Skip to content

GitLab

Switch branch/tag
  1. 16 Oct, 2019 1 commit
  3. 14 Oct, 2019 5 commits
  5. 11 Oct, 2019 3 commits
  7. 09 Oct, 2019 5 commits
  9. 06 Oct, 2019 8 commits
  11. 02 Oct, 2019 2 commits
  13. 28 Sep, 2019 1 commit
  15. 24 Sep, 2019 1 commit
  17. 06 Jun, 2019 7 commits
    • Ye Li's avatar
      MLK-21942 DTS: imx8mq: Fix USB alias issue · 85bdcc79
      Ye Li authored 
      

Since we set the driver's compatible string to dwc3 nodes, should
use the dwc3 nodes as the USB alias, otherwise the alias won't work.
If first usb controller is probed failed, the second controller still
use 0 as seq and cause issue to the driver.

Because alias requires different names on the nodes, in the patch we have
to change the nodes name to dwc3_0 and dwc3_1
Signed-off-by: default avatarYe Li <ye.li@nxp.com>
Reviewed-by: default avatarPeng Fan <peng.fan@nxp.com>
(cherry picked from commit 67f084c9485587802571352de8fb2551e8aefe74)
      85bdcc79
    • Ye Li's avatar
      MLK-21956-3 mx6qdlsabreauto: Enable i2c gpio mux to fix io expander issue · fac0bd64
      Ye Li authored 
      

The i2c gpio mux is not enabled, this causes i2c3 not work and all
io expander devices on i2c3 not work.  So we will meet such as
"usb start" fail due to the VBUS is not power up.

The patch add the i2c mux and i2c gpio mux configs. And remove codes of
controlling i2c steering from board, since i2c gpio mux does same thing.
Signed-off-by: default avatarYe Li <ye.li@nxp.com>
Reviewed-by: default avatarPeng Fan <peng.fan@nxp.com>
(cherry picked from commit 39183dc231e12c3072ffd9a37ac6a6616178e263)
      fac0bd64
    • Ye Li's avatar
      MLK-21956-2 mxc_i2c: Remove i2c_idle_bus from probe · 40c24389
      Ye Li authored 
      

i2c_idle_bus is already used in i2c_init_transfer. So before each tranfer
if the bus is not ready, the i2c_idle_bus will be used to force idle.
It is unnecessary to call it again in probe.

We found a issue when enabling i2c mux with the mxc_i2c. The mxc_i2c is probed
after mux probing. However, at this moment the mux is still in idle state,
won't select port. So if we call i2c_idle_bus in probe, it will fail and cause
mxc_i2c probe failed.
Signed-off-by: default avatarYe Li <ye.li@nxp.com>
(cherry picked from commit e999c45f436f1187ef1185997d3838ad8b28421c)
      40c24389
    • Ye Li's avatar
      MLK-21956-1 i2c-mux-gpio: Fix GPIO request flag issue · a80b1536
      Ye Li authored 
      

When requesting GPIO, the GPIOD_IS_OUT is missed in flag, so the GPIO
is set the input mode not output and cause mux not work.
Signed-off-by: default avatarYe Li <ye.li@nxp.com>
(cherry picked from commit 4fd2f04db1513d4bcd294a4c94193a4d59785b6d)
      a80b1536
    • Peng Fan's avatar
      MLK-21961 imx: fix tee_addr for nand boot · eba51a95
      Peng Fan authored 
      

teeaddr is not set. Align with sd/emmc, use tee_addr and set tee_addr.
Signed-off-by: default avatarPeng Fan <peng.fan@nxp.com>
Reported-by: default avatarJessie Lee <jessie.lee@nxp.com>
Reviewed-by: default avatarYe Li <ye.li@nxp.com>
(cherry picked from commit 5fb43e3f8fba534d49d8066411644285eaa279b4)
      eba51a95
    • Ye Li's avatar
      MLK-21885 lmb: Handle the overlap case for lmb reserve · 5f95f318
      Ye Li authored 
      

lmb reserve is used to reserve some memory so that when loading images
(like kernel, dtb, initrd), images won't be loaded into the reserved memory.

The problem in current lmb is it does not handle the overlap case. When adding
a new reserved memory, if the memory region is overlap with regions already been
added in lmb, it will fail. One example is reserved memory in DTB may overlap with
u-boot relocate address. lmb reserves the u-boot relocate address firstly, so when
adding reserved memory from DTB, we will meet failure.

Actually if we handle the overlap case, we can resolve the overlap by using a max
common region for the overlap regions. So that this case won't fail.
Signed-off-by: default avatarYe Li <ye.li@nxp.com>
Reviewed-by: default avatarPeng Fan <peng.fan@nxp.com>
(cherry picked from commit 37d86c68816dffde3dc8dcda5b9d67a195b2f9c2)
      5f95f318
    • Ye Li's avatar
      MLK-21905 mx7ulp_evk: Change kernel DTB to EVKB board · 2c2f0428
      Ye Li authored 
      

The latest mx7ulp evk board is renamed to EVKB board and has dedicated
DTB files. Modify the kernel DTB names in u-boot to use the new DTB.
Signed-off-by: default avatarYe Li <ye.li@nxp.com>
Acked-by: default avatarPeng Fan <peng.fan@nxp.com>
(cherry picked from commit 297a924ddaf456a5a175d9056c0bc45116cc49ce)
      2c2f0428
  19. 04 Jun, 2019 2 commits
  21. 30 May, 2019 2 commits
  23. 28 May, 2019 3 commits
    • Ye Li's avatar
      MLK-21879-3 fsl_qspi: Fix flash write issue on mx7ulp · a9116440
      Ye Li authored 
      

The page write sequence in spi-mem is changed with orignal spi-flash. When
the max_write_size is set by driver, the orignal sequence is
  1. WREN
  2. write max_write_size data to flash
  3. wait for WIP clean
  4. back to #1 if having data remained in a page.

The new sequence is:
  1. WREN
  2. write (max_write_size - command length) data to flash
  3. back to #2 if having data remained in a page
  4. wait for WIP clean

Since mx7ulp has 64 bytes TX buffer (max_write_size), while other iMX chips
have 512 bytes (page size is 256). So when adapt qspi with the new sequence,
we have to check the WIP before each write on mx7ulp. Otherwise the WIP may set
due to previous write is not completed by flash device, then cause current
write failed.
Signed-off-by: default avatarYe Li <ye.li@nxp.com>
      a9116440
    • Ye Li's avatar
      MLK-21879-2 spi-mem: Fix read data size issue · 42ea2fa0
      Ye Li authored 
      

When slave drivers not set the max_read_size, the spi-mem should directly
use data.nbytes and not limit to any size. But current logic will limit to
the max_write_size.
Signed-off-by: default avatarYe Li <ye.li@nxp.com>
      42ea2fa0
    • Ye Li's avatar
      MLK-21879-1 fsl_qspi: increase timeout of is_controller_busy · d0a917b8
      Ye Li authored 
      

The original codes uses 100ms timeout, but the new is_controller_busy only
try 5 times with each only 1us delay.
When the qspi clock is slower, the controller may need longer time in busy.
On mx7ulp, we meet problem that timeout always happen. So change the codes
to 100ms timeout
Signed-off-by: default avatarYe Li <ye.li@nxp.com>
      d0a917b8