From: Lukasz Majewski <l.majewski@majess.pl>
To: u-boot@lists.denx.de
Subject: [U-Boot] [PATCH] arm: trats: add the sd-card power control function.
Date: Sun, 3 Nov 2013 23:24:32 +0100 [thread overview]
Message-ID: <20131103232432.30ef8d9e@jawa> (raw)
In-Reply-To: <5272FD97.9080300@samsung.com>
Hi Jaehoon,
> Hi Lukasz,
>
> Thanks for your comments.
>
> On 11/01/2013 02:03 AM, Lukasz Majewski wrote:
> > Hi Jaehoon,
> >
> >> To use the sd-card, VTF_2.8V will be enabled.
> >> Before this patch, VTF_2.8V is always disabled.(card can't
> >> initialize.) When card is detected, SD-card power will enable with
> >> sd_ldo_control().
> >>
> >> Signed-off-by: Jaehoon Chung <jh80.chung@samsung.com>
> >> Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
> >> ---
> >> board/samsung/trats/trats.c | 38
> >> ++++++++++++++++++++++++++++++++------ 1 file changed, 32
> >> insertions(+), 6 deletions(-)
> >>
> >> diff --git a/board/samsung/trats/trats.c
> >> b/board/samsung/trats/trats.c index 7f61d17..9b6dd12 100644
> >> --- a/board/samsung/trats/trats.c
> >> +++ b/board/samsung/trats/trats.c
> >> @@ -408,11 +408,31 @@ int checkboard(void)
> >> #endif
> >>
> >> #ifdef CONFIG_GENERIC_MMC
> >> +static int sd_ldo_control(int on)
> >> +{
> >> + struct pmic *p = pmic_get("MAX8997_PMIC");
> >> + u32 val = 0;
> >> + int ret = 0;
> >> + if (pmic_probe(p))
> >> + return -1;
> >
> > Maybe -1 -> -ENODEV
> Will fix..
> >
> >> +
> >> + /* LDO17 VTF: 2.8V */
> >> + val = max8997_reg_ldo(2800000) | (on? EN_LDO : DIS_LDO);
> > ^^^ on ?
> Will fix
> >> + ret |= pmic_reg_write(p, MAX8997_REG_LDO17CTRL, val);
> > ^^^^^ ret =
> Will fix.
> >> +
> >> + if (ret) {
> >> + puts("MAX8997 SD setting error!\n");
> >
> > Please use error() instead of puts()
> Using error().
> >
> >> + return -1;
> >> + }
> >> +
> >> + return 0;
> >> +}
> >> +
> >> int board_mmc_init(bd_t *bis)
> >> {
> >> struct exynos4_gpio_part2 *gpio =
> >> (struct exynos4_gpio_part2
> >> *)samsung_get_base_gpio_part2();
> >> - int err;
> >> + int err, err_sd = 0;
> >
> > I think that the err_sd is not needed here. Please use the err as is
> > was at the original code.
> As like the goni, I think sd_err needs herre.
> We need to consider the cases when eMMC is intiailzed and SD-card is
> failed. If SD-card is failed, then it's always retunred "Fail",
> although eMMC is initialized. how about?
I think that it is perfectly correct when we insert SD card (which is
signaled by change of card detect pin - X3[4]) and when we fail to
initialize it to display the error message and pass the error code out
of the function.
Then user can read the error cause from the console and perform hard
reset with SD card removed.
> After replying your, i will send the fixing patch.
>
> Best Regards,
> Jaehoon Chung
> >
> >>
> >> /* eMMC_EN: SD_0_CDn: GPK0[2] Output High */
> >> s5p_gpio_direction_output(&gpio->k0, 2, 1);
> >> @@ -438,14 +458,20 @@ int board_mmc_init(bd_t *bis)
> >> * GPX3[4] T-flash detect pin
> >> */
> >> if (!s5p_gpio_get_value(&gpio->x3, 4)) {
> >> - err = exynos_pinmux_config(PERIPH_ID_SDMMC2,
> >> PINMUX_FLAG_NONE);
> >> - if (err)
> >> + err_sd = exynos_pinmux_config(PERIPH_ID_SDMMC2,
> >> PINMUX_FLAG_NONE);
> >> + if (err_sd)
> >> debug("SDMMC2 not configured\n");
> >> - else
> >> - err = s5p_mmc_init(2, 4);
> >> + else {
> >> + if (!sd_ldo_control(1)) {
> >> + err_sd = s5p_mmc_init(2, 4);
> >> + if (err_sd) {
> >> + sd_ldo_control(0);
> >> + }
> > Parenthesis with this if aren't
> > needed.
> >> + }
> >> + }
> >> }
> >>
> >> - return err;
> >> + return err & err_sd;
To be honest I'm frowned upon when errors are ANDed. In my opinion they
shall be ORed together.
So I think that for the trats.c we only need to do the following:
if (!s5p_gpio_get_value(&gpio->x3, 4)) {
err |= exynos_pinmux_config(PERIPH_ID_SDMMC2, PINMUX_FLAG_NONE);
if (err) {
debug("SDMMC2 not configured\n");
} else {
if (!sd_ldo_control(1)) {
err |= s5p_mmc_init(2, 4);
if (err) {
sd_ldo_control(0);
error("SD catd init failed\
- remove it\n");
}
}
}
}
return err;
Please consider following scenarios:
1. eMMC OK, card present and broken -> print informative error message
(to recover user will remove the card)
2. eMMC OK + card inserted and init OK = no problem
3. eMMC OK + card not inserted = no problem
4. eMMC initialization broken + card inserted and init OK - we will not
boot the device from eMMC, so we will not be able to use SD card at
u-boot.
Please correct me if I'm wrong and/or I've overlooked some useful
scenario for you.
Best regards,
Lukasz Majewski
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20131103/95b38749/attachment.pgp>
next prev parent reply other threads:[~2013-11-03 22:24 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-10-31 13:37 [U-Boot] [PATCH] arm: trats: add the sd-card power control function Jaehoon Chung
2013-10-31 17:03 ` Lukasz Majewski
2013-11-01 1:02 ` Jaehoon Chung
2013-11-03 22:24 ` Lukasz Majewski [this message]
2013-11-04 1:28 ` Jaehoon Chung
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=20131103232432.30ef8d9e@jawa \
--to=l.majewski@majess.pl \
--cc=u-boot@lists.denx.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