
Hi Tom,
On 28 May 2017 at 05:55, Tom Rini trini@konsulko.com wrote:
On Sat, May 27, 2017 at 04:01:29PM -0600, Simon Glass wrote:
Hi Tom,
This is similar to the last attempt, but without this offending patch which will be replaced by a little series sent today: (I'll pull in that series later)
0a2980b dm: mmc: Avoid probing block devices in find_mmc_device()
The following changes since commit 380e86f361e4e2aef83295972863654fde157560:
Merge git://git.denx.de/u-boot-fsl-qoriq (2017-05-26 11:19:27 -0400)
are available in the git repository at:
git://git.denx.de/u-boot-dm.git
for you to fetch changes up to 40fcab41cbcace3ff5928f146dd15c1ee3bfac65:
sandbox: Move to use live tree (2017-05-27 10:38:13 -0600)
Two problems. First, you need to add this to "dm: gpio: Add live tree support":
diff --git a/board/st/stm32f746-disco/stm32f746-disco.c b/board/st/stm32f746-disco/stm32f746-disco.c index aeaa31146a5a..addf82bbae8c 100644 --- a/board/st/stm32f746-disco/stm32f746-disco.c +++ b/board/st/stm32f746-disco/stm32f746-disco.c @@ -101,8 +101,8 @@ int board_late_init(void) if (node < 0) return -1;
gpio_request_by_name_nodev(gd->fdt_blob, node, "led-gpio", 0, &gpio,
GPIOD_IS_OUT);
gpio_request_by_name_nodev(offset_to_ofnode(node), "led-gpio", 0,
&gpio, GPIOD_IS_OUT); if (dm_gpio_is_valid(&gpio)) { dm_gpio_set_value(&gpio, 0);
@@ -115,8 +115,8 @@ int board_late_init(void) if (node < 0) return -1;
gpio_request_by_name_nodev(gd->fdt_blob, node, "button-gpio", 0, &gpio,
GPIOD_IS_IN);
gpio_request_by_name_nodev(offset_to_ofnode(node), "button-gpio", 0,
&gpio, GPIOD_IS_IN); if (dm_gpio_is_valid(&gpio)) { if (dm_gpio_get_value(&gpio))
OK, added, thanks. For me that board fails to build since it has something like:
for (int i = ....)
Is that allowed now?
Second, the zynq_zc702 qemu instance is failing (and you need a new enough version of qemu, such as the one in .travis.yml, to be able to run it. I need to upgrade my local qemu so I don't have the failure bisected atm).
OK I found that it was checking for -FDT_ERR_NOTFOUND which really should not have been returned by a DM function.
I sent the two updated patches. Given that RC1 is coming out soon I'll try to do a new pull after testing, but it won't be until the morning.
Regards, Simon