Based on a patch from Arnaud Patard (Rtp) <arnaud.patard@rtp-net.org>
A few words about design choices:
* Two registers, at and t0, are reserved for TCG internal use. They are
useful for bswap and 64-bit ops.
* Most ops supports a constant argument with value 0, which is actually
mapped to the zero register.
* While the at register is available for constant loading, ops only
support a limited range of constants. TCG does a better job doing the
register allocation and constant loading by itself. There are plenty of
registers available anyway.
Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>
Similarly to what is already done in tcg_liveness_analysis() when
USE_LIVENESS_ANALYSIS is not set.
Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>
Currently zero extensions ops are implemented by a and op with a
constant. This is then catched in some backend, and replaced by
a zero extension instruction. While this works well on RISC
machines, this adds a useless register move on non-RISC machines.
Example on x86:
ext16u_i32 r1, r2
is translated into
mov %eax,%ebx
movzwl %bx, %ebx
while the optimized version should be:
movzwl %ax, %ebx
This patch adds ext{8,16,32}u_i{32,64} TCG ops that can be
implemented in the backends to avoid emitting useless register
moves.
Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>
This patch uses sxtb for ext8s_i32 and sxth for ext16s_i32 in ARM back-end.
Signed-off-by: Laurent Desnogues <laurent.desnogues@gmail.com>
Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>
the encode_imm function in tcg/arm/tcg-target.c lacks shift declaration.
Laurent
Signed-off-by: Laurent Desnogues <laurent.desnogues@gmail.com>
Signed-off-by: Andrzej Zaborowski <andrew.zaborowski@intel.com>
this patch handles all possible constants for immediate operand of ALU ops.
I'm not very satisfied by the implementation.
Laurent
Signed-off-by: Laurent Desnogues <laurent.desnogues@gmail.com>
Signed-off-by: Andrzej Zaborowski <andrew.zaborowski@intel.com>
- use movw/movt to load immediate values for ARMv7-A
- implement add/sub/and/or/xor with immediate (only 8-bit)
Laurent
Signed-off-by: Laurent Desnogues <laurent.desnogues@gmail.com>
Signed-off-by: Andrzej Zaborowski <andrew.zaborowski@intel.com>
Re-implement GUEST_BASE support.
Offset guest ddress space by default if the guest binary contains
regions below the host mmap_min_addr.
Implement support for i386, x86-64 and arm hosts.
Signed-off-by: Riku Voipio <riku.voipio@iki.fi>
Signed-off-by: Paul Brook <paul@codesourcery.com>
Include assert.h from qemu-common.h and remove other direct uses.
cpu-all.h still need to include it because of the dyngen-exec.h hacks
Signed-off-by: Paul Brook <paul@codesourcery.com>
This patch allows DEBUG_TCGV to be defined (and also prevents NDEBUG
from being defined) when passing an option to the configure script.
This should help to prevent any accidental changes that enable
DEBUG_TCGV in tcg/tcg.h from being committed in future, and may
help to encourage testing with DEBUG_TCGV enabled.
Signed-off-by: Stuart Brady <stuart.brady@gmail.com>
Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>
git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@7105 c046a42c-6fe2-441c-8c8c-71466251a162
A const function only reads its arguments and does not use TCG
globals variables. Hence a call to such a function does not
save TCG globals variabes back to their canonical location.
Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>
git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@7008 c046a42c-6fe2-441c-8c8c-71466251a162
The beginning of the register allocation order list on the TCG x86_64
target matches the list of clobbered registers. This means that when an
helper is called, there is almost always clobbered registers that have
to be spilled.
The same way register %rsi and %rdi are at the top of the register
allocation order list, while they can't be used for load/store
operations. This means the data and/or address registers are very often
%rsi and %rdi, and their values have to be spilled, and then moved back
to another register.
This patches changes to the allocation order to avoid those effects.
It results in a 8% gain speed in qemu-x86_64 to compress a bzip2 file,
and a 6% gain in qemu-system-mips64 to compile a small application.
Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>
git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@7003 c046a42c-6fe2-441c-8c8c-71466251a162
Aurelien Jarno