Most SIMD instructions use the 66/f2/f3 prefixes to distinguish between
different variants of the same instruction. Usually the encoding is quite
regular, but in some cases (including non-SIMD instructions) the prefixes
generate very different instructions. Examples include XCHG/PAUSE,
MOVQ/MOVDQA/MOVDQU, and MOVBE/CRC32.
Allow the emulator to handle these special cases by splitting such opcodes
into groups, with different decode flags and execution functions for different
prefixes.
Signed-off-by: Avi Kivity <avi@redhat.com>
Currently we store a rep prefix as 1 or 2 depending on whether it is a REPE or
REPNE. Since sse instructions depend on the prefix value, store it as the
original opcode to simplify things further on.
Signed-off-by: Avi Kivity <avi@redhat.com>
Since sse instructions can issue 16-byte mmios, we need to support them. We
can't increase the kvm_run mmio buffer size to 16 bytes without breaking
compatibility, so instead we break the large mmios into two smaller 8-byte
ones. Since the bus is 64-bit we aren't breaking any atomicity guarantees.
Signed-off-by: Avi Kivity <avi@redhat.com>
As Avi recently mentioned, the new standard mechanism for exposing features
is KVM_GET_SUPPORTED_CPUID, not spamming CAPs. For some reason async pf
missed that.
So expose async_pf here.
Signed-off-by: Glauber Costa <glommer@redhat.com>
CC: Gleb Natapov <gleb@redhat.com>
CC: Avi Kivity <avi@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Use vmx_set_nmi_mask() instead of open-coding management of
the hardware bit and the software hint (nmi_known_unmasked).
There's a slight change of behaviour when running without
hardware virtual NMI support - we now clear the NMI mask if
NMI delivery faulted in that case as well. This improves
emulation accuracy.
Signed-off-by: Avi Kivity <avi@redhat.com>
When we haven't injected an interrupt, we don't need to recover
the nmi blocking state (since the guest can't set it by itself).
This allows us to avoid a VMREAD later on.
Signed-off-by: Avi Kivity <avi@redhat.com>
We may read the cpl quite often in the same vmexit (instruction privilege
check, memory access checks for instruction and operands), so we gain
a bit if we cache the value.
Signed-off-by: Avi Kivity <avi@redhat.com>
In long mode, vm86 mode is disallowed, so we need not check for
it. Reading rflags.vm may require a VMREAD, so it is expensive.
Signed-off-by: Avi Kivity <avi@redhat.com>
Some rflags bits are owned by the host, not guest, so we need to use
kvm_get_rflags() to strip those bits away or kvm_set_rflags() to add them
back.
Signed-off-by: Avi Kivity <avi@redhat.com>
We can get memslot id from memslot->id directly
Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
The mmci.h header contained a few registers not clearly marked
as ST Micro only, rectify this and remove the HWFC magic in the
process. The idea is to make the mmci.h header file more ordered
so other vendors with PL180 derivates can see where to put in
their custom register defines.
Includes portions of an earlier patch from Sebastian Rasmussen.
Acked-by: Sebastian Rasmussen <sebastian.rasmussen@stericsson.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
The interrupt printing functionality in the ep93xx gpio debugfs function
does not behave as expected. It prints [interrupt] beside all pins which
are capable of being interrupts, not just those which are currently
configured as interrupts.
The best solution is just to remove the custom ep93xx gpio debugfs
function all together. The generic gpiolib one is good enough.
Signed-off-by: Ryan Mallon <ryan@bluewatersys.com>
Acked-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
According to Documentation/Changes, the kernel should be buildable with
GNU make 3.80+. Commit 88d7be031f (kbuild:
Use a single clean rule for kernel and external modules) introduced the
"$(or" construct, which requires make 3.81. This causes "make clean" to
malfunction when it is used with external modules.
Replace "$(or" with an equivalent "$(if" expression, to restore backward
compatibility.
Signed-off-by: Kevin Cernekee <cernekee@gmail.com>
Cc: stable@kernel.org
Signed-off-by: Michal Marek <mmarek@suse.cz>
Disable irqs during optimized callback, so we dont miss any in-irq kprobes.
The following commands:
# cd /debug/tracing/
# echo "p mutex_unlock" >> kprobe_events
# echo "p _raw_spin_lock" >> kprobe_events
# echo "p smp_apic_timer_interrupt" >> ./kprobe_events
# echo 1 > events/enable
Cause the optimized kprobes to be missed. None is missed
with the fix applied.
Signed-off-by: Jiri Olsa <jolsa@redhat.com>
Acked-by: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Link: http://lkml.kernel.org/r/20110511110613.GB2390@jolsa.brq.redhat.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
When DBG() is used in a simple if-else, the resulting code path
currently depends on the definition of DBG(). Inserting the statement in
a "do { ... } while (0)" prevents this possible misuse.
Signed-off-by: Niels de Vos <ndevos@redhat.com>
[tomi.valkeinen@ti.com: changed the title of the commit msg]
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Rather than always add the omap2 dirs to the build list (and thus
force everyone to generate a useless built-in.o), bind the dirs to
their relevant kconfig symbol.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
DSS_IRQSTATUS register is available only on OMAP3, and reading it in
dss_dump_registers() on OMAP4 seems to cause a crash.
The register is not used for anything, and displaying its value is of no
use, so let's just remove it altogether.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
struct dispc_reg was originally used while migrating from old omapfb to
catch cases where the arguments to dispc_read_reg/dispc_write_reg were in
wrong order, since old omapfb had the arguments in reverse order.
Remove this struct and use u16 instead
Signed-off-by: Archit Taneja <archit@ti.com>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Represent manager/channel specific DISPC registers as inline functions returning
the required dispc_reg struct. This is done since the current method is not
scalable as the number of overlay managers increase in number.
Signed-off-by: Archit Taneja <archit@ti.com>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Move all DISPC register definitions to a new header dispc.h. There are
separate register offset definitions for GFX, VID1 and VID2 pipeline share
register definitions by using an argument. The introduction of VID3 pipeline
on OMAP4 will not let us use the above method since VID3 pipe register offsets
don't map with VID1 and VID2 offsets.
Represent overlay registers as DISPC_OVL_XXXX(plane), where the plane argument
tells the overlay. Register offsets are calculated as:
DISPC_OVL_XXXX(plane) = DISPC_OVL_BASE(plane) + DISPC_XXXX_OFFSET(plane)
Idea suggested by Tomi Valkeinen.
Signed-off-by: Archit Taneja <archit@ti.com>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
OMAP4 requires SCP clocks enabled to access DSI PLL registers and DSI COMPLEXIO
registers. Enable scp clock before accessing the registers and disable it before
exiting dsi_dump_regs().
Signed-off-by: Archit Taneja <archit@ti.com>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Add support for Powertip PH480242T, a LCD 4.3inch (480x242) display
type with 24-bit RGB interface, to panel-generic-dpi.
Tested with IGEP v2 board.
Signed-off-by: Enric Balletbo i Serra <eballetbo@iseebcn.com>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Add support for Seiko 70WVW1TZ3, a LCD 7.0inch WVGA (800x480) display
type with 24-bit RGB interface and Touch-Panel, to panel-generic-dpi
Tested with IGEP v2 board.
Signed-off-by: Enric Balletbo i Serra <eballetbo@iseebcn.com>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
By default, the DSI is not getting enabled for omap4.
OMAP2PLUS does not catch this issue since it has ARCH_OMAP3.
The issue is only seen when using defconfig with ARCH_OMAP4 only.
Signed-off-by: Vikram Pandita <vikram.pandita@ti.com>
Cc: Archit Taneja <archit@ti.com>
Cc: Tomi Valkeinen <tomi.valkeinen@ti.com>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
OMAP4 DSI block has new interrupts for the two new DSI lanes.
Add definitions for those interrupts, and add the interrupts to the CIO
error mask.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
ULPS is a low power state where the DSI lanes are kept at ground. This
patch implements ULPS by having a DSI bus inactivity timer which
triggers the entry to ULPS. ULPS exit will happen automatically when the
driver needs to do something on the DSI lanes.
The ulps_timeout is configurable from board file or via sysfs.
Additionally another sysfs file, "ulps", can be used to check the
current ULPS state, or to manually enter or exit ULPS.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
ESD workqueue will be shared with other functionality also. Rename
"esd_wq" to "workqueue" to better reflect its usage.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Separate panel reset code to a function of its own. This will keep the
code cleaner in the future when panel reset is called from multiple
locations.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
ESD check in Taal driver is currently on/off feature with hardcoded
interval. This patch changes it to a configurable interval, which can be
set from the board file.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
The panel drivers can leave the VDDS_DSI regulator enabled, even when
the panel is disabled, to ensure that the DSI pins are powered.
This patch ensures that VDDS_DSI is disabled on DSI module unload.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Add dsi_mux_pads function pointer to omap_dss_board_info, and use the
function pointer in DSI code to configure the DSI pads either to normal
DSI operation, or to pull down when in ULPS.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Add parameter to omapdss_dsi_display_disable() which the panel driver
can use to tell if the DSI lanes should be put to ULPS before disabling
the interface.
This can be used to skip ULPS entry in cases where the panel doesn't
care about ULPS state, for example when the panel will be reset, or when
the display interface will be enabled again right after the disable.
This will speed up the operation considerably in cases where entering
ULPS would fail with timeout, and the panel driver isn't even interested
in entering ULPS.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Add dsi_cio_wait_tx_clk_esc_reset() function which waits for the
TXCLKESC domains to come out of reset.
Things have worked fine without this, but better be safe than sorry.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Use dsi_enable_scp_clk and dsi_disable_scp_clk in CIO init and uninit,
and improve the CIO init by adding a few status checks and error
handling.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
