From: Ahmad Fatoum <a.fatoum@pengutronix.de>
To: barebox@lists.infradead.org
Cc: Michael Tretter <m.tretter@pengutronix.de>
Subject: Re: [PATCH master] mci: dw_mmc: drop CD from mask programmed into interrupt status register
Date: Mon, 9 Feb 2026 17:45:31 +0100 [thread overview]
Message-ID: <25889b48-22eb-4196-8dd0-8ba3875aab09@pengutronix.de> (raw)
In-Reply-To: <20260209164045.1924618-1-a.fatoum@pengutronix.de>
Hi,
On 2/9/26 5:40 PM, Ahmad Fatoum wrote:
> We don't usually care for the card detect in barebox and just
> unconditionally try to access the card. If it fails, a probe can be
> attempted later via setting mmcX.probe=1.
>
> The dw_mmc driver diverges from this and enables all kinds
> of interrupts in the interrupt status polled after submitting commands,
> including card-detect.
>
> This appears to be problematic, when we have a GPIO card detect and are
> not using the card detect hardware function.
>
> Align behavior with Linux and skip polling for hardware card-detect when
> we have a GPIO card detect. The current behavior was inherited from
> U-Boot, which still does it this way and it's possible it would be affected
> by this issue as well, once U-Boot updates to use DTs >= v6.19-rc1, provided
> that the ROCK 5 support in U-Boot uses CONFIG_OF_UPSTREAM.
>
> Fixes: a3bf6c16f77d ("dts: update to v6.19-rc1")
> Reported-by: Michael Tretter <m.tretter@pengutronix.de>
> Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
> ---
I should have added: This is only commpile-tested. Would be great if
someone (Michael?) could give this a boot on a Radxa ROCK 5.
Thanks,
Ahmad
> drivers/mci/dw_mmc.c | 33 ++++++++++++++++++++++++++++++++-
> drivers/mci/dw_mmc.h | 1 +
> 2 files changed, 33 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/mci/dw_mmc.c b/drivers/mci/dw_mmc.c
> index 30e5d56944b6..9ddf931eb513 100644
> --- a/drivers/mci/dw_mmc.c
> +++ b/drivers/mci/dw_mmc.c
> @@ -12,6 +12,7 @@
> #include <driver.h>
> #include <malloc.h>
> #include <clock.h>
> +#include <linux/gpio/consumer.h>
> #include <init.h>
> #include <mci.h>
> #include <io.h>
> @@ -23,6 +24,20 @@
> #include <errno.h>
> #include "dw_mmc.h"
>
> +/* Most importantly, it doesn't contain DWMCI_INTMSK_CD as to allow
> + * for GPIO card detect, unlike DWMCI_INTMSK_ALL.
> + * The list is taken from Linux v6.19 with addition of DWMCI_INTMSK_FRUN
> + * as barebox had inherited this into DWMCI_DATA_ERR from U-Boot
> + */
> +#define DW_MCI_DATA_ERROR_FLAGS (DWMCI_INTMSK_DRTO | DWMCI_INTMSK_DCRC | \
> + DWMCI_INTMSK_HTO | DWMCI_INTMSK_SBE | \
> + DWMCI_INTMSK_EBE | DWMCI_INTMSK_HLE | \
> + DWMCI_INTMSK_FRUN)
> +#define DW_MCI_CMD_ERROR_FLAGS (DWMCI_INTMSK_RTO | DWMCI_INTMSK_RCRC | \
> + DWMCI_INTMSK_RE | DWMCI_INTMSK_HLE)
> +
> +#define DW_MCI_ERROR_FLAGS (DW_MCI_DATA_ERROR_FLAGS | \
> + DW_MCI_CMD_ERROR_FLAGS)
>
> struct dwmci_host {
> struct mci_host mci;
> @@ -37,6 +52,7 @@ struct dwmci_host {
> u32 fifoth_val;
> u32 pwren_value;
> dma_addr_t idmac_dma;
> + struct gpio_desc *cd_gpio;
> };
>
> struct dwmci_idmac {
> @@ -285,7 +301,13 @@ dwmci_cmd(struct mci_host *mci, struct mci_cmd *cmd, struct mci_data *data)
> }
> }
>
> - dwmci_writel(host, DWMCI_RINTSTS, DWMCI_INTMSK_ALL);
> + /*
> + * Enable interrupts for command done, data over, data empty,
> + * receive ready and error such as transmit, receive timeout, crc error
> + */
> + dwmci_writel(host, DWMCI_RINTSTS, DWMCI_INTMSK_CDONE | DWMCI_INTMSK_DTO |
> + DWMCI_INTMSK_TXDR | DWMCI_INTMSK_RXDR |
> + DW_MCI_ERROR_FLAGS);
>
> if (data) {
>
> @@ -586,6 +608,15 @@ static int dw_mmc_probe(struct device *dev)
> reset_control_deassert(rst);
> }
>
> + /*
> + * We don't actually care for the cd-gpios and it's usually connected
> + * just to a dumb lever shorting to ground, but let's mux as GPIO
> + * to be on the safe side.
> + */
> + host->cd_gpio = gpiod_get_optional(dev, "cd", GPIOD_IN);
> + if (IS_ERR(host->cd_gpio))
> + dev_dbg(dev, "Failed to get cd-gpios: %pe (ignoring)\n", host->cd_gpio);
> +
> iores = dev_request_mem_resource(dev, 0);
> if (IS_ERR(iores))
> return PTR_ERR(iores);
> diff --git a/drivers/mci/dw_mmc.h b/drivers/mci/dw_mmc.h
> index 23fa116d7514..eb5331d1ccb2 100644
> --- a/drivers/mci/dw_mmc.h
> +++ b/drivers/mci/dw_mmc.h
> @@ -53,6 +53,7 @@
>
> /* Interrupt Mask register */
> #define DWMCI_INTMSK_ALL 0xffffffff
> +#define DWMCI_INTMSK_CD BIT(0)
> #define DWMCI_INTMSK_RE BIT(1)
> #define DWMCI_INTMSK_CDONE BIT(2)
> #define DWMCI_INTMSK_DTO BIT(3)
--
Pengutronix e.K. | |
Steuerwalder Str. 21 | http://www.pengutronix.de/ |
31137 Hildesheim, Germany | Phone: +49-5121-206917-0 |
Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 |
next prev parent reply other threads:[~2026-02-09 16:46 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-02-09 16:40 Ahmad Fatoum
2026-02-09 16:45 ` Ahmad Fatoum [this message]
2026-02-10 8:29 ` Michael Tretter
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=25889b48-22eb-4196-8dd0-8ba3875aab09@pengutronix.de \
--to=a.fatoum@pengutronix.de \
--cc=barebox@lists.infradead.org \
--cc=m.tretter@pengutronix.de \
/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