OMAP2/3: VENC hwmod: add OCPIF_SWSUP_IDLE flag to interface
authorPaul Walmsley <paul@pwsan.com>
Fri, 4 Mar 2011 06:02:15 +0000 (06:02 +0000)
committerTony Lindgren <tony@atomide.com>
Wed, 9 Mar 2011 21:03:15 +0000 (13:03 -0800)
commitc39bee8ac4aca750c85609ba0847599b36feb693
tree45157eb8ecc3eaa781b10fcc3528434d4b71003a
parent0640b436e410290193f554dcfd777bcdeee59697
OMAP2/3: VENC hwmod: add OCPIF_SWSUP_IDLE flag to interface

According to the hwmod interface data, the DSS submodule "VENC" uses a
clock, "dss_54m_fck"/"dss_tv_fck", which the PRCM cannot autoidle.  By
default, the hwmod code assumes that interface clocks can be autoidled
by the PRCM.  When the interface clock can't be autoidled by the PRCM,
those interfaces must be marked with the OCPIF_SWSUP_IDLE flag.
Otherwise, the "interface clock" will always have a non-zero use
count, and the device won't enter idle.  This problem was observed on
N8x0.

Fix the immediate problem by marking the VENC interface with the
OCPIF_SWSUP_IDLE flag.  But it's not clear that
"dss_54m_fck"/"dss_tv_fck" is really the correct interface clock for
VENC.  It may be that the VENC interface should use a
hardware-autoidling interface clock.  This is the situation on OMAP4,
which uses "l3_div_ck" as the VENC interface clock, which can be
autoidled by the PRCM.  Clarification from TI is needed.

Problem found and patch tested on N8x0 by Tony Lindgren
<tony@atomide.com>.

Signed-off-by: Paul Walmsley <paul@pwsan.com>
Cc: Tony Lindgren <tony@atomide.com>
Cc: Senthilvadivu Guruswamy <svadivu@ti.com>
Cc: Sumit Semwal <sumit.semwal@ti.com>
Cc: Kevin Hilman <khilman@ti.com>
Cc: BenoƮt Cousson <b-cousson@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
arch/arm/mach-omap2/omap_hwmod_2420_data.c
arch/arm/mach-omap2/omap_hwmod_2430_data.c
arch/arm/mach-omap2/omap_hwmod_3xxx_data.c