1. 20 Mar, 2015 1 commit
    • MLK-10439: arm: imx: fix system hang after resume back under low busfreq · 91cf351a
      Bai Ping authored
      System will be no response after resume back under low busfreq mode. The
      root cause of this issue is that when the system is under ultra_low_bus_freq
      mode on i.MX6SL, resume process has a low bus mode deamon in background,
      cause system enter low bus mode twice, busfreq driver will notify thermal
      driver to turn off PLL3 twice, and cause PLL3 usecount incorrect and UART
      no message out.
      Signed-off-by: 's avatarBai Ping <b51503@freescale.com>
  2. 19 Mar, 2015 1 commit
  3. 16 Mar, 2015 1 commit
    • MLK-10423: Capture: System hang if capture test app been killed · 899414fc
      Sandor Yu authored
      System will hang if csi unit test mx6s_v4l2_capture.out process
      been killed.
      
      In csi capture driver function mx6s_csi_enable/disable is called
      by vidioc_streamon/streamoff function.
      But when csi unit test process is killed, vidioc_streamon/streamoff
      will not been called, and csi function still working even the csi
      clock is gated.
      
      Move mx6s_csi_enable/disable function from
      mx6s_vidioc_streamon/streamoff function to
      mx6s_start_streaming/ mx6s_stop_streaming function to resolve the
      issue.
      Signed-off-by: 's avatarSandor Yu <R01008@freescale.com>
      (cherry picked from commit fe3700b2dc97952ef5a055a48e1d5ba89e11b7dd)
  4. 06 Mar, 2015 2 commits
    • MLK-10210 WIFI: ath6kl: Fixed the issue that warning messages were printed when… · c33b14da
      Shenwei Wang authored
      MLK-10210 WIFI: ath6kl: Fixed the issue that warning messages were printed when uninstalling ath6kl_sdio after iperf test.
      
      When handing the command "ifconfig wlan0 down", the driver handler did
      not notify the disconnected event to the upper layer stack cfg80211,
      which would cause the relating resources unable to be released. A
      notification of disconnet event to cfg80211 was added when the driver
      was changed to the disconnect state.
      Signed-off-by: 's avatarShenwei Wang <shenwei.wang@freescale.com>
      (cherry picked from commit 0b7b203a563ac98f5f0c5695a5149ec80d2114b4)
    • MLK-10211 can: flexcan: fix resume failed issue · 34b7539e
      Dong Aisheng authored
      We can observe the following error during system resume.
      dpm_run_callback(): platform_pm_resume+0x0/0x4c returns -110
      PM: Device 2090000.can failed to resume: error -110
      
      It is caused by the flexcan is in disable state and the clock is disabled
      by default while the CAN is not up. During resume, the calling of
      flexcan_chip_enable will obviously fail due to no clock.
      
      The issue only happened when CAN device is not up. If CAN is up, clock is there
      and no issue will happen.
      
      Because the CAN is always in disabled state by default if it's not up.
      We don't have to disable it again during suspend and enable it after resume.
      This patch simply the suspend/resume to only enter/exit disabled mode if CAN
      is up.
      Signed-off-by: 's avatarDong Aisheng <b29396@freescale.com>
      (cherry picked from commit 1a8a68cb11ce2de085efb05aa0262a7eca07aba6)
  5. 05 Mar, 2015 3 commits
  6. 04 Mar, 2015 1 commit
  7. 03 Mar, 2015 2 commits
  8. 02 Mar, 2015 1 commit
    • MLK-10349 ARM: dts: imx6dql: fix the clock for MIPI CSI2 · 9dd1ed06
      Robby Cai authored
      Fix the clock index for cfg clock and use MACRO instead of hard-codes.
      
      This patch fixes the following issue.
      -----------------------------------------------------------
      root@imx6qdlsolo:~# /unit_tests/mxc_v4l2_capture.out -d /dev/video1 1.yuv
      in_width = 176, in_height = 144
      out_width = 176, out_height = 144
      top = 0, left = 0
      mipi csi2 can not receive sensor clk!
      ...
      
      ERROR: v4l2 capture: mxc_v4l_dqueue timeout enc_counter 0
      VIDIOC_DQBUF failed.ERROR: v4l2 capture: VIDIOC_QBUF: buffer already queued
      -----------------------------------------------------------
      Signed-off-by: 's avatarRobby Cai <r63905@freescale.com>
      (cherry picked from commit 6e4ee449de591d3cfb93575ca639ca32944832bc)
  9. 27 Feb, 2015 2 commits
    • MLK-10279-2 usb: chipidea: imx: only do charger detection if we enable it · 8af72e14
      Peter Chen authored
      We should only do charger detection if both we have enabled
      usb charger at kernel configuration and set usb charger enable
      at dts. Otherwise, we will do useless things, like toggle dp.
      Signed-off-by: 's avatarPeter Chen <peter.chen@freescale.com>
      (cherry picked from commit 6b677c09dbda0274c6b2e1d61ea0d7f1cc73f621)
    • MLK-10279-1 usb: common: otg-fsm: only signal connect after switching to peripheral · 0fb0fc89
      Peter Chen authored
      We should signal connect (pull up dp) after we have already
      at peripheral mode, otherwise, the dp may be toggled due to
      we reset controller or do disconnect during the initialization
      for peripheral, then, the host may be confused during the
      enumeration, eg, it finds the reset can't succeed, but the
      device is still there, see below error message.
      
      hub 1-0:1.0: USB hub found
      hub 1-0:1.0: 1 port detected
      hub 1-0:1.0: cannot reset port 1 (err = -32)
      hub 1-0:1.0: cannot reset port 1 (err = -32)
      hub 1-0:1.0: cannot reset port 1 (err = -32)
      hub 1-0:1.0: cannot reset port 1 (err = -32)
      hub 1-0:1.0: cannot reset port 1 (err = -32)
      hub 1-0:1.0: Cannot enable port 1.  Maybe the USB cable is bad?
      hub 1-0:1.0: cannot reset port 1 (err = -32)
      hub 1-0:1.0: cannot reset port 1 (err = -32)
      hub 1-0:1.0: cannot reset port 1 (err = -32)
      hub 1-0:1.0: cannot reset port 1 (err = -32)
      hub 1-0:1.0: cannot reset port 1 (err = -32)
      hub 1-0:1.0: Cannot enable port 1.  Maybe the USB cable is bad?
      hub 1-0:1.0: cannot reset port 1 (err = -32)
      hub 1-0:1.0: cannot reset port 1 (err = -32)
      hub 1-0:1.0: cannot reset port 1 (err = -32)
      hub 1-0:1.0: cannot reset port 1 (err = -32)
      hub 1-0:1.0: cannot reset port 1 (err = -32)
      hub 1-0:1.0: Cannot enable port 1.  Maybe the USB cable is bad?
      hub 1-0:1.0: cannot reset port 1 (err = -32)
      hub 1-0:1.0: cannot reset port 1 (err = -32)
      hub 1-0:1.0: cannot reset port 1 (err = -32)
      hub 1-0:1.0: cannot reset port 1 (err = -32)
      hub 1-0:1.0: cannot reset port 1 (err = -32)
      hub 1-0:1.0: Cannot enable port 1.  Maybe the USB cable is bad?
      hub 1-0:1.0: unable to enumerate USB device on port 1
      Signed-off-by: 's avatarPeter Chen <peter.chen@freescale.com>
      (cherry picked from commit b4b5728425454c1713d152855f09af9c9c12d78d)
  10. 25 Feb, 2015 2 commits
  11. 15 Feb, 2015 1 commit
    • MLK-10248 video: ipuv3-fb: fix hang up with cmdline option for CLAA WVGA · 6b4144ac
      Robby Cai authored
      The kernel boot stops at "Starting kernel ..." if assign cmdline
      "video=mxcfb0:dev=lcd,if=RGB565".
      
      The cause is that there's no 'mode_str' property for mxcfb0 in DTS file,
      so of_property_read_string() does not touch it and 'mode_str' could be used
      uninitialized in mxcfb_get_of_property(), where it's assigned to
      plat_data->mode_str and then used uninitialized in fb_find_mode().
      This may cause buffer overflow. This patch initializes 'mode_str' to NULL
      to fix it.
      Signed-off-by: 's avatarRobby Cai <r63905@freescale.com>
      (cherry picked from commit 115d11804b7b56d058ee6be8496b13f57869be93)
  12. 13 Feb, 2015 1 commit
    • MLK-10243 usb: chipidea: udc: enable and disable BSV irq only for ID change · a95e6f53
      Li Jun authored
      Since BSV irq is only used for B-device, this patch correct the setting of
      BSVIE of otgsc only for case of ID change, that is:
      - In otg fsm mode, ID change from 0 to 1 means otg state from A_IDLE to
        B_IDLE; ID change from 1 to 0 means otg state from B_IDLE to A_IDLE, so
        only disable BSVIE if the current state is B_IDLE(to be update to A_IDLE),
        only enable BSVIE if the current state is A_IDLE(to be update to B_IDLE).
      - In non-otg-fsm mode, it's already correct.
      
      This fix the bug of MLK-10243.
      Signed-off-by: 's avatarLi Jun <jun.li@freescale.com>
      Signed-off-by: 's avatarPeter Chen <peter.chen@freescale.com>
      (cherry picked from commit c6151e8cc4a01128cbab4b26c09f859cfcd7ff39)
  13. 12 Feb, 2015 5 commits
  14. 11 Feb, 2015 17 commits