| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <CAGXv+5Hw+13J6eU6X3EPa+Vr4kmyNH5L83Zj2VLqx=_Z+XtOzw@mail.gmail.com>
Date: Tue, 1 Apr 2025 18:38:28 +0800
From: Chen-Yu Tsai <wenst@...omium.org>
To: AngeloGioacchino Del Regno <angelogioacchino.delregno@...labora.com>
Cc: Mark Brown <broonie@...nel.org>, Jaroslav Kysela <perex@...ex.cz>, Takashi Iwai <tiwai@...e.com>,
Matthias Brugger <matthias.bgg@...il.com>, linux-sound@...r.kernel.org,
linux-mediatek@...ts.infradead.org, linux-arm-kernel@...ts.infradead.org,
linux-kernel@...r.kernel.org, treapking@...omium.org, yuanhsinte@...omium.org
Subject: Re: [PATCH] ASoC: mediatek: re-enable buffer pre-allocation on some platforms
On Tue, Apr 1, 2025 at 6:12 PM AngeloGioacchino Del Regno
<angelogioacchino.delregno@...labora.com> wrote:
>
> Il 01/04/25 10:56, Chen-Yu Tsai ha scritto:
> > In commit 32c9c06adb5b ("ASoC: mediatek: disable buffer pre-allocation")
> > buffer pre-allocation was disabled to accommodate newer platforms that
> > have a limited reserved memory region for the audio frontend.
> >
> > Turns out disabling pre-allocation across the board impacts platforms
> > that don't have this reserved memory region. Buffer allocation failures
> > have been observed on MT8173 and MT8183 based Chromebooks under low
> > memory conditions, which results in no audio playback for the user.
> >
> > Reinstate the original policy of pre-allocating audio buffers at probe
> > time on MT8173, MT8183, and MT8186 platforms. These platforms do not
> > have reserved memory for the audio frontend.
> >
> > Fixes: 32c9c06adb5b ("ASoC: mediatek: disable buffer pre-allocation")
> > Signed-off-by: Chen-Yu Tsai <wenst@...omium.org>
>
> Have you tried to add the reserved memory regions to the MT8173/83/86 AFE?
I haven't tried it, though I'm fairly certain that would work as well.
> This means not only devicetree but also the driver - that should be simple
> enough, and would allow to eventually add a layer of further commonization
> between the AFE PCM drivers of all those MediaTek SoCs.
I think no driver changes are necessary? It's a single memory region,
so it would automatically be picked by the driver core for all coherent
DMA allocations.
I picked the driver route because it restores old behavior.
ChenYu
> Cheers,
> Angelo
>
> > ---
> > sound/soc/mediatek/common/mtk-afe-platform-driver.c | 4 +++-
> > sound/soc/mediatek/common/mtk-base-afe.h | 1 +
> > sound/soc/mediatek/mt8173/mt8173-afe-pcm.c | 1 +
> > sound/soc/mediatek/mt8183/mt8183-afe-pcm.c | 1 +
> > sound/soc/mediatek/mt8186/mt8186-afe-pcm.c | 1 +
> > 5 files changed, 7 insertions(+), 1 deletion(-)
> >
> > diff --git a/sound/soc/mediatek/common/mtk-afe-platform-driver.c b/sound/soc/mediatek/common/mtk-afe-platform-driver.c
> > index 6b6330583941..70fd05d5ff48 100644
> > --- a/sound/soc/mediatek/common/mtk-afe-platform-driver.c
> > +++ b/sound/soc/mediatek/common/mtk-afe-platform-driver.c
> > @@ -120,7 +120,9 @@ int mtk_afe_pcm_new(struct snd_soc_component *component,
> > struct mtk_base_afe *afe = snd_soc_component_get_drvdata(component);
> >
> > size = afe->mtk_afe_hardware->buffer_bytes_max;
> > - snd_pcm_set_managed_buffer_all(pcm, SNDRV_DMA_TYPE_DEV, afe->dev, 0, size);
> > + snd_pcm_set_managed_buffer_all(pcm, SNDRV_DMA_TYPE_DEV, afe->dev,
> > + afe->preallocate_buffers ? size : 0,
> > + size);
> >
> > return 0;
> > }
> > diff --git a/sound/soc/mediatek/common/mtk-base-afe.h b/sound/soc/mediatek/common/mtk-base-afe.h
> > index f51578b6c50a..a406f2e3e7a8 100644
> > --- a/sound/soc/mediatek/common/mtk-base-afe.h
> > +++ b/sound/soc/mediatek/common/mtk-base-afe.h
> > @@ -117,6 +117,7 @@ struct mtk_base_afe {
> > struct mtk_base_afe_irq *irqs;
> > int irqs_size;
> > int memif_32bit_supported;
> > + bool preallocate_buffers;
> >
> > struct list_head sub_dais;
> > struct snd_soc_dai_driver *dai_drivers;
> > diff --git a/sound/soc/mediatek/mt8173/mt8173-afe-pcm.c b/sound/soc/mediatek/mt8173/mt8173-afe-pcm.c
> > index 04ed0cfec174..37b20ec80829 100644
> > --- a/sound/soc/mediatek/mt8173/mt8173-afe-pcm.c
> > +++ b/sound/soc/mediatek/mt8173/mt8173-afe-pcm.c
> > @@ -1114,6 +1114,7 @@ static int mt8173_afe_pcm_dev_probe(struct platform_device *pdev)
> > afe->mtk_afe_hardware = &mt8173_afe_hardware;
> > afe->memif_fs = mt8173_memif_fs;
> > afe->irq_fs = mt8173_irq_fs;
> > + afe->preallocate_buffers = true;
> >
> > platform_set_drvdata(pdev, afe);
> >
> > diff --git a/sound/soc/mediatek/mt8183/mt8183-afe-pcm.c b/sound/soc/mediatek/mt8183/mt8183-afe-pcm.c
> > index d083b4bf0f95..0ef07fb2898d 100644
> > --- a/sound/soc/mediatek/mt8183/mt8183-afe-pcm.c
> > +++ b/sound/soc/mediatek/mt8183/mt8183-afe-pcm.c
> > @@ -1214,6 +1214,7 @@ static int mt8183_afe_pcm_dev_probe(struct platform_device *pdev)
> > afe->mtk_afe_hardware = &mt8183_afe_hardware;
> > afe->memif_fs = mt8183_memif_fs;
> > afe->irq_fs = mt8183_irq_fs;
> > + afe->preallocate_buffers = true;
> >
> > afe->runtime_resume = mt8183_afe_runtime_resume;
> > afe->runtime_suspend = mt8183_afe_runtime_suspend;
> > diff --git a/sound/soc/mediatek/mt8186/mt8186-afe-pcm.c b/sound/soc/mediatek/mt8186/mt8186-afe-pcm.c
> > index db7c93401bee..c588a68d672d 100644
> > --- a/sound/soc/mediatek/mt8186/mt8186-afe-pcm.c
> > +++ b/sound/soc/mediatek/mt8186/mt8186-afe-pcm.c
> > @@ -2935,6 +2935,7 @@ static int mt8186_afe_pcm_dev_probe(struct platform_device *pdev)
> > afe->irq_fs = mt8186_irq_fs;
> > afe->get_dai_fs = mt8186_get_dai_fs;
> > afe->get_memif_pbuf_size = mt8186_get_memif_pbuf_size;
> > + afe->preallocate_buffers = true;
> >
> > afe->runtime_resume = mt8186_afe_runtime_resume;
> > afe->runtime_suspend = mt8186_afe_runtime_suspend;
>
>
Powered by blists - more mailing lists