From: Adrian Hunter <adrian.hunter@intel.com>
To: Hans de Goede <hdegoede@redhat.com>,
Ulf Hansson <ulf.hansson@linaro.org>
Cc: "russianneuromancer @ ya . ru" <russianneuromancer@ya.ru>,
linux-mmc@vger.kernel.org
Subject: Re: [PATCH 1/2] mmc: sdhci-acpi: Disable 1.8V modes on external microSD on Lenovo Miix 320
Date: Wed, 15 Jan 2020 14:57:40 +0200 [thread overview]
Message-ID: <61bc9265-ece0-eeb6-d4a1-4631138ecf29@intel.com> (raw)
In-Reply-To: <20200108093903.57620-2-hdegoede@redhat.com>
On 8/01/20 11:39 am, Hans de Goede wrote:
> Based on a sample of 7 DSDTs from Cherry Trail devices using an AXP288
> PMIC depending on the design one of 2 possible LDOs on the PMIC is used
> for the MMC signalling voltage, either DLDO3 or GPIO1LDO (GPIO1 pin in
> low noise LDO mode).
>
> The Lenovo Miix 320-10ICR uses GPIO1LDO in the SHC1 ACPI device's DSM
> methods to set 3.3 or 1.8 signalling voltage and this appears to work
> as advertised, so presumably the device is actually using GPIO1LDO for
> the external microSD signalling voltage.
>
> But this device has a bug in the _PS0 method of the SHC1 ACPI device,
> the DSM remembers the last set signalling voltage and the _PS0 restores
> this after a (runtime) suspend-resume cycle, but it "restores" the voltage
> on DLDO3 instead of setting it on GPIO1LDO as the DSM method does. DLDO3
> is used for the LCD and setting it to 1.8V causes the LCD to go black.
>
> This issue can be worked around by setting the SDHCI_QUIRK2_NO_1_8_V
> quirk on the sdhci_host so that the DSM never gets used to program the
> signalling voltage to 1.8V.
Could you instead call the 3.3V DSM at runtime suspend time, then the _PS0
would not "restore" the 1.8V value? That should allow you to use 1.8V UHS-I
speed modes with SD cards that support them.
>
> So far we have mostly been able to avoid using device specific quirks in
> the sdhci-acpi code, but given that this issue is specific to this one
> model and we certainly do not want to disable 1.8V modes everywhere I
> see no other option.
>
> This commit adds a new mechanism for setting sdhci-acpi specific quirks
> and a matching sdhci-acpi.quirks module parameter to make testing quirks /
> similar issues on other devices easier.
>
> The first quirk supported by this mechanism is SDHCI_ACPI_QUIRK_SD_NO_1_8V,
> which when set causes any slots with the SDHCI_ACPI_SD_CD flag to get the
> SDHCI_QUIRK2_NO_1_8_V quirk set on their sdhci_host.
>
> This commit also adds a DMI table for specifying default quirks for some
> models and adds an entry for the Lenovo Miix 320-10ICR which enables the
> SDHCI_QUIRK2_NO_1_8_V by default on this model, fixing the LCD going black
> when the external microSD slot is used.
>
> BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=111294
> BugLink: https://gitlab.freedesktop.org/drm/intel/issues/355
> Reported-by: russianneuromancer <russianneuromancer@ya.ru>
> Signed-off-by: Hans de Goede <hdegoede@redhat.com>
> ---
> drivers/mmc/host/sdhci-acpi.c | 39 +++++++++++++++++++++++++++++++++++
> 1 file changed, 39 insertions(+)
>
> diff --git a/drivers/mmc/host/sdhci-acpi.c b/drivers/mmc/host/sdhci-acpi.c
> index 105e73d4a3b9..9f150c73e958 100644
> --- a/drivers/mmc/host/sdhci-acpi.c
> +++ b/drivers/mmc/host/sdhci-acpi.c
> @@ -23,6 +23,7 @@
> #include <linux/pm.h>
> #include <linux/pm_runtime.h>
> #include <linux/delay.h>
> +#include <linux/dmi.h>
>
> #include <linux/mmc/host.h>
> #include <linux/mmc/pm.h>
> @@ -75,6 +76,14 @@ struct sdhci_acpi_host {
> unsigned long private[0] ____cacheline_aligned;
> };
>
> +enum {
> + SDHCI_ACPI_QUIRK_SD_NO_1_8V = BIT(0),
> +};
> +
> +static int quirks = -1;
> +module_param(quirks, int, 0444);
> +MODULE_PARM_DESC(quirks, "Override sdhci-acpi specific quirks");
Why is a module parameter needed?
> +
> static inline void *sdhci_acpi_priv(struct sdhci_acpi_host *c)
> {
> return (void *)c->private;
> @@ -647,6 +656,24 @@ static const struct acpi_device_id sdhci_acpi_ids[] = {
> };
> MODULE_DEVICE_TABLE(acpi, sdhci_acpi_ids);
>
> +static const struct dmi_system_id sdhci_acpi_quirks[] = {
> + {
> + /*
> + * The Lenovo Miix 320-10ICR has a bug in the _PS0 method of
> + * the SHC1 ACPI device, this bug causes it to reprogram the
> + * wrong LDO (DLDO3) to 1.8V if 1.8V modes are used and the
> + * card is (runtime) suspended + resumed. DLDO3 is used for
> + * the LCD and setting it to 1.8V causes the LCD to go black.
> + */
> + .matches = {
> + DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"),
> + DMI_MATCH(DMI_PRODUCT_VERSION, "Lenovo MIIX 320-10ICR"),
> + },
> + .driver_data = (void *)SDHCI_ACPI_QUIRK_SD_NO_1_8V,
> + },
> + {} /* Terminating entry */
> +};
> +
> static const struct sdhci_acpi_slot *sdhci_acpi_get_slot(struct acpi_device *adev)
> {
> const struct sdhci_acpi_uid_slot *u;
> @@ -663,6 +690,7 @@ static int sdhci_acpi_probe(struct platform_device *pdev)
> struct device *dev = &pdev->dev;
> const struct sdhci_acpi_slot *slot;
> struct acpi_device *device, *child;
> + const struct dmi_system_id *id;
> struct sdhci_acpi_host *c;
> struct sdhci_host *host;
> struct resource *iomem;
> @@ -670,6 +698,14 @@ static int sdhci_acpi_probe(struct platform_device *pdev)
> size_t priv_size;
> int err;
>
> + if (quirks == -1) {
> + id = dmi_first_match(sdhci_acpi_quirks);
> + if (id)
> + quirks = (long)id->driver_data;
> + else
> + quirks = 0;
> + }
> +
> device = ACPI_COMPANION(dev);
> if (!device)
> return -ENODEV;
> @@ -759,6 +795,9 @@ static int sdhci_acpi_probe(struct platform_device *pdev)
> dev_warn(dev, "failed to setup card detect gpio\n");
> c->use_runtime_pm = false;
> }
> +
> + if (quirks & SDHCI_ACPI_QUIRK_SD_NO_1_8V)
> + host->quirks2 |= SDHCI_QUIRK2_NO_1_8_V;
> }
>
> err = sdhci_setup_host(host);
>
next prev parent reply other threads:[~2020-01-15 12:58 UTC|newest]
Thread overview: 24+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-01-08 9:39 [PATCH 0/2] mmc: sdhci-acpi: Introduce device specific quirks, fix issues on 2 device models Hans de Goede
2020-01-08 9:39 ` [PATCH 1/2] mmc: sdhci-acpi: Disable 1.8V modes on external microSD on Lenovo Miix 320 Hans de Goede
2020-01-15 12:57 ` Adrian Hunter [this message]
2020-01-15 13:31 ` Hans de Goede
2020-01-15 13:48 ` Adrian Hunter
2020-01-15 15:31 ` Hans de Goede
2020-01-16 7:59 ` Adrian Hunter
2020-01-16 11:05 ` [FSL P5020 P5040 PPC] Onboard SD card doesn't work anymore after the 'mmc-v5.4-2' updates Christian Zigotzky
2020-01-16 15:46 ` Ulf Hansson
2020-01-20 9:17 ` Christian Zigotzky
2020-01-20 11:18 ` Ulf Hansson
2020-01-24 11:42 ` Michael Ellerman
2020-01-25 13:26 ` Christian Zigotzky
2020-01-28 11:55 ` Michael Ellerman
2020-01-28 7:58 ` [PASEMI PA6T PPC] Onboard CF card device with new SanDisk High (>8G) CF cards Christian Zigotzky
2020-01-28 8:08 ` Christoph Hellwig
2020-01-28 14:16 ` Rob Herring
2020-01-28 14:48 ` Christian Zigotzky
2020-01-16 13:26 ` [PATCH 1/2] mmc: sdhci-acpi: Disable 1.8V modes on external microSD on Lenovo Miix 320 Hans de Goede
2020-01-17 9:16 ` Adrian Hunter
2020-03-06 14:07 ` Hans de Goede
2020-03-06 14:10 ` Hans de Goede
2020-01-08 9:39 ` [PATCH 2/2] mmc: sdhci-acpi: Disable write protect detection on Acer Aspire Switch 10 (SW5-012) Hans de Goede
2020-01-15 12:51 ` [PATCH 0/2] mmc: sdhci-acpi: Introduce device specific quirks, fix issues on 2 device models Adrian Hunter
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=61bc9265-ece0-eeb6-d4a1-4631138ecf29@intel.com \
--to=adrian.hunter@intel.com \
--cc=hdegoede@redhat.com \
--cc=linux-mmc@vger.kernel.org \
--cc=russianneuromancer@ya.ru \
--cc=ulf.hansson@linaro.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox