This hack was introduced in issue#287 which later becomes endless maintainance pain.
=====
Our previous check_exit_request use `brcond` in the middle of a TranslationBlock which
breaks the assumptions and thus a hack to liveness_pass_1 is used for _all_ brcond instructions
which causes issues for MIPS and many other scenarios.
=====
This patch also resolves PC not sync-ed when no memory hooks are installed, finally. Now
Unicorn will always have correct PC no matter what happens.
* qemu/tcg: fix UC_HOOK_MEM_READ on aarch64.
Directly jump into the slow path when there is any hookmem enabled. This
fixes#1908.
Signed-off-by: Glenn Baker <glenn.baker@gmx.com>
* qemu/tcg: fix UC_HOOK_MEM_READ on ppc64.
Directly jump into the slow path when there is any hookmem enabled.
Signed-off-by: Glenn Baker <glenn.baker@gmx.com>
* qemu/tcg: check for UC_HOOK_MEM_READ_AFTER.
Use has_hookmem() helper to determine wether "slow-path" TLB read is
needed. Add this helper to x86 architecture as well so that to check for
all hookmem.
Signed-off-by: Glenn Baker <glenn.baker@gmx.com>
* qemu/tcg: factor out has_hookmem().
It's the same implementation for all architectures, so factor out
has_hookmem() into tcg_uc_has_hookmem().
Signed-off-by: Glenn Baker <glenn.baker@gmx.com>
---------
Signed-off-by: Glenn Baker <glenn.baker@gmx.com>
* Remove global variable from aarch64 tcg target
This obviously breaks trying to run two unicorn instances at once on
aarch64. It appears a similar variable had already been moved to the
state struct for i386 tcg target.
* Reenable writing to jit region while calling tb_add_jump
On arm macs, every place that writes to jit code needs to have
tb_exec_unlock called first. This is already in most necessary places,
but not this one.
* Don't forget to call restore_jit_state in uc_context_restore
Every time UC_INIT is used, restore_jit_state must be used on the return
path, or occasional assertion failures will pop up on arm macs.
* Restore pc before calling into tlb fill hook
In my application it is important to have correct pc values available
from this hook.
tcg: Fix do_nonatomic_op_* vs signed operations
The smin/smax/umin/umax operations require the operands to be
properly sign extended. Do not drop the MO_SIGN bit from the
load, and additionally extend the val input.
1. Add cmake support in CMakeLists.txt according to https://developer.android.com/ndk/guides/other_build_systems
2. Resolve symbols errors
3. Backport fixes from 438ed42311
> QEMU relies on two optimization for ppc64 and arm:
>
> 1. if(0) /* optimized code */
> 2. assert(0); /* optimized code */
>
> But the assert on mingw32 doesn't have noreturn attribute which prevents
> the second optimization and some code is reverted to the original code
> to fit in the first optimization.
>
> The assert implementation is copied from glib as qemu did.
Unfortunately, NDK also doesn't have an assert implementation qemu prefers.
* fix finding python path which only has python3.
* fix#1246, arm host issue.
* skip assembler tests on non-x86 host.
* update macro of dummy value.
* fix MSVC macro.
* update dummy array value macro.
* restore to original qemu code.
