The enum members of 'dss_clk_source' have clock source names specific to
OMAP2/3. Change the names to more generic terms such that they now describe
where the clocks come from and what they are used for.

Also, change the enum member names to have "DSS_CLK_SRC" instead of "DSS_SRC"
for more clarity.
Signed-off-by: NArchit Taneja <archit@ti.com>
Signed-off-by: NTomi Valkeinen <tomi.valkeinen@ti.com>

The maximum supported frequency for DSS has increased from 173 to 186 Mhz on
OMAP4.

Introduce a dss feature function to get the max_fck to replace DISPC_MAX_FCK
macro.
Signed-off-by: NArchit Taneja <archit@ti.com>
Signed-off-by: NTomi Valkeinen <tomi.valkeinen@ti.com>

Add dss_features and register fields to incorporate changes in DISPC pipelines
between OMAP3 and OMAP4.

Register fields added: FEAT_REG_HORIZONTALACCU, FEAT_REG_VERTICALACCU
DSS Features added: FEAT_LINEBUFFERSPLIT, FEAT_ROWREPEATENABLE, FEAT_RESIZECONF

_dispc_set_scaling() and _dispc_set_rotation_attrs() have been cleaned up.
Signed-off-by: NArchit Taneja <archit@ti.com>
Signed-off-by: NTomi Valkeinen <tomi.valkeinen@ti.com>

There is a linker error from lcd_2430sdp.c if CONFIG_TWL4030_CORE is not
set. This can be triggered on OMAP2 builds when OMAP3 or OMAP4 are not set.

drivers/built-in.o: In function `sdp2430_panel_disable':
drivers/video/omap/lcd_2430sdp.c:123: undefined reference to `twl_i2c_write_u8'
drivers/video/omap/lcd_2430sdp.c:124: undefined reference to `twl_i2c_write_u8'
drivers/built-in.o: In function `sdp2430_panel_enable':
drivers/video/omap/lcd_2430sdp.c:110: undefined reference to `twl_i2c_write_u8'
drivers/video/omap/lcd_2430sdp.c:112: undefined reference to `twl_i2c_write_u8'

Fix this by selecting the TWL4030_CORE for MACH_OMAP_2430SDP when building
with CONFIG_FB_OMAP as there is no own Kconfig entry for lcd_2430 and it is
compiled always when both MACH_OMAP_2430SDP and FB_OMAP are set.
Signed-off-by: NJarkko Nikula <jhnikula@gmail.com>
Signed-off-by: NTomi Valkeinen <tomi.valkeinen@ti.com>

hwmod databases provide information about which optional clocks are available
for a given platform. This is available via a function pointer opt_clock_enable
in pdata.

Use this information during get/enable/disable/put of clocks.
Signed-off-by: NSumit Semwal <sumit.semwal@ti.com>
Signed-off-by: NTomi Valkeinen <tomi.valkeinen@ti.com>

dsi_reset_tx_fifo() was not used. Furthermore, OMAP errata states that
TX FIFO flush is not functional, so the function wouldn't even have
worked.
Signed-off-by: NTomi Valkeinen <tomi.valkeinen@ti.com>

There was an unused list defined in display.c. Remove it.
Signed-off-by: NTomi Valkeinen <tomi.valkeinen@ti.com>

CONFIG_FB_OMAP_LCD_VGA option is present in drivers/video/omap
There is no explaination about what this flag does.
Lets add information about it.

FB_OMAP_LCD_VGA flag sets resolution of display to VGA (640 X 480).
The default resolution of 3430 LDP is 320 X 240.
Signed-off-by: NMayuresh Janorkar <mayur@ti.com>
Signed-off-by: NTomi Valkeinen <tomi.valkeinen@ti.com>

Only OMAP 3430 hardware has SDI support. The availability of SDI HW can
be found out by checking if the LCD channel supports SDI displays.

This patch checks for SDI HW support before accessing SDI registers,
which fixes a crash on OMAP4 when SDI SW support is compiled in.
Signed-off-by: NTomi Valkeinen <tomi.valkeinen@ti.com>

OMAP2 doesn't support SDI or DSI. Remove them from
omap2_dss_supported_displays.
Signed-off-by: NTomi Valkeinen <tomi.valkeinen@ti.com>

OMAP 3630 does not support SDI. Split omap3_dss_supported_displays into
3430 and 3630 entries, and remove the SDI from 3630 entry.
Signed-off-by: NTomi Valkeinen <tomi.valkeinen@ti.com>

Support for the display interface was checked in a separate switch-case.
There's no reason for that, and this patch handles the fail code path in
the same switch-case where the display initialization is done.
Signed-off-by: NTomi Valkeinen <tomi.valkeinen@ti.com>

Currently, the core DSS platform device requests for an irq line for OMAP2 and
OMAP3. Make DISPC and DSI platform devices request for a shared IRQ line.

On OMAP3, the logical OR of DSI and DISPC interrupt lines goes to the MPU. There
is a register DSS_IRQSTATUS which tells if the interrupt came from DISPC or DSI.

On OMAP2, there is no DSI, only DISPC interrupts goto the MPU. There is no
DSS_IRQSTATUS register.

Hence, it makes more sense to have separate irq handlers corresponding to the
DSS sub modules instead of having a common handler.

Since on OMAP3 the logical OR of the lines goes to MPU, the irq line is shared
among the IRQ handlers.

The hwmod irq info has been removed for DSS to DISPC and DSI for OMAP2 and OMAP3
hwmod databases. The Probes of DISPC and DSI now request for irq handlers.
Signed-off-by: NArchit Taneja <archit@ti.com>
Signed-off-by: NTomi Valkeinen <tomi.valkeinen@ti.com>

When omapfb.mode is passed through bootargs, when omapfb is setting mode,
it would check if timings passed are fine for panel attached to it.
It makes use of check_timing API provided by the panel.

In current code if check_timing API is not available for attached panel,
OMAPFB would return -EINVAL and BPP sent via bootargs will not have any effect.

In case of panels like TAAL panel, omapfb or any other driver should not be allowed to
change the timings. So bpps sent via bootargs will not have an effect.

In such case we can check only the x_res and y_res with the panels resolution
and if they match go ahead and set the bpps.
The bpp value sent via bootarg would have an effect.
Signed-off-by: NMayuresh Janorkar <mayur@ti.com>
Signed-off-by: NTomi Valkeinen <tomi.valkeinen@ti.com>

Add support for Samsung LTE430WQ-F0C to panel-generic-dpi.
Signed-off-by: NTomi Valkeinen <tomi.valkeinen@ti.com>

DSS submodules DPI/SDI/DSI/VENC require a regulator to function.
However, if the board doesn't use, say, SDI, the board shouldn't need to
configure vdds_sdi regulator required by the SDI module.

Currently the regulators are acquired when the DSS driver is loaded.
This means that if the kernel is configured with SDI, vdds_sdi regulator
is needed for all boards.

This patch changes the DSS driver to acquire the regulators only when a
display of particular type is initialized. For example, vdds_sdi is
acquired when sdi_init_display() is called.
Signed-off-by: NTomi Valkeinen <tomi.valkeinen@ti.com>

MODULE_PARM_DESC() takes the name of the actual module parameter, not the
name of the variable, as input. Fix the module parameter description for
def_disp.
Signed-off-by: NJani Nikula <jani@nikula.org>
Signed-off-by: NTomi Valkeinen <tomi.valkeinen@ti.com>

Enable DSS2 and OMAPFB for OMAP4 in Kconfig
Signed-off-by: NMayuresh Janorkar <mayur@ti.com>
Signed-off-by: NTomi Valkeinen <tomi.valkeinen@ti.com>

The dss struct in dss.c has omap2/3 specific clock names. Making them generic,
to increase readability and extendability.
Signed-off-by: NSumit Semwal <sumit.semwal@ti.com>
Signed-off-by: NArchit Taneja <archit@ti.com>
Signed-off-by: NTomi Valkeinen <tomi.valkeinen@ti.com>

enum dss_clock structure is replaced with generic names that
could be used across OMAP2420, 2430, 3xxx, 44xx platforms.
Signed-off-by: NSumit Semwal <sumit.semwal@ti.com>
Signed-off-by: NArchit Taneja <archit@ti.com>
Signed-off-by: NTomi Valkeinen <tomi.valkeinen@ti.com>

Currently, clock database has <dev, clock-name> tuples for DSS2. Because of
this, the clock names are different across different OMAP platforms.

This patch aligns the DSS2 clock names and roles across OMAP 2420, 2430, 3xxx,
44xx platforms in the clock databases, hwmod databases for opt-clocks, and DSS
clock handling.

This ensures that clk_get/put/enable/disable APIs in DSS can use uniform role
names.
Signed-off-by: NSumit Semwal <sumit.semwal@ti.com>
Acked-by: NPaul Walmsley <paul@pwsan.com>
Signed-off-by: NTomi Valkeinen <tomi.valkeinen@ti.com>

DSS IRQ number can be obtained from platform_get_irq().  This API in turn
picks the right IRQ number belonging to HW IP from the hwmod database.
So hardcoding of IRQ number could be removed.

This IRQ is stored in dss_irq as part of dss structure, and freed it in
dss_exit().
Reviewed-by: NPaul Walmsley <paul@pwsan.com>
Reviewed-by: NKevin Hilman <khilman@ti.com>
Tested-by: NKevin Hilman <khilman@ti.com>
Signed-off-by: NSenthilvadivu Guruswamy <svadivu@ti.com>
Signed-off-by: NSumit Semwal <sumit.semwal@ti.com>
Signed-off-by: NTomi Valkeinen <tomi.valkeinen@ti.com>

DSS, DISPC, DSI, RFBI, VENC baseaddr can be obtained from platform_get_resource().
This API in turn picks the right silicon baseaddr from the hwmod database.
So hardcoding of base addr could be removed.
Reviewed-by: NPaul Walmsley <paul@pwsan.com>
Reviewed-by: NKevin Hilman <khilman@ti.com>
Tested-by: NKevin Hilman <khilman@ti.com>
Signed-off-by: NSumit Semwal <sumit.semwal@ti.com>
Signed-off-by: NSenthilvadivu Guruswamy <svadivu@ti.com>
Signed-off-by: NTomi Valkeinen <tomi.valkeinen@ti.com>

This patch replaces printk's in the init/probe functions to dev_dbg
for boot time optimization.
Reviewed-by: NKevin Hilman <khilman@ti.com>
Tested-by: NKevin Hilman <khilman@ti.com>
Signed-off-by: NSumit Semwal <sumit.semwal@ti.com>
Signed-off-by: NTomi Valkeinen <tomi.valkeinen@ti.com>

Hwmod adaptation design requires each of the DSS HW IP to be a platform driver.
So a platform_driver for DSI is created and init exit methods are moved from core.c
to its driver probe,remove. pdev member has to be maintained by its own drivers.

Also, vdds_dsi regulator handling is copied to dsi.c, since vdds_dsi regulator is
needed by dpi_init() too. Board files are updated accordingly to add 2 instances of
vdds_dsi regulator.

DSI platform driver is registered from inside omap_dss_probe, in the order desired.
Signed-off-by: NSenthilvadivu Guruswamy <svadivu@ti.com>
Signed-off-by: NSumit Semwal <sumit.semwal@ti.com>
Signed-off-by: NTomi Valkeinen <tomi.valkeinen@ti.com>

Hwmod adaptation design requires each of the DSS HW IP to be a platform driver.
So a platform_driver for VENC is created and init exit methods are moved from core.c
to its driver probe,remove. pdev member has to be maintained by its own drivers.

Also, venc_vdda_dac reading is moved to venc.c.

VENC platform driver is registered from inside omap_dss_probe, in the order desired.
Signed-off-by: NSenthilvadivu Guruswamy <svadivu@ti.com>
Signed-off-by: NSumit Semwal <sumit.semwal@ti.com>
Signed-off-by: NTomi Valkeinen <tomi.valkeinen@ti.com>

Hwmod adaptation design requires each of the DSS HW IP to be a platform driver.
So a platform_driver for DISPC is created and init exit methods are moved from core.c
to its driver probe,remove. pdev member has to be maintained by its own drivers.

DISPC platform driver is registered from inside omap_dss_probe, in the order desired.
Signed-off-by: NSenthilvadivu Guruswamy <svadivu@ti.com>
Signed-off-by: NSumit Semwal <sumit.semwal@ti.com>
Signed-off-by: NTomi Valkeinen <tomi.valkeinen@ti.com>

Hwmod adaptation design requires each of the DSS HW IP to be a platform driver.
So a platform_driver for RFBI is created and init exit methods are moved from core.c
to its driver probe,remove. pdev member has to be maintained by its own drivers.

RFBI platform driver is registered from inside omap_dss_probe, in the order desired.
Signed-off-by: NSenthilvadivu Guruswamy <svadivu@ti.com>
Signed-off-by: NSumit Semwal <sumit.semwal@ti.com>
Signed-off-by: NTomi Valkeinen <tomi.valkeinen@ti.com>

All clock management is moved to dss platform driver. clk_get/put APIs use
dss device instead of core platform device.

Hwmod adaptation design requires each of the DSS HW IP to be a platform driver.
So the device name is changed from omapdss to omapdss_dss in 2420, 2430,
3xxx clock database files. Now the core driver "omapdss" only takes care
of panel registration with the custom bus.
core driver also uses the clk_enable() / clk_disable() APIs exposed by DSS for
clock management.
DSS driver would do clock management of clocks needed by DISPC, RFBI, DSI, VENC

TODO:  The clock content would be adapted to omap_hwmod in a seperate series.
Signed-off-by: NSenthilvadivu Guruswamy <svadivu@ti.com>
Signed-off-by: NSumit Semwal <sumit.semwal@ti.com>
Signed-off-by: NTomi Valkeinen <tomi.valkeinen@ti.com>

Hwmod adaptation design requires each of the DSS HW IP to be a platform driver.
So a platform_driver of DSS is created and init exit methods are moved from core.c
to its driver probe,remove. pdev member has to be maintained by its own drivers.

DSS platform driver is registered from inside omap_dss_probe, in the order desired.
Signed-off-by: NSenthilvadivu Guruswamy <svadivu@ti.com>
Signed-off-by: NSumit Semwal <sumit.semwal@ti.com>
Signed-off-by: NTomi Valkeinen <tomi.valkeinen@ti.com>

As part of omap hwmod changes, DSS will not be the only controller of its
clocks. hwmod initialization also enables the interface clocks, and
manages them.
So, when DSS is built as a module, omap_dss_remove doesn't try to disable
all clocks that have a higher usecount.
Reviewed-by: NKevin Hilman <khilman@ti.com>
Tested-by: NKevin Hilman <khilman@ti.com>
Signed-off-by: NSumit Semwal <sumit.semwal@ti.com>
Signed-off-by: NTomi Valkeinen <tomi.valkeinen@ti.com>

Index i was already used in thhe first loop
Signed-off-by: NRoel Kluin <roel.kluin@gmail.com>
Signed-off-by: NDave Airlie <airlied@redhat.com>

This fixes a bug introduced by 807e8e40 ("mmc: Fix sd/sdio/mmc
initialization frequency retries") that prevented SDIO drivers from
performing SDIO commands in their probe routines -- the above patch
called mmc_claim_host() before sdio_add_func(), which causes a deadlock
if an external SDIO driver calls sdio_claim_host().

Fix tested on an OLPC XO-1.75 with libertas on SDIO.
Signed-off-by: NDmitry Shmidt <dimitrysh@google.com>
Reviewed-and-Tested-by: NChris Ball <cjb@laptop.org>
Signed-off-by: NChris Ball <cjb@laptop.org>

The per-vm mutex doesn't prevent this completely, a flush coming from the
BAR VM could potentially happen at the same time as one for the channel
VM.  Not to mention that if/when we get per-client/channel VM, this will
happen far more frequently.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
Signed-off-by: NDave Airlie <airlied@redhat.com>

TTM assumes an error condition from man->func->get_node() means that
something went horribly wrong, and causes it to bail.

The driver is supposed to return 0, and leave mm_node == NULL to
signal that it couldn't allocate any memory.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
Signed-off-by: NDave Airlie <airlied@redhat.com>

Early gen3 and gen2 chipset do not have the relaxed per-surface tiling
constraints of the later chipsets, so we need to check that the GTT
alignment is correct for the new tiling. If it is not, we need to
rebind.
Reported-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>

Andi Kleen narrowed his GPU hangs on his Sugar Bay (SNB desktop) rev 09
down to the use of GPU semaphores, and we already know that they appear
broken up to Huron River (mobile) rev 08. (I'm optimistic that disabling
GPU semaphores is simply hiding another bug by the latency and
side-effects of the additional device interaction it introduces...)

However, use of semaphores is a massive performance improvement... Only
as long as the system remains stable. Enable at your peril.
Reported-by: NAndi Kleen <andi-fd@firstfloor.org>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=33921Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>

If a virtio-console device gets unplugged while a port is open, a
subsequent close() call on the port accesses vqs to free up buffers.
This can lead to a crash.

The buffers are already freed up as a result of the call to
unplug_ports() from virtcons_remove().  The fix is to simply not access
vq information if port->portdev is NULL.
Reported-by: Njuzhang <juzhang@redhat.com>
CC: stable@kernel.org
Signed-off-by: NAmit Shah <amit.shah@redhat.com>
Signed-off-by: NRusty Russell <rusty@rustcorp.com.au>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Whilst the GT is powered down (rc6), writes to MMADDR are placed in a
FIFO by the System Agent. This is a limited resource, only 64 entries, of
which 20 are reserved for Display and PCH writes, and so we must take
care not to queue up too many writes. To avoid this, there is counter
which we can poll to ensure there are sufficient free entries in the
fifo.

"Issuing a write to a full FIFO is not supported; at worst it could
result in corruption or a system hang."
Reported-and-Tested-by: NMatt Turner <mattst88@gmail.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=34056Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>

This reverts commit c2e0eb16.

As it turns out, userspace already depends upon being able to enable
tiling on existing bo which it promises to be large enough for its
purposes i.e. it will not access beyond the end of the last full-tile
row.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=35016Reported-and-tested-by: NKamal Mostafa <kamal@canonical.com>
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>