Skip to contentSkip to conversation

U-Boot v2018.03-rc3 released

U-Boot Tom Rini 3 participants 4 messages Feb 20, 2018 – Feb 21, 2018 MMC, SD and SDIO

lore
Share & tools
lore

Tom Rini

author
#0 lore raw

Hey all,

So, we're a day late. That's because I wanted the MMC PR to come in for some important fixes, and it was too late in my day to then push a release. I think we probably have a few more fixes that need to come in, but hopefully no more new regressions to be found. That said, if you have your environment, well, anywhere, a little bit of hardware testing wouldn't hurt. I think we're still on track for -rc4 on March 5th and the release on March 12th. But if we aren't, the release will get delayed.

And on a related note, there has been more feedback on getting going back to a longer release cycle. I'm thinking about it and maybe trying a longer cycle for a few releases to get a feel for everything again.

On a different note, I'm thinking that boards and SoC families, etc, that have been orphaned for more than a year should probably get dropped. I've contacted a few people off-list (which is why we've had a few pickups of late) for things, but if you know someone that would be interested in something, please speak up.

Thanks all!

Signature · 8 lines
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20180220/545ca032/attachment.sig>
#1 lore raw
Quoted 11 lines · Tom Rini

Hey all,

So, we're a day late. That's because I wanted the MMC PR to come in for some important fixes, and it was too late in my day to then push a release. I think we probably have a few more fixes that need to come in, but hopefully no more new regressions to be found. That said, if you have your environment, well, anywhere, a little bit of hardware testing wouldn't hurt. I think we're still on track for -rc4 on March 5th and the release on March 12th. But if we aren't, the release will get delayed.

Something's broken my board (AM335x, not quite a BBB):

MMC: mmc at 481d8000 - probe failed: -1

Looks to be this commit:

2d7482cf793fe4d8a98906002708d6e1fa2c5ba3

Still looking, but I'd guess I'm now missing something from my DTB which I didn't need previously.

Signature · 1 line
Alex Kiernan
#2 lore raw
Quoted 15 lines · Alex Kiernan
Quoted 11 lines · Tom Rini

Hey all,

So, we're a day late. That's because I wanted the MMC PR to come in for some important fixes, and it was too late in my day to then push a release. I think we probably have a few more fixes that need to come in, but hopefully no more new regressions to be found. That said, if you have your environment, well, anywhere, a little bit of hardware testing wouldn't hurt. I think we're still on track for -rc4 on March 5th and the release on March 12th. But if we aren't, the release will get delayed.

Something's broken my board (AM335x, not quite a BBB):

Quoted 9 lines

MMC: mmc at 481d8000 - probe failed: -1

Looks to be this commit:

2d7482cf793fe4d8a98906002708d6e1fa2c5ba3

Still looking, but I'd guess I'm now missing something from my DTB which I didn't need previously.

mmc_of_parse says:

        /* f_max is obtained from the optional "max-frequency" property */
        dev_read_u32(dev, "max-frequency", &cfg->f_max);

I don't have a max-frequency (pretty much all my config comes direct from am33xx.dtsi), so if I add a max-frequency to my tree I get a working board, but the comment above suggests I shouldn't need to, so do we need something like this added (or a refactor of mmc_of_parse so it takes a default?)

drivers/mmc/omap_hsmmc.c +0 -0
index 02970f2..b87ff01 100644
@@ -1825,6 +1825,8 @@ static int omap_hsmmc_ofdata_to_platdata(struct

udevice *dev)

        if (ret < 0)
                return ret;

+ if (cfg->f_max == 0)

+               cfg->f_max = 52000000;
        cfg->host_caps |= MMC_MODE_HS_52MHz | MMC_MODE_HS;
        cfg->f_min = 400000;
        cfg->voltages = MMC_VDD_32_33 | MMC_VDD_33_34 | MMC_VDD_165_195;

I'm also seeing MMC_MODE_1BIT (which I didn't previously have set) alongside MMC_MODE_4BIT and MMC_MODE_8BIT (which I previously did), but I'm guessing that doesn't matter?

Signature · 1 line
Alex Kiernan
#3 lore raw
Quoted 40 lines · Alex Kiernan
Quoted 14 lines · Alex Kiernan
Quoted 11 lines · Tom Rini

Hey all,

So, we're a day late. That's because I wanted the MMC PR to come in for some important fixes, and it was too late in my day to then push a release. I think we probably have a few more fixes that need to come in, but hopefully no more new regressions to be found. That said, if you have your environment, well, anywhere, a little bit of hardware testing wouldn't hurt. I think we're still on track for -rc4 on March 5th and the release on March 12th. But if we aren't, the release will get delayed.

Something's broken my board (AM335x, not quite a BBB):

Quoted 9 lines

MMC: mmc at 481d8000 - probe failed: -1

Looks to be this commit:

2d7482cf793fe4d8a98906002708d6e1fa2c5ba3

Still looking, but I'd guess I'm now missing something from my DTB which I didn't need previously.

mmc_of_parse says:

         /* f_max is obtained from the optional "max-frequency" property */
         dev_read_u32(dev, "max-frequency", &cfg->f_max);

I don't have a max-frequency (pretty much all my config comes direct from am33xx.dtsi), so if I add a max-frequency to my tree I get a working board, but the comment above suggests I shouldn't need to, so do we need something like this added (or a refactor of mmc_of_parse so it takes a default?)

drivers/mmc/omap_hsmmc.c +0 -0
index 02970f2..b87ff01 100644

drivers/mmc/omap_hsmmc.c static int omap_hsmmc_ofdata_to_platdata(struct · line 1830

Quoted 6 lines
drivers/mmc/omap_hsmmc.c +2 -0
@@ -1825,6 +1825,8 @@ static int omap_hsmmc_ofdata_to_platdata(struct
 udevice *dev)
         if (ret < 0)
                 return ret;
 
+       if (cfg->f_max == 0)
+               cfg->f_max = 52000000;

Thanks for catching this.
I'll send a series soon related to am335x_evm and mmc. I'll include your fix.

Quoted 7 lines
drivers/mmc/omap_hsmmc.c +0 -0
         cfg->host_caps |= MMC_MODE_HS_52MHz | MMC_MODE_HS;
         cfg->f_min = 400000;
         cfg->voltages = MMC_VDD_32_33 | MMC_VDD_33_34 | MMC_VDD_165_195;

I'm also seeing MMC_MODE_1BIT (which I didn't previously have set) alongside MMC_MODE_4BIT and MMC_MODE_8BIT (which I previously did), but I'm guessing that doesn't matter?

1Bit is always supported. So it doesn't matter.
JJ

Quoted 1 line