kmalloc allocates physically contiguous memory and
may fail for larger allocations due to fragmentation.
The large allocations are caused by the fact that the
scatterlist structure is 24 bytes and the array size
is proportional to the number of pages being mapped.
The tools that process cff dumps expect a linear
memory region, but the start address of that region can
be configured. As long as there is only a single
pagetable (so that there aren't duplicate virtual
addresses in the dump), dumps captured with the
mmu on are easier to deal with than reconfiguring
to turn the mmu off.
Make the framework for reporting per-process memory statistics a little bit
more generic. This should make it easier to keep track of more external
memory sources as they are added.
Certain memory allocations are not properly tracked by kmemleak tool,
which makes it to incorrectly detect memory leak. Notify the tool by using
kmemleak_not_leak() to ignore the memory allocation so that incorrect leaks
report are avoided.
Poking once during adreno_idle is not enough; a GPU hang may still happen.
Seen on 7x27A. Write a few times during the wait timeout, to ensure that
the WPTR is updated properly.
The existing timestamp_cmp function returns a different
result depending on the order of the input parameters due to
having an asymetric valid window. When no rollover is
detected the window is 2^31 but when a rollover is detected
the window is 25000. This change makes the rollover window
symmetric at 2^31.
This function was incorrectly reporting hangs when an
error such as ERESTARTSYS was returned by
__wait_event_interruptible_timeout().
msm: kgsl: Make sure WPTR reg is updated properly
Sometimes writes to WPTR register do not take effect, causing a
3D core hang. Make sure the WPTR is updated properly when waiting.
msm: kgsl: Set default value of wait_timeout in the adreno_dev struct
Set the initalization value of wait_timeout at compile time in the
declaration of the adreno_device struct instead of at runtime in
adreno_probe.
This function is supposed to return the memdesc that
contains the range gpuaddr to gpuaddr + size. One of the
lookups was using sizeof(unsigned int) instead of size,
which could cause false positive results from this function
and possibly kernel panics in the snapshot or postmortem
code, which rely on it to do bounds checking for them.
Some hangs are fooling the postmortem dump code into
running off the end of a buffer. Fix this by making
its bounds check logic work better by reusing the
logic from kgsl_find_region().
Userspace will set a flag in the context if preambles are in use. If
they are, we can safely skip save and restore commands for the
context. GMEM save/restore is still required. To improve performance,
preamble commands are skipped when the context hasn't changed since
the last issueibcmds.
from Code Aurora
