Louis Bossart authored
As suggested by Takashi, move this header file to make it easier to include from e.g. the Intel Skylake driver in follow-up patches Signed-off-by:
Mark Brown <broonie@kernel.org>
-
Takashi Iwai authored
Currently the jackpoll_ms option value is passed indirectly by referring to an array in chip->jackpoll_ms although each card needs to see only the assigned value. Also, the sanity check is done at each time in get_jackpoll_interval() although basically jackpoll_ms option is a read-only, hence we need to evaluate only once at probe time. This patch is the code simplification about the above points: the jack polling interval is directly set to chip->jackpoll_interval so that it can be simply copied to each codec. No functional change but only code reduction. Signed-off-by:
-
- Aug 28, 2018
-
-
Pierre-Louis Bossart authored
As suggested by Takashi, move this header file to make it easier to include from e.g. the Intel Skylake driver in follow-up patches Signed-off-by:
-
Takashi Iwai authored
Since we dropped the memory page fiddling in the own allocators in hda_intel.c, the substream allocation and free ops in both hda_intel.c and hda_tegra.c became nothing but the simple calls of the standard snd_pcm_lib helpers. As both are identical, there is no longer need for indirect calls via ops; it's a good opportunity for removing ops and simplifying the codes. Signed-off-by: -
Takashi Iwai authored
Now the ALSA memory allocator helper supports the new non-cached pages, let's use the new type, SNDRV_DMA_TYPE_DEV_UC_SG, for HD-audio driver. This allows us to reduce lots of codes. As another positive side-effect by this patch, the long-standing issue with non-snoop mode playing in the non-mmap mode is fixed. The core memalloc helper does the proper pgprot setup for non-cached pages for vmap(), which was missing in the past. Reported-and-tested-by:
Hans Hu <HansHu@zhaoxin.com> Signed-off-by:
-
Takashi Iwai authored
Introduce a new flag, uc_buffer, to indicate that the controller requires the non-cached pages for stream buffers, either as a chip-specific requirement or specified via snoop=0 option. This improves the code-readability. Also, this patch fixes the incorrect behavior for C-Media chip where the stream buffers were never handled as non-cached due to the check of driver_type even if you pass snoop=0 option. Signed-off-by:
-
- Aug 02, 2018
-
-
Hans de Goede authored
I added the subsys product-id for the HDMI HDA device rather then for the PCH one, this commit fixes this. BugLink: https://bugzilla.redhat.com/show_bug.cgi?id=1525104 Cc: stable@vger.kernel.org Signed-off-by:
-
- Jul 17, 2018
-
-
Jim Qu authored
On modern laptop, there are more and more platforms have two GPUs, and each of them maybe have audio codec for HDMP/DP output. For some dGPU which is no output, audio codec usually is disabled. In currect HDA audio driver, it will set all codec as VGA_SWITCHEROO_DIS, the audio which is binded to UMA will be suspended if user use debugfs to contorl power In HDA driver side, it is difficult to know which GPU the audio has binded to. So set the bound gpu pci dev to vga_switcheroo. if the audio client is not the third registration, audio id will set in vga_switcheroo enable function. if the audio client is the last registration when vga_switcheroo _ready() get true, we should get audio client id from bound GPU directly. Signed-off-by:
Jim Qu <Jim.Qu@amd.com> Reviewed-by:
-
- Jul 16, 2018
-
-
Jim Qu authored
Except PCI_CLASS_DISPLAY_VGA, some PCI class is sometimes PCI_CLASS_DISPLAY_3D or PCI_CLASS_DISPLAY_OTHER. Signed-off-by:
-
- Jun 28, 2018
-
-
Kai-Heng Feng authored
This patch can make audio controller in AMD Raven Ridge gets runtime suspended to D3, to save ~1W power when it's not in use. Cc: Vijendar Mukunda <Vijendar.Mukunda@amd.com> Signed-off-by:
-
- May 29, 2018
-
-
Hans de Goede authored
Power-saving is causing a humming sound when active on the Intel NUC5i7RY, add it to the blacklist. BugLink: https://bugzilla.kernel.org/show_bug.cgi?id=199607 Signed-off-by:
-
- May 23, 2018
-
-
Hans de Goede authored
Power-saving is causing plops on audio start/stop on ASRock H81M-HDS machines, add these to the power_save blacklist. BugLink: https://bugzilla.redhat.com/show_bug.cgi?id=1525104 Signed-off-by:
-
Hans de Goede authored
Power-saving is causing plops on audio start/stop on Gigabyte P55A-UD3 and Gigabyte Z87-D3HP machines, add these to the power_save blacklist. Note these 2 boards both use 1458:a002 as subsystem ids, so they share a single entry. BugLink: https://bugzilla.redhat.com/show_bug.cgi?id=1525104 Signed-off-by:
-
Hans de Goede authored
Power-saving is causing a plop and silences the first 2 seconds (give or take) of audio, silencing notifications sounds on Medion / Clevo W35xSS_370SS laptops. Add the Clevo W35xSS_370SS to the power_save blacklist. BugLink: https://bugzilla.redhat.com/show_bug.cgi?id=1581607 Signed-off-by:
-
Hans de Goede authored
Power-saving is causing a humming sound when active on the Intel NUC7i3BNB, add it to the blacklist. BugLink: https://bugzilla.redhat.com/show_bug.cgi?id=1520902 Signed-off-by:
-
- May 13, 2018
-
-
Hans de Goede authored
Power-saving is causing loud plops on the Lenovo C50 All in one, add it to the blacklist. BugLink: https://bugzilla.redhat.com/show_bug.cgi?id=1572975 Signed-off-by:
-
- Apr 16, 2018
-
-
David Wang authored
This patch is used to tell kernel that new VIA HDAC controller also support no-snoop path. [ minor coding style fix by tiwai ] Signed-off-by:
David Wang <davidwang@zhaoxin.com> Cc: <stable@vger.kernel.org> Signed-off-by:
-
- Mar 29, 2018
-
-
Lukas Wunner authored
The system sleep PM ops azx_suspend() and azx_resume() were previously called by vga_switcheroo, but commit 07f4f97d ("vga_switcheroo: Use device link for HDA controller") removed their invocation. Unfortunately the commit neglected to update the #ifdef surrounding the two functions, so if CONFIG_PM_SLEEP is *not* enabled but all three of CONFIG_PM, CONFIG_VGA_SWITCHEROO and CONFIG_SND_HDA_CODEC_HDMI *are* enabled, the compiler now emits the following warning: sound/pci/hda/hda_intel.c:1024:12: warning: 'azx_resume' defined but not used [-Wunused-function] static int azx_resume(struct device *dev) ^~~~~~~~~~ sound/pci/hda/hda_intel.c:989:12: warning: 'azx_suspend' defined but not used [-Wunused-function] static int azx_suspend(struct device *dev) ^~~~~~~~~~~ Silence by updating the #ifdef. Because the #ifdef block now uses the same condition as the one immediately succeeding it, the two blocks can be collapsed together, shaving off another two lines. Fixes: 07f4f97d ("vga_switcheroo: Use device link for HDA controller") Reviewed-by:
Arnd Bergmann <arnd@arndb.de> Signed-off-by:
-
- Mar 21, 2018
-
-
Takashi Iwai authored
We've observed too long probe time with Coffee Lake (CFL) machines, and the likely cause is some communication problem between the HD-audio controller and the codec chips. While the controller expects an IRQ wakeup for each codec response, it seems sometimes missing, and it takes one second for the controller driver to time out and read the response in the polling mode. Although we aren't sure about the real culprit yet, in this patch, we put a workaround by forcing the polling mode as default for CFL machines; the polling mode itself isn't too heavy, and much better than other workarounds initially suggested (e.g. disabling power-save), at least. Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=199007 Fixes: e79b0006 ("ALSA: hda - Add Coffelake PCI ID") Reported-and-tested-by:
-
- Mar 13, 2018
-
-
Lukas Wunner authored
Back in 2013, runtime PM for GPUs with integrated HDA controller was introduced with commits 0d69704a ("gpu/vga_switcheroo: add driver control power feature. (v3)") and 246efa4a ("snd/hda: add runtime suspend/resume on optimus support (v4)"). Briefly, the idea was that the HDA controller is forced on and off in unison with the GPU. The original code is mostly still in place even though it was never a 100% perfect solution: E.g. on access to the HDA controller, the GPU is powered up via vga_switcheroo_runtime_resume_hdmi_audio() but there are no provisions to keep it resumed until access to the HDA controller has ceased: The GPU autosuspends after 5 seconds, rendering the HDA controller inaccessible. Additionally, a kludge is required when hda_intel.c probes: It has to check whether the GPU is powered down (check_hdmi_disabled()) and defer probing if so. However in the meantime (in v4.10) the driver core has gained a feature called device links which promises to solve such issues in a clean way: It allows us to declare a dependency from the HDA controller (consumer) to the GPU (supplier). The PM core then automagically ensures that the GPU is runtime resumed as long as the HDA controller's ->probe hook is executed and whenever the HDA controller is accessed. By default, the HDA controller has a dependency on its parent, a PCIe Root Port. Adding a device link creates another dependency on its sibling: PCIe Root Port ^ ^ | | | | HDA ===> GPU The device link is not only used for runtime PM, it also guarantees that on system sleep, the HDA controller suspends before the GPU and resumes after the GPU, and on system shutdown the HDA controller's ->shutdown hook is executed before the one of the GPU. It is a complete solution. Using this functionality is as simple as calling device_link_add(), which results in a dmesg entry like this: pci 0000:01:00.1: Linked as a consumer to 0000:01:00.0 The code for the GPU-governed audio power management can thus be removed (except where it's still needed for legacy manual power control). The device link is added in a PCI quirk rather than in hda_intel.c. It is therefore legal for the GPU to runtime suspend to D3cold even if the HDA controller is not bound to a driver or if CONFIG_SND_HDA_INTEL is not enabled, for accesses to the HDA controller will cause the GPU to wake up regardless if they're occurring outside of hda_intel.c (think config space readout via sysfs). Contrary to the previous implementation, the HDA controller's power state is now self-governed, rather than GPU-governed, whereas the GPU's power state is no longer fully self-governed. (The HDA controller needs to runtime suspend before the GPU can.) It is thus crucial that runtime PM is always activated on the HDA controller even if CONFIG_SND_HDA_POWER_SAVE_DEFAULT is set to 0 (which is the default), lest the GPU stays awake. This is achieved by setting the auto_runtime_pm flag on every codec and the AZX_DCAPS_PM_RUNTIME flag on the HDA controller. A side effect is that power consumption might be reduced if the GPU is in use but the HDA controller is not, because the HDA controller is now allowed to go to D3hot. Before, it was forced to stay in D0 as long as the GPU was in use. (There is no reduction in power consumption on my Nvidia GK107, but there might be on other chips.) The code paths for legacy manual power control are adjusted such that runtime PM is disabled during power off, thereby preventing the PM core from resuming the HDA controller. Note that the device link is not only added on vga_switcheroo capable systems, but for *any* GPU with integrated HDA controller. The idea is that the HDA controller streams audio via connectors located on the GPU, so the GPU needs to be on for the HDA controller to do anything useful. This commit implicitly fixes an unbalanced runtime PM ref upon unbind of hda_intel.c: On ->probe, a runtime PM ref was previously released under the condition "azx_has_pm_runtime(chip) || hda->use_vga_switcheroo", but on ->remove a runtime PM ref was only acquired under the first of those conditions. Thus, binding and unbinding the driver twice on a vga_switcheroo capable system caused the runtime PM refcount to drop below zero. The issue is resolved because the AZX_DCAPS_PM_RUNTIME flag is now always set if use_vga_switcheroo is true. For more information on device links please refer to: https://www.kernel.org/doc/html/latest/driver-api/device_link.html Documentation/driver-api/device_link.rst Cc: Dave Airlie <airlied@redhat.com> Cc: Ben Skeggs <bskeggs@redhat.com> Cc: Alex Deucher <alexander.deucher@amd.com> Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by:
Bjorn Helgaas <bhelgaas@google.com> Reviewed-by:
Peter Wu <peter@lekensteyn.nl> Tested-by: Kai Heng Feng <kai.heng.feng@canonical.com> # AMD PowerXpress Tested-by: Mike Lothian <mike@fireburn.co.uk> # AMD PowerXpress Tested-by: Denis Lisov <dennis.lissov@gmail.com> # Nvidia Optimus Tested-by: Peter Wu <peter@lekensteyn.nl> # Nvidia Optimus Tested-by: Lukas Wunner <lukas@wunner.de> # MacBook Pro Signed-off-by:
-
Guneshwor Singh authored
Icelake is a next generation Intel platform. Add PCI ID for it. Signed-off-by:
Guneshwor Singh <guneshwor.o.singh@intel.com> Signed-off-by:
-
- Mar 12, 2018
-
-
Takashi Iwai authored
With the commit 1ba8f9d3 ("ALSA: hda: Add a power_save blacklist"), we changed the default value of power_save option to -1 for processing the power-save blacklist. Unfortunately, this seems breaking user-space applications that actually read the power_save parameter value via sysfs and judge / adjust the power-saving status. They see the value -1 as if the power-save is turned off, although the actual value is taken from CONFIG_SND_HDA_POWER_SAVE_DEFAULT and it can be a positive. So, overall, passing -1 there was no good idea. Let's partially revert it -- at least for power_save option default value is restored again to CONFIG_SND_HDA_POWER_SAVE_DEFAULT. Meanwhile, in this patch, we keep the blacklist behavior and make is adjustable via the new option, pm_blacklist. Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=199073 Fixes: 1ba8f9d3 ("ALSA: hda: Add a power_save blacklist") Acked-by:
-
