Changes v4-v5:
* Incorprate stack realignment into frame size, instead of pushing r12.
* Disassemble prologues with -d out_asm.
Changes v3-v4:
* Local stack frame fix has been reworked to be independant of the rest
of the change set, making it possible to cherry-pick to stable.
* While doing that, I've dropped the patch that pulled the epilogue
into the exit_tb expansion, as the stated reason (one insn in the
epilogue) is no longer true -- there's the stack pop as well. This
obviated several of the changes that PMM suggested.
* I've dropped all of the changes to goto_tb, as Aurelien tried the
same thing a while ago, and there were reports of no real speedup,
ascribed to Dcache pollution.
Hopefully I've addressed all of the outstanding comments to which there
is still code that applies. The patch set is available at
git://github.com/rth7680/qemu.git tcg-arm
r~
Richard Henderson (19):
tcg-arm: Fix local stack frame
tcg: Log the contents of the prologue with -d out_asm
tcg-arm: Use bic to implement and with constant
tcg-arm: Handle negated constant arguments to and/sub
tcg-arm: Allow constant first argument to sub
tcg-arm: Use tcg_out_dat_rIN for compares
tcg-arm: Handle constant arguments to add2/sub2
tcg-arm: Improve constant generation
tcg-arm: Implement deposit for armv7
tcg-arm: Implement division instructions
tcg-arm: Use TCG_REG_TMP name for the tcg temporary
tcg-arm: Use R12 for the tcg temporary
tcg-arm: Cleanup multiply subroutines
tcg-arm: Cleanup most primitive load store subroutines
tcg-arm: Split out tcg_out_tlb_read
tcg-arm: Improve scheduling of tcg_out_tlb_read
tcg-arm: Use movi32 + blx for calls on v7
tcg-arm: Convert to CONFIG_QEMU_LDST_OPTIMIZATION
tcg-arm: Tidy exit_tb
configure | 2 +-
disas/arm.c | 4 +
include/exec/exec-all.h | 17 +
tcg/arm/tcg-target.c | 1470 +++++++++++++++++++++++++++--------------------
tcg/arm/tcg-target.h | 14 +-
tcg/tcg.c | 10 +
6 files changed, 875 insertions(+), 642 deletions(-)