Skip to content

L
linux-davinci
Switch branch/tag
  1. 12 May, 2009 5 commits
    • Paul Walmsley's avatar
      OMAP3 SDRC: initialize SDRC_POWER at boot · 98cfe5ab
      Paul Walmsley authored 
      Initialize SDRC_POWER to a known-good setting when the kernel boots.
Necessary since some bootloaders don't initialize SDRC_POWER properly.
Signed-off-by: default avatarPaul Walmsley <paul@pwsan.com>
      98cfe5ab
    • Paul Walmsley's avatar
      OMAP3 SRAM: clear the SDRC PWRENA bit during SDRC frequency change · fa0406a8
      Paul Walmsley authored 
      Clear the SDRC_POWER.PWRENA bit before putting the SDRAM into self-refresh
mode.  This prevents the SDRC from attempting to power off the SDRAM,
which can cause the system to hang.
Signed-off-by: default avatarPaul Walmsley <paul@pwsan.com>
      fa0406a8
    • Paul Walmsley's avatar
      OMAP3 clock: add interconnect barriers to CORE DPLL M2 change · d75d9e73
      Paul Walmsley authored 
      Where necessary, add interconnect barriers to force posted writes to
complete before continuing.
Signed-off-by: default avatarPaul Walmsley <paul@pwsan.com>
      d75d9e73
    • Paul Walmsley's avatar
      OMAP3 SRAM: add ARM barriers to omap3_sram_configure_core_dpll · 69d4255b
      Paul Walmsley authored 
      Add more barriers in the SRAM CORE DPLL M2 divider change code.

- Add a DSB SY after the function's entry point to flush all cached
  and buffered writes and wait for the interconnect to claim that they
  have completed[1].  The idea here is to force all delayed write
  traffic going to the SDRAM to at least post to the L3 interconnect
  before continuing.  If these writes are allowed to occur after the
  SDRC is idled, the writes will not be acknowledged and the ARM will
  stall.

  Note that in this case, it does not matter if the writes actually
  complete to the SDRAM - it is only necessary for the writes to leave
  the ARM itself.  If the writes are posted by the interconnect when
  the SDRC goes into idle, the writes will be delayed until the SDRC
  returns from idle[2].  If the SDRC is in the middle of a write when
  it is requested to enter idle, the SDRC will not acknowledge the
  idle request until the writes complete to the SDRAM.[3]

  The old-style DMB in sdram_in_selfrefresh is now superfluous, so,
  remove it.

- Add an ISB before the function's exit point to prevent the ARM from
  speculatively executing into SDRAM before the SDRAM is enabled[4].

...

1. ARMv7 ARM (DDI 0406A) A3-47, A3-48.

2. Private communication with Richard Woodruff <r-woodruff2@ti.com>.

3. Private communication with Richard Woodruff <r-woodruff2@ti.com>.

4. ARMv7 ARM (DDI 0406A) A3-48.
Signed-off-by: default avatarPaul Walmsley <paul@pwsan.com>
Cc: Richard Woodruff <r-woodruff2@ti.com>
      69d4255b
    • Paul Walmsley's avatar
      OMAP3 SRAM: mark OCM RAM as Non-cacheable Normal memory · d9295746
      Paul Walmsley authored 
      Mark the SRAM (aka OCM RAM) as Non-cacheable Normal memory[1].  This
is to prevent the ARM from evicting existing cache lines to SDRAM
while code is executing from the SRAM.  Necessary since one of the
primary uses for the SRAM is to hold the code and data for the CORE
DPLL M2 divider reprogramming code, which must execute while the SDRC
is idled.  If the ARM attempts to write cache lines back to the while
the SRAM code is running, the ARM will stall[2].

TI deals with this problem in the CDP kernel by marking the SRAM as
Strongly-ordered memory.

Tero Kristo <tero.kristo@nokia.com> caught a bug in an earlier version of
this patch - thanks Tero.

...

1. ARMv7 ARM (DDI 0406A) pp. A3-30, A3-31, B3-32.

2. Private communication with Richard Woodruff <r-woodruff2@ti.com>
Signed-off-by: default avatarPaul Walmsley <paul@pwsan.com>
Cc: Tero Kristo <tero.kristo@nokia.com>
Cc: Richard Woodruff <r-woodruff2@ti.com>
      d9295746
  3. 11 May, 2009 12 commits
  5. 10 May, 2009 12 commits
  7. 09 May, 2009 11 commits