The following changes since commit 627634031092e1514f363fd8659a579398de0f0e: Merge tag 'buildsys-qom-qdev-ui-20230227' of https://github.com/philmd/qemu into staging (2023-02-28 15:09:18 +0000) are available in the Git repository at: https://gitlab.com/rth7680/qemu.git tags/pull-tcg-20230228 for you to fetch changes up to c7fbf10db8718d2eba87712bc3410b671157a377: tcg: Update docs/devel/tcg-ops.rst for temporary changes (2023-02-28 10:36:19 -1000) ---------------------------------------------------------------- helper-head: Add fpu/softfloat-types.h softmmu: Use memmove in flatview_write_continue tcg: Add sign param to probe_access_flags, probe_access_full tcg: Convert TARGET_TB_PCREL to CF_PCREL tcg: Simplify temporary lifetimes for translators ---------------------------------------------------------------- Akihiko Odaki (1): softmmu: Use memmove in flatview_write_continue Anton Johansson via (27): include/exec: Introduce `CF_PCREL` target/i386: set `CF_PCREL` in `x86_cpu_realizefn` target/arm: set `CF_PCREL` in `arm_cpu_realizefn` accel/tcg: Replace `TARGET_TB_PCREL` with `CF_PCREL` include/exec: Replace `TARGET_TB_PCREL` with `CF_PCREL` target/arm: Replace `TARGET_TB_PCREL` with `CF_PCREL` target/i386: Replace `TARGET_TB_PCREL` with `CF_PCREL` include/exec: Remove `TARGET_TB_PCREL` define target/arm: Remove `TARGET_TB_PCREL` define target/i386: Remove `TARGET_TB_PCREL` define accel/tcg: Move jmp-cache `CF_PCREL` checks to caller accel/tcg: Replace `tb_pc()` with `tb->pc` target/tricore: Replace `tb_pc()` with `tb->pc` target/sparc: Replace `tb_pc()` with `tb->pc` target/sh4: Replace `tb_pc()` with `tb->pc` target/rx: Replace `tb_pc()` with `tb->pc` target/riscv: Replace `tb_pc()` with `tb->pc` target/openrisc: Replace `tb_pc()` with `tb->pc` target/mips: Replace `tb_pc()` with `tb->pc` target/microblaze: Replace `tb_pc()` with `tb->pc` target/loongarch: Replace `tb_pc()` with `tb->pc` target/i386: Replace `tb_pc()` with `tb->pc` target/hppa: Replace `tb_pc()` with `tb->pc` target/hexagon: Replace `tb_pc()` with `tb->pc` target/avr: Replace `tb_pc()` with `tb->pc` target/arm: Replace `tb_pc()` with `tb->pc` include/exec: Remove `tb_pc()` Daniel Henrique Barboza (1): accel/tcg: Add 'size' param to probe_access_flags() Philippe Mathieu-Daudé (1): exec/helper-head: Include missing "fpu/softfloat-types.h" header Richard Henderson (32): accel/tcg: Add 'size' param to probe_access_full tcg: Adjust TCGContext.temps_in_use check accel/tcg: Pass max_insn to gen_intermediate_code by pointer accel/tcg: Use more accurate max_insns for tb_overflow tcg: Remove branch-to-next regardless of reference count tcg: Rename TEMP_LOCAL to TEMP_TB tcg: Use noinline for major tcg_gen_code subroutines tcg: Add liveness_pass_0 tcg: Remove TEMP_NORMAL tcg: Pass TCGTempKind to tcg_temp_new_internal tcg: Use tcg_constant_i32 in tcg_gen_io_start tcg: Add tcg_gen_movi_ptr tcg: Add tcg_temp_ebb_new_{i32,i64,ptr} tcg: Use tcg_temp_ebb_new_* in tcg/ tcg: Use tcg_constant_ptr in do_dup accel/tcg/plugin: Use tcg_temp_ebb_* accel/tcg/plugin: Tidy plugin_gen_disable_mem_helpers tcg: Don't re-use TEMP_TB temporaries tcg: Change default temp lifetime to TEMP_TB target/arm: Drop copies in gen_sve_{ldr,str} target/arm: Don't use tcg_temp_local_new_* target/cris: Don't use tcg_temp_local_new target/hexagon: Don't use tcg_temp_local_new_* target/hexagon/idef-parser: Drop gen_tmp_local target/hppa: Don't use tcg_temp_local_new target/i386: Don't use tcg_temp_local_new target/mips: Don't use tcg_temp_local_new target/ppc: Don't use tcg_temp_local_new target/xtensa: Don't use tcg_temp_local_new_* exec/gen-icount: Don't use tcg_temp_local_new_i32 tcg: Remove tcg_temp_local_new_*, tcg_const_local_* tcg: Update docs/devel/tcg-ops.rst for temporary changes docs/devel/tcg-ops.rst | 230 +++++++++++++---------- target/hexagon/idef-parser/README.rst | 4 +- accel/tcg/internal.h | 10 +- accel/tcg/tb-jmp-cache.h | 42 +---- include/exec/cpu-defs.h | 3 - include/exec/exec-all.h | 26 +-- include/exec/gen-icount.h | 12 +- include/exec/helper-head.h | 2 + include/exec/translator.h | 4 +- include/tcg/tcg-op.h | 7 +- include/tcg/tcg.h | 64 ++++--- target/arm/cpu-param.h | 2 - target/arm/tcg/translate-a64.h | 1 - target/arm/tcg/translate.h | 2 +- target/hexagon/gen_tcg.h | 4 +- target/i386/cpu-param.h | 4 - accel/stubs/tcg-stub.c | 2 +- accel/tcg/cpu-exec.c | 62 ++++-- accel/tcg/cputlb.c | 21 ++- accel/tcg/perf.c | 2 +- accel/tcg/plugin-gen.c | 32 ++-- accel/tcg/tb-maint.c | 10 +- accel/tcg/translate-all.c | 18 +- accel/tcg/translator.c | 6 +- accel/tcg/user-exec.c | 5 +- semihosting/uaccess.c | 2 +- softmmu/physmem.c | 2 +- target/alpha/translate.c | 2 +- target/arm/cpu.c | 17 +- target/arm/ptw.c | 4 +- target/arm/tcg/mte_helper.c | 4 +- target/arm/tcg/sve_helper.c | 4 +- target/arm/tcg/translate-a64.c | 16 +- target/arm/tcg/translate-sve.c | 38 +--- target/arm/tcg/translate.c | 14 +- target/avr/cpu.c | 3 +- target/avr/translate.c | 2 +- target/cris/translate.c | 8 +- target/hexagon/cpu.c | 4 +- target/hexagon/genptr.c | 16 +- target/hexagon/idef-parser/parser-helpers.c | 26 +-- target/hexagon/translate.c | 4 +- target/hppa/cpu.c | 8 +- target/hppa/translate.c | 5 +- target/i386/cpu.c | 5 + target/i386/helper.c | 2 +- target/i386/tcg/sysemu/excp_helper.c | 4 +- target/i386/tcg/tcg-cpu.c | 8 +- target/i386/tcg/translate.c | 55 +++--- target/loongarch/cpu.c | 6 +- target/loongarch/translate.c | 2 +- target/m68k/translate.c | 2 +- target/microblaze/cpu.c | 4 +- target/microblaze/translate.c | 2 +- target/mips/tcg/exception.c | 3 +- target/mips/tcg/sysemu/special_helper.c | 2 +- target/mips/tcg/translate.c | 59 ++---- target/nios2/translate.c | 2 +- target/openrisc/cpu.c | 4 +- target/openrisc/translate.c | 2 +- target/ppc/translate.c | 8 +- target/riscv/cpu.c | 7 +- target/riscv/translate.c | 2 +- target/rx/cpu.c | 3 +- target/rx/translate.c | 2 +- target/s390x/tcg/mem_helper.c | 2 +- target/s390x/tcg/translate.c | 2 +- target/sh4/cpu.c | 6 +- target/sh4/translate.c | 2 +- target/sparc/cpu.c | 4 +- target/sparc/translate.c | 2 +- target/tricore/cpu.c | 3 +- target/tricore/translate.c | 2 +- target/xtensa/translate.c | 18 +- tcg/optimize.c | 2 +- tcg/tcg-op-gvec.c | 189 ++++++++++--------- tcg/tcg-op.c | 258 ++++++++++++------------- tcg/tcg.c | 280 ++++++++++++++++------------ target/cris/translate_v10.c.inc | 10 +- target/mips/tcg/nanomips_translate.c.inc | 4 +- target/ppc/translate/spe-impl.c.inc | 8 +- target/ppc/translate/vmx-impl.c.inc | 4 +- target/hexagon/README | 8 +- target/hexagon/gen_tcg_funcs.py | 18 +- 84 files changed, 870 insertions(+), 890 deletions(-) From: Philippe Mathieu-Daudé <philmd@linaro.org> 'dh_ctype_f32' is defined as 'float32', itself declared in "fpu/softfloat-types.h". Include this header to avoid when refactoring other headers: In file included from include/exec/helper-proto.h:7, from include/tcg/tcg-op.h:29, from ../../tcg/tcg-op-vec.c:22: include/exec/helper-head.h:44:22: error: unknown type name ‘float32’; did you mean ‘_Float32’? 44 | #define dh_ctype_f32 float32 | ^~~~~~~ Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> Message-Id: <20221216225202.25664-1-philmd@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- include/exec/helper-head.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/include/exec/helper-head.h b/include/exec/helper-head.h index XXXXXXX..XXXXXXX 100644 --- a/include/exec/helper-head.h +++ b/include/exec/helper-head.h @@ -XXX,XX +XXX,XX @@ #ifndef EXEC_HELPER_HEAD_H #define EXEC_HELPER_HEAD_H +#include "fpu/softfloat-types.h" + #define HELPER(name) glue(helper_, name) /* Some types that make sense in C, but not for TCG. */ -- 2.34.1 From: Akihiko Odaki <akihiko.odaki@daynix.com> We found a case where the source passed to flatview_write_continue() may overlap with the destination when fuzzing igb, a new proposed network device with sanitizers. igb uses pci_dma_map() to get Tx packet, and pci_dma_write() to write Rx buffer. While pci_dma_write() is usually used to write data from memory not mapped to the guest, if igb is configured to perform loopback, the data will be sourced from the guest memory. The source and destination can overlap and the usage of memcpy() will be invalid in such a case. While we do not really have to deal with such an invalid request for igb, detecting the overlap in igb code beforehand requires complex code, and only covers this specific case. Instead, just replace memcpy() with memmove() to tolerate overlaps. Using memmove() will slightly damage the performance as it will need to check overlaps before using SIMD instructions for copying, but the cost should be negligible, considering the inherent complexity of flatview_write_continue(). The test cases generated by the fuzzer is available at: https://patchew.org/QEMU/20230129053316.1071513-1-alxndr@bu.edu/ The fixed test case is: fuzz/crash_47dfe62d9f911bf523ff48cd441b61c0013ed805 Signed-off-by: Akihiko Odaki <akihiko.odaki@daynix.com> Acked-by: Alexander Bulekov <alxndr@bu.edu> Acked-by: David Hildenbrand <david@redhat.com> Message-Id: <20230131030155.18932-1-akihiko.odaki@daynix.com> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- softmmu/physmem.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/softmmu/physmem.c b/softmmu/physmem.c index XXXXXXX..XXXXXXX 100644 --- a/softmmu/physmem.c +++ b/softmmu/physmem.c @@ -XXX,XX +XXX,XX @@ static MemTxResult flatview_write_continue(FlatView *fv, hwaddr addr, } else { /* RAM case */ ram_ptr = qemu_ram_ptr_length(mr->ram_block, addr1, &l, false); - memcpy(ram_ptr, buf, l); + memmove(ram_ptr, buf, l); invalidate_and_set_dirty(mr, addr1, l); } -- 2.34.1 From: Daniel Henrique Barboza <dbarboza@ventanamicro.com> probe_access_flags() as it is today uses probe_access_full(), which in turn uses probe_access_internal() with size = 0. probe_access_internal() then uses the size to call the tlb_fill() callback for the given CPU. This size param ('fault_size' as probe_access_internal() calls it) is ignored by most existing .tlb_fill callback implementations, e.g. arm_cpu_tlb_fill(), ppc_cpu_tlb_fill(), x86_cpu_tlb_fill() and mips_cpu_tlb_fill() to name a few. But RISC-V riscv_cpu_tlb_fill() actually uses it. The 'size' parameter is used to check for PMP (Physical Memory Protection) access. This is necessary because PMP does not make any guarantees about all the bytes of the same page having the same permissions, i.e. the same page can have different PMP properties, so we're forced to make sub-page range checks. To allow RISC-V emulation to do a probe_acess_flags() that covers PMP, we need to either add a 'size' param to the existing probe_acess_flags() or create a new interface (e.g. probe_access_range_flags). There are quite a few probe_* APIs already, so let's add a 'size' param to probe_access_flags() and re-use this API. This is done by open coding what probe_access_full() does inside probe_acess_flags() and passing the 'size' param to probe_acess_internal(). Existing probe_access_flags() callers use size = 0 to not change their current API usage. 'size' is asserted to enforce single page access like probe_access() already does. No behavioral changes intended. Signed-off-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Message-Id: <20230223234427.521114-2-dbarboza@ventanamicro.com> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- include/exec/exec-all.h | 3 ++- accel/stubs/tcg-stub.c | 2 +- accel/tcg/cputlb.c | 17 ++++++++++++++--- accel/tcg/user-exec.c | 5 +++-- semihosting/uaccess.c | 2 +- target/arm/ptw.c | 2 +- target/arm/tcg/sve_helper.c | 2 +- target/s390x/tcg/mem_helper.c | 2 +- 8 files changed, 24 insertions(+), 11 deletions(-) diff --git a/include/exec/exec-all.h b/include/exec/exec-all.h index XXXXXXX..XXXXXXX 100644 --- a/include/exec/exec-all.h +++ b/include/exec/exec-all.h @@ -XXX,XX +XXX,XX @@ static inline void *probe_read(CPUArchState *env, target_ulong addr, int size, * probe_access_flags: * @env: CPUArchState * @addr: guest virtual address to look up + * @size: size of the access * @access_type: read, write or execute permission * @mmu_idx: MMU index to use for lookup * @nonfault: suppress the fault @@ -XXX,XX +XXX,XX @@ static inline void *probe_read(CPUArchState *env, target_ulong addr, int size, * Do handle clean pages, so exclude TLB_NOTDIRY from the returned flags. * For simplicity, all "mmio-like" flags are folded to TLB_MMIO. */ -int probe_access_flags(CPUArchState *env, target_ulong addr, +int probe_access_flags(CPUArchState *env, target_ulong addr, int size, MMUAccessType access_type, int mmu_idx, bool nonfault, void **phost, uintptr_t retaddr); diff --git a/accel/stubs/tcg-stub.c b/accel/stubs/tcg-stub.c index XXXXXXX..XXXXXXX 100644 --- a/accel/stubs/tcg-stub.c +++ b/accel/stubs/tcg-stub.c @@ -XXX,XX +XXX,XX @@ void tcg_flush_jmp_cache(CPUState *cpu) { } -int probe_access_flags(CPUArchState *env, target_ulong addr, +int probe_access_flags(CPUArchState *env, target_ulong addr, int size, MMUAccessType access_type, int mmu_idx, bool nonfault, void **phost, uintptr_t retaddr) { diff --git a/accel/tcg/cputlb.c b/accel/tcg/cputlb.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/cputlb.c +++ b/accel/tcg/cputlb.c @@ -XXX,XX +XXX,XX @@ int probe_access_full(CPUArchState *env, target_ulong addr, return flags; } -int probe_access_flags(CPUArchState *env, target_ulong addr, +int probe_access_flags(CPUArchState *env, target_ulong addr, int size, MMUAccessType access_type, int mmu_idx, bool nonfault, void **phost, uintptr_t retaddr) { CPUTLBEntryFull *full; + int flags; - return probe_access_full(env, addr, access_type, mmu_idx, - nonfault, phost, &full, retaddr); + g_assert(-(addr | TARGET_PAGE_MASK) >= size); + + flags = probe_access_internal(env, addr, size, access_type, mmu_idx, + nonfault, phost, &full, retaddr); + + /* Handle clean RAM pages. */ + if (unlikely(flags & TLB_NOTDIRTY)) { + notdirty_write(env_cpu(env), addr, 1, full, retaddr); + flags &= ~TLB_NOTDIRTY; + } + + return flags; } void *probe_access(CPUArchState *env, target_ulong addr, int size, diff --git a/accel/tcg/user-exec.c b/accel/tcg/user-exec.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/user-exec.c +++ b/accel/tcg/user-exec.c @@ -XXX,XX +XXX,XX @@ static int probe_access_internal(CPUArchState *env, target_ulong addr, cpu_loop_exit_sigsegv(env_cpu(env), addr, access_type, maperr, ra); } -int probe_access_flags(CPUArchState *env, target_ulong addr, +int probe_access_flags(CPUArchState *env, target_ulong addr, int size, MMUAccessType access_type, int mmu_idx, bool nonfault, void **phost, uintptr_t ra) { int flags; - flags = probe_access_internal(env, addr, 0, access_type, nonfault, ra); + g_assert(-(addr | TARGET_PAGE_MASK) >= size); + flags = probe_access_internal(env, addr, size, access_type, nonfault, ra); *phost = flags ? NULL : g2h(env_cpu(env), addr); return flags; } diff --git a/semihosting/uaccess.c b/semihosting/uaccess.c index XXXXXXX..XXXXXXX 100644 --- a/semihosting/uaccess.c +++ b/semihosting/uaccess.c @@ -XXX,XX +XXX,XX @@ ssize_t softmmu_strlen_user(CPUArchState *env, target_ulong addr) /* Find the number of bytes remaining in the page. */ left_in_page = TARGET_PAGE_SIZE - (addr & ~TARGET_PAGE_MASK); - flags = probe_access_flags(env, addr, MMU_DATA_LOAD, + flags = probe_access_flags(env, addr, 0, MMU_DATA_LOAD, mmu_idx, true, &h, 0); if (flags & TLB_INVALID_MASK) { return -1; diff --git a/target/arm/ptw.c b/target/arm/ptw.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/ptw.c +++ b/target/arm/ptw.c @@ -XXX,XX +XXX,XX @@ static uint64_t arm_casq_ptw(CPUARMState *env, uint64_t old_val, void *discard; env->tlb_fi = fi; - flags = probe_access_flags(env, ptw->out_virt, MMU_DATA_STORE, + flags = probe_access_flags(env, ptw->out_virt, 0, MMU_DATA_STORE, arm_to_core_mmu_idx(ptw->in_ptw_idx), true, &discard, 0); env->tlb_fi = NULL; diff --git a/target/arm/tcg/sve_helper.c b/target/arm/tcg/sve_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/sve_helper.c +++ b/target/arm/tcg/sve_helper.c @@ -XXX,XX +XXX,XX @@ bool sve_probe_page(SVEHostPage *info, bool nofault, CPUARMState *env, addr = useronly_clean_ptr(addr); #ifdef CONFIG_USER_ONLY - flags = probe_access_flags(env, addr, access_type, mmu_idx, nofault, + flags = probe_access_flags(env, addr, 0, access_type, mmu_idx, nofault, &info->host, retaddr); #else CPUTLBEntryFull *full; diff --git a/target/s390x/tcg/mem_helper.c b/target/s390x/tcg/mem_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/s390x/tcg/mem_helper.c +++ b/target/s390x/tcg/mem_helper.c @@ -XXX,XX +XXX,XX @@ static inline int s390_probe_access(CPUArchState *env, target_ulong addr, int mmu_idx, bool nonfault, void **phost, uintptr_t ra) { - int flags = probe_access_flags(env, addr, access_type, mmu_idx, + int flags = probe_access_flags(env, addr, 0, access_type, mmu_idx, nonfault, phost, ra); if (unlikely(flags & TLB_INVALID_MASK)) { -- 2.34.1 Change to match the recent change to probe_access_flags. All existing callers updated to supply 0, so no change in behaviour. Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- include/exec/exec-all.h | 2 +- accel/tcg/cputlb.c | 4 ++-- target/arm/ptw.c | 2 +- target/arm/tcg/mte_helper.c | 4 ++-- target/arm/tcg/sve_helper.c | 2 +- target/arm/tcg/translate-a64.c | 2 +- target/i386/tcg/sysemu/excp_helper.c | 4 ++-- 7 files changed, 10 insertions(+), 10 deletions(-) diff --git a/include/exec/exec-all.h b/include/exec/exec-all.h index XXXXXXX..XXXXXXX 100644 --- a/include/exec/exec-all.h +++ b/include/exec/exec-all.h @@ -XXX,XX +XXX,XX @@ int probe_access_flags(CPUArchState *env, target_ulong addr, int size, * and must be consumed or copied immediately, before any further * access or changes to TLB @mmu_idx. */ -int probe_access_full(CPUArchState *env, target_ulong addr, +int probe_access_full(CPUArchState *env, target_ulong addr, int size, MMUAccessType access_type, int mmu_idx, bool nonfault, void **phost, CPUTLBEntryFull **pfull, uintptr_t retaddr); diff --git a/accel/tcg/cputlb.c b/accel/tcg/cputlb.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/cputlb.c +++ b/accel/tcg/cputlb.c @@ -XXX,XX +XXX,XX @@ static int probe_access_internal(CPUArchState *env, target_ulong addr, return flags; } -int probe_access_full(CPUArchState *env, target_ulong addr, +int probe_access_full(CPUArchState *env, target_ulong addr, int size, MMUAccessType access_type, int mmu_idx, bool nonfault, void **phost, CPUTLBEntryFull **pfull, uintptr_t retaddr) { - int flags = probe_access_internal(env, addr, 0, access_type, mmu_idx, + int flags = probe_access_internal(env, addr, size, access_type, mmu_idx, nonfault, phost, pfull, retaddr); /* Handle clean RAM pages. */ diff --git a/target/arm/ptw.c b/target/arm/ptw.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/ptw.c +++ b/target/arm/ptw.c @@ -XXX,XX +XXX,XX @@ static bool S1_ptw_translate(CPUARMState *env, S1Translate *ptw, int flags; env->tlb_fi = fi; - flags = probe_access_full(env, addr, MMU_DATA_LOAD, + flags = probe_access_full(env, addr, 0, MMU_DATA_LOAD, arm_to_core_mmu_idx(s2_mmu_idx), true, &ptw->out_host, &full, 0); env->tlb_fi = NULL; diff --git a/target/arm/tcg/mte_helper.c b/target/arm/tcg/mte_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/mte_helper.c +++ b/target/arm/tcg/mte_helper.c @@ -XXX,XX +XXX,XX @@ static uint8_t *allocation_tag_mem(CPUARMState *env, int ptr_mmu_idx, * valid. Indicate to probe_access_flags no-fault, then assert that * we received a valid page. */ - flags = probe_access_full(env, ptr, ptr_access, ptr_mmu_idx, + flags = probe_access_full(env, ptr, 0, ptr_access, ptr_mmu_idx, ra == 0, &host, &full, ra); assert(!(flags & TLB_INVALID_MASK)); @@ -XXX,XX +XXX,XX @@ static uint8_t *allocation_tag_mem(CPUARMState *env, int ptr_mmu_idx, */ in_page = -(ptr | TARGET_PAGE_MASK); if (unlikely(ptr_size > in_page)) { - flags |= probe_access_full(env, ptr + in_page, ptr_access, + flags |= probe_access_full(env, ptr + in_page, 0, ptr_access, ptr_mmu_idx, ra == 0, &host, &full, ra); assert(!(flags & TLB_INVALID_MASK)); } diff --git a/target/arm/tcg/sve_helper.c b/target/arm/tcg/sve_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/sve_helper.c +++ b/target/arm/tcg/sve_helper.c @@ -XXX,XX +XXX,XX @@ bool sve_probe_page(SVEHostPage *info, bool nofault, CPUARMState *env, &info->host, retaddr); #else CPUTLBEntryFull *full; - flags = probe_access_full(env, addr, access_type, mmu_idx, nofault, + flags = probe_access_full(env, addr, 0, access_type, mmu_idx, nofault, &info->host, &full, retaddr); #endif info->flags = flags; diff --git a/target/arm/tcg/translate-a64.c b/target/arm/tcg/translate-a64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/translate-a64.c +++ b/target/arm/tcg/translate-a64.c @@ -XXX,XX +XXX,XX @@ static bool is_guarded_page(CPUARMState *env, DisasContext *s) * that the TLB entry must be present and valid, and thus this * access will never raise an exception. */ - flags = probe_access_full(env, addr, MMU_INST_FETCH, mmu_idx, + flags = probe_access_full(env, addr, 0, MMU_INST_FETCH, mmu_idx, false, &host, &full, 0); assert(!(flags & TLB_INVALID_MASK)); diff --git a/target/i386/tcg/sysemu/excp_helper.c b/target/i386/tcg/sysemu/excp_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/i386/tcg/sysemu/excp_helper.c +++ b/target/i386/tcg/sysemu/excp_helper.c @@ -XXX,XX +XXX,XX @@ static bool ptw_translate(PTETranslate *inout, hwaddr addr) int flags; inout->gaddr = addr; - flags = probe_access_full(inout->env, addr, MMU_DATA_STORE, + flags = probe_access_full(inout->env, addr, 0, MMU_DATA_STORE, inout->ptw_idx, true, &inout->haddr, &full, 0); if (unlikely(flags & TLB_INVALID_MASK)) { @@ -XXX,XX +XXX,XX @@ do_check_protect_pse36: CPUTLBEntryFull *full; int flags, nested_page_size; - flags = probe_access_full(env, paddr, access_type, + flags = probe_access_full(env, paddr, 0, access_type, MMU_NESTED_IDX, true, &pte_trans.haddr, &full, 0); if (unlikely(flags & TLB_INVALID_MASK)) { -- 2.34.1 From: Anton Johansson via <qemu-devel@nongnu.org> Adds a new field to TranslationBlock.cflags denoting whether or not the instructions of a given translation block are pc-relative. This field aims to replace the macro `TARGET_TB_PCREL`. Signed-off-by: Anton Johansson <anjo@rev.ng> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Message-Id: <20230227135202.9710-2-anjo@rev.ng> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- include/exec/exec-all.h | 1 + 1 file changed, 1 insertion(+) diff --git a/include/exec/exec-all.h b/include/exec/exec-all.h index XXXXXXX..XXXXXXX 100644 --- a/include/exec/exec-all.h +++ b/include/exec/exec-all.h @@ -XXX,XX +XXX,XX @@ struct TranslationBlock { #define CF_INVALID 0x00040000 /* TB is stale. Set with @jmp_lock held */ #define CF_PARALLEL 0x00080000 /* Generate code for a parallel context */ #define CF_NOIRQ 0x00100000 /* Generate an uninterruptible TB */ +#define CF_PCREL 0x00200000 /* Opcodes in TB are PC-relative */ #define CF_CLUSTER_MASK 0xff000000 /* Top 8 bits are cluster ID */ #define CF_CLUSTER_SHIFT 24 -- 2.34.1 From: Anton Johansson via <qemu-devel@nongnu.org> Signed-off-by: Anton Johansson <anjo@rev.ng> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Message-Id: <20230227135202.9710-3-anjo@rev.ng> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/i386/cpu.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/target/i386/cpu.c b/target/i386/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/i386/cpu.c +++ b/target/i386/cpu.c @@ -XXX,XX +XXX,XX @@ static void x86_cpu_realizefn(DeviceState *dev, Error **errp) static bool ht_warned; unsigned requested_lbr_fmt; + /* Use pc-relative instructions in system-mode */ +#ifndef CONFIG_USER_ONLY + cs->tcg_cflags |= CF_PCREL; +#endif + if (cpu->apic_id == UNASSIGNED_APIC_ID) { error_setg(errp, "apic-id property was not initialized properly"); return; -- 2.34.1 From: Anton Johansson via <qemu-devel@nongnu.org> Signed-off-by: Anton Johansson <anjo@rev.ng> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Message-Id: <20230227135202.9710-4-anjo@rev.ng> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/arm/cpu.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/target/arm/cpu.c b/target/arm/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu.c +++ b/target/arm/cpu.c @@ -XXX,XX +XXX,XX @@ static void arm_cpu_realizefn(DeviceState *dev, Error **errp) Error *local_err = NULL; bool no_aa32 = false; + /* Use pc-relative instructions in system-mode */ +#ifndef CONFIG_USER_ONLY + cs->tcg_cflags |= CF_PCREL; +#endif + /* If we needed to query the host kernel for the CPU features * then it's possible that might have failed in the initfn, but * this is the first point where we can report it. -- 2.34.1 From: Anton Johansson via <qemu-devel@nongnu.org> Signed-off-by: Anton Johansson <anjo@rev.ng> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Message-Id: <20230227135202.9710-5-anjo@rev.ng> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- accel/tcg/internal.h | 10 ++++---- accel/tcg/tb-jmp-cache.h | 48 +++++++++++++++++++-------------------- accel/tcg/cpu-exec.c | 8 +++---- accel/tcg/perf.c | 2 +- accel/tcg/tb-maint.c | 8 +++---- accel/tcg/translate-all.c | 14 ++++++------ 6 files changed, 44 insertions(+), 46 deletions(-) diff --git a/accel/tcg/internal.h b/accel/tcg/internal.h index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/internal.h +++ b/accel/tcg/internal.h @@ -XXX,XX +XXX,XX @@ void cpu_restore_state_from_tb(CPUState *cpu, TranslationBlock *tb, /* Return the current PC from CPU, which may be cached in TB. */ static inline target_ulong log_pc(CPUState *cpu, const TranslationBlock *tb) { -#if TARGET_TB_PCREL - return cpu->cc->get_pc(cpu); -#else - return tb_pc(tb); -#endif + if (tb_cflags(tb) & CF_PCREL) { + return cpu->cc->get_pc(cpu); + } else { + return tb_pc(tb); + } } extern int64_t max_delay; diff --git a/accel/tcg/tb-jmp-cache.h b/accel/tcg/tb-jmp-cache.h index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/tb-jmp-cache.h +++ b/accel/tcg/tb-jmp-cache.h @@ -XXX,XX +XXX,XX @@ /* * Accessed in parallel; all accesses to 'tb' must be atomic. - * For TARGET_TB_PCREL, accesses to 'pc' must be protected by - * a load_acquire/store_release to 'tb'. + * For CF_PCREL, accesses to 'pc' must be protected by a + * load_acquire/store_release to 'tb'. */ struct CPUJumpCache { struct rcu_head rcu; struct { TranslationBlock *tb; -#if TARGET_TB_PCREL target_ulong pc; -#endif } array[TB_JMP_CACHE_SIZE]; }; static inline TranslationBlock * -tb_jmp_cache_get_tb(CPUJumpCache *jc, uint32_t hash) +tb_jmp_cache_get_tb(CPUJumpCache *jc, uint32_t cflags, uint32_t hash) { -#if TARGET_TB_PCREL - /* Use acquire to ensure current load of pc from jc. */ - return qatomic_load_acquire(&jc->array[hash].tb); -#else - /* Use rcu_read to ensure current load of pc from *tb. */ - return qatomic_rcu_read(&jc->array[hash].tb); -#endif + if (cflags & CF_PCREL) { + /* Use acquire to ensure current load of pc from jc. */ + return qatomic_load_acquire(&jc->array[hash].tb); + } else { + /* Use rcu_read to ensure current load of pc from *tb. */ + return qatomic_rcu_read(&jc->array[hash].tb); + } } static inline target_ulong tb_jmp_cache_get_pc(CPUJumpCache *jc, uint32_t hash, TranslationBlock *tb) { -#if TARGET_TB_PCREL - return jc->array[hash].pc; -#else - return tb_pc(tb); -#endif + if (tb_cflags(tb) & CF_PCREL) { + return jc->array[hash].pc; + } else { + return tb_pc(tb); + } } static inline void tb_jmp_cache_set(CPUJumpCache *jc, uint32_t hash, TranslationBlock *tb, target_ulong pc) { -#if TARGET_TB_PCREL - jc->array[hash].pc = pc; - /* Use store_release on tb to ensure pc is written first. */ - qatomic_store_release(&jc->array[hash].tb, tb); -#else - /* Use the pc value already stored in tb->pc. */ - qatomic_set(&jc->array[hash].tb, tb); -#endif + if (tb_cflags(tb) & CF_PCREL) { + jc->array[hash].pc = pc; + /* Use store_release on tb to ensure pc is written first. */ + qatomic_store_release(&jc->array[hash].tb, tb); + } else{ + /* Use the pc value already stored in tb->pc. */ + qatomic_set(&jc->array[hash].tb, tb); + } } #endif /* ACCEL_TCG_TB_JMP_CACHE_H */ diff --git a/accel/tcg/cpu-exec.c b/accel/tcg/cpu-exec.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/cpu-exec.c +++ b/accel/tcg/cpu-exec.c @@ -XXX,XX +XXX,XX @@ static bool tb_lookup_cmp(const void *p, const void *d) const TranslationBlock *tb = p; const struct tb_desc *desc = d; - if ((TARGET_TB_PCREL || tb_pc(tb) == desc->pc) && + if ((tb_cflags(tb) & CF_PCREL || tb_pc(tb) == desc->pc) && tb_page_addr0(tb) == desc->page_addr0 && tb->cs_base == desc->cs_base && tb->flags == desc->flags && @@ -XXX,XX +XXX,XX @@ static TranslationBlock *tb_htable_lookup(CPUState *cpu, target_ulong pc, return NULL; } desc.page_addr0 = phys_pc; - h = tb_hash_func(phys_pc, (TARGET_TB_PCREL ? 0 : pc), + h = tb_hash_func(phys_pc, (cflags & CF_PCREL ? 0 : pc), flags, cflags, *cpu->trace_dstate); return qht_lookup_custom(&tb_ctx.htable, &desc, h, tb_lookup_cmp); } @@ -XXX,XX +XXX,XX @@ static inline TranslationBlock *tb_lookup(CPUState *cpu, target_ulong pc, hash = tb_jmp_cache_hash_func(pc); jc = cpu->tb_jmp_cache; - tb = tb_jmp_cache_get_tb(jc, hash); + tb = tb_jmp_cache_get_tb(jc, cflags, hash); if (likely(tb && tb_jmp_cache_get_pc(jc, hash, tb) == pc && @@ -XXX,XX +XXX,XX @@ cpu_tb_exec(CPUState *cpu, TranslationBlock *itb, int *tb_exit) if (cc->tcg_ops->synchronize_from_tb) { cc->tcg_ops->synchronize_from_tb(cpu, last_tb); } else { - assert(!TARGET_TB_PCREL); + tcg_debug_assert(!(tb_cflags(last_tb) & CF_PCREL)); assert(cc->set_pc); cc->set_pc(cpu, tb_pc(last_tb)); } diff --git a/accel/tcg/perf.c b/accel/tcg/perf.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/perf.c +++ b/accel/tcg/perf.c @@ -XXX,XX +XXX,XX @@ void perf_report_code(uint64_t guest_pc, TranslationBlock *tb, for (insn = 0; insn < tb->icount; insn++) { /* FIXME: This replicates the restore_state_to_opc() logic. */ q[insn].address = tcg_ctx->gen_insn_data[insn][0]; - if (TARGET_TB_PCREL) { + if (tb_cflags(tb) & CF_PCREL) { q[insn].address |= (guest_pc & TARGET_PAGE_MASK); } else { #if defined(TARGET_I386) diff --git a/accel/tcg/tb-maint.c b/accel/tcg/tb-maint.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/tb-maint.c +++ b/accel/tcg/tb-maint.c @@ -XXX,XX +XXX,XX @@ static bool tb_cmp(const void *ap, const void *bp) const TranslationBlock *a = ap; const TranslationBlock *b = bp; - return ((TARGET_TB_PCREL || tb_pc(a) == tb_pc(b)) && + return ((tb_cflags(a) & CF_PCREL || tb_pc(a) == tb_pc(b)) && a->cs_base == b->cs_base && a->flags == b->flags && (tb_cflags(a) & ~CF_INVALID) == (tb_cflags(b) & ~CF_INVALID) && @@ -XXX,XX +XXX,XX @@ static void tb_jmp_cache_inval_tb(TranslationBlock *tb) { CPUState *cpu; - if (TARGET_TB_PCREL) { + if (tb_cflags(tb) & CF_PCREL) { /* A TB may be at any virtual address */ CPU_FOREACH(cpu) { tcg_flush_jmp_cache(cpu); @@ -XXX,XX +XXX,XX @@ static void do_tb_phys_invalidate(TranslationBlock *tb, bool rm_from_page_list) /* remove the TB from the hash list */ phys_pc = tb_page_addr0(tb); - h = tb_hash_func(phys_pc, (TARGET_TB_PCREL ? 0 : tb_pc(tb)), + h = tb_hash_func(phys_pc, (orig_cflags & CF_PCREL ? 0 : tb_pc(tb)), tb->flags, orig_cflags, tb->trace_vcpu_dstate); if (!qht_remove(&tb_ctx.htable, tb, h)) { return; @@ -XXX,XX +XXX,XX @@ TranslationBlock *tb_link_page(TranslationBlock *tb, tb_page_addr_t phys_pc, tb_record(tb, p, p2); /* add in the hash table */ - h = tb_hash_func(phys_pc, (TARGET_TB_PCREL ? 0 : tb_pc(tb)), + h = tb_hash_func(phys_pc, (tb->cflags & CF_PCREL ? 0 : tb_pc(tb)), tb->flags, tb->cflags, tb->trace_vcpu_dstate); qht_insert(&tb_ctx.htable, tb, h, &existing_tb); diff --git a/accel/tcg/translate-all.c b/accel/tcg/translate-all.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/translate-all.c +++ b/accel/tcg/translate-all.c @@ -XXX,XX +XXX,XX @@ static int encode_search(TranslationBlock *tb, uint8_t *block) for (j = 0; j < TARGET_INSN_START_WORDS; ++j) { if (i == 0) { - prev = (!TARGET_TB_PCREL && j == 0 ? tb_pc(tb) : 0); + prev = (!(tb_cflags(tb) & CF_PCREL) && j == 0 ? tb_pc(tb) : 0); } else { prev = tcg_ctx->gen_insn_data[i - 1][j]; } @@ -XXX,XX +XXX,XX @@ static int cpu_unwind_data_from_tb(TranslationBlock *tb, uintptr_t host_pc, } memset(data, 0, sizeof(uint64_t) * TARGET_INSN_START_WORDS); - if (!TARGET_TB_PCREL) { + if (!(tb_cflags(tb) & CF_PCREL)) { data[0] = tb_pc(tb); } @@ -XXX,XX +XXX,XX @@ TranslationBlock *tb_gen_code(CPUState *cpu, gen_code_buf = tcg_ctx->code_gen_ptr; tb->tc.ptr = tcg_splitwx_to_rx(gen_code_buf); -#if !TARGET_TB_PCREL - tb->pc = pc; -#endif + if (!(cflags & CF_PCREL)) { + tb->pc = pc; + } tb->cs_base = cs_base; tb->flags = flags; tb->cflags = cflags; @@ -XXX,XX +XXX,XX @@ TranslationBlock *tb_gen_code(CPUState *cpu, tb->tc.size = gen_code_size; /* - * For TARGET_TB_PCREL, attribute all executions of the generated - * code to its first mapping. + * For CF_PCREL, attribute all executions of the generated code + * to its first mapping. */ perf_report_code(pc, tb, tcg_splitwx_to_rx(gen_code_buf)); -- 2.34.1 From: Anton Johansson via <qemu-devel@nongnu.org> Signed-off-by: Anton Johansson <anjo@rev.ng> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Message-Id: <20230227135202.9710-6-anjo@rev.ng> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- include/exec/exec-all.h | 27 +++++++++++---------------- 1 file changed, 11 insertions(+), 16 deletions(-) diff --git a/include/exec/exec-all.h b/include/exec/exec-all.h index XXXXXXX..XXXXXXX 100644 --- a/include/exec/exec-all.h +++ b/include/exec/exec-all.h @@ -XXX,XX +XXX,XX @@ struct tb_tc { }; struct TranslationBlock { -#if !TARGET_TB_PCREL /* * Guest PC corresponding to this block. This must be the true * virtual address. Therefore e.g. x86 stores EIP + CS_BASE, and * targets like Arm, MIPS, HP-PA, which reuse low bits for ISA or * privilege, must store those bits elsewhere. * - * If TARGET_TB_PCREL, the opcodes for the TranslationBlock are - * written such that the TB is associated only with the physical - * page and may be run in any virtual address context. In this case, - * PC must always be taken from ENV in a target-specific manner. + * If CF_PCREL, the opcodes for the TranslationBlock are written + * such that the TB is associated only with the physical page and + * may be run in any virtual address context. In this case, PC + * must always be taken from ENV in a target-specific manner. * Unwind information is taken as offsets from the page, to be * deposited into the "current" PC. */ target_ulong pc; -#endif /* * Target-specific data associated with the TranslationBlock, e.g.: @@ -XXX,XX +XXX,XX @@ struct TranslationBlock { uintptr_t jmp_dest[2]; }; -/* Hide the read to avoid ifdefs for TARGET_TB_PCREL. */ -static inline target_ulong tb_pc(const TranslationBlock *tb) -{ -#if TARGET_TB_PCREL - qemu_build_not_reached(); -#else - return tb->pc; -#endif -} - /* Hide the qatomic_read to make code a little easier on the eyes */ static inline uint32_t tb_cflags(const TranslationBlock *tb) { return qatomic_read(&tb->cflags); } +/* Hide the read to avoid ifdefs for CF_PCREL. */ +static inline target_ulong tb_pc(const TranslationBlock *tb) +{ + assert(!(tb_cflags(tb) & CF_PCREL)); + return tb->pc; +} + static inline tb_page_addr_t tb_page_addr0(const TranslationBlock *tb) { #ifdef CONFIG_USER_ONLY -- 2.34.1 From: Anton Johansson via <qemu-devel@nongnu.org> Signed-off-by: Anton Johansson <anjo@rev.ng> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Message-Id: <20230227135202.9710-7-anjo@rev.ng> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/arm/tcg/translate.h | 2 +- target/arm/cpu.c | 8 ++++---- target/arm/tcg/translate-a64.c | 8 ++++---- target/arm/tcg/translate.c | 6 +++--- 4 files changed, 12 insertions(+), 12 deletions(-) diff --git a/target/arm/tcg/translate.h b/target/arm/tcg/translate.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/translate.h +++ b/target/arm/tcg/translate.h @@ -XXX,XX +XXX,XX @@ typedef struct DisasContext { /* The address of the current instruction being translated. */ target_ulong pc_curr; /* - * For TARGET_TB_PCREL, the full value of cpu_pc is not known + * For CF_PCREL, the full value of cpu_pc is not known * (although the page offset is known). For convenience, the * translation loop uses the full virtual address that triggered * the translation, from base.pc_start through pc_curr. diff --git a/target/arm/cpu.c b/target/arm/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu.c +++ b/target/arm/cpu.c @@ -XXX,XX +XXX,XX @@ static vaddr arm_cpu_get_pc(CPUState *cs) void arm_cpu_synchronize_from_tb(CPUState *cs, const TranslationBlock *tb) { - /* The program counter is always up to date with TARGET_TB_PCREL. */ - if (!TARGET_TB_PCREL) { + /* The program counter is always up to date with CF_PCREL. */ + if (!(tb_cflags(tb) & CF_PCREL)) { CPUARMState *env = cs->env_ptr; /* * It's OK to look at env for the current mode here, because it's @@ -XXX,XX +XXX,XX @@ void arm_restore_state_to_opc(CPUState *cs, CPUARMState *env = cs->env_ptr; if (is_a64(env)) { - if (TARGET_TB_PCREL) { + if (tb_cflags(tb) & CF_PCREL) { env->pc = (env->pc & TARGET_PAGE_MASK) | data[0]; } else { env->pc = data[0]; @@ -XXX,XX +XXX,XX @@ void arm_restore_state_to_opc(CPUState *cs, env->condexec_bits = 0; env->exception.syndrome = data[2] << ARM_INSN_START_WORD2_SHIFT; } else { - if (TARGET_TB_PCREL) { + if (tb_cflags(tb) & CF_PCREL) { env->regs[15] = (env->regs[15] & TARGET_PAGE_MASK) | data[0]; } else { env->regs[15] = data[0]; diff --git a/target/arm/tcg/translate-a64.c b/target/arm/tcg/translate-a64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/translate-a64.c +++ b/target/arm/tcg/translate-a64.c @@ -XXX,XX +XXX,XX @@ static void reset_btype(DisasContext *s) static void gen_pc_plus_diff(DisasContext *s, TCGv_i64 dest, target_long diff) { assert(s->pc_save != -1); - if (TARGET_TB_PCREL) { + if (tb_cflags(s->base.tb) & CF_PCREL) { tcg_gen_addi_i64(dest, cpu_pc, (s->pc_curr - s->pc_save) + diff); } else { tcg_gen_movi_i64(dest, s->pc_curr + diff); @@ -XXX,XX +XXX,XX @@ static void gen_goto_tb(DisasContext *s, int n, int64_t diff) * update to pc to the unlinked path. A long chain of links * can thus avoid many updates to the PC. */ - if (TARGET_TB_PCREL) { + if (tb_cflags(s->base.tb) & CF_PCREL) { gen_a64_update_pc(s, diff); tcg_gen_goto_tb(n); } else { @@ -XXX,XX +XXX,XX @@ static void disas_pc_rel_adr(DisasContext *s, uint32_t insn) if (page) { /* ADRP (page based) */ offset <<= 12; - /* The page offset is ok for TARGET_TB_PCREL. */ + /* The page offset is ok for CF_PCREL. */ offset -= s->pc_curr & 0xfff; } @@ -XXX,XX +XXX,XX @@ static void aarch64_tr_insn_start(DisasContextBase *dcbase, CPUState *cpu) DisasContext *dc = container_of(dcbase, DisasContext, base); target_ulong pc_arg = dc->base.pc_next; - if (TARGET_TB_PCREL) { + if (tb_cflags(dcbase->tb) & CF_PCREL) { pc_arg &= ~TARGET_PAGE_MASK; } tcg_gen_insn_start(pc_arg, 0, 0); diff --git a/target/arm/tcg/translate.c b/target/arm/tcg/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/translate.c +++ b/target/arm/tcg/translate.c @@ -XXX,XX +XXX,XX @@ static target_long jmp_diff(DisasContext *s, target_long diff) static void gen_pc_plus_diff(DisasContext *s, TCGv_i32 var, target_long diff) { assert(s->pc_save != -1); - if (TARGET_TB_PCREL) { + if (tb_cflags(s->base.tb) & CF_PCREL) { tcg_gen_addi_i32(var, cpu_R[15], (s->pc_curr - s->pc_save) + diff); } else { tcg_gen_movi_i32(var, s->pc_curr + diff); @@ -XXX,XX +XXX,XX @@ static void gen_goto_tb(DisasContext *s, int n, target_long diff) * update to pc to the unlinked path. A long chain of links * can thus avoid many updates to the PC. */ - if (TARGET_TB_PCREL) { + if (tb_cflags(s->base.tb) & CF_PCREL) { gen_update_pc(s, diff); tcg_gen_goto_tb(n); } else { @@ -XXX,XX +XXX,XX @@ static void arm_tr_insn_start(DisasContextBase *dcbase, CPUState *cpu) uint32_t condexec_bits; target_ulong pc_arg = dc->base.pc_next; - if (TARGET_TB_PCREL) { + if (tb_cflags(dcbase->tb) & CF_PCREL) { pc_arg &= ~TARGET_PAGE_MASK; } if (dc->eci) { -- 2.34.1 From: Anton Johansson via <qemu-devel@nongnu.org> Signed-off-by: Anton Johansson <anjo@rev.ng> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Message-Id: <20230227135202.9710-8-anjo@rev.ng> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/i386/helper.c | 2 +- target/i386/tcg/tcg-cpu.c | 6 +++--- target/i386/tcg/translate.c | 26 +++++++++++++------------- 3 files changed, 17 insertions(+), 17 deletions(-) diff --git a/target/i386/helper.c b/target/i386/helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/i386/helper.c +++ b/target/i386/helper.c @@ -XXX,XX +XXX,XX @@ static inline target_ulong get_memio_eip(CPUX86State *env) } /* Per x86_restore_state_to_opc. */ - if (TARGET_TB_PCREL) { + if (cs->tcg_cflags & CF_PCREL) { return (env->eip & TARGET_PAGE_MASK) | data[0]; } else { return data[0] - env->segs[R_CS].base; diff --git a/target/i386/tcg/tcg-cpu.c b/target/i386/tcg/tcg-cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/i386/tcg/tcg-cpu.c +++ b/target/i386/tcg/tcg-cpu.c @@ -XXX,XX +XXX,XX @@ static void x86_cpu_exec_exit(CPUState *cs) static void x86_cpu_synchronize_from_tb(CPUState *cs, const TranslationBlock *tb) { - /* The instruction pointer is always up to date with TARGET_TB_PCREL. */ - if (!TARGET_TB_PCREL) { + /* The instruction pointer is always up to date with CF_PCREL. */ + if (!(tb_cflags(tb) & CF_PCREL)) { CPUX86State *env = cs->env_ptr; env->eip = tb_pc(tb) - tb->cs_base; } @@ -XXX,XX +XXX,XX @@ static void x86_restore_state_to_opc(CPUState *cs, CPUX86State *env = &cpu->env; int cc_op = data[1]; - if (TARGET_TB_PCREL) { + if (tb_cflags(tb) & CF_PCREL) { env->eip = (env->eip & TARGET_PAGE_MASK) | data[0]; } else { env->eip = data[0] - tb->cs_base; diff --git a/target/i386/tcg/translate.c b/target/i386/tcg/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/i386/tcg/translate.c +++ b/target/i386/tcg/translate.c @@ -XXX,XX +XXX,XX @@ static inline void gen_op_st_rm_T0_A0(DisasContext *s, int idx, int d) static void gen_update_eip_cur(DisasContext *s) { assert(s->pc_save != -1); - if (TARGET_TB_PCREL) { + if (tb_cflags(s->base.tb) & CF_PCREL) { tcg_gen_addi_tl(cpu_eip, cpu_eip, s->base.pc_next - s->pc_save); } else { tcg_gen_movi_tl(cpu_eip, s->base.pc_next - s->cs_base); @@ -XXX,XX +XXX,XX @@ static void gen_update_eip_cur(DisasContext *s) static void gen_update_eip_next(DisasContext *s) { assert(s->pc_save != -1); - if (TARGET_TB_PCREL) { + if (tb_cflags(s->base.tb) & CF_PCREL) { tcg_gen_addi_tl(cpu_eip, cpu_eip, s->pc - s->pc_save); } else { tcg_gen_movi_tl(cpu_eip, s->pc - s->cs_base); @@ -XXX,XX +XXX,XX @@ static TCGv_i32 eip_next_i32(DisasContext *s) if (CODE64(s)) { return tcg_constant_i32(-1); } - if (TARGET_TB_PCREL) { + if (tb_cflags(s->base.tb) & CF_PCREL) { TCGv_i32 ret = tcg_temp_new_i32(); tcg_gen_trunc_tl_i32(ret, cpu_eip); tcg_gen_addi_i32(ret, ret, s->pc - s->pc_save); @@ -XXX,XX +XXX,XX @@ static TCGv_i32 eip_next_i32(DisasContext *s) static TCGv eip_next_tl(DisasContext *s) { assert(s->pc_save != -1); - if (TARGET_TB_PCREL) { + if (tb_cflags(s->base.tb) & CF_PCREL) { TCGv ret = tcg_temp_new(); tcg_gen_addi_tl(ret, cpu_eip, s->pc - s->pc_save); return ret; @@ -XXX,XX +XXX,XX @@ static TCGv eip_next_tl(DisasContext *s) static TCGv eip_cur_tl(DisasContext *s) { assert(s->pc_save != -1); - if (TARGET_TB_PCREL) { + if (tb_cflags(s->base.tb) & CF_PCREL) { TCGv ret = tcg_temp_new(); tcg_gen_addi_tl(ret, cpu_eip, s->base.pc_next - s->pc_save); return ret; @@ -XXX,XX +XXX,XX @@ static void gen_rot_rm_T1(DisasContext *s, MemOp ot, int op1, int is_right) tcg_temp_free_i32(t0); tcg_temp_free_i32(t1); - /* The CC_OP value is no longer predictable. */ + /* The CC_OP value is no longer predictable. */ set_cc_op(s, CC_OP_DYNAMIC); } @@ -XXX,XX +XXX,XX @@ static void gen_rotc_rm_T1(DisasContext *s, MemOp ot, int op1, gen_op_ld_v(s, ot, s->T0, s->A0); else gen_op_mov_v_reg(s, ot, s->T0, op1); - + if (is_right) { switch (ot) { case MO_8: @@ -XXX,XX +XXX,XX @@ static TCGv gen_lea_modrm_1(DisasContext *s, AddressParts a, bool is_vsib) ea = cpu_regs[a.base]; } if (!ea) { - if (TARGET_TB_PCREL && a.base == -2) { + if (tb_cflags(s->base.tb) & CF_PCREL && a.base == -2) { /* With cpu_eip ~= pc_save, the expression is pc-relative. */ tcg_gen_addi_tl(s->A0, cpu_eip, a.disp - s->pc_save); } else { @@ -XXX,XX +XXX,XX @@ static void gen_jmp_rel(DisasContext *s, MemOp ot, int diff, int tb_num) if (!CODE64(s)) { if (ot == MO_16) { mask = 0xffff; - if (TARGET_TB_PCREL && CODE32(s)) { + if (tb_cflags(s->base.tb) & CF_PCREL && CODE32(s)) { use_goto_tb = false; } } else { @@ -XXX,XX +XXX,XX @@ static void gen_jmp_rel(DisasContext *s, MemOp ot, int diff, int tb_num) gen_update_cc_op(s); set_cc_op(s, CC_OP_DYNAMIC); - if (TARGET_TB_PCREL) { + if (tb_cflags(s->base.tb) & CF_PCREL) { tcg_gen_addi_tl(cpu_eip, cpu_eip, new_pc - s->pc_save); /* * If we can prove the branch does not leave the page and we have @@ -XXX,XX +XXX,XX @@ static void gen_jmp_rel(DisasContext *s, MemOp ot, int diff, int tb_num) translator_use_goto_tb(&s->base, new_eip + s->cs_base)) { /* jump to same page: we can use a direct jump */ tcg_gen_goto_tb(tb_num); - if (!TARGET_TB_PCREL) { + if (!(tb_cflags(s->base.tb) & CF_PCREL)) { tcg_gen_movi_tl(cpu_eip, new_eip); } tcg_gen_exit_tb(s->base.tb, tb_num); s->base.is_jmp = DISAS_NORETURN; } else { - if (!TARGET_TB_PCREL) { + if (!(tb_cflags(s->base.tb) & CF_PCREL)) { tcg_gen_movi_tl(cpu_eip, new_eip); } if (s->jmp_opt) { @@ -XXX,XX +XXX,XX @@ static void i386_tr_insn_start(DisasContextBase *dcbase, CPUState *cpu) target_ulong pc_arg = dc->base.pc_next; dc->prev_insn_end = tcg_last_op(); - if (TARGET_TB_PCREL) { + if (tb_cflags(dcbase->tb) & CF_PCREL) { pc_arg -= dc->cs_base; pc_arg &= ~TARGET_PAGE_MASK; } -- 2.34.1 From: Anton Johansson via <qemu-devel@nongnu.org> Signed-off-by: Anton Johansson <anjo@rev.ng> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Message-Id: <20230227135202.9710-9-anjo@rev.ng> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- include/exec/cpu-defs.h | 3 --- 1 file changed, 3 deletions(-) diff --git a/include/exec/cpu-defs.h b/include/exec/cpu-defs.h index XXXXXXX..XXXXXXX 100644 --- a/include/exec/cpu-defs.h +++ b/include/exec/cpu-defs.h @@ -XXX,XX +XXX,XX @@ # error TARGET_PAGE_BITS must be defined in cpu-param.h # endif #endif -#ifndef TARGET_TB_PCREL -# define TARGET_TB_PCREL 0 -#endif #define TARGET_LONG_SIZE (TARGET_LONG_BITS / 8) -- 2.34.1 From: Anton Johansson via <qemu-devel@nongnu.org> Signed-off-by: Anton Johansson <anjo@rev.ng> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Message-Id: <20230227135202.9710-10-anjo@rev.ng> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/arm/cpu-param.h | 2 -- 1 file changed, 2 deletions(-) diff --git a/target/arm/cpu-param.h b/target/arm/cpu-param.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu-param.h +++ b/target/arm/cpu-param.h @@ -XXX,XX +XXX,XX @@ # define TARGET_PAGE_BITS_VARY # define TARGET_PAGE_BITS_MIN 10 -# define TARGET_TB_PCREL 1 - /* * Cache the attrs and shareability fields from the page table entry. * -- 2.34.1 From: Anton Johansson via <qemu-devel@nongnu.org> Signed-off-by: Anton Johansson <anjo@rev.ng> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Message-Id: <20230227135202.9710-11-anjo@rev.ng> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/i386/cpu-param.h | 4 ---- 1 file changed, 4 deletions(-) diff --git a/target/i386/cpu-param.h b/target/i386/cpu-param.h index XXXXXXX..XXXXXXX 100644 --- a/target/i386/cpu-param.h +++ b/target/i386/cpu-param.h @@ -XXX,XX +XXX,XX @@ #define TARGET_PAGE_BITS 12 #define NB_MMU_MODES 5 -#ifndef CONFIG_USER_ONLY -# define TARGET_TB_PCREL 1 -#endif - #endif -- 2.34.1 From: Anton Johansson via <qemu-devel@nongnu.org> tb-jmp-cache.h contains a few small functions that only exist to hide a CF_PCREL check, however the caller often already performs such a check. This patch moves CF_PCREL checks from the callee to the caller, and also removes these functions which now only hide an access of the jmp-cache. Signed-off-by: Anton Johansson <anjo@rev.ng> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Message-Id: <20230227135202.9710-12-anjo@rev.ng> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- accel/tcg/tb-jmp-cache.h | 36 --------------------------- accel/tcg/cpu-exec.c | 54 +++++++++++++++++++++++++++++----------- 2 files changed, 40 insertions(+), 50 deletions(-) diff --git a/accel/tcg/tb-jmp-cache.h b/accel/tcg/tb-jmp-cache.h index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/tb-jmp-cache.h +++ b/accel/tcg/tb-jmp-cache.h @@ -XXX,XX +XXX,XX @@ struct CPUJumpCache { } array[TB_JMP_CACHE_SIZE]; }; -static inline TranslationBlock * -tb_jmp_cache_get_tb(CPUJumpCache *jc, uint32_t cflags, uint32_t hash) -{ - if (cflags & CF_PCREL) { - /* Use acquire to ensure current load of pc from jc. */ - return qatomic_load_acquire(&jc->array[hash].tb); - } else { - /* Use rcu_read to ensure current load of pc from *tb. */ - return qatomic_rcu_read(&jc->array[hash].tb); - } -} - -static inline target_ulong -tb_jmp_cache_get_pc(CPUJumpCache *jc, uint32_t hash, TranslationBlock *tb) -{ - if (tb_cflags(tb) & CF_PCREL) { - return jc->array[hash].pc; - } else { - return tb_pc(tb); - } -} - -static inline void -tb_jmp_cache_set(CPUJumpCache *jc, uint32_t hash, - TranslationBlock *tb, target_ulong pc) -{ - if (tb_cflags(tb) & CF_PCREL) { - jc->array[hash].pc = pc; - /* Use store_release on tb to ensure pc is written first. */ - qatomic_store_release(&jc->array[hash].tb, tb); - } else{ - /* Use the pc value already stored in tb->pc. */ - qatomic_set(&jc->array[hash].tb, tb); - } -} - #endif /* ACCEL_TCG_TB_JMP_CACHE_H */ diff --git a/accel/tcg/cpu-exec.c b/accel/tcg/cpu-exec.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/cpu-exec.c +++ b/accel/tcg/cpu-exec.c @@ -XXX,XX +XXX,XX @@ static inline TranslationBlock *tb_lookup(CPUState *cpu, target_ulong pc, hash = tb_jmp_cache_hash_func(pc); jc = cpu->tb_jmp_cache; - tb = tb_jmp_cache_get_tb(jc, cflags, hash); - if (likely(tb && - tb_jmp_cache_get_pc(jc, hash, tb) == pc && - tb->cs_base == cs_base && - tb->flags == flags && - tb->trace_vcpu_dstate == *cpu->trace_dstate && - tb_cflags(tb) == cflags)) { - return tb; + if (cflags & CF_PCREL) { + /* Use acquire to ensure current load of pc from jc. */ + tb = qatomic_load_acquire(&jc->array[hash].tb); + + if (likely(tb && + jc->array[hash].pc == pc && + tb->cs_base == cs_base && + tb->flags == flags && + tb->trace_vcpu_dstate == *cpu->trace_dstate && + tb_cflags(tb) == cflags)) { + return tb; + } + tb = tb_htable_lookup(cpu, pc, cs_base, flags, cflags); + if (tb == NULL) { + return NULL; + } + jc->array[hash].pc = pc; + /* Use store_release on tb to ensure pc is written first. */ + qatomic_store_release(&jc->array[hash].tb, tb); + } else { + /* Use rcu_read to ensure current load of pc from *tb. */ + tb = qatomic_rcu_read(&jc->array[hash].tb); + + if (likely(tb && + tb_pc(tb) == pc && + tb->cs_base == cs_base && + tb->flags == flags && + tb->trace_vcpu_dstate == *cpu->trace_dstate && + tb_cflags(tb) == cflags)) { + return tb; + } + tb = tb_htable_lookup(cpu, pc, cs_base, flags, cflags); + if (tb == NULL) { + return NULL; + } + /* Use the pc value already stored in tb->pc. */ + qatomic_set(&jc->array[hash].tb, tb); } - tb = tb_htable_lookup(cpu, pc, cs_base, flags, cflags); - if (tb == NULL) { - return NULL; - } - tb_jmp_cache_set(jc, hash, tb, pc); + return tb; } @@ -XXX,XX +XXX,XX @@ cpu_exec_loop(CPUState *cpu, SyncClocks *sc) * for the fast lookup */ h = tb_jmp_cache_hash_func(pc); - tb_jmp_cache_set(cpu->tb_jmp_cache, h, tb, pc); + /* Use the pc value already stored in tb->pc. */ + qatomic_set(&cpu->tb_jmp_cache->array[h].tb, tb); } #ifndef CONFIG_USER_ONLY -- 2.34.1 From: Anton Johansson via <qemu-devel@nongnu.org> Signed-off-by: Anton Johansson <anjo@rev.ng> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Message-Id: <20230227135202.9710-13-anjo@rev.ng> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- accel/tcg/internal.h | 2 +- accel/tcg/cpu-exec.c | 6 +++--- accel/tcg/tb-maint.c | 8 ++++---- accel/tcg/translate-all.c | 4 ++-- 4 files changed, 10 insertions(+), 10 deletions(-) diff --git a/accel/tcg/internal.h b/accel/tcg/internal.h index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/internal.h +++ b/accel/tcg/internal.h @@ -XXX,XX +XXX,XX @@ static inline target_ulong log_pc(CPUState *cpu, const TranslationBlock *tb) if (tb_cflags(tb) & CF_PCREL) { return cpu->cc->get_pc(cpu); } else { - return tb_pc(tb); + return tb->pc; } } diff --git a/accel/tcg/cpu-exec.c b/accel/tcg/cpu-exec.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/cpu-exec.c +++ b/accel/tcg/cpu-exec.c @@ -XXX,XX +XXX,XX @@ static bool tb_lookup_cmp(const void *p, const void *d) const TranslationBlock *tb = p; const struct tb_desc *desc = d; - if ((tb_cflags(tb) & CF_PCREL || tb_pc(tb) == desc->pc) && + if ((tb_cflags(tb) & CF_PCREL || tb->pc == desc->pc) && tb_page_addr0(tb) == desc->page_addr0 && tb->cs_base == desc->cs_base && tb->flags == desc->flags && @@ -XXX,XX +XXX,XX @@ static inline TranslationBlock *tb_lookup(CPUState *cpu, target_ulong pc, tb = qatomic_rcu_read(&jc->array[hash].tb); if (likely(tb && - tb_pc(tb) == pc && + tb->pc == pc && tb->cs_base == cs_base && tb->flags == flags && tb->trace_vcpu_dstate == *cpu->trace_dstate && @@ -XXX,XX +XXX,XX @@ cpu_tb_exec(CPUState *cpu, TranslationBlock *itb, int *tb_exit) } else { tcg_debug_assert(!(tb_cflags(last_tb) & CF_PCREL)); assert(cc->set_pc); - cc->set_pc(cpu, tb_pc(last_tb)); + cc->set_pc(cpu, last_tb->pc); } if (qemu_loglevel_mask(CPU_LOG_EXEC)) { target_ulong pc = log_pc(cpu, last_tb); diff --git a/accel/tcg/tb-maint.c b/accel/tcg/tb-maint.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/tb-maint.c +++ b/accel/tcg/tb-maint.c @@ -XXX,XX +XXX,XX @@ static bool tb_cmp(const void *ap, const void *bp) const TranslationBlock *a = ap; const TranslationBlock *b = bp; - return ((tb_cflags(a) & CF_PCREL || tb_pc(a) == tb_pc(b)) && + return ((tb_cflags(a) & CF_PCREL || a->pc == b->pc) && a->cs_base == b->cs_base && a->flags == b->flags && (tb_cflags(a) & ~CF_INVALID) == (tb_cflags(b) & ~CF_INVALID) && @@ -XXX,XX +XXX,XX @@ static void tb_jmp_cache_inval_tb(TranslationBlock *tb) tcg_flush_jmp_cache(cpu); } } else { - uint32_t h = tb_jmp_cache_hash_func(tb_pc(tb)); + uint32_t h = tb_jmp_cache_hash_func(tb->pc); CPU_FOREACH(cpu) { CPUJumpCache *jc = cpu->tb_jmp_cache; @@ -XXX,XX +XXX,XX @@ static void do_tb_phys_invalidate(TranslationBlock *tb, bool rm_from_page_list) /* remove the TB from the hash list */ phys_pc = tb_page_addr0(tb); - h = tb_hash_func(phys_pc, (orig_cflags & CF_PCREL ? 0 : tb_pc(tb)), + h = tb_hash_func(phys_pc, (orig_cflags & CF_PCREL ? 0 : tb->pc), tb->flags, orig_cflags, tb->trace_vcpu_dstate); if (!qht_remove(&tb_ctx.htable, tb, h)) { return; @@ -XXX,XX +XXX,XX @@ TranslationBlock *tb_link_page(TranslationBlock *tb, tb_page_addr_t phys_pc, tb_record(tb, p, p2); /* add in the hash table */ - h = tb_hash_func(phys_pc, (tb->cflags & CF_PCREL ? 0 : tb_pc(tb)), + h = tb_hash_func(phys_pc, (tb->cflags & CF_PCREL ? 0 : tb->pc), tb->flags, tb->cflags, tb->trace_vcpu_dstate); qht_insert(&tb_ctx.htable, tb, h, &existing_tb); diff --git a/accel/tcg/translate-all.c b/accel/tcg/translate-all.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/translate-all.c +++ b/accel/tcg/translate-all.c @@ -XXX,XX +XXX,XX @@ static int encode_search(TranslationBlock *tb, uint8_t *block) for (j = 0; j < TARGET_INSN_START_WORDS; ++j) { if (i == 0) { - prev = (!(tb_cflags(tb) & CF_PCREL) && j == 0 ? tb_pc(tb) : 0); + prev = (!(tb_cflags(tb) & CF_PCREL) && j == 0 ? tb->pc : 0); } else { prev = tcg_ctx->gen_insn_data[i - 1][j]; } @@ -XXX,XX +XXX,XX @@ static int cpu_unwind_data_from_tb(TranslationBlock *tb, uintptr_t host_pc, memset(data, 0, sizeof(uint64_t) * TARGET_INSN_START_WORDS); if (!(tb_cflags(tb) & CF_PCREL)) { - data[0] = tb_pc(tb); + data[0] = tb->pc; } /* -- 2.34.1 From: Anton Johansson via <qemu-devel@nongnu.org> Signed-off-by: Anton Johansson <anjo@rev.ng> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Message-Id: <20230227135202.9710-14-anjo@rev.ng> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/tricore/cpu.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/target/tricore/cpu.c b/target/tricore/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/tricore/cpu.c +++ b/target/tricore/cpu.c @@ -XXX,XX +XXX,XX @@ static void tricore_cpu_synchronize_from_tb(CPUState *cs, TriCoreCPU *cpu = TRICORE_CPU(cs); CPUTriCoreState *env = &cpu->env; - env->PC = tb_pc(tb); + tcg_debug_assert(!(cs->tcg_cflags & CF_PCREL)); + env->PC = tb->pc; } static void tricore_restore_state_to_opc(CPUState *cs, -- 2.34.1 From: Anton Johansson via <qemu-devel@nongnu.org> Signed-off-by: Anton Johansson <anjo@rev.ng> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Message-Id: <20230227135202.9710-15-anjo@rev.ng> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/sparc/cpu.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/target/sparc/cpu.c b/target/sparc/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/sparc/cpu.c +++ b/target/sparc/cpu.c @@ -XXX,XX +XXX,XX @@ #include "exec/exec-all.h" #include "hw/qdev-properties.h" #include "qapi/visitor.h" +#include "tcg/tcg.h" //#define DEBUG_FEATURES @@ -XXX,XX +XXX,XX @@ static void sparc_cpu_synchronize_from_tb(CPUState *cs, { SPARCCPU *cpu = SPARC_CPU(cs); - cpu->env.pc = tb_pc(tb); + tcg_debug_assert(!(cs->tcg_cflags & CF_PCREL)); + cpu->env.pc = tb->pc; cpu->env.npc = tb->cs_base; } -- 2.34.1 From: Anton Johansson via <qemu-devel@nongnu.org> Signed-off-by: Anton Johansson <anjo@rev.ng> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Message-Id: <20230227135202.9710-16-anjo@rev.ng> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/sh4/cpu.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/target/sh4/cpu.c b/target/sh4/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/sh4/cpu.c +++ b/target/sh4/cpu.c @@ -XXX,XX +XXX,XX @@ #include "migration/vmstate.h" #include "exec/exec-all.h" #include "fpu/softfloat-helpers.h" +#include "tcg/tcg.h" static void superh_cpu_set_pc(CPUState *cs, vaddr value) { @@ -XXX,XX +XXX,XX @@ static void superh_cpu_synchronize_from_tb(CPUState *cs, { SuperHCPU *cpu = SUPERH_CPU(cs); - cpu->env.pc = tb_pc(tb); + tcg_debug_assert(!(cs->tcg_cflags & CF_PCREL)); + cpu->env.pc = tb->pc; cpu->env.flags = tb->flags & TB_FLAG_ENVFLAGS_MASK; } @@ -XXX,XX +XXX,XX @@ static bool superh_io_recompile_replay_branch(CPUState *cs, CPUSH4State *env = &cpu->env; if ((env->flags & (TB_FLAG_DELAY_SLOT | TB_FLAG_DELAY_SLOT_COND)) - && env->pc != tb_pc(tb)) { + && !(cs->tcg_cflags & CF_PCREL) && env->pc != tb->pc) { env->pc -= 2; env->flags &= ~(TB_FLAG_DELAY_SLOT | TB_FLAG_DELAY_SLOT_COND); return true; -- 2.34.1 From: Anton Johansson via <qemu-devel@nongnu.org> Signed-off-by: Anton Johansson <anjo@rev.ng> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Message-Id: <20230227135202.9710-17-anjo@rev.ng> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/rx/cpu.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/target/rx/cpu.c b/target/rx/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/rx/cpu.c +++ b/target/rx/cpu.c @@ -XXX,XX +XXX,XX @@ static void rx_cpu_synchronize_from_tb(CPUState *cs, { RXCPU *cpu = RX_CPU(cs); - cpu->env.pc = tb_pc(tb); + tcg_debug_assert(!(cs->tcg_cflags & CF_PCREL)); + cpu->env.pc = tb->pc; } static void rx_restore_state_to_opc(CPUState *cs, -- 2.34.1 From: Anton Johansson via <qemu-devel@nongnu.org> Signed-off-by: Anton Johansson <anjo@rev.ng> Reviewed-by: Palmer Dabbelt <palmer@rivosinc.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Message-Id: <20230227135202.9710-18-anjo@rev.ng> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/riscv/cpu.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/cpu.c +++ b/target/riscv/cpu.c @@ -XXX,XX +XXX,XX @@ #include "fpu/softfloat-helpers.h" #include "sysemu/kvm.h" #include "kvm_riscv.h" +#include "tcg/tcg.h" /* RISC-V CPU definitions */ @@ -XXX,XX +XXX,XX @@ static void riscv_cpu_synchronize_from_tb(CPUState *cs, CPURISCVState *env = &cpu->env; RISCVMXL xl = FIELD_EX32(tb->flags, TB_FLAGS, XL); + tcg_debug_assert(!(cs->tcg_cflags & CF_PCREL)); + if (xl == MXL_RV32) { - env->pc = (int32_t)tb_pc(tb); + env->pc = (int32_t) tb->pc; } else { - env->pc = tb_pc(tb); + env->pc = tb->pc; } } -- 2.34.1 From: Anton Johansson via <qemu-devel@nongnu.org> Signed-off-by: Anton Johansson <anjo@rev.ng> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Message-Id: <20230227135202.9710-19-anjo@rev.ng> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/openrisc/cpu.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/target/openrisc/cpu.c b/target/openrisc/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/openrisc/cpu.c +++ b/target/openrisc/cpu.c @@ -XXX,XX +XXX,XX @@ #include "qemu/qemu-print.h" #include "cpu.h" #include "exec/exec-all.h" +#include "tcg/tcg.h" static void openrisc_cpu_set_pc(CPUState *cs, vaddr value) { @@ -XXX,XX +XXX,XX @@ static void openrisc_cpu_synchronize_from_tb(CPUState *cs, { OpenRISCCPU *cpu = OPENRISC_CPU(cs); - cpu->env.pc = tb_pc(tb); + tcg_debug_assert(!(cs->tcg_cflags & CF_PCREL)); + cpu->env.pc = tb->pc; } static void openrisc_restore_state_to_opc(CPUState *cs, -- 2.34.1 From: Anton Johansson via <qemu-devel@nongnu.org> Signed-off-by: Anton Johansson <anjo@rev.ng> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Message-Id: <20230227135202.9710-20-anjo@rev.ng> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/mips/tcg/exception.c | 3 ++- target/mips/tcg/sysemu/special_helper.c | 2 +- 2 files changed, 3 insertions(+), 2 deletions(-) diff --git a/target/mips/tcg/exception.c b/target/mips/tcg/exception.c index XXXXXXX..XXXXXXX 100644 --- a/target/mips/tcg/exception.c +++ b/target/mips/tcg/exception.c @@ -XXX,XX +XXX,XX @@ void mips_cpu_synchronize_from_tb(CPUState *cs, const TranslationBlock *tb) MIPSCPU *cpu = MIPS_CPU(cs); CPUMIPSState *env = &cpu->env; - env->active_tc.PC = tb_pc(tb); + tcg_debug_assert(!(cs->tcg_cflags & CF_PCREL)); + env->active_tc.PC = tb->pc; env->hflags &= ~MIPS_HFLAG_BMASK; env->hflags |= tb->flags & MIPS_HFLAG_BMASK; } diff --git a/target/mips/tcg/sysemu/special_helper.c b/target/mips/tcg/sysemu/special_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/mips/tcg/sysemu/special_helper.c +++ b/target/mips/tcg/sysemu/special_helper.c @@ -XXX,XX +XXX,XX @@ bool mips_io_recompile_replay_branch(CPUState *cs, const TranslationBlock *tb) CPUMIPSState *env = &cpu->env; if ((env->hflags & MIPS_HFLAG_BMASK) != 0 - && env->active_tc.PC != tb_pc(tb)) { + && !(cs->tcg_cflags & CF_PCREL) && env->active_tc.PC != tb->pc) { env->active_tc.PC -= (env->hflags & MIPS_HFLAG_B16 ? 2 : 4); env->hflags &= ~MIPS_HFLAG_BMASK; return true; -- 2.34.1 From: Anton Johansson via <qemu-devel@nongnu.org> Signed-off-by: Anton Johansson <anjo@rev.ng> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Message-Id: <20230227135202.9710-21-anjo@rev.ng> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/microblaze/cpu.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/target/microblaze/cpu.c b/target/microblaze/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/microblaze/cpu.c +++ b/target/microblaze/cpu.c @@ -XXX,XX +XXX,XX @@ #include "exec/exec-all.h" #include "exec/gdbstub.h" #include "fpu/softfloat-helpers.h" +#include "tcg/tcg.h" static const struct { const char *name; @@ -XXX,XX +XXX,XX @@ static void mb_cpu_synchronize_from_tb(CPUState *cs, { MicroBlazeCPU *cpu = MICROBLAZE_CPU(cs); - cpu->env.pc = tb_pc(tb); + tcg_debug_assert(!(cs->tcg_cflags & CF_PCREL)); + cpu->env.pc = tb->pc; cpu->env.iflags = tb->flags & IFLAGS_TB_MASK; } -- 2.34.1 From: Anton Johansson via <qemu-devel@nongnu.org> Signed-off-by: Anton Johansson <anjo@rev.ng> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Message-Id: <20230227135202.9710-22-anjo@rev.ng> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/loongarch/cpu.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/target/loongarch/cpu.c b/target/loongarch/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/loongarch/cpu.c +++ b/target/loongarch/cpu.c @@ -XXX,XX +XXX,XX @@ #include "fpu/softfloat-helpers.h" #include "cpu-csr.h" #include "sysemu/reset.h" +#include "tcg/tcg.h" const char * const regnames[32] = { "r0", "r1", "r2", "r3", "r4", "r5", "r6", "r7", @@ -XXX,XX +XXX,XX @@ static void loongarch_cpu_synchronize_from_tb(CPUState *cs, LoongArchCPU *cpu = LOONGARCH_CPU(cs); CPULoongArchState *env = &cpu->env; - env->pc = tb_pc(tb); + tcg_debug_assert(!(cs->tcg_cflags & CF_PCREL)); + env->pc = tb->pc; } static void loongarch_restore_state_to_opc(CPUState *cs, @@ -XXX,XX +XXX,XX @@ static ObjectClass *loongarch_cpu_class_by_name(const char *cpu_model) oc = object_class_by_name(cpu_model); if (!oc) { - g_autofree char *typename + g_autofree char *typename = g_strdup_printf(LOONGARCH_CPU_TYPE_NAME("%s"), cpu_model); oc = object_class_by_name(typename); if (!oc) { -- 2.34.1 From: Anton Johansson via <qemu-devel@nongnu.org> Signed-off-by: Anton Johansson <anjo@rev.ng> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Message-Id: <20230227135202.9710-23-anjo@rev.ng> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/i386/tcg/tcg-cpu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/target/i386/tcg/tcg-cpu.c b/target/i386/tcg/tcg-cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/i386/tcg/tcg-cpu.c +++ b/target/i386/tcg/tcg-cpu.c @@ -XXX,XX +XXX,XX @@ static void x86_cpu_synchronize_from_tb(CPUState *cs, /* The instruction pointer is always up to date with CF_PCREL. */ if (!(tb_cflags(tb) & CF_PCREL)) { CPUX86State *env = cs->env_ptr; - env->eip = tb_pc(tb) - tb->cs_base; + env->eip = tb->pc - tb->cs_base; } } -- 2.34.1 From: Anton Johansson via <qemu-devel@nongnu.org> Signed-off-by: Anton Johansson <anjo@rev.ng> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Message-Id: <20230227135202.9710-24-anjo@rev.ng> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/hppa/cpu.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/target/hppa/cpu.c b/target/hppa/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/hppa/cpu.c +++ b/target/hppa/cpu.c @@ -XXX,XX +XXX,XX @@ #include "qemu/module.h" #include "exec/exec-all.h" #include "fpu/softfloat.h" - +#include "tcg/tcg.h" static void hppa_cpu_set_pc(CPUState *cs, vaddr value) { @@ -XXX,XX +XXX,XX @@ static void hppa_cpu_synchronize_from_tb(CPUState *cs, { HPPACPU *cpu = HPPA_CPU(cs); + tcg_debug_assert(!(cs->tcg_cflags & CF_PCREL)); + #ifdef CONFIG_USER_ONLY - cpu->env.iaoq_f = tb_pc(tb); + cpu->env.iaoq_f = tb->pc; cpu->env.iaoq_b = tb->cs_base; #else /* Recover the IAOQ values from the GVA + PRIV. */ @@ -XXX,XX +XXX,XX @@ static void hppa_cpu_synchronize_from_tb(CPUState *cs, int32_t diff = cs_base; cpu->env.iasq_f = iasq_f; - cpu->env.iaoq_f = (tb_pc(tb) & ~iasq_f) + priv; + cpu->env.iaoq_f = (tb->pc & ~iasq_f) + priv; if (diff) { cpu->env.iaoq_b = cpu->env.iaoq_f + diff; } -- 2.34.1 From: Anton Johansson via <qemu-devel@nongnu.org> Signed-off-by: Anton Johansson <anjo@rev.ng> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Taylor Simpson <tsimpson@quicinc.com> Message-Id: <20230227135202.9710-25-anjo@rev.ng> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/hexagon/cpu.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/target/hexagon/cpu.c b/target/hexagon/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/hexagon/cpu.c +++ b/target/hexagon/cpu.c @@ -XXX,XX +XXX,XX @@ #include "qapi/error.h" #include "hw/qdev-properties.h" #include "fpu/softfloat-helpers.h" +#include "tcg/tcg.h" static void hexagon_v67_cpu_init(Object *obj) { @@ -XXX,XX +XXX,XX @@ static void hexagon_cpu_synchronize_from_tb(CPUState *cs, { HexagonCPU *cpu = HEXAGON_CPU(cs); CPUHexagonState *env = &cpu->env; - env->gpr[HEX_REG_PC] = tb_pc(tb); + tcg_debug_assert(!(cs->tcg_cflags & CF_PCREL)); + env->gpr[HEX_REG_PC] = tb->pc; } static bool hexagon_cpu_has_work(CPUState *cs) -- 2.34.1 From: Anton Johansson via <qemu-devel@nongnu.org> Signed-off-by: Anton Johansson <anjo@rev.ng> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Message-Id: <20230227135202.9710-26-anjo@rev.ng> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/avr/cpu.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/target/avr/cpu.c b/target/avr/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/avr/cpu.c +++ b/target/avr/cpu.c @@ -XXX,XX +XXX,XX @@ static void avr_cpu_synchronize_from_tb(CPUState *cs, AVRCPU *cpu = AVR_CPU(cs); CPUAVRState *env = &cpu->env; - env->pc_w = tb_pc(tb) / 2; /* internally PC points to words */ + tcg_debug_assert(!(cs->tcg_cflags & CF_PCREL)); + env->pc_w = tb->pc / 2; /* internally PC points to words */ } static void avr_restore_state_to_opc(CPUState *cs, -- 2.34.1 From: Anton Johansson via <qemu-devel@nongnu.org> Signed-off-by: Anton Johansson <anjo@rev.ng> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Message-Id: <20230227135202.9710-27-anjo@rev.ng> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/arm/cpu.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/target/arm/cpu.c b/target/arm/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu.c +++ b/target/arm/cpu.c @@ -XXX,XX +XXX,XX @@ void arm_cpu_synchronize_from_tb(CPUState *cs, * never possible for an AArch64 TB to chain to an AArch32 TB. */ if (is_a64(env)) { - env->pc = tb_pc(tb); + env->pc = tb->pc; } else { - env->regs[15] = tb_pc(tb); + env->regs[15] = tb->pc; } } } -- 2.34.1 From: Anton Johansson via <qemu-devel@nongnu.org> Signed-off-by: Anton Johansson <anjo@rev.ng> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Message-Id: <20230227135202.9710-28-anjo@rev.ng> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- include/exec/exec-all.h | 7 ------- 1 file changed, 7 deletions(-) diff --git a/include/exec/exec-all.h b/include/exec/exec-all.h index XXXXXXX..XXXXXXX 100644 --- a/include/exec/exec-all.h +++ b/include/exec/exec-all.h @@ -XXX,XX +XXX,XX @@ static inline uint32_t tb_cflags(const TranslationBlock *tb) return qatomic_read(&tb->cflags); } -/* Hide the read to avoid ifdefs for CF_PCREL. */ -static inline target_ulong tb_pc(const TranslationBlock *tb) -{ - assert(!(tb_cflags(tb) & CF_PCREL)); - return tb->pc; -} - static inline tb_page_addr_t tb_page_addr0(const TranslationBlock *tb) { #ifdef CONFIG_USER_ONLY -- 2.34.1 Change the temps_in_use check to use assert not fprintf. Move the assert for double-free before the check for count, since that is the more immediate problem. Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/tcg.c | 12 +++++------- 1 file changed, 5 insertions(+), 7 deletions(-) diff --git a/tcg/tcg.c b/tcg/tcg.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/tcg.c +++ b/tcg/tcg.c @@ -XXX,XX +XXX,XX @@ void tcg_temp_free_internal(TCGTemp *ts) g_assert_not_reached(); } -#if defined(CONFIG_DEBUG_TCG) - s->temps_in_use--; - if (s->temps_in_use < 0) { - fprintf(stderr, "More temporaries freed than allocated!\n"); - } -#endif - tcg_debug_assert(ts->temp_allocated != 0); ts->temp_allocated = 0; +#if defined(CONFIG_DEBUG_TCG) + assert(s->temps_in_use > 0); + s->temps_in_use--; +#endif + idx = temp_idx(ts); k = ts->base_type + (ts->kind == TEMP_NORMAL ? 0 : TCG_TYPE_COUNT); set_bit(idx, s->free_temps[k].l); -- 2.34.1 In preparation for returning the number of insns generated via the same pointer. Adjust only the prototypes so far. Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- include/exec/translator.h | 4 ++-- accel/tcg/translate-all.c | 2 +- accel/tcg/translator.c | 4 ++-- target/alpha/translate.c | 2 +- target/arm/tcg/translate.c | 2 +- target/avr/translate.c | 2 +- target/cris/translate.c | 2 +- target/hexagon/translate.c | 2 +- target/hppa/translate.c | 2 +- target/i386/tcg/translate.c | 2 +- target/loongarch/translate.c | 2 +- target/m68k/translate.c | 2 +- target/microblaze/translate.c | 2 +- target/mips/tcg/translate.c | 2 +- target/nios2/translate.c | 2 +- target/openrisc/translate.c | 2 +- target/ppc/translate.c | 2 +- target/riscv/translate.c | 2 +- target/rx/translate.c | 2 +- target/s390x/tcg/translate.c | 2 +- target/sh4/translate.c | 2 +- target/sparc/translate.c | 2 +- target/tricore/translate.c | 2 +- target/xtensa/translate.c | 2 +- 24 files changed, 26 insertions(+), 26 deletions(-) diff --git a/include/exec/translator.h b/include/exec/translator.h index XXXXXXX..XXXXXXX 100644 --- a/include/exec/translator.h +++ b/include/exec/translator.h @@ -XXX,XX +XXX,XX @@ * This function must be provided by the target, which should create * the target-specific DisasContext, and then invoke translator_loop. */ -void gen_intermediate_code(CPUState *cpu, TranslationBlock *tb, int max_insns, +void gen_intermediate_code(CPUState *cpu, TranslationBlock *tb, int *max_insns, target_ulong pc, void *host_pc); /** @@ -XXX,XX +XXX,XX @@ typedef struct TranslatorOps { * - When single-stepping is enabled (system-wide or on the current vCPU). * - When too many instructions have been translated. */ -void translator_loop(CPUState *cpu, TranslationBlock *tb, int max_insns, +void translator_loop(CPUState *cpu, TranslationBlock *tb, int *max_insns, target_ulong pc, void *host_pc, const TranslatorOps *ops, DisasContextBase *db); diff --git a/accel/tcg/translate-all.c b/accel/tcg/translate-all.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/translate-all.c +++ b/accel/tcg/translate-all.c @@ -XXX,XX +XXX,XX @@ static int setjmp_gen_code(CPUArchState *env, TranslationBlock *tb, tcg_func_start(tcg_ctx); tcg_ctx->cpu = env_cpu(env); - gen_intermediate_code(env_cpu(env), tb, *max_insns, pc, host_pc); + gen_intermediate_code(env_cpu(env), tb, max_insns, pc, host_pc); assert(tb->size != 0); tcg_ctx->cpu = NULL; *max_insns = tb->icount; diff --git a/accel/tcg/translator.c b/accel/tcg/translator.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/translator.c +++ b/accel/tcg/translator.c @@ -XXX,XX +XXX,XX @@ bool translator_use_goto_tb(DisasContextBase *db, target_ulong dest) return ((db->pc_first ^ dest) & TARGET_PAGE_MASK) == 0; } -void translator_loop(CPUState *cpu, TranslationBlock *tb, int max_insns, +void translator_loop(CPUState *cpu, TranslationBlock *tb, int *max_insns, target_ulong pc, void *host_pc, const TranslatorOps *ops, DisasContextBase *db) { @@ -XXX,XX +XXX,XX @@ void translator_loop(CPUState *cpu, TranslationBlock *tb, int max_insns, db->pc_next = pc; db->is_jmp = DISAS_NEXT; db->num_insns = 0; - db->max_insns = max_insns; + db->max_insns = *max_insns; db->singlestep_enabled = cflags & CF_SINGLE_STEP; db->host_addr[0] = host_pc; db->host_addr[1] = NULL; diff --git a/target/alpha/translate.c b/target/alpha/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/alpha/translate.c +++ b/target/alpha/translate.c @@ -XXX,XX +XXX,XX @@ static const TranslatorOps alpha_tr_ops = { .disas_log = alpha_tr_disas_log, }; -void gen_intermediate_code(CPUState *cpu, TranslationBlock *tb, int max_insns, +void gen_intermediate_code(CPUState *cpu, TranslationBlock *tb, int *max_insns, target_ulong pc, void *host_pc) { DisasContext dc; diff --git a/target/arm/tcg/translate.c b/target/arm/tcg/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/translate.c +++ b/target/arm/tcg/translate.c @@ -XXX,XX +XXX,XX @@ static const TranslatorOps thumb_translator_ops = { }; /* generate intermediate code for basic block 'tb'. */ -void gen_intermediate_code(CPUState *cpu, TranslationBlock *tb, int max_insns, +void gen_intermediate_code(CPUState *cpu, TranslationBlock *tb, int *max_insns, target_ulong pc, void *host_pc) { DisasContext dc = { }; diff --git a/target/avr/translate.c b/target/avr/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/avr/translate.c +++ b/target/avr/translate.c @@ -XXX,XX +XXX,XX @@ static const TranslatorOps avr_tr_ops = { .disas_log = avr_tr_disas_log, }; -void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int max_insns, +void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int *max_insns, target_ulong pc, void *host_pc) { DisasContext dc = { }; diff --git a/target/cris/translate.c b/target/cris/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/cris/translate.c +++ b/target/cris/translate.c @@ -XXX,XX +XXX,XX @@ static const TranslatorOps cris_tr_ops = { .disas_log = cris_tr_disas_log, }; -void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int max_insns, +void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int *max_insns, target_ulong pc, void *host_pc) { DisasContext dc; diff --git a/target/hexagon/translate.c b/target/hexagon/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/hexagon/translate.c +++ b/target/hexagon/translate.c @@ -XXX,XX +XXX,XX @@ static const TranslatorOps hexagon_tr_ops = { .disas_log = hexagon_tr_disas_log, }; -void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int max_insns, +void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int *max_insns, target_ulong pc, void *host_pc) { DisasContext ctx; diff --git a/target/hppa/translate.c b/target/hppa/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/hppa/translate.c +++ b/target/hppa/translate.c @@ -XXX,XX +XXX,XX @@ static const TranslatorOps hppa_tr_ops = { .disas_log = hppa_tr_disas_log, }; -void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int max_insns, +void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int *max_insns, target_ulong pc, void *host_pc) { DisasContext ctx; diff --git a/target/i386/tcg/translate.c b/target/i386/tcg/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/i386/tcg/translate.c +++ b/target/i386/tcg/translate.c @@ -XXX,XX +XXX,XX @@ static const TranslatorOps i386_tr_ops = { }; /* generate intermediate code for basic block 'tb'. */ -void gen_intermediate_code(CPUState *cpu, TranslationBlock *tb, int max_insns, +void gen_intermediate_code(CPUState *cpu, TranslationBlock *tb, int *max_insns, target_ulong pc, void *host_pc) { DisasContext dc; diff --git a/target/loongarch/translate.c b/target/loongarch/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/loongarch/translate.c +++ b/target/loongarch/translate.c @@ -XXX,XX +XXX,XX @@ static const TranslatorOps loongarch_tr_ops = { .disas_log = loongarch_tr_disas_log, }; -void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int max_insns, +void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int *max_insns, target_ulong pc, void *host_pc) { DisasContext ctx; diff --git a/target/m68k/translate.c b/target/m68k/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/m68k/translate.c +++ b/target/m68k/translate.c @@ -XXX,XX +XXX,XX @@ static const TranslatorOps m68k_tr_ops = { .disas_log = m68k_tr_disas_log, }; -void gen_intermediate_code(CPUState *cpu, TranslationBlock *tb, int max_insns, +void gen_intermediate_code(CPUState *cpu, TranslationBlock *tb, int *max_insns, target_ulong pc, void *host_pc) { DisasContext dc; diff --git a/target/microblaze/translate.c b/target/microblaze/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/microblaze/translate.c +++ b/target/microblaze/translate.c @@ -XXX,XX +XXX,XX @@ static const TranslatorOps mb_tr_ops = { .disas_log = mb_tr_disas_log, }; -void gen_intermediate_code(CPUState *cpu, TranslationBlock *tb, int max_insns, +void gen_intermediate_code(CPUState *cpu, TranslationBlock *tb, int *max_insns, target_ulong pc, void *host_pc) { DisasContext dc; diff --git a/target/mips/tcg/translate.c b/target/mips/tcg/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/mips/tcg/translate.c +++ b/target/mips/tcg/translate.c @@ -XXX,XX +XXX,XX @@ static const TranslatorOps mips_tr_ops = { .disas_log = mips_tr_disas_log, }; -void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int max_insns, +void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int *max_insns, target_ulong pc, void *host_pc) { DisasContext ctx; diff --git a/target/nios2/translate.c b/target/nios2/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/nios2/translate.c +++ b/target/nios2/translate.c @@ -XXX,XX +XXX,XX @@ static const TranslatorOps nios2_tr_ops = { .disas_log = nios2_tr_disas_log, }; -void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int max_insns, +void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int *max_insns, target_ulong pc, void *host_pc) { DisasContext dc; diff --git a/target/openrisc/translate.c b/target/openrisc/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/openrisc/translate.c +++ b/target/openrisc/translate.c @@ -XXX,XX +XXX,XX @@ static const TranslatorOps openrisc_tr_ops = { .disas_log = openrisc_tr_disas_log, }; -void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int max_insns, +void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int *max_insns, target_ulong pc, void *host_pc) { DisasContext ctx; diff --git a/target/ppc/translate.c b/target/ppc/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/ppc/translate.c +++ b/target/ppc/translate.c @@ -XXX,XX +XXX,XX @@ static const TranslatorOps ppc_tr_ops = { .disas_log = ppc_tr_disas_log, }; -void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int max_insns, +void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int *max_insns, target_ulong pc, void *host_pc) { DisasContext ctx; diff --git a/target/riscv/translate.c b/target/riscv/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/translate.c +++ b/target/riscv/translate.c @@ -XXX,XX +XXX,XX @@ static const TranslatorOps riscv_tr_ops = { .disas_log = riscv_tr_disas_log, }; -void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int max_insns, +void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int *max_insns, target_ulong pc, void *host_pc) { DisasContext ctx; diff --git a/target/rx/translate.c b/target/rx/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/rx/translate.c +++ b/target/rx/translate.c @@ -XXX,XX +XXX,XX @@ static const TranslatorOps rx_tr_ops = { .disas_log = rx_tr_disas_log, }; -void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int max_insns, +void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int *max_insns, target_ulong pc, void *host_pc) { DisasContext dc; diff --git a/target/s390x/tcg/translate.c b/target/s390x/tcg/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/s390x/tcg/translate.c +++ b/target/s390x/tcg/translate.c @@ -XXX,XX +XXX,XX @@ static const TranslatorOps s390x_tr_ops = { .disas_log = s390x_tr_disas_log, }; -void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int max_insns, +void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int *max_insns, target_ulong pc, void *host_pc) { DisasContext dc; diff --git a/target/sh4/translate.c b/target/sh4/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/sh4/translate.c +++ b/target/sh4/translate.c @@ -XXX,XX +XXX,XX @@ static const TranslatorOps sh4_tr_ops = { .disas_log = sh4_tr_disas_log, }; -void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int max_insns, +void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int *max_insns, target_ulong pc, void *host_pc) { DisasContext ctx; diff --git a/target/sparc/translate.c b/target/sparc/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/sparc/translate.c +++ b/target/sparc/translate.c @@ -XXX,XX +XXX,XX @@ static const TranslatorOps sparc_tr_ops = { .disas_log = sparc_tr_disas_log, }; -void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int max_insns, +void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int *max_insns, target_ulong pc, void *host_pc) { DisasContext dc = {}; diff --git a/target/tricore/translate.c b/target/tricore/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/tricore/translate.c +++ b/target/tricore/translate.c @@ -XXX,XX +XXX,XX @@ static const TranslatorOps tricore_tr_ops = { }; -void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int max_insns, +void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int *max_insns, target_ulong pc, void *host_pc) { DisasContext ctx; diff --git a/target/xtensa/translate.c b/target/xtensa/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/xtensa/translate.c +++ b/target/xtensa/translate.c @@ -XXX,XX +XXX,XX @@ static const TranslatorOps xtensa_translator_ops = { .disas_log = xtensa_tr_disas_log, }; -void gen_intermediate_code(CPUState *cpu, TranslationBlock *tb, int max_insns, +void gen_intermediate_code(CPUState *cpu, TranslationBlock *tb, int *max_insns, target_ulong pc, void *host_pc) { DisasContext dc = {}; -- 2.34.1 Write back the number of insns that we attempt to translate, so that if we longjmp out we have a more accurate limit for the next attempt. This results in fewer restarts when some limit is consumed by few instructions. Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- accel/tcg/translator.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/accel/tcg/translator.c b/accel/tcg/translator.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/translator.c +++ b/accel/tcg/translator.c @@ -XXX,XX +XXX,XX @@ void translator_loop(CPUState *cpu, TranslationBlock *tb, int *max_insns, plugin_enabled = plugin_gen_tb_start(cpu, db, cflags & CF_MEMI_ONLY); while (true) { - db->num_insns++; + *max_insns = ++db->num_insns; ops->insn_start(db, cpu); tcg_debug_assert(db->is_jmp == DISAS_NEXT); /* no early exit */ -- 2.34.1 Just because the label reference count is more than 1 does not mean we cannot remove a branch-to-next. By doing this first, the label reference count may drop to 0, and then the label itself gets removed as before. Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/tcg.c | 33 +++++++++++++++++---------------- 1 file changed, 17 insertions(+), 16 deletions(-) diff --git a/tcg/tcg.c b/tcg/tcg.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/tcg.c +++ b/tcg/tcg.c @@ -XXX,XX +XXX,XX @@ TCGOp *tcg_op_insert_after(TCGContext *s, TCGOp *old_op, /* Reachable analysis : remove unreachable code. */ static void reachable_code_pass(TCGContext *s) { - TCGOp *op, *op_next; + TCGOp *op, *op_next, *op_prev; bool dead = false; QTAILQ_FOREACH_SAFE(op, &s->ops, link, op_next) { @@ -XXX,XX +XXX,XX @@ static void reachable_code_pass(TCGContext *s) switch (op->opc) { case INDEX_op_set_label: label = arg_label(op->args[0]); + + /* + * Optimization can fold conditional branches to unconditional. + * If we find a label which is preceded by an unconditional + * branch to next, remove the branch. We couldn't do this when + * processing the branch because any dead code between the branch + * and label had not yet been removed. + */ + op_prev = QTAILQ_PREV(op, link); + if (op_prev->opc == INDEX_op_br && + label == arg_label(op_prev->args[0])) { + tcg_op_remove(s, op_prev); + /* Fall through means insns become live again. */ + dead = false; + } + if (label->refs == 0) { /* * While there is an occasional backward branch, virtually @@ -XXX,XX +XXX,XX @@ static void reachable_code_pass(TCGContext *s) /* Once we see a label, insns become live again. */ dead = false; remove = false; - - /* - * Optimization can fold conditional branches to unconditional. - * If we find a label with one reference which is preceded by - * an unconditional branch to it, remove both. This needed to - * wait until the dead code in between them was removed. - */ - if (label->refs == 1) { - TCGOp *op_prev = QTAILQ_PREV(op, link); - if (op_prev->opc == INDEX_op_br && - label == arg_label(op_prev->args[0])) { - tcg_op_remove(s, op_prev); - remove = true; - } - } } break; -- 2.34.1 Use TEMP_TB as that is more explicit about the default lifetime of the data. While "global" and "local" used to be contrasting, we have more lifetimes than that now. Do not yet rename tcg_temp_local_new_*, just the enum. Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- include/tcg/tcg.h | 12 ++++++++---- tcg/optimize.c | 2 +- tcg/tcg.c | 18 +++++++++--------- 3 files changed, 18 insertions(+), 14 deletions(-) diff --git a/include/tcg/tcg.h b/include/tcg/tcg.h index XXXXXXX..XXXXXXX 100644 --- a/include/tcg/tcg.h +++ b/include/tcg/tcg.h @@ -XXX,XX +XXX,XX @@ typedef enum TCGTempVal { typedef enum TCGTempKind { /* Temp is dead at the end of all basic blocks. */ TEMP_NORMAL, - /* Temp is live across conditional branch, but dead otherwise. */ + /* + * Temp is dead at the end of the extended basic block (EBB), + * the single-entry multiple-exit region that falls through + * conditional branches. + */ TEMP_EBB, - /* Temp is saved across basic blocks but dead at the end of TBs. */ - TEMP_LOCAL, - /* Temp is saved across both basic blocks and translation blocks. */ + /* Temp is live across the entire translation block, but dead at end. */ + TEMP_TB, + /* Temp is live across the entire translation block, and between them. */ TEMP_GLOBAL, /* Temp is in a fixed register. */ TEMP_FIXED, diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static TCGTemp *find_better_copy(TCGContext *s, TCGTemp *ts) } else if (i->kind > ts->kind) { if (i->kind == TEMP_GLOBAL) { g = i; - } else if (i->kind == TEMP_LOCAL) { + } else if (i->kind == TEMP_TB) { l = i; } } diff --git a/tcg/tcg.c b/tcg/tcg.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/tcg.c +++ b/tcg/tcg.c @@ -XXX,XX +XXX,XX @@ TCGTemp *tcg_global_mem_new_internal(TCGType type, TCGv_ptr base, TCGTemp *tcg_temp_new_internal(TCGType type, bool temp_local) { TCGContext *s = tcg_ctx; - TCGTempKind kind = temp_local ? TEMP_LOCAL : TEMP_NORMAL; + TCGTempKind kind = temp_local ? TEMP_TB : TEMP_NORMAL; TCGTemp *ts; int idx, k; @@ -XXX,XX +XXX,XX @@ void tcg_temp_free_internal(TCGTemp *ts) */ return; case TEMP_NORMAL: - case TEMP_LOCAL: + case TEMP_TB: break; default: g_assert_not_reached(); @@ -XXX,XX +XXX,XX @@ static void tcg_reg_alloc_start(TCGContext *s) case TEMP_EBB: val = TEMP_VAL_DEAD; /* fall through */ - case TEMP_LOCAL: + case TEMP_TB: ts->mem_allocated = 0; break; default: @@ -XXX,XX +XXX,XX @@ static char *tcg_get_arg_str_ptr(TCGContext *s, char *buf, int buf_size, case TEMP_GLOBAL: pstrcpy(buf, buf_size, ts->name); break; - case TEMP_LOCAL: + case TEMP_TB: snprintf(buf, buf_size, "loc%d", idx - s->nb_globals); break; case TEMP_EBB: @@ -XXX,XX +XXX,XX @@ static void la_bb_end(TCGContext *s, int ng, int nt) switch (ts->kind) { case TEMP_FIXED: case TEMP_GLOBAL: - case TEMP_LOCAL: + case TEMP_TB: state = TS_DEAD | TS_MEM; break; case TEMP_NORMAL: @@ -XXX,XX +XXX,XX @@ static void la_bb_sync(TCGContext *s, int ng, int nt) int state; switch (ts->kind) { - case TEMP_LOCAL: + case TEMP_TB: state = ts->state; ts->state = state | TS_MEM; if (state != TS_DEAD) { @@ -XXX,XX +XXX,XX @@ static void temp_free_or_dead(TCGContext *s, TCGTemp *ts, int free_or_dead) case TEMP_FIXED: return; case TEMP_GLOBAL: - case TEMP_LOCAL: + case TEMP_TB: new_type = TEMP_VAL_MEM; break; case TEMP_NORMAL: @@ -XXX,XX +XXX,XX @@ static void tcg_reg_alloc_bb_end(TCGContext *s, TCGRegSet allocated_regs) TCGTemp *ts = &s->temps[i]; switch (ts->kind) { - case TEMP_LOCAL: + case TEMP_TB: temp_save(s, ts, allocated_regs); break; case TEMP_NORMAL: @@ -XXX,XX +XXX,XX @@ static void tcg_reg_alloc_cbranch(TCGContext *s, TCGRegSet allocated_regs) * Keep tcg_debug_asserts for safety. */ switch (ts->kind) { - case TEMP_LOCAL: + case TEMP_TB: tcg_debug_assert(ts->val_type != TEMP_VAL_REG || ts->mem_coherent); break; case TEMP_NORMAL: -- 2.34.1 This makes it easier to assign blame with perf. Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/tcg.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/tcg/tcg.c b/tcg/tcg.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/tcg.c +++ b/tcg/tcg.c @@ -XXX,XX +XXX,XX @@ TCGOp *tcg_op_insert_after(TCGContext *s, TCGOp *old_op, } /* Reachable analysis : remove unreachable code. */ -static void reachable_code_pass(TCGContext *s) +static void __attribute__((noinline)) +reachable_code_pass(TCGContext *s) { TCGOp *op, *op_next, *op_prev; bool dead = false; @@ -XXX,XX +XXX,XX @@ static void la_cross_call(TCGContext *s, int nt) /* Liveness analysis : update the opc_arg_life array to tell if a given input arguments is dead. Instructions updating dead temporaries are removed. */ -static void liveness_pass_1(TCGContext *s) +static void __attribute__((noinline)) +liveness_pass_1(TCGContext *s) { int nb_globals = s->nb_globals; int nb_temps = s->nb_temps; @@ -XXX,XX +XXX,XX @@ static void liveness_pass_1(TCGContext *s) } /* Liveness analysis: Convert indirect regs to direct temporaries. */ -static bool liveness_pass_2(TCGContext *s) +static bool __attribute__((noinline)) +liveness_pass_2(TCGContext *s) { int nb_globals = s->nb_globals; int nb_temps, i; -- 2.34.1 Attempt to reduce the lifetime of TEMP_TB. Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/tcg.c | 70 +++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 70 insertions(+) diff --git a/tcg/tcg.c b/tcg/tcg.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/tcg.c +++ b/tcg/tcg.c @@ -XXX,XX +XXX,XX @@ static void la_cross_call(TCGContext *s, int nt) } } +/* + * Liveness analysis: Verify the lifetime of TEMP_TB, and reduce + * to TEMP_EBB, if possible. + */ +static void __attribute__((noinline)) +liveness_pass_0(TCGContext *s) +{ + void * const multiple_ebb = (void *)(uintptr_t)-1; + int nb_temps = s->nb_temps; + TCGOp *op, *ebb; + + for (int i = s->nb_globals; i < nb_temps; ++i) { + s->temps[i].state_ptr = NULL; + } + + /* + * Represent each EBB by the op at which it begins. In the case of + * the first EBB, this is the first op, otherwise it is a label. + * Collect the uses of each TEMP_TB: NULL for unused, EBB for use + * within a single EBB, else MULTIPLE_EBB. + */ + ebb = QTAILQ_FIRST(&s->ops); + QTAILQ_FOREACH(op, &s->ops, link) { + const TCGOpDef *def; + int nb_oargs, nb_iargs; + + switch (op->opc) { + case INDEX_op_set_label: + ebb = op; + continue; + case INDEX_op_discard: + continue; + case INDEX_op_call: + nb_oargs = TCGOP_CALLO(op); + nb_iargs = TCGOP_CALLI(op); + break; + default: + def = &tcg_op_defs[op->opc]; + nb_oargs = def->nb_oargs; + nb_iargs = def->nb_iargs; + break; + } + + for (int i = 0; i < nb_oargs + nb_iargs; ++i) { + TCGTemp *ts = arg_temp(op->args[i]); + + if (ts->kind != TEMP_TB) { + continue; + } + if (ts->state_ptr == NULL) { + ts->state_ptr = ebb; + } else if (ts->state_ptr != ebb) { + ts->state_ptr = multiple_ebb; + } + } + } + + /* + * For TEMP_TB that turned out not to be used beyond one EBB, + * reduce the liveness to TEMP_EBB. + */ + for (int i = s->nb_globals; i < nb_temps; ++i) { + TCGTemp *ts = &s->temps[i]; + if (ts->kind == TEMP_TB && ts->state_ptr != multiple_ebb) { + ts->kind = TEMP_EBB; + } + } +} + /* Liveness analysis : update the opc_arg_life array to tell if a given input arguments is dead. Instructions updating dead temporaries are removed. */ @@ -XXX,XX +XXX,XX @@ int tcg_gen_code(TCGContext *s, TranslationBlock *tb, target_ulong pc_start) #endif reachable_code_pass(s); + liveness_pass_0(s); liveness_pass_1(s); if (s->nb_indirects > 0) { -- 2.34.1 TEMP_NORMAL is a subset of TEMP_EBB. Promote single basic block temps to single extended basic block. Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- include/tcg/tcg.h | 2 -- tcg/tcg.c | 19 +++---------------- 2 files changed, 3 insertions(+), 18 deletions(-) diff --git a/include/tcg/tcg.h b/include/tcg/tcg.h index XXXXXXX..XXXXXXX 100644 --- a/include/tcg/tcg.h +++ b/include/tcg/tcg.h @@ -XXX,XX +XXX,XX @@ typedef enum TCGTempVal { } TCGTempVal; typedef enum TCGTempKind { - /* Temp is dead at the end of all basic blocks. */ - TEMP_NORMAL, /* * Temp is dead at the end of the extended basic block (EBB), * the single-entry multiple-exit region that falls through diff --git a/tcg/tcg.c b/tcg/tcg.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/tcg.c +++ b/tcg/tcg.c @@ -XXX,XX +XXX,XX @@ TCGTemp *tcg_global_mem_new_internal(TCGType type, TCGv_ptr base, TCGTemp *tcg_temp_new_internal(TCGType type, bool temp_local) { TCGContext *s = tcg_ctx; - TCGTempKind kind = temp_local ? TEMP_TB : TEMP_NORMAL; + TCGTempKind kind = temp_local ? TEMP_TB : TEMP_EBB; TCGTemp *ts; int idx, k; @@ -XXX,XX +XXX,XX @@ void tcg_temp_free_internal(TCGTemp *ts) * silently ignore free. */ return; - case TEMP_NORMAL: + case TEMP_EBB: case TEMP_TB: break; default: @@ -XXX,XX +XXX,XX @@ void tcg_temp_free_internal(TCGTemp *ts) #endif idx = temp_idx(ts); - k = ts->base_type + (ts->kind == TEMP_NORMAL ? 0 : TCG_TYPE_COUNT); + k = ts->base_type + (ts->kind == TEMP_EBB ? 0 : TCG_TYPE_COUNT); set_bit(idx, s->free_temps[k].l); } @@ -XXX,XX +XXX,XX @@ static void tcg_reg_alloc_start(TCGContext *s) break; case TEMP_GLOBAL: break; - case TEMP_NORMAL: case TEMP_EBB: val = TEMP_VAL_DEAD; /* fall through */ @@ -XXX,XX +XXX,XX @@ static char *tcg_get_arg_str_ptr(TCGContext *s, char *buf, int buf_size, snprintf(buf, buf_size, "loc%d", idx - s->nb_globals); break; case TEMP_EBB: - snprintf(buf, buf_size, "ebb%d", idx - s->nb_globals); - break; - case TEMP_NORMAL: snprintf(buf, buf_size, "tmp%d", idx - s->nb_globals); break; case TEMP_CONST: @@ -XXX,XX +XXX,XX @@ static void la_bb_end(TCGContext *s, int ng, int nt) case TEMP_TB: state = TS_DEAD | TS_MEM; break; - case TEMP_NORMAL: case TEMP_EBB: case TEMP_CONST: state = TS_DEAD; @@ -XXX,XX +XXX,XX @@ static void la_bb_sync(TCGContext *s, int ng, int nt) continue; } break; - case TEMP_NORMAL: - s->temps[i].state = TS_DEAD; - break; case TEMP_EBB: case TEMP_CONST: continue; @@ -XXX,XX +XXX,XX @@ static void temp_free_or_dead(TCGContext *s, TCGTemp *ts, int free_or_dead) case TEMP_TB: new_type = TEMP_VAL_MEM; break; - case TEMP_NORMAL: case TEMP_EBB: new_type = free_or_dead < 0 ? TEMP_VAL_MEM : TEMP_VAL_DEAD; break; @@ -XXX,XX +XXX,XX @@ static void tcg_reg_alloc_bb_end(TCGContext *s, TCGRegSet allocated_regs) case TEMP_TB: temp_save(s, ts, allocated_regs); break; - case TEMP_NORMAL: case TEMP_EBB: /* The liveness analysis already ensures that temps are dead. Keep an tcg_debug_assert for safety. */ @@ -XXX,XX +XXX,XX @@ static void tcg_reg_alloc_cbranch(TCGContext *s, TCGRegSet allocated_regs) case TEMP_TB: tcg_debug_assert(ts->val_type != TEMP_VAL_REG || ts->mem_coherent); break; - case TEMP_NORMAL: - tcg_debug_assert(ts->val_type == TEMP_VAL_DEAD); - break; case TEMP_EBB: case TEMP_CONST: break; -- 2.34.1 While the argument can only be TEMP_EBB or TEMP_TB, it's more obvious this way. Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- include/tcg/tcg.h | 18 +++++++++--------- tcg/tcg.c | 8 ++++---- 2 files changed, 13 insertions(+), 13 deletions(-) diff --git a/include/tcg/tcg.h b/include/tcg/tcg.h index XXXXXXX..XXXXXXX 100644 --- a/include/tcg/tcg.h +++ b/include/tcg/tcg.h @@ -XXX,XX +XXX,XX @@ void tcg_set_frame(TCGContext *s, TCGReg reg, intptr_t start, intptr_t size); TCGTemp *tcg_global_mem_new_internal(TCGType, TCGv_ptr, intptr_t, const char *); -TCGTemp *tcg_temp_new_internal(TCGType, bool); +TCGTemp *tcg_temp_new_internal(TCGType, TCGTempKind); void tcg_temp_free_internal(TCGTemp *); TCGv_vec tcg_temp_new_vec(TCGType type); TCGv_vec tcg_temp_new_vec_matching(TCGv_vec match); @@ -XXX,XX +XXX,XX @@ static inline TCGv_i32 tcg_global_mem_new_i32(TCGv_ptr reg, intptr_t offset, static inline TCGv_i32 tcg_temp_new_i32(void) { - TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_I32, false); + TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_I32, TEMP_EBB); return temp_tcgv_i32(t); } static inline TCGv_i32 tcg_temp_local_new_i32(void) { - TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_I32, true); + TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_I32, TEMP_TB); return temp_tcgv_i32(t); } @@ -XXX,XX +XXX,XX @@ static inline TCGv_i64 tcg_global_mem_new_i64(TCGv_ptr reg, intptr_t offset, static inline TCGv_i64 tcg_temp_new_i64(void) { - TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_I64, false); + TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_I64, TEMP_EBB); return temp_tcgv_i64(t); } static inline TCGv_i64 tcg_temp_local_new_i64(void) { - TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_I64, true); + TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_I64, TEMP_TB); return temp_tcgv_i64(t); } static inline TCGv_i128 tcg_temp_new_i128(void) { - TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_I128, false); + TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_I128, TEMP_EBB); return temp_tcgv_i128(t); } static inline TCGv_i128 tcg_temp_local_new_i128(void) { - TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_I128, true); + TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_I128, TEMP_TB); return temp_tcgv_i128(t); } @@ -XXX,XX +XXX,XX @@ static inline TCGv_ptr tcg_global_mem_new_ptr(TCGv_ptr reg, intptr_t offset, static inline TCGv_ptr tcg_temp_new_ptr(void) { - TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_PTR, false); + TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_PTR, TEMP_EBB); return temp_tcgv_ptr(t); } static inline TCGv_ptr tcg_temp_local_new_ptr(void) { - TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_PTR, true); + TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_PTR, TEMP_TB); return temp_tcgv_ptr(t); } diff --git a/tcg/tcg.c b/tcg/tcg.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/tcg.c +++ b/tcg/tcg.c @@ -XXX,XX +XXX,XX @@ TCGTemp *tcg_global_mem_new_internal(TCGType type, TCGv_ptr base, return ts; } -TCGTemp *tcg_temp_new_internal(TCGType type, bool temp_local) +TCGTemp *tcg_temp_new_internal(TCGType type, TCGTempKind kind) { TCGContext *s = tcg_ctx; - TCGTempKind kind = temp_local ? TEMP_TB : TEMP_EBB; + bool temp_local = kind == TEMP_TB; TCGTemp *ts; int idx, k; @@ -XXX,XX +XXX,XX @@ TCGv_vec tcg_temp_new_vec(TCGType type) } #endif - t = tcg_temp_new_internal(type, 0); + t = tcg_temp_new_internal(type, TEMP_EBB); return temp_tcgv_vec(t); } @@ -XXX,XX +XXX,XX @@ TCGv_vec tcg_temp_new_vec_matching(TCGv_vec match) tcg_debug_assert(t->temp_allocated != 0); - t = tcg_temp_new_internal(t->base_type, 0); + t = tcg_temp_new_internal(t->base_type, TEMP_EBB); return temp_tcgv_vec(t); } -- 2.34.1 Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- include/exec/gen-icount.h | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/include/exec/gen-icount.h b/include/exec/gen-icount.h index XXXXXXX..XXXXXXX 100644 --- a/include/exec/gen-icount.h +++ b/include/exec/gen-icount.h @@ -XXX,XX +XXX,XX @@ static TCGOp *icount_start_insn; static inline void gen_io_start(void) { - TCGv_i32 tmp = tcg_const_i32(1); - tcg_gen_st_i32(tmp, cpu_env, + tcg_gen_st_i32(tcg_constant_i32(1), cpu_env, offsetof(ArchCPU, parent_obj.can_do_io) - offsetof(ArchCPU, env)); - tcg_temp_free_i32(tmp); } static inline void gen_tb_start(const TranslationBlock *tb) -- 2.34.1 Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- include/tcg/tcg-op.h | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/include/tcg/tcg-op.h b/include/tcg/tcg-op.h index XXXXXXX..XXXXXXX 100644 --- a/include/tcg/tcg-op.h +++ b/include/tcg/tcg-op.h @@ -XXX,XX +XXX,XX @@ static inline void tcg_gen_mov_ptr(TCGv_ptr d, TCGv_ptr s) glue(tcg_gen_mov_,PTR)((NAT)d, (NAT)s); } +static inline void tcg_gen_movi_ptr(TCGv_ptr d, intptr_t s) +{ + glue(tcg_gen_movi_,PTR)((NAT)d, s); +} + static inline void tcg_gen_brcondi_ptr(TCGCond cond, TCGv_ptr a, intptr_t b, TCGLabel *label) { -- 2.34.1 TCG internals will want to be able to allocate and reuse explicitly life-limited temporaries. Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- include/tcg/tcg.h | 28 ++++++++++++++++++++++++++++ 1 file changed, 28 insertions(+) diff --git a/include/tcg/tcg.h b/include/tcg/tcg.h index XXXXXXX..XXXXXXX 100644 --- a/include/tcg/tcg.h +++ b/include/tcg/tcg.h @@ -XXX,XX +XXX,XX @@ static inline TCGv_i32 tcg_global_mem_new_i32(TCGv_ptr reg, intptr_t offset, return temp_tcgv_i32(t); } +/* Used only by tcg infrastructure: tcg-op.c or plugin-gen.c */ +static inline TCGv_i32 tcg_temp_ebb_new_i32(void) +{ + TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_I32, TEMP_EBB); + return temp_tcgv_i32(t); +} + static inline TCGv_i32 tcg_temp_new_i32(void) { TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_I32, TEMP_EBB); @@ -XXX,XX +XXX,XX @@ static inline TCGv_i64 tcg_global_mem_new_i64(TCGv_ptr reg, intptr_t offset, return temp_tcgv_i64(t); } +/* Used only by tcg infrastructure: tcg-op.c or plugin-gen.c */ +static inline TCGv_i64 tcg_temp_ebb_new_i64(void) +{ + TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_I64, TEMP_EBB); + return temp_tcgv_i64(t); +} + static inline TCGv_i64 tcg_temp_new_i64(void) { TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_I64, TEMP_EBB); @@ -XXX,XX +XXX,XX @@ static inline TCGv_i64 tcg_temp_local_new_i64(void) return temp_tcgv_i64(t); } +/* Used only by tcg infrastructure: tcg-op.c or plugin-gen.c */ +static inline TCGv_i128 tcg_temp_ebb_new_i128(void) +{ + TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_I128, TEMP_EBB); + return temp_tcgv_i128(t); +} + static inline TCGv_i128 tcg_temp_new_i128(void) { TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_I128, TEMP_EBB); @@ -XXX,XX +XXX,XX @@ static inline TCGv_ptr tcg_global_mem_new_ptr(TCGv_ptr reg, intptr_t offset, return temp_tcgv_ptr(t); } +/* Used only by tcg infrastructure: tcg-op.c or plugin-gen.c */ +static inline TCGv_ptr tcg_temp_ebb_new_ptr(void) +{ + TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_PTR, TEMP_EBB); + return temp_tcgv_ptr(t); +} + static inline TCGv_ptr tcg_temp_new_ptr(void) { TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_PTR, TEMP_EBB); -- 2.34.1 All of these have obvious and quite local scope. Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/tcg-op-gvec.c | 186 ++++++++++++++++----------------- tcg/tcg-op.c | 258 +++++++++++++++++++++++----------------------- tcg/tcg.c | 2 +- 3 files changed, 223 insertions(+), 223 deletions(-) diff --git a/tcg/tcg-op-gvec.c b/tcg/tcg-op-gvec.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/tcg-op-gvec.c +++ b/tcg/tcg-op-gvec.c @@ -XXX,XX +XXX,XX @@ void tcg_gen_gvec_2_ool(uint32_t dofs, uint32_t aofs, TCGv_ptr a0, a1; TCGv_i32 desc = tcg_constant_i32(simd_desc(oprsz, maxsz, data)); - a0 = tcg_temp_new_ptr(); - a1 = tcg_temp_new_ptr(); + a0 = tcg_temp_ebb_new_ptr(); + a1 = tcg_temp_ebb_new_ptr(); tcg_gen_addi_ptr(a0, cpu_env, dofs); tcg_gen_addi_ptr(a1, cpu_env, aofs); @@ -XXX,XX +XXX,XX @@ void tcg_gen_gvec_2i_ool(uint32_t dofs, uint32_t aofs, TCGv_i64 c, TCGv_ptr a0, a1; TCGv_i32 desc = tcg_constant_i32(simd_desc(oprsz, maxsz, data)); - a0 = tcg_temp_new_ptr(); - a1 = tcg_temp_new_ptr(); + a0 = tcg_temp_ebb_new_ptr(); + a1 = tcg_temp_ebb_new_ptr(); tcg_gen_addi_ptr(a0, cpu_env, dofs); tcg_gen_addi_ptr(a1, cpu_env, aofs); @@ -XXX,XX +XXX,XX @@ void tcg_gen_gvec_3_ool(uint32_t dofs, uint32_t aofs, uint32_t bofs, TCGv_ptr a0, a1, a2; TCGv_i32 desc = tcg_constant_i32(simd_desc(oprsz, maxsz, data)); - a0 = tcg_temp_new_ptr(); - a1 = tcg_temp_new_ptr(); - a2 = tcg_temp_new_ptr(); + a0 = tcg_temp_ebb_new_ptr(); + a1 = tcg_temp_ebb_new_ptr(); + a2 = tcg_temp_ebb_new_ptr(); tcg_gen_addi_ptr(a0, cpu_env, dofs); tcg_gen_addi_ptr(a1, cpu_env, aofs); @@ -XXX,XX +XXX,XX @@ void tcg_gen_gvec_4_ool(uint32_t dofs, uint32_t aofs, uint32_t bofs, TCGv_ptr a0, a1, a2, a3; TCGv_i32 desc = tcg_constant_i32(simd_desc(oprsz, maxsz, data)); - a0 = tcg_temp_new_ptr(); - a1 = tcg_temp_new_ptr(); - a2 = tcg_temp_new_ptr(); - a3 = tcg_temp_new_ptr(); + a0 = tcg_temp_ebb_new_ptr(); + a1 = tcg_temp_ebb_new_ptr(); + a2 = tcg_temp_ebb_new_ptr(); + a3 = tcg_temp_ebb_new_ptr(); tcg_gen_addi_ptr(a0, cpu_env, dofs); tcg_gen_addi_ptr(a1, cpu_env, aofs); @@ -XXX,XX +XXX,XX @@ void tcg_gen_gvec_5_ool(uint32_t dofs, uint32_t aofs, uint32_t bofs, TCGv_ptr a0, a1, a2, a3, a4; TCGv_i32 desc = tcg_constant_i32(simd_desc(oprsz, maxsz, data)); - a0 = tcg_temp_new_ptr(); - a1 = tcg_temp_new_ptr(); - a2 = tcg_temp_new_ptr(); - a3 = tcg_temp_new_ptr(); - a4 = tcg_temp_new_ptr(); + a0 = tcg_temp_ebb_new_ptr(); + a1 = tcg_temp_ebb_new_ptr(); + a2 = tcg_temp_ebb_new_ptr(); + a3 = tcg_temp_ebb_new_ptr(); + a4 = tcg_temp_ebb_new_ptr(); tcg_gen_addi_ptr(a0, cpu_env, dofs); tcg_gen_addi_ptr(a1, cpu_env, aofs); @@ -XXX,XX +XXX,XX @@ void tcg_gen_gvec_2_ptr(uint32_t dofs, uint32_t aofs, TCGv_ptr a0, a1; TCGv_i32 desc = tcg_constant_i32(simd_desc(oprsz, maxsz, data)); - a0 = tcg_temp_new_ptr(); - a1 = tcg_temp_new_ptr(); + a0 = tcg_temp_ebb_new_ptr(); + a1 = tcg_temp_ebb_new_ptr(); tcg_gen_addi_ptr(a0, cpu_env, dofs); tcg_gen_addi_ptr(a1, cpu_env, aofs); @@ -XXX,XX +XXX,XX @@ void tcg_gen_gvec_3_ptr(uint32_t dofs, uint32_t aofs, uint32_t bofs, TCGv_ptr a0, a1, a2; TCGv_i32 desc = tcg_constant_i32(simd_desc(oprsz, maxsz, data)); - a0 = tcg_temp_new_ptr(); - a1 = tcg_temp_new_ptr(); - a2 = tcg_temp_new_ptr(); + a0 = tcg_temp_ebb_new_ptr(); + a1 = tcg_temp_ebb_new_ptr(); + a2 = tcg_temp_ebb_new_ptr(); tcg_gen_addi_ptr(a0, cpu_env, dofs); tcg_gen_addi_ptr(a1, cpu_env, aofs); @@ -XXX,XX +XXX,XX @@ void tcg_gen_gvec_4_ptr(uint32_t dofs, uint32_t aofs, uint32_t bofs, TCGv_ptr a0, a1, a2, a3; TCGv_i32 desc = tcg_constant_i32(simd_desc(oprsz, maxsz, data)); - a0 = tcg_temp_new_ptr(); - a1 = tcg_temp_new_ptr(); - a2 = tcg_temp_new_ptr(); - a3 = tcg_temp_new_ptr(); + a0 = tcg_temp_ebb_new_ptr(); + a1 = tcg_temp_ebb_new_ptr(); + a2 = tcg_temp_ebb_new_ptr(); + a3 = tcg_temp_ebb_new_ptr(); tcg_gen_addi_ptr(a0, cpu_env, dofs); tcg_gen_addi_ptr(a1, cpu_env, aofs); @@ -XXX,XX +XXX,XX @@ void tcg_gen_gvec_5_ptr(uint32_t dofs, uint32_t aofs, uint32_t bofs, TCGv_ptr a0, a1, a2, a3, a4; TCGv_i32 desc = tcg_constant_i32(simd_desc(oprsz, maxsz, data)); - a0 = tcg_temp_new_ptr(); - a1 = tcg_temp_new_ptr(); - a2 = tcg_temp_new_ptr(); - a3 = tcg_temp_new_ptr(); - a4 = tcg_temp_new_ptr(); + a0 = tcg_temp_ebb_new_ptr(); + a1 = tcg_temp_ebb_new_ptr(); + a2 = tcg_temp_ebb_new_ptr(); + a3 = tcg_temp_ebb_new_ptr(); + a4 = tcg_temp_ebb_new_ptr(); tcg_gen_addi_ptr(a0, cpu_env, dofs); tcg_gen_addi_ptr(a1, cpu_env, aofs); @@ -XXX,XX +XXX,XX @@ static void do_dup(unsigned vece, uint32_t dofs, uint32_t oprsz, be simple enough. */ if (TCG_TARGET_REG_BITS == 64 && (vece != MO_32 || !check_size_impl(oprsz, 4))) { - t_64 = tcg_temp_new_i64(); + t_64 = tcg_temp_ebb_new_i64(); tcg_gen_extu_i32_i64(t_64, in_32); tcg_gen_dup_i64(vece, t_64, t_64); } else { - t_32 = tcg_temp_new_i32(); + t_32 = tcg_temp_ebb_new_i32(); tcg_gen_dup_i32(vece, t_32, in_32); } } else if (in_64) { /* We are given a 64-bit variable input. */ - t_64 = tcg_temp_new_i64(); + t_64 = tcg_temp_ebb_new_i64(); tcg_gen_dup_i64(vece, t_64, in_64); } else { /* We are given a constant input. */ @@ -XXX,XX +XXX,XX @@ static void do_dup(unsigned vece, uint32_t dofs, uint32_t oprsz, } /* Otherwise implement out of line. */ - t_ptr = tcg_temp_new_ptr(); + t_ptr = tcg_temp_ebb_new_ptr(); tcg_gen_addi_ptr(t_ptr, cpu_env, dofs); /* @@ -XXX,XX +XXX,XX @@ static void do_dup(unsigned vece, uint32_t dofs, uint32_t oprsz, if (in_32) { t_val = in_32; } else if (in_64) { - t_val = tcg_temp_new_i32(); + t_val = tcg_temp_ebb_new_i32(); tcg_gen_extrl_i64_i32(t_val, in_64); } else { t_val = tcg_constant_i32(in_c); @@ -XXX,XX +XXX,XX @@ static void do_dup(unsigned vece, uint32_t dofs, uint32_t oprsz, if (in_32) { fns[vece](t_ptr, t_desc, in_32); } else if (in_64) { - t_32 = tcg_temp_new_i32(); + t_32 = tcg_temp_ebb_new_i32(); tcg_gen_extrl_i64_i32(t_32, in_64); fns[vece](t_ptr, t_desc, t_32); tcg_temp_free_i32(t_32); @@ -XXX,XX +XXX,XX @@ void tcg_gen_gvec_dup_mem(unsigned vece, uint32_t dofs, uint32_t aofs, do_dup_store(type, dofs, oprsz, maxsz, t_vec); tcg_temp_free_vec(t_vec); } else if (vece <= MO_32) { - TCGv_i32 in = tcg_temp_new_i32(); + TCGv_i32 in = tcg_temp_ebb_new_i32(); switch (vece) { case MO_8: tcg_gen_ld8u_i32(in, cpu_env, aofs); @@ -XXX,XX +XXX,XX @@ void tcg_gen_gvec_dup_mem(unsigned vece, uint32_t dofs, uint32_t aofs, do_dup(vece, dofs, oprsz, maxsz, in, NULL, 0); tcg_temp_free_i32(in); } else { - TCGv_i64 in = tcg_temp_new_i64(); + TCGv_i64 in = tcg_temp_ebb_new_i64(); tcg_gen_ld_i64(in, cpu_env, aofs); do_dup(vece, dofs, oprsz, maxsz, NULL, in, 0); tcg_temp_free_i64(in); @@ -XXX,XX +XXX,XX @@ void tcg_gen_gvec_dup_mem(unsigned vece, uint32_t dofs, uint32_t aofs, } tcg_temp_free_vec(in); } else { - TCGv_i64 in0 = tcg_temp_new_i64(); - TCGv_i64 in1 = tcg_temp_new_i64(); + TCGv_i64 in0 = tcg_temp_ebb_new_i64(); + TCGv_i64 in1 = tcg_temp_ebb_new_i64(); tcg_gen_ld_i64(in0, cpu_env, aofs); tcg_gen_ld_i64(in1, cpu_env, aofs + 8); @@ -XXX,XX +XXX,XX @@ void tcg_gen_gvec_dup_mem(unsigned vece, uint32_t dofs, uint32_t aofs, int j; for (j = 0; j < 4; ++j) { - in[j] = tcg_temp_new_i64(); + in[j] = tcg_temp_ebb_new_i64(); tcg_gen_ld_i64(in[j], cpu_env, aofs + j * 8); } for (i = (aofs == dofs) * 32; i < oprsz; i += 32) { @@ -XXX,XX +XXX,XX @@ void tcg_gen_gvec_not(unsigned vece, uint32_t dofs, uint32_t aofs, the 64-bit operation. */ static void gen_addv_mask(TCGv_i64 d, TCGv_i64 a, TCGv_i64 b, TCGv_i64 m) { - TCGv_i64 t1 = tcg_temp_new_i64(); - TCGv_i64 t2 = tcg_temp_new_i64(); - TCGv_i64 t3 = tcg_temp_new_i64(); + TCGv_i64 t1 = tcg_temp_ebb_new_i64(); + TCGv_i64 t2 = tcg_temp_ebb_new_i64(); + TCGv_i64 t3 = tcg_temp_ebb_new_i64(); tcg_gen_andc_i64(t1, a, m); tcg_gen_andc_i64(t2, b, m); @@ -XXX,XX +XXX,XX @@ void tcg_gen_vec_add8_i64(TCGv_i64 d, TCGv_i64 a, TCGv_i64 b) void tcg_gen_vec_add8_i32(TCGv_i32 d, TCGv_i32 a, TCGv_i32 b) { TCGv_i32 m = tcg_constant_i32((int32_t)dup_const(MO_8, 0x80)); - TCGv_i32 t1 = tcg_temp_new_i32(); - TCGv_i32 t2 = tcg_temp_new_i32(); - TCGv_i32 t3 = tcg_temp_new_i32(); + TCGv_i32 t1 = tcg_temp_ebb_new_i32(); + TCGv_i32 t2 = tcg_temp_ebb_new_i32(); + TCGv_i32 t3 = tcg_temp_ebb_new_i32(); tcg_gen_andc_i32(t1, a, m); tcg_gen_andc_i32(t2, b, m); @@ -XXX,XX +XXX,XX @@ void tcg_gen_vec_add16_i64(TCGv_i64 d, TCGv_i64 a, TCGv_i64 b) void tcg_gen_vec_add16_i32(TCGv_i32 d, TCGv_i32 a, TCGv_i32 b) { - TCGv_i32 t1 = tcg_temp_new_i32(); - TCGv_i32 t2 = tcg_temp_new_i32(); + TCGv_i32 t1 = tcg_temp_ebb_new_i32(); + TCGv_i32 t2 = tcg_temp_ebb_new_i32(); tcg_gen_andi_i32(t1, a, ~0xffff); tcg_gen_add_i32(t2, a, b); @@ -XXX,XX +XXX,XX @@ void tcg_gen_vec_add16_i32(TCGv_i32 d, TCGv_i32 a, TCGv_i32 b) void tcg_gen_vec_add32_i64(TCGv_i64 d, TCGv_i64 a, TCGv_i64 b) { - TCGv_i64 t1 = tcg_temp_new_i64(); - TCGv_i64 t2 = tcg_temp_new_i64(); + TCGv_i64 t1 = tcg_temp_ebb_new_i64(); + TCGv_i64 t2 = tcg_temp_ebb_new_i64(); tcg_gen_andi_i64(t1, a, ~0xffffffffull); tcg_gen_add_i64(t2, a, b); @@ -XXX,XX +XXX,XX @@ void tcg_gen_gvec_subs(unsigned vece, uint32_t dofs, uint32_t aofs, Compare gen_addv_mask above. */ static void gen_subv_mask(TCGv_i64 d, TCGv_i64 a, TCGv_i64 b, TCGv_i64 m) { - TCGv_i64 t1 = tcg_temp_new_i64(); - TCGv_i64 t2 = tcg_temp_new_i64(); - TCGv_i64 t3 = tcg_temp_new_i64(); + TCGv_i64 t1 = tcg_temp_ebb_new_i64(); + TCGv_i64 t2 = tcg_temp_ebb_new_i64(); + TCGv_i64 t3 = tcg_temp_ebb_new_i64(); tcg_gen_or_i64(t1, a, m); tcg_gen_andc_i64(t2, b, m); @@ -XXX,XX +XXX,XX @@ void tcg_gen_vec_sub8_i64(TCGv_i64 d, TCGv_i64 a, TCGv_i64 b) void tcg_gen_vec_sub8_i32(TCGv_i32 d, TCGv_i32 a, TCGv_i32 b) { TCGv_i32 m = tcg_constant_i32((int32_t)dup_const(MO_8, 0x80)); - TCGv_i32 t1 = tcg_temp_new_i32(); - TCGv_i32 t2 = tcg_temp_new_i32(); - TCGv_i32 t3 = tcg_temp_new_i32(); + TCGv_i32 t1 = tcg_temp_ebb_new_i32(); + TCGv_i32 t2 = tcg_temp_ebb_new_i32(); + TCGv_i32 t3 = tcg_temp_ebb_new_i32(); tcg_gen_or_i32(t1, a, m); tcg_gen_andc_i32(t2, b, m); @@ -XXX,XX +XXX,XX @@ void tcg_gen_vec_sub16_i64(TCGv_i64 d, TCGv_i64 a, TCGv_i64 b) void tcg_gen_vec_sub16_i32(TCGv_i32 d, TCGv_i32 a, TCGv_i32 b) { - TCGv_i32 t1 = tcg_temp_new_i32(); - TCGv_i32 t2 = tcg_temp_new_i32(); + TCGv_i32 t1 = tcg_temp_ebb_new_i32(); + TCGv_i32 t2 = tcg_temp_ebb_new_i32(); tcg_gen_andi_i32(t1, b, ~0xffff); tcg_gen_sub_i32(t2, a, b); @@ -XXX,XX +XXX,XX @@ void tcg_gen_vec_sub16_i32(TCGv_i32 d, TCGv_i32 a, TCGv_i32 b) void tcg_gen_vec_sub32_i64(TCGv_i64 d, TCGv_i64 a, TCGv_i64 b) { - TCGv_i64 t1 = tcg_temp_new_i64(); - TCGv_i64 t2 = tcg_temp_new_i64(); + TCGv_i64 t1 = tcg_temp_ebb_new_i64(); + TCGv_i64 t2 = tcg_temp_ebb_new_i64(); tcg_gen_andi_i64(t1, b, ~0xffffffffull); tcg_gen_sub_i64(t2, a, b); @@ -XXX,XX +XXX,XX @@ void tcg_gen_gvec_umax(unsigned vece, uint32_t dofs, uint32_t aofs, Compare gen_subv_mask above. */ static void gen_negv_mask(TCGv_i64 d, TCGv_i64 b, TCGv_i64 m) { - TCGv_i64 t2 = tcg_temp_new_i64(); - TCGv_i64 t3 = tcg_temp_new_i64(); + TCGv_i64 t2 = tcg_temp_ebb_new_i64(); + TCGv_i64 t3 = tcg_temp_ebb_new_i64(); tcg_gen_andc_i64(t3, m, b); tcg_gen_andc_i64(t2, b, m); @@ -XXX,XX +XXX,XX @@ void tcg_gen_vec_neg16_i64(TCGv_i64 d, TCGv_i64 b) void tcg_gen_vec_neg32_i64(TCGv_i64 d, TCGv_i64 b) { - TCGv_i64 t1 = tcg_temp_new_i64(); - TCGv_i64 t2 = tcg_temp_new_i64(); + TCGv_i64 t1 = tcg_temp_ebb_new_i64(); + TCGv_i64 t2 = tcg_temp_ebb_new_i64(); tcg_gen_andi_i64(t1, b, ~0xffffffffull); tcg_gen_neg_i64(t2, b); @@ -XXX,XX +XXX,XX @@ void tcg_gen_gvec_neg(unsigned vece, uint32_t dofs, uint32_t aofs, static void gen_absv_mask(TCGv_i64 d, TCGv_i64 b, unsigned vece) { - TCGv_i64 t = tcg_temp_new_i64(); + TCGv_i64 t = tcg_temp_ebb_new_i64(); int nbit = 8 << vece; /* Create -1 for each negative element. */ @@ -XXX,XX +XXX,XX @@ static const GVecGen2s gop_ands = { void tcg_gen_gvec_ands(unsigned vece, uint32_t dofs, uint32_t aofs, TCGv_i64 c, uint32_t oprsz, uint32_t maxsz) { - TCGv_i64 tmp = tcg_temp_new_i64(); + TCGv_i64 tmp = tcg_temp_ebb_new_i64(); tcg_gen_dup_i64(vece, tmp, c); tcg_gen_gvec_2s(dofs, aofs, oprsz, maxsz, tmp, &gop_ands); tcg_temp_free_i64(tmp); @@ -XXX,XX +XXX,XX @@ static const GVecGen2s gop_xors = { void tcg_gen_gvec_xors(unsigned vece, uint32_t dofs, uint32_t aofs, TCGv_i64 c, uint32_t oprsz, uint32_t maxsz) { - TCGv_i64 tmp = tcg_temp_new_i64(); + TCGv_i64 tmp = tcg_temp_ebb_new_i64(); tcg_gen_dup_i64(vece, tmp, c); tcg_gen_gvec_2s(dofs, aofs, oprsz, maxsz, tmp, &gop_xors); tcg_temp_free_i64(tmp); @@ -XXX,XX +XXX,XX @@ static const GVecGen2s gop_ors = { void tcg_gen_gvec_ors(unsigned vece, uint32_t dofs, uint32_t aofs, TCGv_i64 c, uint32_t oprsz, uint32_t maxsz) { - TCGv_i64 tmp = tcg_temp_new_i64(); + TCGv_i64 tmp = tcg_temp_ebb_new_i64(); tcg_gen_dup_i64(vece, tmp, c); tcg_gen_gvec_2s(dofs, aofs, oprsz, maxsz, tmp, &gop_ors); tcg_temp_free_i64(tmp); @@ -XXX,XX +XXX,XX @@ void tcg_gen_vec_sar8i_i64(TCGv_i64 d, TCGv_i64 a, int64_t c) { uint64_t s_mask = dup_const(MO_8, 0x80 >> c); uint64_t c_mask = dup_const(MO_8, 0xff >> c); - TCGv_i64 s = tcg_temp_new_i64(); + TCGv_i64 s = tcg_temp_ebb_new_i64(); tcg_gen_shri_i64(d, a, c); tcg_gen_andi_i64(s, d, s_mask); /* isolate (shifted) sign bit */ @@ -XXX,XX +XXX,XX @@ void tcg_gen_vec_sar16i_i64(TCGv_i64 d, TCGv_i64 a, int64_t c) { uint64_t s_mask = dup_const(MO_16, 0x8000 >> c); uint64_t c_mask = dup_const(MO_16, 0xffff >> c); - TCGv_i64 s = tcg_temp_new_i64(); + TCGv_i64 s = tcg_temp_ebb_new_i64(); tcg_gen_shri_i64(d, a, c); tcg_gen_andi_i64(s, d, s_mask); /* isolate (shifted) sign bit */ @@ -XXX,XX +XXX,XX @@ void tcg_gen_vec_sar8i_i32(TCGv_i32 d, TCGv_i32 a, int32_t c) { uint32_t s_mask = dup_const(MO_8, 0x80 >> c); uint32_t c_mask = dup_const(MO_8, 0xff >> c); - TCGv_i32 s = tcg_temp_new_i32(); + TCGv_i32 s = tcg_temp_ebb_new_i32(); tcg_gen_shri_i32(d, a, c); tcg_gen_andi_i32(s, d, s_mask); /* isolate (shifted) sign bit */ @@ -XXX,XX +XXX,XX @@ void tcg_gen_vec_sar16i_i32(TCGv_i32 d, TCGv_i32 a, int32_t c) { uint32_t s_mask = dup_const(MO_16, 0x8000 >> c); uint32_t c_mask = dup_const(MO_16, 0xffff >> c); - TCGv_i32 s = tcg_temp_new_i32(); + TCGv_i32 s = tcg_temp_ebb_new_i32(); tcg_gen_shri_i32(d, a, c); tcg_gen_andi_i32(s, d, s_mask); /* isolate (shifted) sign bit */ @@ -XXX,XX +XXX,XX @@ do_gvec_shifts(unsigned vece, uint32_t dofs, uint32_t aofs, TCGv_i32 shift, TCGv_vec v_shift = tcg_temp_new_vec(type); if (vece == MO_64) { - TCGv_i64 sh64 = tcg_temp_new_i64(); + TCGv_i64 sh64 = tcg_temp_ebb_new_i64(); tcg_gen_extu_i32_i64(sh64, shift); tcg_gen_dup_i64_vec(MO_64, v_shift, sh64); tcg_temp_free_i64(sh64); @@ -XXX,XX +XXX,XX @@ do_gvec_shifts(unsigned vece, uint32_t dofs, uint32_t aofs, TCGv_i32 shift, if (vece == MO_32 && check_size_impl(oprsz, 4)) { expand_2s_i32(dofs, aofs, oprsz, shift, false, g->fni4); } else if (vece == MO_64 && check_size_impl(oprsz, 8)) { - TCGv_i64 sh64 = tcg_temp_new_i64(); + TCGv_i64 sh64 = tcg_temp_ebb_new_i64(); tcg_gen_extu_i32_i64(sh64, shift); expand_2s_i64(dofs, aofs, oprsz, sh64, false, g->fni8); tcg_temp_free_i64(sh64); } else { - TCGv_ptr a0 = tcg_temp_new_ptr(); - TCGv_ptr a1 = tcg_temp_new_ptr(); - TCGv_i32 desc = tcg_temp_new_i32(); + TCGv_ptr a0 = tcg_temp_ebb_new_ptr(); + TCGv_ptr a1 = tcg_temp_ebb_new_ptr(); + TCGv_i32 desc = tcg_temp_ebb_new_i32(); tcg_gen_shli_i32(desc, shift, SIMD_DATA_SHIFT); tcg_gen_ori_i32(desc, desc, simd_desc(oprsz, maxsz, 0)); @@ -XXX,XX +XXX,XX @@ static void tcg_gen_shlv_mod_vec(unsigned vece, TCGv_vec d, static void tcg_gen_shl_mod_i32(TCGv_i32 d, TCGv_i32 a, TCGv_i32 b) { - TCGv_i32 t = tcg_temp_new_i32(); + TCGv_i32 t = tcg_temp_ebb_new_i32(); tcg_gen_andi_i32(t, b, 31); tcg_gen_shl_i32(d, a, t); @@ -XXX,XX +XXX,XX @@ static void tcg_gen_shl_mod_i32(TCGv_i32 d, TCGv_i32 a, TCGv_i32 b) static void tcg_gen_shl_mod_i64(TCGv_i64 d, TCGv_i64 a, TCGv_i64 b) { - TCGv_i64 t = tcg_temp_new_i64(); + TCGv_i64 t = tcg_temp_ebb_new_i64(); tcg_gen_andi_i64(t, b, 63); tcg_gen_shl_i64(d, a, t); @@ -XXX,XX +XXX,XX @@ static void tcg_gen_shrv_mod_vec(unsigned vece, TCGv_vec d, static void tcg_gen_shr_mod_i32(TCGv_i32 d, TCGv_i32 a, TCGv_i32 b) { - TCGv_i32 t = tcg_temp_new_i32(); + TCGv_i32 t = tcg_temp_ebb_new_i32(); tcg_gen_andi_i32(t, b, 31); tcg_gen_shr_i32(d, a, t); @@ -XXX,XX +XXX,XX @@ static void tcg_gen_shr_mod_i32(TCGv_i32 d, TCGv_i32 a, TCGv_i32 b) static void tcg_gen_shr_mod_i64(TCGv_i64 d, TCGv_i64 a, TCGv_i64 b) { - TCGv_i64 t = tcg_temp_new_i64(); + TCGv_i64 t = tcg_temp_ebb_new_i64(); tcg_gen_andi_i64(t, b, 63); tcg_gen_shr_i64(d, a, t); @@ -XXX,XX +XXX,XX @@ static void tcg_gen_sarv_mod_vec(unsigned vece, TCGv_vec d, static void tcg_gen_sar_mod_i32(TCGv_i32 d, TCGv_i32 a, TCGv_i32 b) { - TCGv_i32 t = tcg_temp_new_i32(); + TCGv_i32 t = tcg_temp_ebb_new_i32(); tcg_gen_andi_i32(t, b, 31); tcg_gen_sar_i32(d, a, t); @@ -XXX,XX +XXX,XX @@ static void tcg_gen_sar_mod_i32(TCGv_i32 d, TCGv_i32 a, TCGv_i32 b) static void tcg_gen_sar_mod_i64(TCGv_i64 d, TCGv_i64 a, TCGv_i64 b) { - TCGv_i64 t = tcg_temp_new_i64(); + TCGv_i64 t = tcg_temp_ebb_new_i64(); tcg_gen_andi_i64(t, b, 63); tcg_gen_sar_i64(d, a, t); @@ -XXX,XX +XXX,XX @@ static void tcg_gen_rotlv_mod_vec(unsigned vece, TCGv_vec d, static void tcg_gen_rotl_mod_i32(TCGv_i32 d, TCGv_i32 a, TCGv_i32 b) { - TCGv_i32 t = tcg_temp_new_i32(); + TCGv_i32 t = tcg_temp_ebb_new_i32(); tcg_gen_andi_i32(t, b, 31); tcg_gen_rotl_i32(d, a, t); @@ -XXX,XX +XXX,XX @@ static void tcg_gen_rotl_mod_i32(TCGv_i32 d, TCGv_i32 a, TCGv_i32 b) static void tcg_gen_rotl_mod_i64(TCGv_i64 d, TCGv_i64 a, TCGv_i64 b) { - TCGv_i64 t = tcg_temp_new_i64(); + TCGv_i64 t = tcg_temp_ebb_new_i64(); tcg_gen_andi_i64(t, b, 63); tcg_gen_rotl_i64(d, a, t); @@ -XXX,XX +XXX,XX @@ static void tcg_gen_rotrv_mod_vec(unsigned vece, TCGv_vec d, static void tcg_gen_rotr_mod_i32(TCGv_i32 d, TCGv_i32 a, TCGv_i32 b) { - TCGv_i32 t = tcg_temp_new_i32(); + TCGv_i32 t = tcg_temp_ebb_new_i32(); tcg_gen_andi_i32(t, b, 31); tcg_gen_rotr_i32(d, a, t); @@ -XXX,XX +XXX,XX @@ static void tcg_gen_rotr_mod_i32(TCGv_i32 d, TCGv_i32 a, TCGv_i32 b) static void tcg_gen_rotr_mod_i64(TCGv_i64 d, TCGv_i64 a, TCGv_i64 b) { - TCGv_i64 t = tcg_temp_new_i64(); + TCGv_i64 t = tcg_temp_ebb_new_i64(); tcg_gen_andi_i64(t, b, 63); tcg_gen_rotr_i64(d, a, t); @@ -XXX,XX +XXX,XX @@ void tcg_gen_gvec_rotrv(unsigned vece, uint32_t dofs, uint32_t aofs, static void expand_cmp_i32(uint32_t dofs, uint32_t aofs, uint32_t bofs, uint32_t oprsz, TCGCond cond) { - TCGv_i32 t0 = tcg_temp_new_i32(); - TCGv_i32 t1 = tcg_temp_new_i32(); + TCGv_i32 t0 = tcg_temp_ebb_new_i32(); + TCGv_i32 t1 = tcg_temp_ebb_new_i32(); uint32_t i; for (i = 0; i < oprsz; i += 4) { @@ -XXX,XX +XXX,XX @@ static void expand_cmp_i32(uint32_t dofs, uint32_t aofs, uint32_t bofs, static void expand_cmp_i64(uint32_t dofs, uint32_t aofs, uint32_t bofs, uint32_t oprsz, TCGCond cond) { - TCGv_i64 t0 = tcg_temp_new_i64(); - TCGv_i64 t1 = tcg_temp_new_i64(); + TCGv_i64 t0 = tcg_temp_ebb_new_i64(); + TCGv_i64 t1 = tcg_temp_ebb_new_i64(); uint32_t i; for (i = 0; i < oprsz; i += 8) { @@ -XXX,XX +XXX,XX @@ void tcg_gen_gvec_cmp(TCGCond cond, unsigned vece, uint32_t dofs, static void tcg_gen_bitsel_i64(TCGv_i64 d, TCGv_i64 a, TCGv_i64 b, TCGv_i64 c) { - TCGv_i64 t = tcg_temp_new_i64(); + TCGv_i64 t = tcg_temp_ebb_new_i64(); tcg_gen_and_i64(t, b, a); tcg_gen_andc_i64(d, c, a); diff --git a/tcg/tcg-op.c b/tcg/tcg-op.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/tcg-op.c +++ b/tcg/tcg-op.c @@ -XXX,XX +XXX,XX @@ void tcg_gen_div_i32(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2) if (TCG_TARGET_HAS_div_i32) { tcg_gen_op3_i32(INDEX_op_div_i32, ret, arg1, arg2); } else if (TCG_TARGET_HAS_div2_i32) { - TCGv_i32 t0 = tcg_temp_new_i32(); + TCGv_i32 t0 = tcg_temp_ebb_new_i32(); tcg_gen_sari_i32(t0, arg1, 31); tcg_gen_op5_i32(INDEX_op_div2_i32, ret, t0, arg1, t0, arg2); tcg_temp_free_i32(t0); @@ -XXX,XX +XXX,XX @@ void tcg_gen_rem_i32(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2) if (TCG_TARGET_HAS_rem_i32) { tcg_gen_op3_i32(INDEX_op_rem_i32, ret, arg1, arg2); } else if (TCG_TARGET_HAS_div_i32) { - TCGv_i32 t0 = tcg_temp_new_i32(); + TCGv_i32 t0 = tcg_temp_ebb_new_i32(); tcg_gen_op3_i32(INDEX_op_div_i32, t0, arg1, arg2); tcg_gen_mul_i32(t0, t0, arg2); tcg_gen_sub_i32(ret, arg1, t0); tcg_temp_free_i32(t0); } else if (TCG_TARGET_HAS_div2_i32) { - TCGv_i32 t0 = tcg_temp_new_i32(); + TCGv_i32 t0 = tcg_temp_ebb_new_i32(); tcg_gen_sari_i32(t0, arg1, 31); tcg_gen_op5_i32(INDEX_op_div2_i32, t0, ret, arg1, t0, arg2); tcg_temp_free_i32(t0); @@ -XXX,XX +XXX,XX @@ void tcg_gen_divu_i32(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2) if (TCG_TARGET_HAS_div_i32) { tcg_gen_op3_i32(INDEX_op_divu_i32, ret, arg1, arg2); } else if (TCG_TARGET_HAS_div2_i32) { - TCGv_i32 t0 = tcg_temp_new_i32(); + TCGv_i32 t0 = tcg_temp_ebb_new_i32(); tcg_gen_movi_i32(t0, 0); tcg_gen_op5_i32(INDEX_op_divu2_i32, ret, t0, arg1, t0, arg2); tcg_temp_free_i32(t0); @@ -XXX,XX +XXX,XX @@ void tcg_gen_remu_i32(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2) if (TCG_TARGET_HAS_rem_i32) { tcg_gen_op3_i32(INDEX_op_remu_i32, ret, arg1, arg2); } else if (TCG_TARGET_HAS_div_i32) { - TCGv_i32 t0 = tcg_temp_new_i32(); + TCGv_i32 t0 = tcg_temp_ebb_new_i32(); tcg_gen_op3_i32(INDEX_op_divu_i32, t0, arg1, arg2); tcg_gen_mul_i32(t0, t0, arg2); tcg_gen_sub_i32(ret, arg1, t0); tcg_temp_free_i32(t0); } else if (TCG_TARGET_HAS_div2_i32) { - TCGv_i32 t0 = tcg_temp_new_i32(); + TCGv_i32 t0 = tcg_temp_ebb_new_i32(); tcg_gen_movi_i32(t0, 0); tcg_gen_op5_i32(INDEX_op_divu2_i32, t0, ret, arg1, t0, arg2); tcg_temp_free_i32(t0); @@ -XXX,XX +XXX,XX @@ void tcg_gen_andc_i32(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2) if (TCG_TARGET_HAS_andc_i32) { tcg_gen_op3_i32(INDEX_op_andc_i32, ret, arg1, arg2); } else { - TCGv_i32 t0 = tcg_temp_new_i32(); + TCGv_i32 t0 = tcg_temp_ebb_new_i32(); tcg_gen_not_i32(t0, arg2); tcg_gen_and_i32(ret, arg1, t0); tcg_temp_free_i32(t0); @@ -XXX,XX +XXX,XX @@ void tcg_gen_orc_i32(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2) if (TCG_TARGET_HAS_orc_i32) { tcg_gen_op3_i32(INDEX_op_orc_i32, ret, arg1, arg2); } else { - TCGv_i32 t0 = tcg_temp_new_i32(); + TCGv_i32 t0 = tcg_temp_ebb_new_i32(); tcg_gen_not_i32(t0, arg2); tcg_gen_or_i32(ret, arg1, t0); tcg_temp_free_i32(t0); @@ -XXX,XX +XXX,XX @@ void tcg_gen_clz_i32(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2) if (TCG_TARGET_HAS_clz_i32) { tcg_gen_op3_i32(INDEX_op_clz_i32, ret, arg1, arg2); } else if (TCG_TARGET_HAS_clz_i64) { - TCGv_i64 t1 = tcg_temp_new_i64(); - TCGv_i64 t2 = tcg_temp_new_i64(); + TCGv_i64 t1 = tcg_temp_ebb_new_i64(); + TCGv_i64 t2 = tcg_temp_ebb_new_i64(); tcg_gen_extu_i32_i64(t1, arg1); tcg_gen_extu_i32_i64(t2, arg2); tcg_gen_addi_i64(t2, t2, 32); @@ -XXX,XX +XXX,XX @@ void tcg_gen_ctz_i32(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2) if (TCG_TARGET_HAS_ctz_i32) { tcg_gen_op3_i32(INDEX_op_ctz_i32, ret, arg1, arg2); } else if (TCG_TARGET_HAS_ctz_i64) { - TCGv_i64 t1 = tcg_temp_new_i64(); - TCGv_i64 t2 = tcg_temp_new_i64(); + TCGv_i64 t1 = tcg_temp_ebb_new_i64(); + TCGv_i64 t2 = tcg_temp_ebb_new_i64(); tcg_gen_extu_i32_i64(t1, arg1); tcg_gen_extu_i32_i64(t2, arg2); tcg_gen_ctz_i64(t1, t1, t2); @@ -XXX,XX +XXX,XX @@ void tcg_gen_ctz_i32(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2) || TCG_TARGET_HAS_ctpop_i64 || TCG_TARGET_HAS_clz_i32 || TCG_TARGET_HAS_clz_i64) { - TCGv_i32 z, t = tcg_temp_new_i32(); + TCGv_i32 z, t = tcg_temp_ebb_new_i32(); if (TCG_TARGET_HAS_ctpop_i32 || TCG_TARGET_HAS_ctpop_i64) { tcg_gen_subi_i32(t, arg1, 1); @@ -XXX,XX +XXX,XX @@ void tcg_gen_ctzi_i32(TCGv_i32 ret, TCGv_i32 arg1, uint32_t arg2) { if (!TCG_TARGET_HAS_ctz_i32 && TCG_TARGET_HAS_ctpop_i32 && arg2 == 32) { /* This equivalence has the advantage of not requiring a fixup. */ - TCGv_i32 t = tcg_temp_new_i32(); + TCGv_i32 t = tcg_temp_ebb_new_i32(); tcg_gen_subi_i32(t, arg1, 1); tcg_gen_andc_i32(t, t, arg1); tcg_gen_ctpop_i32(ret, t); @@ -XXX,XX +XXX,XX @@ void tcg_gen_ctzi_i32(TCGv_i32 ret, TCGv_i32 arg1, uint32_t arg2) void tcg_gen_clrsb_i32(TCGv_i32 ret, TCGv_i32 arg) { if (TCG_TARGET_HAS_clz_i32) { - TCGv_i32 t = tcg_temp_new_i32(); + TCGv_i32 t = tcg_temp_ebb_new_i32(); tcg_gen_sari_i32(t, arg, 31); tcg_gen_xor_i32(t, t, arg); tcg_gen_clzi_i32(t, t, 32); @@ -XXX,XX +XXX,XX @@ void tcg_gen_ctpop_i32(TCGv_i32 ret, TCGv_i32 arg1) if (TCG_TARGET_HAS_ctpop_i32) { tcg_gen_op2_i32(INDEX_op_ctpop_i32, ret, arg1); } else if (TCG_TARGET_HAS_ctpop_i64) { - TCGv_i64 t = tcg_temp_new_i64(); + TCGv_i64 t = tcg_temp_ebb_new_i64(); tcg_gen_extu_i32_i64(t, arg1); tcg_gen_ctpop_i64(t, t); tcg_gen_extrl_i64_i32(ret, t); @@ -XXX,XX +XXX,XX @@ void tcg_gen_rotl_i32(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2) } else { TCGv_i32 t0, t1; - t0 = tcg_temp_new_i32(); - t1 = tcg_temp_new_i32(); + t0 = tcg_temp_ebb_new_i32(); + t1 = tcg_temp_ebb_new_i32(); tcg_gen_shl_i32(t0, arg1, arg2); tcg_gen_subfi_i32(t1, 32, arg2); tcg_gen_shr_i32(t1, arg1, t1); @@ -XXX,XX +XXX,XX @@ void tcg_gen_rotli_i32(TCGv_i32 ret, TCGv_i32 arg1, int32_t arg2) tcg_gen_rotl_i32(ret, arg1, tcg_constant_i32(arg2)); } else { TCGv_i32 t0, t1; - t0 = tcg_temp_new_i32(); - t1 = tcg_temp_new_i32(); + t0 = tcg_temp_ebb_new_i32(); + t1 = tcg_temp_ebb_new_i32(); tcg_gen_shli_i32(t0, arg1, arg2); tcg_gen_shri_i32(t1, arg1, 32 - arg2); tcg_gen_or_i32(ret, t0, t1); @@ -XXX,XX +XXX,XX @@ void tcg_gen_rotr_i32(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2) } else { TCGv_i32 t0, t1; - t0 = tcg_temp_new_i32(); - t1 = tcg_temp_new_i32(); + t0 = tcg_temp_ebb_new_i32(); + t1 = tcg_temp_ebb_new_i32(); tcg_gen_shr_i32(t0, arg1, arg2); tcg_gen_subfi_i32(t1, 32, arg2); tcg_gen_shl_i32(t1, arg1, t1); @@ -XXX,XX +XXX,XX @@ void tcg_gen_deposit_i32(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2, return; } - t1 = tcg_temp_new_i32(); + t1 = tcg_temp_ebb_new_i32(); if (TCG_TARGET_HAS_extract2_i32) { if (ofs + len == 32) { @@ -XXX,XX +XXX,XX @@ void tcg_gen_extract2_i32(TCGv_i32 ret, TCGv_i32 al, TCGv_i32 ah, } else if (TCG_TARGET_HAS_extract2_i32) { tcg_gen_op4i_i32(INDEX_op_extract2_i32, ret, al, ah, ofs); } else { - TCGv_i32 t0 = tcg_temp_new_i32(); + TCGv_i32 t0 = tcg_temp_ebb_new_i32(); tcg_gen_shri_i32(t0, al, ofs); tcg_gen_deposit_i32(ret, t0, ah, 32 - ofs, ofs); tcg_temp_free_i32(t0); @@ -XXX,XX +XXX,XX @@ void tcg_gen_movcond_i32(TCGCond cond, TCGv_i32 ret, TCGv_i32 c1, } else if (TCG_TARGET_HAS_movcond_i32) { tcg_gen_op6i_i32(INDEX_op_movcond_i32, ret, c1, c2, v1, v2, cond); } else { - TCGv_i32 t0 = tcg_temp_new_i32(); - TCGv_i32 t1 = tcg_temp_new_i32(); + TCGv_i32 t0 = tcg_temp_ebb_new_i32(); + TCGv_i32 t1 = tcg_temp_ebb_new_i32(); tcg_gen_setcond_i32(cond, t0, c1, c2); tcg_gen_neg_i32(t0, t0); tcg_gen_and_i32(t1, v1, t0); @@ -XXX,XX +XXX,XX @@ void tcg_gen_add2_i32(TCGv_i32 rl, TCGv_i32 rh, TCGv_i32 al, if (TCG_TARGET_HAS_add2_i32) { tcg_gen_op6_i32(INDEX_op_add2_i32, rl, rh, al, ah, bl, bh); } else { - TCGv_i64 t0 = tcg_temp_new_i64(); - TCGv_i64 t1 = tcg_temp_new_i64(); + TCGv_i64 t0 = tcg_temp_ebb_new_i64(); + TCGv_i64 t1 = tcg_temp_ebb_new_i64(); tcg_gen_concat_i32_i64(t0, al, ah); tcg_gen_concat_i32_i64(t1, bl, bh); tcg_gen_add_i64(t0, t0, t1); @@ -XXX,XX +XXX,XX @@ void tcg_gen_sub2_i32(TCGv_i32 rl, TCGv_i32 rh, TCGv_i32 al, if (TCG_TARGET_HAS_sub2_i32) { tcg_gen_op6_i32(INDEX_op_sub2_i32, rl, rh, al, ah, bl, bh); } else { - TCGv_i64 t0 = tcg_temp_new_i64(); - TCGv_i64 t1 = tcg_temp_new_i64(); + TCGv_i64 t0 = tcg_temp_ebb_new_i64(); + TCGv_i64 t1 = tcg_temp_ebb_new_i64(); tcg_gen_concat_i32_i64(t0, al, ah); tcg_gen_concat_i32_i64(t1, bl, bh); tcg_gen_sub_i64(t0, t0, t1); @@ -XXX,XX +XXX,XX @@ void tcg_gen_mulu2_i32(TCGv_i32 rl, TCGv_i32 rh, TCGv_i32 arg1, TCGv_i32 arg2) if (TCG_TARGET_HAS_mulu2_i32) { tcg_gen_op4_i32(INDEX_op_mulu2_i32, rl, rh, arg1, arg2); } else if (TCG_TARGET_HAS_muluh_i32) { - TCGv_i32 t = tcg_temp_new_i32(); + TCGv_i32 t = tcg_temp_ebb_new_i32(); tcg_gen_op3_i32(INDEX_op_mul_i32, t, arg1, arg2); tcg_gen_op3_i32(INDEX_op_muluh_i32, rh, arg1, arg2); tcg_gen_mov_i32(rl, t); tcg_temp_free_i32(t); } else if (TCG_TARGET_REG_BITS == 64) { - TCGv_i64 t0 = tcg_temp_new_i64(); - TCGv_i64 t1 = tcg_temp_new_i64(); + TCGv_i64 t0 = tcg_temp_ebb_new_i64(); + TCGv_i64 t1 = tcg_temp_ebb_new_i64(); tcg_gen_extu_i32_i64(t0, arg1); tcg_gen_extu_i32_i64(t1, arg2); tcg_gen_mul_i64(t0, t0, t1); @@ -XXX,XX +XXX,XX @@ void tcg_gen_muls2_i32(TCGv_i32 rl, TCGv_i32 rh, TCGv_i32 arg1, TCGv_i32 arg2) if (TCG_TARGET_HAS_muls2_i32) { tcg_gen_op4_i32(INDEX_op_muls2_i32, rl, rh, arg1, arg2); } else if (TCG_TARGET_HAS_mulsh_i32) { - TCGv_i32 t = tcg_temp_new_i32(); + TCGv_i32 t = tcg_temp_ebb_new_i32(); tcg_gen_op3_i32(INDEX_op_mul_i32, t, arg1, arg2); tcg_gen_op3_i32(INDEX_op_mulsh_i32, rh, arg1, arg2); tcg_gen_mov_i32(rl, t); tcg_temp_free_i32(t); } else if (TCG_TARGET_REG_BITS == 32) { - TCGv_i32 t0 = tcg_temp_new_i32(); - TCGv_i32 t1 = tcg_temp_new_i32(); - TCGv_i32 t2 = tcg_temp_new_i32(); - TCGv_i32 t3 = tcg_temp_new_i32(); + TCGv_i32 t0 = tcg_temp_ebb_new_i32(); + TCGv_i32 t1 = tcg_temp_ebb_new_i32(); + TCGv_i32 t2 = tcg_temp_ebb_new_i32(); + TCGv_i32 t3 = tcg_temp_ebb_new_i32(); tcg_gen_mulu2_i32(t0, t1, arg1, arg2); /* Adjust for negative inputs. */ tcg_gen_sari_i32(t2, arg1, 31); @@ -XXX,XX +XXX,XX @@ void tcg_gen_muls2_i32(TCGv_i32 rl, TCGv_i32 rh, TCGv_i32 arg1, TCGv_i32 arg2) tcg_temp_free_i32(t2); tcg_temp_free_i32(t3); } else { - TCGv_i64 t0 = tcg_temp_new_i64(); - TCGv_i64 t1 = tcg_temp_new_i64(); + TCGv_i64 t0 = tcg_temp_ebb_new_i64(); + TCGv_i64 t1 = tcg_temp_ebb_new_i64(); tcg_gen_ext_i32_i64(t0, arg1); tcg_gen_ext_i32_i64(t1, arg2); tcg_gen_mul_i64(t0, t0, t1); @@ -XXX,XX +XXX,XX @@ void tcg_gen_muls2_i32(TCGv_i32 rl, TCGv_i32 rh, TCGv_i32 arg1, TCGv_i32 arg2) void tcg_gen_mulsu2_i32(TCGv_i32 rl, TCGv_i32 rh, TCGv_i32 arg1, TCGv_i32 arg2) { if (TCG_TARGET_REG_BITS == 32) { - TCGv_i32 t0 = tcg_temp_new_i32(); - TCGv_i32 t1 = tcg_temp_new_i32(); - TCGv_i32 t2 = tcg_temp_new_i32(); + TCGv_i32 t0 = tcg_temp_ebb_new_i32(); + TCGv_i32 t1 = tcg_temp_ebb_new_i32(); + TCGv_i32 t2 = tcg_temp_ebb_new_i32(); tcg_gen_mulu2_i32(t0, t1, arg1, arg2); /* Adjust for negative input for the signed arg1. */ tcg_gen_sari_i32(t2, arg1, 31); @@ -XXX,XX +XXX,XX @@ void tcg_gen_mulsu2_i32(TCGv_i32 rl, TCGv_i32 rh, TCGv_i32 arg1, TCGv_i32 arg2) tcg_temp_free_i32(t1); tcg_temp_free_i32(t2); } else { - TCGv_i64 t0 = tcg_temp_new_i64(); - TCGv_i64 t1 = tcg_temp_new_i64(); + TCGv_i64 t0 = tcg_temp_ebb_new_i64(); + TCGv_i64 t1 = tcg_temp_ebb_new_i64(); tcg_gen_ext_i32_i64(t0, arg1); tcg_gen_extu_i32_i64(t1, arg2); tcg_gen_mul_i64(t0, t0, t1); @@ -XXX,XX +XXX,XX @@ void tcg_gen_bswap16_i32(TCGv_i32 ret, TCGv_i32 arg, int flags) if (TCG_TARGET_HAS_bswap16_i32) { tcg_gen_op3i_i32(INDEX_op_bswap16_i32, ret, arg, flags); } else { - TCGv_i32 t0 = tcg_temp_new_i32(); - TCGv_i32 t1 = tcg_temp_new_i32(); + TCGv_i32 t0 = tcg_temp_ebb_new_i32(); + TCGv_i32 t1 = tcg_temp_ebb_new_i32(); tcg_gen_shri_i32(t0, arg, 8); if (!(flags & TCG_BSWAP_IZ)) { @@ -XXX,XX +XXX,XX @@ void tcg_gen_bswap32_i32(TCGv_i32 ret, TCGv_i32 arg) if (TCG_TARGET_HAS_bswap32_i32) { tcg_gen_op3i_i32(INDEX_op_bswap32_i32, ret, arg, 0); } else { - TCGv_i32 t0 = tcg_temp_new_i32(); - TCGv_i32 t1 = tcg_temp_new_i32(); + TCGv_i32 t0 = tcg_temp_ebb_new_i32(); + TCGv_i32 t1 = tcg_temp_ebb_new_i32(); TCGv_i32 t2 = tcg_constant_i32(0x00ff00ff); /* arg = abcd */ @@ -XXX,XX +XXX,XX @@ void tcg_gen_umax_i32(TCGv_i32 ret, TCGv_i32 a, TCGv_i32 b) void tcg_gen_abs_i32(TCGv_i32 ret, TCGv_i32 a) { - TCGv_i32 t = tcg_temp_new_i32(); + TCGv_i32 t = tcg_temp_ebb_new_i32(); tcg_gen_sari_i32(t, a, 31); tcg_gen_xor_i32(ret, a, t); @@ -XXX,XX +XXX,XX @@ void tcg_gen_mul_i64(TCGv_i64 ret, TCGv_i64 arg1, TCGv_i64 arg2) TCGv_i64 t0; TCGv_i32 t1; - t0 = tcg_temp_new_i64(); - t1 = tcg_temp_new_i32(); + t0 = tcg_temp_ebb_new_i64(); + t1 = tcg_temp_ebb_new_i32(); tcg_gen_mulu2_i32(TCGV_LOW(t0), TCGV_HIGH(t0), TCGV_LOW(arg1), TCGV_LOW(arg2)); @@ -XXX,XX +XXX,XX @@ static inline void tcg_gen_shifti_i64(TCGv_i64 ret, TCGv_i64 arg1, tcg_gen_extract2_i32(TCGV_HIGH(ret), TCGV_LOW(arg1), TCGV_HIGH(arg1), 32 - c); } else { - TCGv_i32 t0 = tcg_temp_new_i32(); + TCGv_i32 t0 = tcg_temp_ebb_new_i32(); tcg_gen_shri_i32(t0, TCGV_LOW(arg1), 32 - c); tcg_gen_deposit_i32(TCGV_HIGH(ret), t0, TCGV_HIGH(arg1), c, 32 - c); @@ -XXX,XX +XXX,XX @@ void tcg_gen_div_i64(TCGv_i64 ret, TCGv_i64 arg1, TCGv_i64 arg2) if (TCG_TARGET_HAS_div_i64) { tcg_gen_op3_i64(INDEX_op_div_i64, ret, arg1, arg2); } else if (TCG_TARGET_HAS_div2_i64) { - TCGv_i64 t0 = tcg_temp_new_i64(); + TCGv_i64 t0 = tcg_temp_ebb_new_i64(); tcg_gen_sari_i64(t0, arg1, 63); tcg_gen_op5_i64(INDEX_op_div2_i64, ret, t0, arg1, t0, arg2); tcg_temp_free_i64(t0); @@ -XXX,XX +XXX,XX @@ void tcg_gen_rem_i64(TCGv_i64 ret, TCGv_i64 arg1, TCGv_i64 arg2) if (TCG_TARGET_HAS_rem_i64) { tcg_gen_op3_i64(INDEX_op_rem_i64, ret, arg1, arg2); } else if (TCG_TARGET_HAS_div_i64) { - TCGv_i64 t0 = tcg_temp_new_i64(); + TCGv_i64 t0 = tcg_temp_ebb_new_i64(); tcg_gen_op3_i64(INDEX_op_div_i64, t0, arg1, arg2); tcg_gen_mul_i64(t0, t0, arg2); tcg_gen_sub_i64(ret, arg1, t0); tcg_temp_free_i64(t0); } else if (TCG_TARGET_HAS_div2_i64) { - TCGv_i64 t0 = tcg_temp_new_i64(); + TCGv_i64 t0 = tcg_temp_ebb_new_i64(); tcg_gen_sari_i64(t0, arg1, 63); tcg_gen_op5_i64(INDEX_op_div2_i64, t0, ret, arg1, t0, arg2); tcg_temp_free_i64(t0); @@ -XXX,XX +XXX,XX @@ void tcg_gen_divu_i64(TCGv_i64 ret, TCGv_i64 arg1, TCGv_i64 arg2) if (TCG_TARGET_HAS_div_i64) { tcg_gen_op3_i64(INDEX_op_divu_i64, ret, arg1, arg2); } else if (TCG_TARGET_HAS_div2_i64) { - TCGv_i64 t0 = tcg_temp_new_i64(); + TCGv_i64 t0 = tcg_temp_ebb_new_i64(); tcg_gen_movi_i64(t0, 0); tcg_gen_op5_i64(INDEX_op_divu2_i64, ret, t0, arg1, t0, arg2); tcg_temp_free_i64(t0); @@ -XXX,XX +XXX,XX @@ void tcg_gen_remu_i64(TCGv_i64 ret, TCGv_i64 arg1, TCGv_i64 arg2) if (TCG_TARGET_HAS_rem_i64) { tcg_gen_op3_i64(INDEX_op_remu_i64, ret, arg1, arg2); } else if (TCG_TARGET_HAS_div_i64) { - TCGv_i64 t0 = tcg_temp_new_i64(); + TCGv_i64 t0 = tcg_temp_ebb_new_i64(); tcg_gen_op3_i64(INDEX_op_divu_i64, t0, arg1, arg2); tcg_gen_mul_i64(t0, t0, arg2); tcg_gen_sub_i64(ret, arg1, t0); tcg_temp_free_i64(t0); } else if (TCG_TARGET_HAS_div2_i64) { - TCGv_i64 t0 = tcg_temp_new_i64(); + TCGv_i64 t0 = tcg_temp_ebb_new_i64(); tcg_gen_movi_i64(t0, 0); tcg_gen_op5_i64(INDEX_op_divu2_i64, t0, ret, arg1, t0, arg2); tcg_temp_free_i64(t0); @@ -XXX,XX +XXX,XX @@ void tcg_gen_bswap16_i64(TCGv_i64 ret, TCGv_i64 arg, int flags) } else if (TCG_TARGET_HAS_bswap16_i64) { tcg_gen_op3i_i64(INDEX_op_bswap16_i64, ret, arg, flags); } else { - TCGv_i64 t0 = tcg_temp_new_i64(); - TCGv_i64 t1 = tcg_temp_new_i64(); + TCGv_i64 t0 = tcg_temp_ebb_new_i64(); + TCGv_i64 t1 = tcg_temp_ebb_new_i64(); tcg_gen_shri_i64(t0, arg, 8); if (!(flags & TCG_BSWAP_IZ)) { @@ -XXX,XX +XXX,XX @@ void tcg_gen_bswap32_i64(TCGv_i64 ret, TCGv_i64 arg, int flags) } else if (TCG_TARGET_HAS_bswap32_i64) { tcg_gen_op3i_i64(INDEX_op_bswap32_i64, ret, arg, flags); } else { - TCGv_i64 t0 = tcg_temp_new_i64(); - TCGv_i64 t1 = tcg_temp_new_i64(); + TCGv_i64 t0 = tcg_temp_ebb_new_i64(); + TCGv_i64 t1 = tcg_temp_ebb_new_i64(); TCGv_i64 t2 = tcg_constant_i64(0x00ff00ff); /* arg = xxxxabcd */ @@ -XXX,XX +XXX,XX @@ void tcg_gen_bswap64_i64(TCGv_i64 ret, TCGv_i64 arg) { if (TCG_TARGET_REG_BITS == 32) { TCGv_i32 t0, t1; - t0 = tcg_temp_new_i32(); - t1 = tcg_temp_new_i32(); + t0 = tcg_temp_ebb_new_i32(); + t1 = tcg_temp_ebb_new_i32(); tcg_gen_bswap32_i32(t0, TCGV_LOW(arg)); tcg_gen_bswap32_i32(t1, TCGV_HIGH(arg)); @@ -XXX,XX +XXX,XX @@ void tcg_gen_bswap64_i64(TCGv_i64 ret, TCGv_i64 arg) } else if (TCG_TARGET_HAS_bswap64_i64) { tcg_gen_op3i_i64(INDEX_op_bswap64_i64, ret, arg, 0); } else { - TCGv_i64 t0 = tcg_temp_new_i64(); - TCGv_i64 t1 = tcg_temp_new_i64(); - TCGv_i64 t2 = tcg_temp_new_i64(); + TCGv_i64 t0 = tcg_temp_ebb_new_i64(); + TCGv_i64 t1 = tcg_temp_ebb_new_i64(); + TCGv_i64 t2 = tcg_temp_ebb_new_i64(); /* arg = abcdefgh */ tcg_gen_movi_i64(t2, 0x00ff00ff00ff00ffull); @@ -XXX,XX +XXX,XX @@ void tcg_gen_bswap64_i64(TCGv_i64 ret, TCGv_i64 arg) void tcg_gen_hswap_i64(TCGv_i64 ret, TCGv_i64 arg) { uint64_t m = 0x0000ffff0000ffffull; - TCGv_i64 t0 = tcg_temp_new_i64(); - TCGv_i64 t1 = tcg_temp_new_i64(); + TCGv_i64 t0 = tcg_temp_ebb_new_i64(); + TCGv_i64 t1 = tcg_temp_ebb_new_i64(); /* See include/qemu/bitops.h, hswap64. */ tcg_gen_rotli_i64(t1, arg, 32); @@ -XXX,XX +XXX,XX @@ void tcg_gen_andc_i64(TCGv_i64 ret, TCGv_i64 arg1, TCGv_i64 arg2) } else if (TCG_TARGET_HAS_andc_i64) { tcg_gen_op3_i64(INDEX_op_andc_i64, ret, arg1, arg2); } else { - TCGv_i64 t0 = tcg_temp_new_i64(); + TCGv_i64 t0 = tcg_temp_ebb_new_i64(); tcg_gen_not_i64(t0, arg2); tcg_gen_and_i64(ret, arg1, t0); tcg_temp_free_i64(t0); @@ -XXX,XX +XXX,XX @@ void tcg_gen_orc_i64(TCGv_i64 ret, TCGv_i64 arg1, TCGv_i64 arg2) } else if (TCG_TARGET_HAS_orc_i64) { tcg_gen_op3_i64(INDEX_op_orc_i64, ret, arg1, arg2); } else { - TCGv_i64 t0 = tcg_temp_new_i64(); + TCGv_i64 t0 = tcg_temp_ebb_new_i64(); tcg_gen_not_i64(t0, arg2); tcg_gen_or_i64(ret, arg1, t0); tcg_temp_free_i64(t0); @@ -XXX,XX +XXX,XX @@ void tcg_gen_clzi_i64(TCGv_i64 ret, TCGv_i64 arg1, uint64_t arg2) if (TCG_TARGET_REG_BITS == 32 && TCG_TARGET_HAS_clz_i32 && arg2 <= 0xffffffffu) { - TCGv_i32 t = tcg_temp_new_i32(); + TCGv_i32 t = tcg_temp_ebb_new_i32(); tcg_gen_clzi_i32(t, TCGV_LOW(arg1), arg2 - 32); tcg_gen_addi_i32(t, t, 32); tcg_gen_clz_i32(TCGV_LOW(ret), TCGV_HIGH(arg1), t); @@ -XXX,XX +XXX,XX @@ void tcg_gen_ctz_i64(TCGv_i64 ret, TCGv_i64 arg1, TCGv_i64 arg2) if (TCG_TARGET_HAS_ctz_i64) { tcg_gen_op3_i64(INDEX_op_ctz_i64, ret, arg1, arg2); } else if (TCG_TARGET_HAS_ctpop_i64 || TCG_TARGET_HAS_clz_i64) { - TCGv_i64 z, t = tcg_temp_new_i64(); + TCGv_i64 z, t = tcg_temp_ebb_new_i64(); if (TCG_TARGET_HAS_ctpop_i64) { tcg_gen_subi_i64(t, arg1, 1); @@ -XXX,XX +XXX,XX @@ void tcg_gen_ctzi_i64(TCGv_i64 ret, TCGv_i64 arg1, uint64_t arg2) if (TCG_TARGET_REG_BITS == 32 && TCG_TARGET_HAS_ctz_i32 && arg2 <= 0xffffffffu) { - TCGv_i32 t32 = tcg_temp_new_i32(); + TCGv_i32 t32 = tcg_temp_ebb_new_i32(); tcg_gen_ctzi_i32(t32, TCGV_HIGH(arg1), arg2 - 32); tcg_gen_addi_i32(t32, t32, 32); tcg_gen_ctz_i32(TCGV_LOW(ret), TCGV_LOW(arg1), t32); @@ -XXX,XX +XXX,XX @@ void tcg_gen_ctzi_i64(TCGv_i64 ret, TCGv_i64 arg1, uint64_t arg2) && TCG_TARGET_HAS_ctpop_i64 && arg2 == 64) { /* This equivalence has the advantage of not requiring a fixup. */ - TCGv_i64 t = tcg_temp_new_i64(); + TCGv_i64 t = tcg_temp_ebb_new_i64(); tcg_gen_subi_i64(t, arg1, 1); tcg_gen_andc_i64(t, t, arg1); tcg_gen_ctpop_i64(ret, t); @@ -XXX,XX +XXX,XX @@ void tcg_gen_ctzi_i64(TCGv_i64 ret, TCGv_i64 arg1, uint64_t arg2) void tcg_gen_clrsb_i64(TCGv_i64 ret, TCGv_i64 arg) { if (TCG_TARGET_HAS_clz_i64 || TCG_TARGET_HAS_clz_i32) { - TCGv_i64 t = tcg_temp_new_i64(); + TCGv_i64 t = tcg_temp_ebb_new_i64(); tcg_gen_sari_i64(t, arg, 63); tcg_gen_xor_i64(t, t, arg); tcg_gen_clzi_i64(t, t, 64); @@ -XXX,XX +XXX,XX @@ void tcg_gen_rotl_i64(TCGv_i64 ret, TCGv_i64 arg1, TCGv_i64 arg2) tcg_gen_op3_i64(INDEX_op_rotl_i64, ret, arg1, arg2); } else { TCGv_i64 t0, t1; - t0 = tcg_temp_new_i64(); - t1 = tcg_temp_new_i64(); + t0 = tcg_temp_ebb_new_i64(); + t1 = tcg_temp_ebb_new_i64(); tcg_gen_shl_i64(t0, arg1, arg2); tcg_gen_subfi_i64(t1, 64, arg2); tcg_gen_shr_i64(t1, arg1, t1); @@ -XXX,XX +XXX,XX @@ void tcg_gen_rotli_i64(TCGv_i64 ret, TCGv_i64 arg1, int64_t arg2) tcg_gen_rotl_i64(ret, arg1, tcg_constant_i64(arg2)); } else { TCGv_i64 t0, t1; - t0 = tcg_temp_new_i64(); - t1 = tcg_temp_new_i64(); + t0 = tcg_temp_ebb_new_i64(); + t1 = tcg_temp_ebb_new_i64(); tcg_gen_shli_i64(t0, arg1, arg2); tcg_gen_shri_i64(t1, arg1, 64 - arg2); tcg_gen_or_i64(ret, t0, t1); @@ -XXX,XX +XXX,XX @@ void tcg_gen_rotr_i64(TCGv_i64 ret, TCGv_i64 arg1, TCGv_i64 arg2) tcg_gen_op3_i64(INDEX_op_rotr_i64, ret, arg1, arg2); } else { TCGv_i64 t0, t1; - t0 = tcg_temp_new_i64(); - t1 = tcg_temp_new_i64(); + t0 = tcg_temp_ebb_new_i64(); + t1 = tcg_temp_ebb_new_i64(); tcg_gen_shr_i64(t0, arg1, arg2); tcg_gen_subfi_i64(t1, 64, arg2); tcg_gen_shl_i64(t1, arg1, t1); @@ -XXX,XX +XXX,XX @@ void tcg_gen_deposit_i64(TCGv_i64 ret, TCGv_i64 arg1, TCGv_i64 arg2, } } - t1 = tcg_temp_new_i64(); + t1 = tcg_temp_ebb_new_i64(); if (TCG_TARGET_HAS_extract2_i64) { if (ofs + len == 64) { @@ -XXX,XX +XXX,XX @@ void tcg_gen_sextract_i64(TCGv_i64 ret, TCGv_i64 arg, tcg_gen_sextract_i32(TCGV_HIGH(ret), TCGV_HIGH(arg), 0, len - 32); return; } else if (len > 32) { - TCGv_i32 t = tcg_temp_new_i32(); + TCGv_i32 t = tcg_temp_ebb_new_i32(); /* Extract the bits for the high word normally. */ tcg_gen_sextract_i32(t, TCGV_HIGH(arg), ofs + 32, len - 32); /* Shift the field down for the low part. */ @@ -XXX,XX +XXX,XX @@ void tcg_gen_extract2_i64(TCGv_i64 ret, TCGv_i64 al, TCGv_i64 ah, } else if (TCG_TARGET_HAS_extract2_i64) { tcg_gen_op4i_i64(INDEX_op_extract2_i64, ret, al, ah, ofs); } else { - TCGv_i64 t0 = tcg_temp_new_i64(); + TCGv_i64 t0 = tcg_temp_ebb_new_i64(); tcg_gen_shri_i64(t0, al, ofs); tcg_gen_deposit_i64(ret, t0, ah, 64 - ofs, ofs); tcg_temp_free_i64(t0); @@ -XXX,XX +XXX,XX @@ void tcg_gen_movcond_i64(TCGCond cond, TCGv_i64 ret, TCGv_i64 c1, } else if (cond == TCG_COND_NEVER) { tcg_gen_mov_i64(ret, v2); } else if (TCG_TARGET_REG_BITS == 32) { - TCGv_i32 t0 = tcg_temp_new_i32(); - TCGv_i32 t1 = tcg_temp_new_i32(); + TCGv_i32 t0 = tcg_temp_ebb_new_i32(); + TCGv_i32 t1 = tcg_temp_ebb_new_i32(); tcg_gen_op6i_i32(INDEX_op_setcond2_i32, t0, TCGV_LOW(c1), TCGV_HIGH(c1), TCGV_LOW(c2), TCGV_HIGH(c2), cond); @@ -XXX,XX +XXX,XX @@ void tcg_gen_movcond_i64(TCGCond cond, TCGv_i64 ret, TCGv_i64 c1, } else if (TCG_TARGET_HAS_movcond_i64) { tcg_gen_op6i_i64(INDEX_op_movcond_i64, ret, c1, c2, v1, v2, cond); } else { - TCGv_i64 t0 = tcg_temp_new_i64(); - TCGv_i64 t1 = tcg_temp_new_i64(); + TCGv_i64 t0 = tcg_temp_ebb_new_i64(); + TCGv_i64 t1 = tcg_temp_ebb_new_i64(); tcg_gen_setcond_i64(cond, t0, c1, c2); tcg_gen_neg_i64(t0, t0); tcg_gen_and_i64(t1, v1, t0); @@ -XXX,XX +XXX,XX @@ void tcg_gen_add2_i64(TCGv_i64 rl, TCGv_i64 rh, TCGv_i64 al, if (TCG_TARGET_HAS_add2_i64) { tcg_gen_op6_i64(INDEX_op_add2_i64, rl, rh, al, ah, bl, bh); } else { - TCGv_i64 t0 = tcg_temp_new_i64(); - TCGv_i64 t1 = tcg_temp_new_i64(); + TCGv_i64 t0 = tcg_temp_ebb_new_i64(); + TCGv_i64 t1 = tcg_temp_ebb_new_i64(); tcg_gen_add_i64(t0, al, bl); tcg_gen_setcond_i64(TCG_COND_LTU, t1, t0, al); tcg_gen_add_i64(rh, ah, bh); @@ -XXX,XX +XXX,XX @@ void tcg_gen_sub2_i64(TCGv_i64 rl, TCGv_i64 rh, TCGv_i64 al, if (TCG_TARGET_HAS_sub2_i64) { tcg_gen_op6_i64(INDEX_op_sub2_i64, rl, rh, al, ah, bl, bh); } else { - TCGv_i64 t0 = tcg_temp_new_i64(); - TCGv_i64 t1 = tcg_temp_new_i64(); + TCGv_i64 t0 = tcg_temp_ebb_new_i64(); + TCGv_i64 t1 = tcg_temp_ebb_new_i64(); tcg_gen_sub_i64(t0, al, bl); tcg_gen_setcond_i64(TCG_COND_LTU, t1, al, bl); tcg_gen_sub_i64(rh, ah, bh); @@ -XXX,XX +XXX,XX @@ void tcg_gen_mulu2_i64(TCGv_i64 rl, TCGv_i64 rh, TCGv_i64 arg1, TCGv_i64 arg2) if (TCG_TARGET_HAS_mulu2_i64) { tcg_gen_op4_i64(INDEX_op_mulu2_i64, rl, rh, arg1, arg2); } else if (TCG_TARGET_HAS_muluh_i64) { - TCGv_i64 t = tcg_temp_new_i64(); + TCGv_i64 t = tcg_temp_ebb_new_i64(); tcg_gen_op3_i64(INDEX_op_mul_i64, t, arg1, arg2); tcg_gen_op3_i64(INDEX_op_muluh_i64, rh, arg1, arg2); tcg_gen_mov_i64(rl, t); tcg_temp_free_i64(t); } else { - TCGv_i64 t0 = tcg_temp_new_i64(); + TCGv_i64 t0 = tcg_temp_ebb_new_i64(); tcg_gen_mul_i64(t0, arg1, arg2); gen_helper_muluh_i64(rh, arg1, arg2); tcg_gen_mov_i64(rl, t0); @@ -XXX,XX +XXX,XX @@ void tcg_gen_muls2_i64(TCGv_i64 rl, TCGv_i64 rh, TCGv_i64 arg1, TCGv_i64 arg2) if (TCG_TARGET_HAS_muls2_i64) { tcg_gen_op4_i64(INDEX_op_muls2_i64, rl, rh, arg1, arg2); } else if (TCG_TARGET_HAS_mulsh_i64) { - TCGv_i64 t = tcg_temp_new_i64(); + TCGv_i64 t = tcg_temp_ebb_new_i64(); tcg_gen_op3_i64(INDEX_op_mul_i64, t, arg1, arg2); tcg_gen_op3_i64(INDEX_op_mulsh_i64, rh, arg1, arg2); tcg_gen_mov_i64(rl, t); tcg_temp_free_i64(t); } else if (TCG_TARGET_HAS_mulu2_i64 || TCG_TARGET_HAS_muluh_i64) { - TCGv_i64 t0 = tcg_temp_new_i64(); - TCGv_i64 t1 = tcg_temp_new_i64(); - TCGv_i64 t2 = tcg_temp_new_i64(); - TCGv_i64 t3 = tcg_temp_new_i64(); + TCGv_i64 t0 = tcg_temp_ebb_new_i64(); + TCGv_i64 t1 = tcg_temp_ebb_new_i64(); + TCGv_i64 t2 = tcg_temp_ebb_new_i64(); + TCGv_i64 t3 = tcg_temp_ebb_new_i64(); tcg_gen_mulu2_i64(t0, t1, arg1, arg2); /* Adjust for negative inputs. */ tcg_gen_sari_i64(t2, arg1, 63); @@ -XXX,XX +XXX,XX @@ void tcg_gen_muls2_i64(TCGv_i64 rl, TCGv_i64 rh, TCGv_i64 arg1, TCGv_i64 arg2) tcg_temp_free_i64(t2); tcg_temp_free_i64(t3); } else { - TCGv_i64 t0 = tcg_temp_new_i64(); + TCGv_i64 t0 = tcg_temp_ebb_new_i64(); tcg_gen_mul_i64(t0, arg1, arg2); gen_helper_mulsh_i64(rh, arg1, arg2); tcg_gen_mov_i64(rl, t0); @@ -XXX,XX +XXX,XX @@ void tcg_gen_muls2_i64(TCGv_i64 rl, TCGv_i64 rh, TCGv_i64 arg1, TCGv_i64 arg2) void tcg_gen_mulsu2_i64(TCGv_i64 rl, TCGv_i64 rh, TCGv_i64 arg1, TCGv_i64 arg2) { - TCGv_i64 t0 = tcg_temp_new_i64(); - TCGv_i64 t1 = tcg_temp_new_i64(); - TCGv_i64 t2 = tcg_temp_new_i64(); + TCGv_i64 t0 = tcg_temp_ebb_new_i64(); + TCGv_i64 t1 = tcg_temp_ebb_new_i64(); + TCGv_i64 t2 = tcg_temp_ebb_new_i64(); tcg_gen_mulu2_i64(t0, t1, arg1, arg2); /* Adjust for negative input for the signed arg1. */ tcg_gen_sari_i64(t2, arg1, 63); @@ -XXX,XX +XXX,XX @@ void tcg_gen_umax_i64(TCGv_i64 ret, TCGv_i64 a, TCGv_i64 b) void tcg_gen_abs_i64(TCGv_i64 ret, TCGv_i64 a) { - TCGv_i64 t = tcg_temp_new_i64(); + TCGv_i64 t = tcg_temp_ebb_new_i64(); tcg_gen_sari_i64(t, a, 63); tcg_gen_xor_i64(ret, a, t); @@ -XXX,XX +XXX,XX @@ void tcg_gen_extrh_i64_i32(TCGv_i32 ret, TCGv_i64 arg) tcg_gen_op2(INDEX_op_extrh_i64_i32, tcgv_i32_arg(ret), tcgv_i64_arg(arg)); } else { - TCGv_i64 t = tcg_temp_new_i64(); + TCGv_i64 t = tcg_temp_ebb_new_i64(); tcg_gen_shri_i64(t, arg, 32); tcg_gen_mov_i32(ret, (TCGv_i32)t); tcg_temp_free_i64(t); @@ -XXX,XX +XXX,XX @@ void tcg_gen_concat_i32_i64(TCGv_i64 dest, TCGv_i32 low, TCGv_i32 high) return; } - tmp = tcg_temp_new_i64(); + tmp = tcg_temp_ebb_new_i64(); /* These extensions are only needed for type correctness. We may be able to do better given target specific information. */ tcg_gen_extu_i32_i64(tmp, high); @@ -XXX,XX +XXX,XX @@ void tcg_gen_lookup_and_goto_ptr(void) } plugin_gen_disable_mem_helpers(); - ptr = tcg_temp_new_ptr(); + ptr = tcg_temp_ebb_new_ptr(); gen_helper_lookup_tb_ptr(ptr, cpu_env); tcg_gen_op1i(INDEX_op_goto_ptr, tcgv_ptr_arg(ptr)); tcg_temp_free_ptr(ptr); @@ -XXX,XX +XXX,XX @@ void tcg_gen_qemu_st_i32(TCGv_i32 val, TCGv addr, TCGArg idx, MemOp memop) oi = make_memop_idx(memop, idx); if (!TCG_TARGET_HAS_MEMORY_BSWAP && (memop & MO_BSWAP)) { - swap = tcg_temp_new_i32(); + swap = tcg_temp_ebb_new_i32(); switch (memop & MO_SIZE) { case MO_16: tcg_gen_bswap16_i32(swap, val, 0); @@ -XXX,XX +XXX,XX @@ void tcg_gen_qemu_st_i64(TCGv_i64 val, TCGv addr, TCGArg idx, MemOp memop) oi = make_memop_idx(memop, idx); if (!TCG_TARGET_HAS_MEMORY_BSWAP && (memop & MO_BSWAP)) { - swap = tcg_temp_new_i64(); + swap = tcg_temp_ebb_new_i64(); switch (memop & MO_SIZE) { case MO_16: tcg_gen_bswap16_i64(swap, val, 0); @@ -XXX,XX +XXX,XX @@ void tcg_gen_qemu_st_i128(TCGv_i128 val, TCGv addr, TCGArg idx, MemOp memop) addr_p8 = tcg_temp_new(); if ((mop[0] ^ memop) & MO_BSWAP) { - TCGv_i64 t = tcg_temp_new_i64(); + TCGv_i64 t = tcg_temp_ebb_new_i64(); tcg_gen_bswap64_i64(t, x); gen_ldst_i64(INDEX_op_qemu_st_i64, t, addr, mop[0], idx); @@ -XXX,XX +XXX,XX @@ static void * const table_cmpxchg[(MO_SIZE | MO_BSWAP) + 1] = { void tcg_gen_nonatomic_cmpxchg_i32(TCGv_i32 retv, TCGv addr, TCGv_i32 cmpv, TCGv_i32 newv, TCGArg idx, MemOp memop) { - TCGv_i32 t1 = tcg_temp_new_i32(); - TCGv_i32 t2 = tcg_temp_new_i32(); + TCGv_i32 t1 = tcg_temp_ebb_new_i32(); + TCGv_i32 t2 = tcg_temp_ebb_new_i32(); tcg_gen_ext_i32(t2, cmpv, memop & MO_SIZE); @@ -XXX,XX +XXX,XX @@ void tcg_gen_nonatomic_cmpxchg_i64(TCGv_i64 retv, TCGv addr, TCGv_i64 cmpv, return; } - t1 = tcg_temp_new_i64(); - t2 = tcg_temp_new_i64(); + t1 = tcg_temp_ebb_new_i64(); + t2 = tcg_temp_ebb_new_i64(); tcg_gen_ext_i64(t2, cmpv, memop & MO_SIZE); @@ -XXX,XX +XXX,XX @@ void tcg_gen_atomic_cmpxchg_i64(TCGv_i64 retv, TCGv addr, TCGv_i64 cmpv, tcg_gen_movi_i32(TCGV_HIGH(retv), 0); } } else { - TCGv_i32 c32 = tcg_temp_new_i32(); - TCGv_i32 n32 = tcg_temp_new_i32(); - TCGv_i32 r32 = tcg_temp_new_i32(); + TCGv_i32 c32 = tcg_temp_ebb_new_i32(); + TCGv_i32 n32 = tcg_temp_ebb_new_i32(); + TCGv_i32 r32 = tcg_temp_ebb_new_i32(); tcg_gen_extrl_i64_i32(c32, cmpv); tcg_gen_extrl_i64_i32(n32, newv); @@ -XXX,XX +XXX,XX @@ void tcg_gen_nonatomic_cmpxchg_i128(TCGv_i128 retv, TCGv addr, TCGv_i128 cmpv, gen(retv, cpu_env, addr, cmpv, newv, tcg_constant_i32(oi)); } else { - TCGv_i128 oldv = tcg_temp_new_i128(); - TCGv_i128 tmpv = tcg_temp_new_i128(); - TCGv_i64 t0 = tcg_temp_new_i64(); - TCGv_i64 t1 = tcg_temp_new_i64(); + TCGv_i128 oldv = tcg_temp_ebb_new_i128(); + TCGv_i128 tmpv = tcg_temp_ebb_new_i128(); + TCGv_i64 t0 = tcg_temp_ebb_new_i64(); + TCGv_i64 t1 = tcg_temp_ebb_new_i64(); TCGv_i64 z = tcg_constant_i64(0); tcg_gen_qemu_ld_i128(oldv, addr, idx, memop); @@ -XXX,XX +XXX,XX @@ static void do_nonatomic_op_i32(TCGv_i32 ret, TCGv addr, TCGv_i32 val, TCGArg idx, MemOp memop, bool new_val, void (*gen)(TCGv_i32, TCGv_i32, TCGv_i32)) { - TCGv_i32 t1 = tcg_temp_new_i32(); - TCGv_i32 t2 = tcg_temp_new_i32(); + TCGv_i32 t1 = tcg_temp_ebb_new_i32(); + TCGv_i32 t2 = tcg_temp_ebb_new_i32(); memop = tcg_canonicalize_memop(memop, 0, 0); @@ -XXX,XX +XXX,XX @@ static void do_nonatomic_op_i64(TCGv_i64 ret, TCGv addr, TCGv_i64 val, TCGArg idx, MemOp memop, bool new_val, void (*gen)(TCGv_i64, TCGv_i64, TCGv_i64)) { - TCGv_i64 t1 = tcg_temp_new_i64(); - TCGv_i64 t2 = tcg_temp_new_i64(); + TCGv_i64 t1 = tcg_temp_ebb_new_i64(); + TCGv_i64 t2 = tcg_temp_ebb_new_i64(); memop = tcg_canonicalize_memop(memop, 1, 0); @@ -XXX,XX +XXX,XX @@ static void do_atomic_op_i64(TCGv_i64 ret, TCGv addr, TCGv_i64 val, tcg_gen_movi_i64(ret, 0); #endif /* CONFIG_ATOMIC64 */ } else { - TCGv_i32 v32 = tcg_temp_new_i32(); - TCGv_i32 r32 = tcg_temp_new_i32(); + TCGv_i32 v32 = tcg_temp_ebb_new_i32(); + TCGv_i32 r32 = tcg_temp_ebb_new_i32(); tcg_gen_extrl_i64_i32(v32, val); do_atomic_op_i32(r32, addr, v32, idx, memop & ~MO_SIGN, table); diff --git a/tcg/tcg.c b/tcg/tcg.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/tcg.c +++ b/tcg/tcg.c @@ -XXX,XX +XXX,XX @@ void tcg_gen_callN(void *func, TCGTemp *ret, int nargs, TCGTemp **args) case TCG_CALL_ARG_EXTEND_U: case TCG_CALL_ARG_EXTEND_S: { - TCGv_i64 temp = tcg_temp_new_i64(); + TCGv_i64 temp = tcg_temp_ebb_new_i64(); TCGv_i32 orig = temp_tcgv_i32(ts); if (loc->kind == TCG_CALL_ARG_EXTEND_S) { -- 2.34.1 Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/tcg-op-gvec.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/tcg/tcg-op-gvec.c b/tcg/tcg-op-gvec.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/tcg-op-gvec.c +++ b/tcg/tcg-op-gvec.c @@ -XXX,XX +XXX,XX @@ static void do_dup(unsigned vece, uint32_t dofs, uint32_t oprsz, * stores through to memset. */ if (oprsz == maxsz && vece == MO_8) { - TCGv_ptr t_size = tcg_const_ptr(oprsz); + TCGv_ptr t_size = tcg_constant_ptr(oprsz); TCGv_i32 t_val; if (in_32) { @@ -XXX,XX +XXX,XX @@ static void do_dup(unsigned vece, uint32_t dofs, uint32_t oprsz, if (in_64) { tcg_temp_free_i32(t_val); } - tcg_temp_free_ptr(t_size); tcg_temp_free_ptr(t_ptr); return; } -- 2.34.1 All of these uses have quite local scope. Avoid tcg_const_*, because we haven't added a corresponding interface for TEMP_EBB. Use explicit tcg_gen_movi_* instead. Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- accel/tcg/plugin-gen.c | 24 ++++++++++++++---------- 1 file changed, 14 insertions(+), 10 deletions(-) diff --git a/accel/tcg/plugin-gen.c b/accel/tcg/plugin-gen.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/plugin-gen.c +++ b/accel/tcg/plugin-gen.c @@ -XXX,XX +XXX,XX @@ void HELPER(plugin_vcpu_mem_cb)(unsigned int vcpu_index, static void do_gen_mem_cb(TCGv vaddr, uint32_t info) { - TCGv_i32 cpu_index = tcg_temp_new_i32(); - TCGv_i32 meminfo = tcg_const_i32(info); - TCGv_i64 vaddr64 = tcg_temp_new_i64(); - TCGv_ptr udata = tcg_const_ptr(NULL); + TCGv_i32 cpu_index = tcg_temp_ebb_new_i32(); + TCGv_i32 meminfo = tcg_temp_ebb_new_i32(); + TCGv_i64 vaddr64 = tcg_temp_ebb_new_i64(); + TCGv_ptr udata = tcg_temp_ebb_new_ptr(); + tcg_gen_movi_i32(meminfo, info); + tcg_gen_movi_ptr(udata, 0); tcg_gen_ld_i32(cpu_index, cpu_env, -offsetof(ArchCPU, env) + offsetof(CPUState, cpu_index)); tcg_gen_extu_tl_i64(vaddr64, vaddr); @@ -XXX,XX +XXX,XX @@ static void do_gen_mem_cb(TCGv vaddr, uint32_t info) static void gen_empty_udata_cb(void) { - TCGv_i32 cpu_index = tcg_temp_new_i32(); - TCGv_ptr udata = tcg_const_ptr(NULL); /* will be overwritten later */ + TCGv_i32 cpu_index = tcg_temp_ebb_new_i32(); + TCGv_ptr udata = tcg_temp_ebb_new_ptr(); + tcg_gen_movi_ptr(udata, 0); tcg_gen_ld_i32(cpu_index, cpu_env, -offsetof(ArchCPU, env) + offsetof(CPUState, cpu_index)); gen_helper_plugin_vcpu_udata_cb(cpu_index, udata); @@ -XXX,XX +XXX,XX @@ static void gen_empty_udata_cb(void) */ static void gen_empty_inline_cb(void) { - TCGv_i64 val = tcg_temp_new_i64(); - TCGv_ptr ptr = tcg_const_ptr(NULL); /* overwritten later */ + TCGv_i64 val = tcg_temp_ebb_new_i64(); + TCGv_ptr ptr = tcg_temp_ebb_new_ptr(); + tcg_gen_movi_ptr(ptr, 0); tcg_gen_ld_i64(val, ptr, 0); /* pass an immediate != 0 so that it doesn't get optimized away */ tcg_gen_addi_i64(val, val, 0xdeadface); @@ -XXX,XX +XXX,XX @@ static void gen_empty_mem_cb(TCGv addr, uint32_t info) */ static void gen_empty_mem_helper(void) { - TCGv_ptr ptr; + TCGv_ptr ptr = tcg_temp_ebb_new_ptr(); - ptr = tcg_const_ptr(NULL); + tcg_gen_movi_ptr(ptr, 0); tcg_gen_st_ptr(ptr, cpu_env, offsetof(CPUState, plugin_mem_cbs) - offsetof(ArchCPU, env)); tcg_temp_free_ptr(ptr); -- 2.34.1 Here we are creating a temp whose value needs to be replaced, but always storing NULL into CPUState.plugin_mem_cbs. Use tcg_constant_ptr(0) explicitly. Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- accel/tcg/plugin-gen.c | 8 ++------ 1 file changed, 2 insertions(+), 6 deletions(-) diff --git a/accel/tcg/plugin-gen.c b/accel/tcg/plugin-gen.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/plugin-gen.c +++ b/accel/tcg/plugin-gen.c @@ -XXX,XX +XXX,XX @@ static void inject_mem_disable_helper(struct qemu_plugin_insn *plugin_insn, /* called before finishing a TB with exit_tb, goto_tb or goto_ptr */ void plugin_gen_disable_mem_helpers(void) { - TCGv_ptr ptr; - /* * We could emit the clearing unconditionally and be done. However, this can * be wasteful if for instance plugins don't track memory accesses, or if @@ -XXX,XX +XXX,XX @@ void plugin_gen_disable_mem_helpers(void) if (!tcg_ctx->plugin_tb->mem_helper) { return; } - ptr = tcg_const_ptr(NULL); - tcg_gen_st_ptr(ptr, cpu_env, offsetof(CPUState, plugin_mem_cbs) - - offsetof(ArchCPU, env)); - tcg_temp_free_ptr(ptr); + tcg_gen_st_ptr(tcg_constant_ptr(NULL), cpu_env, + offsetof(CPUState, plugin_mem_cbs) - offsetof(ArchCPU, env)); } static void plugin_gen_tb_udata(const struct qemu_plugin_tb *ptb, -- 2.34.1 Reusing TEMP_TB interferes with detecting whether the temp can be adjusted to TEMP_EBB. Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- include/tcg/tcg.h | 2 +- tcg/tcg.c | 101 ++++++++++++++++++++++++---------------------- 2 files changed, 53 insertions(+), 50 deletions(-) diff --git a/include/tcg/tcg.h b/include/tcg/tcg.h index XXXXXXX..XXXXXXX 100644 --- a/include/tcg/tcg.h +++ b/include/tcg/tcg.h @@ -XXX,XX +XXX,XX @@ struct TCGContext { #endif GHashTable *const_table[TCG_TYPE_COUNT]; - TCGTempSet free_temps[TCG_TYPE_COUNT * 2]; + TCGTempSet free_temps[TCG_TYPE_COUNT]; TCGTemp temps[TCG_MAX_TEMPS]; /* globals first, temps after */ QTAILQ_HEAD(, TCGOp) ops, free_ops; diff --git a/tcg/tcg.c b/tcg/tcg.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/tcg.c +++ b/tcg/tcg.c @@ -XXX,XX +XXX,XX @@ TCGTemp *tcg_global_mem_new_internal(TCGType type, TCGv_ptr base, TCGTemp *tcg_temp_new_internal(TCGType type, TCGTempKind kind) { TCGContext *s = tcg_ctx; - bool temp_local = kind == TEMP_TB; TCGTemp *ts; - int idx, k; + int n; - k = type + (temp_local ? TCG_TYPE_COUNT : 0); - idx = find_first_bit(s->free_temps[k].l, TCG_MAX_TEMPS); - if (idx < TCG_MAX_TEMPS) { - /* There is already an available temp with the right type. */ - clear_bit(idx, s->free_temps[k].l); + if (kind == TEMP_EBB) { + int idx = find_first_bit(s->free_temps[type].l, TCG_MAX_TEMPS); - ts = &s->temps[idx]; - ts->temp_allocated = 1; - tcg_debug_assert(ts->base_type == type); - tcg_debug_assert(ts->kind == kind); - } else { - int i, n; + if (idx < TCG_MAX_TEMPS) { + /* There is already an available temp with the right type. */ + clear_bit(idx, s->free_temps[type].l); - switch (type) { - case TCG_TYPE_I32: - case TCG_TYPE_V64: - case TCG_TYPE_V128: - case TCG_TYPE_V256: - n = 1; - break; - case TCG_TYPE_I64: - n = 64 / TCG_TARGET_REG_BITS; - break; - case TCG_TYPE_I128: - n = 128 / TCG_TARGET_REG_BITS; - break; - default: - g_assert_not_reached(); + ts = &s->temps[idx]; + ts->temp_allocated = 1; + tcg_debug_assert(ts->base_type == type); + tcg_debug_assert(ts->kind == kind); + goto done; } + } else { + tcg_debug_assert(kind == TEMP_TB); + } - ts = tcg_temp_alloc(s); - ts->base_type = type; - ts->temp_allocated = 1; - ts->kind = kind; + switch (type) { + case TCG_TYPE_I32: + case TCG_TYPE_V64: + case TCG_TYPE_V128: + case TCG_TYPE_V256: + n = 1; + break; + case TCG_TYPE_I64: + n = 64 / TCG_TARGET_REG_BITS; + break; + case TCG_TYPE_I128: + n = 128 / TCG_TARGET_REG_BITS; + break; + default: + g_assert_not_reached(); + } - if (n == 1) { - ts->type = type; - } else { - ts->type = TCG_TYPE_REG; + ts = tcg_temp_alloc(s); + ts->base_type = type; + ts->temp_allocated = 1; + ts->kind = kind; - for (i = 1; i < n; ++i) { - TCGTemp *ts2 = tcg_temp_alloc(s); + if (n == 1) { + ts->type = type; + } else { + ts->type = TCG_TYPE_REG; - tcg_debug_assert(ts2 == ts + i); - ts2->base_type = type; - ts2->type = TCG_TYPE_REG; - ts2->temp_allocated = 1; - ts2->temp_subindex = i; - ts2->kind = kind; - } + for (int i = 1; i < n; ++i) { + TCGTemp *ts2 = tcg_temp_alloc(s); + + tcg_debug_assert(ts2 == ts + i); + ts2->base_type = type; + ts2->type = TCG_TYPE_REG; + ts2->temp_allocated = 1; + ts2->temp_subindex = i; + ts2->kind = kind; } } + done: #if defined(CONFIG_DEBUG_TCG) s->temps_in_use++; #endif @@ -XXX,XX +XXX,XX @@ TCGv_vec tcg_temp_new_vec_matching(TCGv_vec match) void tcg_temp_free_internal(TCGTemp *ts) { TCGContext *s = tcg_ctx; - int k, idx; switch (ts->kind) { case TEMP_CONST: @@ -XXX,XX +XXX,XX @@ void tcg_temp_free_internal(TCGTemp *ts) s->temps_in_use--; #endif - idx = temp_idx(ts); - k = ts->base_type + (ts->kind == TEMP_EBB ? 0 : TCG_TYPE_COUNT); - set_bit(idx, s->free_temps[k].l); + if (ts->kind == TEMP_EBB) { + int idx = temp_idx(ts); + set_bit(idx, s->free_temps[ts->base_type].l); + } } TCGTemp *tcg_constant_internal(TCGType type, int64_t val) -- 2.34.1 Guest front-ends now get temps that span the lifetime of the translation block by default, which avoids accidentally using the temp across branches and invalidating the data. Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- include/tcg/tcg.h | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/include/tcg/tcg.h b/include/tcg/tcg.h index XXXXXXX..XXXXXXX 100644 --- a/include/tcg/tcg.h +++ b/include/tcg/tcg.h @@ -XXX,XX +XXX,XX @@ static inline TCGv_i32 tcg_temp_ebb_new_i32(void) static inline TCGv_i32 tcg_temp_new_i32(void) { - TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_I32, TEMP_EBB); + TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_I32, TEMP_TB); return temp_tcgv_i32(t); } @@ -XXX,XX +XXX,XX @@ static inline TCGv_i64 tcg_temp_ebb_new_i64(void) static inline TCGv_i64 tcg_temp_new_i64(void) { - TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_I64, TEMP_EBB); + TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_I64, TEMP_TB); return temp_tcgv_i64(t); } @@ -XXX,XX +XXX,XX @@ static inline TCGv_i128 tcg_temp_ebb_new_i128(void) static inline TCGv_i128 tcg_temp_new_i128(void) { - TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_I128, TEMP_EBB); + TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_I128, TEMP_TB); return temp_tcgv_i128(t); } @@ -XXX,XX +XXX,XX @@ static inline TCGv_ptr tcg_temp_ebb_new_ptr(void) static inline TCGv_ptr tcg_temp_new_ptr(void) { - TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_PTR, TEMP_EBB); + TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_PTR, TEMP_TB); return temp_tcgv_ptr(t); } -- 2.34.1 Since we now get TEMP_TB temporaries by default, we no longer need to make copies across these loops. These were the only uses of new_tmp_a64_local(), so remove that as well. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/arm/tcg/translate-a64.h | 1 - target/arm/tcg/translate-a64.c | 6 ------ target/arm/tcg/translate-sve.c | 32 -------------------------------- 3 files changed, 39 deletions(-) diff --git a/target/arm/tcg/translate-a64.h b/target/arm/tcg/translate-a64.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/translate-a64.h +++ b/target/arm/tcg/translate-a64.h @@ -XXX,XX +XXX,XX @@ #define TARGET_ARM_TRANSLATE_A64_H TCGv_i64 new_tmp_a64(DisasContext *s); -TCGv_i64 new_tmp_a64_local(DisasContext *s); TCGv_i64 new_tmp_a64_zero(DisasContext *s); TCGv_i64 cpu_reg(DisasContext *s, int reg); TCGv_i64 cpu_reg_sp(DisasContext *s, int reg); diff --git a/target/arm/tcg/translate-a64.c b/target/arm/tcg/translate-a64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/translate-a64.c +++ b/target/arm/tcg/translate-a64.c @@ -XXX,XX +XXX,XX @@ TCGv_i64 new_tmp_a64(DisasContext *s) return s->tmp_a64[s->tmp_a64_count++] = tcg_temp_new_i64(); } -TCGv_i64 new_tmp_a64_local(DisasContext *s) -{ - assert(s->tmp_a64_count < TMP_A64_MAX); - return s->tmp_a64[s->tmp_a64_count++] = tcg_temp_local_new_i64(); -} - TCGv_i64 new_tmp_a64_zero(DisasContext *s) { TCGv_i64 t = new_tmp_a64(s); diff --git a/target/arm/tcg/translate-sve.c b/target/arm/tcg/translate-sve.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/translate-sve.c +++ b/target/arm/tcg/translate-sve.c @@ -XXX,XX +XXX,XX @@ void gen_sve_ldr(DisasContext *s, TCGv_ptr base, int vofs, TCGLabel *loop = gen_new_label(); TCGv_ptr tp, i = tcg_const_local_ptr(0); - /* Copy the clean address into a local temp, live across the loop. */ - t0 = clean_addr; - clean_addr = new_tmp_a64_local(s); - tcg_gen_mov_i64(clean_addr, t0); - - if (base != cpu_env) { - TCGv_ptr b = tcg_temp_local_new_ptr(); - tcg_gen_mov_ptr(b, base); - base = b; - } - gen_set_label(loop); t0 = tcg_temp_new_i64(); @@ -XXX,XX +XXX,XX @@ void gen_sve_ldr(DisasContext *s, TCGv_ptr base, int vofs, tcg_gen_brcondi_ptr(TCG_COND_LTU, i, len_align, loop); tcg_temp_free_ptr(i); - - if (base != cpu_env) { - tcg_temp_free_ptr(base); - assert(len_remain == 0); - } } /* @@ -XXX,XX +XXX,XX @@ void gen_sve_str(DisasContext *s, TCGv_ptr base, int vofs, TCGLabel *loop = gen_new_label(); TCGv_ptr tp, i = tcg_const_local_ptr(0); - /* Copy the clean address into a local temp, live across the loop. */ - t0 = clean_addr; - clean_addr = new_tmp_a64_local(s); - tcg_gen_mov_i64(clean_addr, t0); - - if (base != cpu_env) { - TCGv_ptr b = tcg_temp_local_new_ptr(); - tcg_gen_mov_ptr(b, base); - base = b; - } - gen_set_label(loop); t0 = tcg_temp_new_i64(); @@ -XXX,XX +XXX,XX @@ void gen_sve_str(DisasContext *s, TCGv_ptr base, int vofs, tcg_gen_brcondi_ptr(TCG_COND_LTU, i, len_align, loop); tcg_temp_free_ptr(i); - - if (base != cpu_env) { - tcg_temp_free_ptr(base); - assert(len_remain == 0); - } } /* Predicate register stores can be any multiple of 2. */ -- 2.34.1 Since tcg_temp_new_* is now identical, use those. Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/arm/tcg/translate-sve.c | 6 +++--- target/arm/tcg/translate.c | 6 +++--- 2 files changed, 6 insertions(+), 6 deletions(-) diff --git a/target/arm/tcg/translate-sve.c b/target/arm/tcg/translate-sve.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/translate-sve.c +++ b/target/arm/tcg/translate-sve.c @@ -XXX,XX +XXX,XX @@ static bool do_clast_vector(DisasContext *s, arg_rprr_esz *a, bool before) return true; } - last = tcg_temp_local_new_i32(); + last = tcg_temp_new_i32(); over = gen_new_label(); find_last_active(s, last, esz, a->pg); @@ -XXX,XX +XXX,XX @@ void gen_sve_ldr(DisasContext *s, TCGv_ptr base, int vofs, tcg_temp_free_i64(t0); } else { TCGLabel *loop = gen_new_label(); - TCGv_ptr tp, i = tcg_const_local_ptr(0); + TCGv_ptr tp, i = tcg_const_ptr(0); gen_set_label(loop); @@ -XXX,XX +XXX,XX @@ void gen_sve_str(DisasContext *s, TCGv_ptr base, int vofs, tcg_temp_free_i64(t0); } else { TCGLabel *loop = gen_new_label(); - TCGv_ptr tp, i = tcg_const_local_ptr(0); + TCGv_ptr tp, i = tcg_const_ptr(0); gen_set_label(loop); diff --git a/target/arm/tcg/translate.c b/target/arm/tcg/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/translate.c +++ b/target/arm/tcg/translate.c @@ -XXX,XX +XXX,XX @@ static bool op_strex(DisasContext *s, arg_STREX *a, MemOp mop, bool rel) tcg_gen_mb(TCG_MO_ALL | TCG_BAR_STRL); } - addr = tcg_temp_local_new_i32(); + addr = tcg_temp_new_i32(); load_reg_var(s, addr, a->rn); tcg_gen_addi_i32(addr, addr, a->imm); @@ -XXX,XX +XXX,XX @@ static bool op_ldrex(DisasContext *s, arg_LDREX *a, MemOp mop, bool acq) return true; } - addr = tcg_temp_local_new_i32(); + addr = tcg_temp_new_i32(); load_reg_var(s, addr, a->rn); tcg_gen_addi_i32(addr, addr, a->imm); @@ -XXX,XX +XXX,XX @@ static bool trans_LE(DisasContext *s, arg_LE *a) * Decrement by 1 << (4 - LTPSIZE). We need to use a TCG local * so that decr stays live after the brcondi. */ - TCGv_i32 decr = tcg_temp_local_new_i32(); + TCGv_i32 decr = tcg_temp_new_i32(); TCGv_i32 ltpsize = load_cpu_field(v7m.ltpsize); tcg_gen_sub_i32(decr, tcg_constant_i32(4), ltpsize); tcg_gen_shl_i32(decr, tcg_constant_i32(1), decr); -- 2.34.1 Since tcg_temp_new is now identical, use that. Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/cris/translate.c | 6 +++--- target/cris/translate_v10.c.inc | 10 +++++----- 2 files changed, 8 insertions(+), 8 deletions(-) diff --git a/target/cris/translate.c b/target/cris/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/cris/translate.c +++ b/target/cris/translate.c @@ -XXX,XX +XXX,XX @@ static int dec_bound_r(CPUCRISState *env, DisasContext *dc) LOG_DIS("bound.%c $r%u, $r%u\n", memsize_char(size), dc->op1, dc->op2); cris_cc_mask(dc, CC_MASK_NZ); - l0 = tcg_temp_local_new(); + l0 = tcg_temp_new(); dec_prep_move_r(dc, dc->op1, dc->op2, size, 0, l0); cris_alu(dc, CC_OP_BOUND, cpu_R[dc->op2], cpu_R[dc->op2], l0, 4); tcg_temp_free(l0); @@ -XXX,XX +XXX,XX @@ static int dec_bound_m(CPUCRISState *env, DisasContext *dc) dc->op1, dc->postinc ? "+]" : "]", dc->op2); - l[0] = tcg_temp_local_new(); - l[1] = tcg_temp_local_new(); + l[0] = tcg_temp_new(); + l[1] = tcg_temp_new(); insn_len = dec_prep_alu_m(env, dc, 0, memsize, l[0], l[1]); cris_cc_mask(dc, CC_MASK_NZ); cris_alu(dc, CC_OP_BOUND, cpu_R[dc->op2], l[0], l[1], 4); diff --git a/target/cris/translate_v10.c.inc b/target/cris/translate_v10.c.inc index XXXXXXX..XXXXXXX 100644 --- a/target/cris/translate_v10.c.inc +++ b/target/cris/translate_v10.c.inc @@ -XXX,XX +XXX,XX @@ static void gen_store_v10_conditional(DisasContext *dc, TCGv addr, TCGv val, unsigned int size, int mem_index) { TCGLabel *l1 = gen_new_label(); - TCGv taddr = tcg_temp_local_new(); - TCGv tval = tcg_temp_local_new(); - TCGv t1 = tcg_temp_local_new(); + TCGv taddr = tcg_temp_new(); + TCGv tval = tcg_temp_new(); + TCGv t1 = tcg_temp_new(); dc->postinc = 0; cris_evaluate_flags(dc); @@ -XXX,XX +XXX,XX @@ static void dec10_reg_bound(DisasContext *dc, int size) { TCGv t; - t = tcg_temp_local_new(); + t = tcg_temp_new(); t_gen_zext(t, cpu_R[dc->src], size); cris_alu(dc, CC_OP_BOUND, cpu_R[dc->dst], cpu_R[dc->dst], t, 4); tcg_temp_free(t); @@ -XXX,XX +XXX,XX @@ static int dec10_ind_bound(CPUCRISState *env, DisasContext *dc, int rd = dc->dst; TCGv t; - t = tcg_temp_local_new(); + t = tcg_temp_new(); insn_len += dec10_prep_move_m(env, dc, 0, size, t); cris_alu(dc, CC_OP_BOUND, cpu_R[dc->dst], cpu_R[rd], t, 4); if (dc->dst == 15) { -- 2.34.1 Since tcg_temp_new_* is now identical, use those. Reviewed-by: Taylor Simpson <tsimpson@quicinc.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/hexagon/idef-parser/README.rst | 4 ++-- target/hexagon/gen_tcg.h | 4 ++-- target/hexagon/genptr.c | 16 ++++++++-------- target/hexagon/idef-parser/parser-helpers.c | 4 ++-- target/hexagon/translate.c | 2 +- target/hexagon/README | 8 ++++---- target/hexagon/gen_tcg_funcs.py | 18 +++++++----------- 7 files changed, 26 insertions(+), 30 deletions(-) diff --git a/target/hexagon/idef-parser/README.rst b/target/hexagon/idef-parser/README.rst index XXXXXXX..XXXXXXX 100644 --- a/target/hexagon/idef-parser/README.rst +++ b/target/hexagon/idef-parser/README.rst @@ -XXX,XX +XXX,XX @@ generators the previous declarations are mapped to :: - int var1; -> TCGv_i32 var1 = tcg_temp_local_new_i32(); + int var1; -> TCGv_i32 var1 = tcg_temp_new_i32(); - int var2 = 0; -> TCGv_i32 var1 = tcg_temp_local_new_i32(); + int var2 = 0; -> TCGv_i32 var1 = tcg_temp_new_i32(); tcg_gen_movi_i32(j, ((int64_t) 0ULL)); which are later automatically freed at the end of the function they're declared diff --git a/target/hexagon/gen_tcg.h b/target/hexagon/gen_tcg.h index XXXXXXX..XXXXXXX 100644 --- a/target/hexagon/gen_tcg.h +++ b/target/hexagon/gen_tcg.h @@ -XXX,XX +XXX,XX @@ */ #define fGEN_TCG_PRED_LOAD(GET_EA, PRED, SIZE, SIGN) \ do { \ - TCGv LSB = tcg_temp_local_new(); \ + TCGv LSB = tcg_temp_new(); \ TCGLabel *label = gen_new_label(); \ tcg_gen_movi_tl(EA, 0); \ PRED; \ @@ -XXX,XX +XXX,XX @@ /* Predicated loads into a register pair */ #define fGEN_TCG_PRED_LOAD_PAIR(GET_EA, PRED) \ do { \ - TCGv LSB = tcg_temp_local_new(); \ + TCGv LSB = tcg_temp_new(); \ TCGLabel *label = gen_new_label(); \ tcg_gen_movi_tl(EA, 0); \ PRED; \ diff --git a/target/hexagon/genptr.c b/target/hexagon/genptr.c index XXXXXXX..XXXXXXX 100644 --- a/target/hexagon/genptr.c +++ b/target/hexagon/genptr.c @@ -XXX,XX +XXX,XX @@ static void gen_cond_call(DisasContext *ctx, TCGv pred, TCGCond cond, int pc_off) { TCGv next_PC; - TCGv lsb = tcg_temp_local_new(); + TCGv lsb = tcg_temp_new(); TCGLabel *skip = gen_new_label(); tcg_gen_andi_tl(lsb, pred, 1); gen_write_new_pc_pcrel(ctx, pc_off, cond, lsb); @@ -XXX,XX +XXX,XX @@ static void gen_cond_call(DisasContext *ctx, TCGv pred, static void gen_endloop0(DisasContext *ctx) { - TCGv lpcfg = tcg_temp_local_new(); + TCGv lpcfg = tcg_temp_new(); GET_USR_FIELD(USR_LPCFG, lpcfg); @@ -XXX,XX +XXX,XX @@ static void gen_sar(TCGv dst, TCGv src, TCGv shift_amt) /* Bidirectional shift right with saturation */ static void gen_asr_r_r_sat(TCGv RdV, TCGv RsV, TCGv RtV) { - TCGv shift_amt = tcg_temp_local_new(); + TCGv shift_amt = tcg_temp_new(); TCGLabel *positive = gen_new_label(); TCGLabel *done = gen_new_label(); @@ -XXX,XX +XXX,XX @@ static void gen_asr_r_r_sat(TCGv RdV, TCGv RsV, TCGv RtV) /* Bidirectional shift left with saturation */ static void gen_asl_r_r_sat(TCGv RdV, TCGv RsV, TCGv RtV) { - TCGv shift_amt = tcg_temp_local_new(); + TCGv shift_amt = tcg_temp_new(); TCGLabel *positive = gen_new_label(); TCGLabel *done = gen_new_label(); @@ -XXX,XX +XXX,XX @@ static void gen_log_vreg_write(DisasContext *ctx, intptr_t srcoff, int num, intptr_t dstoff; if (is_predicated) { - TCGv cancelled = tcg_temp_local_new(); + TCGv cancelled = tcg_temp_new(); label_end = gen_new_label(); /* Don't do anything if the slot was cancelled */ @@ -XXX,XX +XXX,XX @@ static void gen_log_qreg_write(intptr_t srcoff, int num, int vnew, intptr_t dstoff; if (is_predicated) { - TCGv cancelled = tcg_temp_local_new(); + TCGv cancelled = tcg_temp_new(); label_end = gen_new_label(); /* Don't do anything if the slot was cancelled */ @@ -XXX,XX +XXX,XX @@ void gen_satu_i64_ovfl(TCGv ovfl, TCGv_i64 dest, TCGv_i64 source, int width) /* Implements the fADDSAT64 macro in TCG */ void gen_add_sat_i64(TCGv_i64 ret, TCGv_i64 a, TCGv_i64 b) { - TCGv_i64 sum = tcg_temp_local_new_i64(); + TCGv_i64 sum = tcg_temp_new_i64(); TCGv_i64 xor = tcg_temp_new_i64(); TCGv_i64 cond1 = tcg_temp_new_i64(); - TCGv_i64 cond2 = tcg_temp_local_new_i64(); + TCGv_i64 cond2 = tcg_temp_new_i64(); TCGv_i64 cond3 = tcg_temp_new_i64(); TCGv_i64 mask = tcg_constant_i64(0x8000000000000000ULL); TCGv_i64 max_pos = tcg_constant_i64(0x7FFFFFFFFFFFFFFFLL); diff --git a/target/hexagon/idef-parser/parser-helpers.c b/target/hexagon/idef-parser/parser-helpers.c index XXXXXXX..XXXXXXX 100644 --- a/target/hexagon/idef-parser/parser-helpers.c +++ b/target/hexagon/idef-parser/parser-helpers.c @@ -XXX,XX +XXX,XX @@ HexValue gen_tmp_local(Context *c, rvalue.is_manual = false; rvalue.tmp.index = c->inst.tmp_count; OUT(c, locp, "TCGv_i", &bit_width, " tmp_", &c->inst.tmp_count, - " = tcg_temp_local_new_i", &bit_width, "();\n"); + " = tcg_temp_new_i", &bit_width, "();\n"); c->inst.tmp_count++; return rvalue; } @@ -XXX,XX +XXX,XX @@ void gen_varid_allocate(Context *c, new_var.signedness = signedness; EMIT_HEAD(c, "TCGv_%s %s", bit_suffix, varid->var.name->str); - EMIT_HEAD(c, " = tcg_temp_local_new_%s();\n", bit_suffix); + EMIT_HEAD(c, " = tcg_temp_new_%s();\n", bit_suffix); g_array_append_val(c->inst.allocated, new_var); } diff --git a/target/hexagon/translate.c b/target/hexagon/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/hexagon/translate.c +++ b/target/hexagon/translate.c @@ -XXX,XX +XXX,XX @@ void process_store(DisasContext *ctx, int slot_num) tcg_temp_free(cancelled); } { - TCGv address = tcg_temp_local_new(); + TCGv address = tcg_temp_new(); tcg_gen_mov_tl(address, hex_store_addr[slot_num]); /* diff --git a/target/hexagon/README b/target/hexagon/README index XXXXXXX..XXXXXXX 100644 --- a/target/hexagon/README +++ b/target/hexagon/README @@ -XXX,XX +XXX,XX @@ tcg_funcs_generated.c.inc Insn *insn, Packet *pkt) { - TCGv RdV = tcg_temp_local_new(); + TCGv RdV = tcg_temp_new(); const int RdN = insn->regno[0]; TCGv RsV = hex_gpr[insn->regno[1]]; TCGv RtV = hex_gpr[insn->regno[2]]; @@ -XXX,XX +XXX,XX @@ istruction. const int VdN = insn->regno[0]; const intptr_t VdV_off = ctx_future_vreg_off(ctx, VdN, 1, true); - TCGv_ptr VdV = tcg_temp_local_new_ptr(); + TCGv_ptr VdV = tcg_temp_new_ptr(); tcg_gen_addi_ptr(VdV, cpu_env, VdV_off); const int VuN = insn->regno[1]; const intptr_t VuV_off = vreg_src_off(ctx, VuN); - TCGv_ptr VuV = tcg_temp_local_new_ptr(); + TCGv_ptr VuV = tcg_temp_new_ptr(); const int VvN = insn->regno[2]; const intptr_t VvV_off = vreg_src_off(ctx, VvN); - TCGv_ptr VvV = tcg_temp_local_new_ptr(); + TCGv_ptr VvV = tcg_temp_new_ptr(); tcg_gen_addi_ptr(VuV, cpu_env, VuV_off); tcg_gen_addi_ptr(VvV, cpu_env, VvV_off); TCGv slot = tcg_constant_tl(insn->slot); diff --git a/target/hexagon/gen_tcg_funcs.py b/target/hexagon/gen_tcg_funcs.py index XXXXXXX..XXXXXXX 100755 --- a/target/hexagon/gen_tcg_funcs.py +++ b/target/hexagon/gen_tcg_funcs.py @@ -XXX,XX +XXX,XX @@ ## Helpers for gen_tcg_func ## def gen_decl_ea_tcg(f, tag): - if ('A_CONDEXEC' in hex_common.attribdict[tag] or - 'A_LOAD' in hex_common.attribdict[tag]): - f.write(" TCGv EA = tcg_temp_local_new();\n") - else: - f.write(" TCGv EA = tcg_temp_new();\n") + f.write(" TCGv EA = tcg_temp_new();\n") def gen_free_ea_tcg(f): f.write(" tcg_temp_free(EA);\n") def genptr_decl_pair_writable(f, tag, regtype, regid, regno): regN="%s%sN" % (regtype,regid) - f.write(" TCGv_i64 %s%sV = tcg_temp_local_new_i64();\n" % \ + f.write(" TCGv_i64 %s%sV = tcg_temp_new_i64();\n" % \ (regtype, regid)) if (regtype == "C"): f.write(" const int %s = insn->regno[%d] + HEX_REG_SA0;\n" % \ @@ -XXX,XX +XXX,XX @@ def genptr_decl_pair_writable(f, tag, regtype, regid, regno): def genptr_decl_writable(f, tag, regtype, regid, regno): regN="%s%sN" % (regtype,regid) - f.write(" TCGv %s%sV = tcg_temp_local_new();\n" % \ + f.write(" TCGv %s%sV = tcg_temp_new();\n" % \ (regtype, regid)) if (regtype == "C"): f.write(" const int %s = insn->regno[%d] + HEX_REG_SA0;\n" % \ @@ -XXX,XX +XXX,XX @@ def genptr_decl(f, tag, regtype, regid, regno): regN="%s%sN" % (regtype,regid) if (regtype == "R"): if (regid in {"ss", "tt"}): - f.write(" TCGv_i64 %s%sV = tcg_temp_local_new_i64();\n" % \ + f.write(" TCGv_i64 %s%sV = tcg_temp_new_i64();\n" % \ (regtype, regid)) f.write(" const int %s = insn->regno[%d];\n" % \ (regN, regno)) @@ -XXX,XX +XXX,XX @@ def genptr_decl(f, tag, regtype, regid, regno): print("Bad register parse: ", regtype, regid) elif (regtype == "C"): if (regid == "ss"): - f.write(" TCGv_i64 %s%sV = tcg_temp_local_new_i64();\n" % \ + f.write(" TCGv_i64 %s%sV = tcg_temp_new_i64();\n" % \ (regtype, regid)) f.write(" const int %s = insn->regno[%d] + HEX_REG_SA0;\n" % \ (regN, regno)) elif (regid == "dd"): genptr_decl_pair_writable(f, tag, regtype, regid, regno) elif (regid == "s"): - f.write(" TCGv %s%sV = tcg_temp_local_new();\n" % \ + f.write(" TCGv %s%sV = tcg_temp_new();\n" % \ (regtype, regid)) f.write(" const int %s%sN = insn->regno[%d] + HEX_REG_SA0;\n" % \ (regtype, regid, regno)) @@ -XXX,XX +XXX,XX @@ def genptr_dst_write_opn(f,regtype, regid, tag): ## We produce: ## static void generate_A2_add(DisasContext *ctx) ## { -## TCGv RdV = tcg_temp_local_new(); +## TCGv RdV = tcg_temp_new(); ## const int RdN = insn->regno[0]; ## TCGv RsV = hex_gpr[insn->regno[1]]; ## TCGv RtV = hex_gpr[insn->regno[2]]; -- 2.34.1 This is now equivalent to gen_tmp. Reviewed-by: Taylor Simpson <tsimpson@quicinc.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/hexagon/idef-parser/parser-helpers.c | 24 ++------------------- 1 file changed, 2 insertions(+), 22 deletions(-) diff --git a/target/hexagon/idef-parser/parser-helpers.c b/target/hexagon/idef-parser/parser-helpers.c index XXXXXXX..XXXXXXX 100644 --- a/target/hexagon/idef-parser/parser-helpers.c +++ b/target/hexagon/idef-parser/parser-helpers.c @@ -XXX,XX +XXX,XX @@ HexValue gen_tmp(Context *c, return rvalue; } -HexValue gen_tmp_local(Context *c, - YYLTYPE *locp, - unsigned bit_width, - HexSignedness signedness) -{ - HexValue rvalue; - assert(bit_width == 32 || bit_width == 64); - memset(&rvalue, 0, sizeof(HexValue)); - rvalue.type = TEMP; - rvalue.bit_width = bit_width; - rvalue.signedness = signedness; - rvalue.is_dotnew = false; - rvalue.is_manual = false; - rvalue.tmp.index = c->inst.tmp_count; - OUT(c, locp, "TCGv_i", &bit_width, " tmp_", &c->inst.tmp_count, - " = tcg_temp_new_i", &bit_width, "();\n"); - c->inst.tmp_count++; - return rvalue; -} - HexValue gen_tmp_value(Context *c, YYLTYPE *locp, const char *value, @@ -XXX,XX +XXX,XX @@ HexValue gen_rvalue_sat(Context *c, YYLTYPE *locp, HexSat *sat, assert_signedness(c, locp, sat->signedness); unsigned_str = (sat->signedness == UNSIGNED) ? "u" : ""; - res = gen_tmp_local(c, locp, value->bit_width, sat->signedness); - ovfl = gen_tmp_local(c, locp, 32, sat->signedness); + res = gen_tmp(c, locp, value->bit_width, sat->signedness); + ovfl = gen_tmp(c, locp, 32, sat->signedness); OUT(c, locp, "gen_sat", unsigned_str, "_", bit_suffix, "_ovfl("); OUT(c, locp, &ovfl, ", ", &res, ", ", value, ", ", &width->imm.value, ");\n"); -- 2.34.1 This wasn't actually used at all, just some unused macro re-definitions. Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/hppa/translate.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/target/hppa/translate.c b/target/hppa/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/hppa/translate.c +++ b/target/hppa/translate.c @@ -XXX,XX +XXX,XX @@ #undef TCGv #undef tcg_temp_new #undef tcg_global_mem_new -#undef tcg_temp_local_new #undef tcg_temp_free #if TARGET_LONG_BITS == 64 @@ -XXX,XX +XXX,XX @@ #define tcg_temp_new tcg_temp_new_i64 #define tcg_global_mem_new tcg_global_mem_new_i64 -#define tcg_temp_local_new tcg_temp_local_new_i64 #define tcg_temp_free tcg_temp_free_i64 #define tcg_gen_movi_reg tcg_gen_movi_i64 @@ -XXX,XX +XXX,XX @@ #define TCGv_reg TCGv_i32 #define tcg_temp_new tcg_temp_new_i32 #define tcg_global_mem_new tcg_global_mem_new_i32 -#define tcg_temp_local_new tcg_temp_local_new_i32 #define tcg_temp_free tcg_temp_free_i32 #define tcg_gen_movi_reg tcg_gen_movi_i32 -- 2.34.1 Since tcg_temp_new is now identical, use that. In some cases we can avoid a copy from A0 or T0. Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/i386/tcg/translate.c | 27 +++++++++------------------ 1 file changed, 9 insertions(+), 18 deletions(-) diff --git a/target/i386/tcg/translate.c b/target/i386/tcg/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/i386/tcg/translate.c +++ b/target/i386/tcg/translate.c @@ -XXX,XX +XXX,XX @@ static bool disas_insn(DisasContext *s, CPUState *cpu) if (mod == 3) { goto illegal_op; } - a0 = tcg_temp_local_new(); - t0 = tcg_temp_local_new(); + a0 = s->A0; + t0 = s->T0; label1 = gen_new_label(); - tcg_gen_mov_tl(a0, s->A0); - tcg_gen_mov_tl(t0, s->T0); - gen_set_label(label1); t1 = tcg_temp_new(); t2 = tcg_temp_new(); @@ -XXX,XX +XXX,XX @@ static bool disas_insn(DisasContext *s, CPUState *cpu) tcg_gen_brcond_tl(TCG_COND_NE, t0, t2, label1); tcg_temp_free(t2); - tcg_temp_free(a0); tcg_gen_neg_tl(s->T0, t0); - tcg_temp_free(t0); } else { tcg_gen_neg_tl(s->T0, s->T0); if (mod != 3) { @@ -XXX,XX +XXX,XX @@ static bool disas_insn(DisasContext *s, CPUState *cpu) #endif { TCGLabel *label1; - TCGv t0, t1, t2, a0; + TCGv t0, t1, t2; if (!PE(s) || VM86(s)) goto illegal_op; - t0 = tcg_temp_local_new(); - t1 = tcg_temp_local_new(); - t2 = tcg_temp_local_new(); + t0 = tcg_temp_new(); + t1 = tcg_temp_new(); + t2 = tcg_temp_new(); ot = MO_16; modrm = x86_ldub_code(env, s); reg = (modrm >> 3) & 7; @@ -XXX,XX +XXX,XX @@ static bool disas_insn(DisasContext *s, CPUState *cpu) if (mod != 3) { gen_lea_modrm(env, s, modrm); gen_op_ld_v(s, ot, t0, s->A0); - a0 = tcg_temp_local_new(); - tcg_gen_mov_tl(a0, s->A0); } else { gen_op_mov_v_reg(s, ot, t0, rm); - a0 = NULL; } gen_op_mov_v_reg(s, ot, t1, reg); tcg_gen_andi_tl(s->tmp0, t0, 3); @@ -XXX,XX +XXX,XX @@ static bool disas_insn(DisasContext *s, CPUState *cpu) tcg_gen_movi_tl(t2, CC_Z); gen_set_label(label1); if (mod != 3) { - gen_op_st_v(s, ot, t0, a0); - tcg_temp_free(a0); + gen_op_st_v(s, ot, t0, s->A0); } else { gen_op_mov_reg_v(s, ot, rm, t0); } @@ -XXX,XX +XXX,XX @@ static bool disas_insn(DisasContext *s, CPUState *cpu) modrm = x86_ldub_code(env, s); reg = ((modrm >> 3) & 7) | REX_R(s); gen_ldst_modrm(env, s, modrm, MO_16, OR_TMP0, 0); - t0 = tcg_temp_local_new(); + t0 = tcg_temp_new(); gen_update_cc_op(s); if (b == 0x102) { gen_helper_lar(t0, cpu_env, s->T0); @@ -XXX,XX +XXX,XX @@ static void i386_tr_init_disas_context(DisasContextBase *dcbase, CPUState *cpu) dc->tmp2_i32 = tcg_temp_new_i32(); dc->tmp3_i32 = tcg_temp_new_i32(); dc->tmp4 = tcg_temp_new(); - dc->cc_srcT = tcg_temp_local_new(); + dc->cc_srcT = tcg_temp_new(); } static void i386_tr_tb_start(DisasContextBase *db, CPUState *cpu) -- 2.34.1 Since tcg_temp_new is now identical, use that. Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/mips/tcg/translate.c | 57 ++++++------------------ target/mips/tcg/nanomips_translate.c.inc | 4 +- 2 files changed, 16 insertions(+), 45 deletions(-) diff --git a/target/mips/tcg/translate.c b/target/mips/tcg/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/mips/tcg/translate.c +++ b/target/mips/tcg/translate.c @@ -XXX,XX +XXX,XX @@ static void gen_arith_imm(DisasContext *ctx, uint32_t opc, switch (opc) { case OPC_ADDI: { - TCGv t0 = tcg_temp_local_new(); + TCGv t0 = tcg_temp_new(); TCGv t1 = tcg_temp_new(); TCGv t2 = tcg_temp_new(); TCGLabel *l1 = gen_new_label(); @@ -XXX,XX +XXX,XX @@ static void gen_arith_imm(DisasContext *ctx, uint32_t opc, #if defined(TARGET_MIPS64) case OPC_DADDI: { - TCGv t0 = tcg_temp_local_new(); + TCGv t0 = tcg_temp_new(); TCGv t1 = tcg_temp_new(); TCGv t2 = tcg_temp_new(); TCGLabel *l1 = gen_new_label(); @@ -XXX,XX +XXX,XX @@ static void gen_arith(DisasContext *ctx, uint32_t opc, switch (opc) { case OPC_ADD: { - TCGv t0 = tcg_temp_local_new(); + TCGv t0 = tcg_temp_new(); TCGv t1 = tcg_temp_new(); TCGv t2 = tcg_temp_new(); TCGLabel *l1 = gen_new_label(); @@ -XXX,XX +XXX,XX @@ static void gen_arith(DisasContext *ctx, uint32_t opc, break; case OPC_SUB: { - TCGv t0 = tcg_temp_local_new(); + TCGv t0 = tcg_temp_new(); TCGv t1 = tcg_temp_new(); TCGv t2 = tcg_temp_new(); TCGLabel *l1 = gen_new_label(); @@ -XXX,XX +XXX,XX @@ static void gen_arith(DisasContext *ctx, uint32_t opc, #if defined(TARGET_MIPS64) case OPC_DADD: { - TCGv t0 = tcg_temp_local_new(); + TCGv t0 = tcg_temp_new(); TCGv t1 = tcg_temp_new(); TCGv t2 = tcg_temp_new(); TCGLabel *l1 = gen_new_label(); @@ -XXX,XX +XXX,XX @@ static void gen_arith(DisasContext *ctx, uint32_t opc, break; case OPC_DSUB: { - TCGv t0 = tcg_temp_local_new(); + TCGv t0 = tcg_temp_new(); TCGv t1 = tcg_temp_new(); TCGv t2 = tcg_temp_new(); TCGLabel *l1 = gen_new_label(); @@ -XXX,XX +XXX,XX @@ static void gen_loongson_integer(DisasContext *ctx, uint32_t opc, return; } - switch (opc) { - case OPC_MULT_G_2E: - case OPC_MULT_G_2F: - case OPC_MULTU_G_2E: - case OPC_MULTU_G_2F: -#if defined(TARGET_MIPS64) - case OPC_DMULT_G_2E: - case OPC_DMULT_G_2F: - case OPC_DMULTU_G_2E: - case OPC_DMULTU_G_2F: -#endif - t0 = tcg_temp_new(); - t1 = tcg_temp_new(); - break; - default: - t0 = tcg_temp_local_new(); - t1 = tcg_temp_local_new(); - break; - } - + t0 = tcg_temp_new(); + t1 = tcg_temp_new(); gen_load_gpr(t0, rs); gen_load_gpr(t1, rt); @@ -XXX,XX +XXX,XX @@ static void gen_loongson_multimedia(DisasContext *ctx, int rd, int rs, int rt) TCGCond cond; opc = MASK_LMMI(ctx->opcode); - switch (opc) { - case OPC_ADD_CP2: - case OPC_SUB_CP2: - case OPC_DADD_CP2: - case OPC_DSUB_CP2: - t0 = tcg_temp_local_new_i64(); - t1 = tcg_temp_local_new_i64(); - break; - default: - t0 = tcg_temp_new_i64(); - t1 = tcg_temp_new_i64(); - break; - } - check_cp1_enabled(ctx); + + t0 = tcg_temp_new_i64(); + t1 = tcg_temp_new_i64(); gen_load_fpr64(ctx, t0, rs); gen_load_fpr64(ctx, t1, rt); @@ -XXX,XX +XXX,XX @@ static void gen_mftr(CPUMIPSState *env, DisasContext *ctx, int rt, int rd, int u, int sel, int h) { int other_tc = env->CP0_VPEControl & (0xff << CP0VPECo_TargTC); - TCGv t0 = tcg_temp_local_new(); + TCGv t0 = tcg_temp_new(); if ((env->CP0_VPEConf0 & (1 << CP0VPEC0_MVP)) == 0 && ((env->tcs[other_tc].CP0_TCBind & (0xf << CP0TCBd_CurVPE)) != @@ -XXX,XX +XXX,XX @@ static void gen_mttr(CPUMIPSState *env, DisasContext *ctx, int rd, int rt, int u, int sel, int h) { int other_tc = env->CP0_VPEControl & (0xff << CP0VPECo_TargTC); - TCGv t0 = tcg_temp_local_new(); + TCGv t0 = tcg_temp_new(); gen_load_gpr(t0, rt); if ((env->CP0_VPEConf0 & (1 << CP0VPEC0_MVP)) == 0 && @@ -XXX,XX +XXX,XX @@ static void gen_flt3_arith(DisasContext *ctx, uint32_t opc, case OPC_ALNV_PS: check_ps(ctx); { - TCGv t0 = tcg_temp_local_new(); + TCGv t0 = tcg_temp_new(); TCGv_i32 fp = tcg_temp_new_i32(); TCGv_i32 fph = tcg_temp_new_i32(); TCGLabel *l1 = gen_new_label(); diff --git a/target/mips/tcg/nanomips_translate.c.inc b/target/mips/tcg/nanomips_translate.c.inc index XXXXXXX..XXXXXXX 100644 --- a/target/mips/tcg/nanomips_translate.c.inc +++ b/target/mips/tcg/nanomips_translate.c.inc @@ -XXX,XX +XXX,XX @@ static void gen_llwp(DisasContext *ctx, uint32_t base, int16_t offset, static void gen_scwp(DisasContext *ctx, uint32_t base, int16_t offset, uint32_t reg1, uint32_t reg2, bool eva) { - TCGv taddr = tcg_temp_local_new(); - TCGv lladdr = tcg_temp_local_new(); + TCGv taddr = tcg_temp_new(); + TCGv lladdr = tcg_temp_new(); TCGv_i64 tval = tcg_temp_new_i64(); TCGv_i64 llval = tcg_temp_new_i64(); TCGv_i64 val = tcg_temp_new_i64(); -- 2.34.1 Since tcg_temp_new is now identical, use that. Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/ppc/translate.c | 6 +++--- target/ppc/translate/spe-impl.c.inc | 8 ++++---- target/ppc/translate/vmx-impl.c.inc | 4 ++-- 3 files changed, 9 insertions(+), 9 deletions(-) diff --git a/target/ppc/translate.c b/target/ppc/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/ppc/translate.c +++ b/target/ppc/translate.c @@ -XXX,XX +XXX,XX @@ static void gen_bcond(DisasContext *ctx, int type) TCGv target; if (type == BCOND_LR || type == BCOND_CTR || type == BCOND_TAR) { - target = tcg_temp_local_new(); + target = tcg_temp_new(); if (type == BCOND_CTR) { tcg_gen_mov_tl(target, cpu_ctr); } else if (type == BCOND_TAR) { @@ -XXX,XX +XXX,XX @@ static inline void gen_405_mulladd_insn(DisasContext *ctx, int opc2, int opc3, { TCGv t0, t1; - t0 = tcg_temp_local_new(); - t1 = tcg_temp_local_new(); + t0 = tcg_temp_new(); + t1 = tcg_temp_new(); switch (opc3 & 0x0D) { case 0x05: diff --git a/target/ppc/translate/spe-impl.c.inc b/target/ppc/translate/spe-impl.c.inc index XXXXXXX..XXXXXXX 100644 --- a/target/ppc/translate/spe-impl.c.inc +++ b/target/ppc/translate/spe-impl.c.inc @@ -XXX,XX +XXX,XX @@ static inline void gen_op_evsrwu(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2) { TCGLabel *l1 = gen_new_label(); TCGLabel *l2 = gen_new_label(); - TCGv_i32 t0 = tcg_temp_local_new_i32(); + TCGv_i32 t0 = tcg_temp_new_i32(); /* No error here: 6 bits are used */ tcg_gen_andi_i32(t0, arg2, 0x3F); @@ -XXX,XX +XXX,XX @@ static inline void gen_op_evsrws(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2) { TCGLabel *l1 = gen_new_label(); TCGLabel *l2 = gen_new_label(); - TCGv_i32 t0 = tcg_temp_local_new_i32(); + TCGv_i32 t0 = tcg_temp_new_i32(); /* No error here: 6 bits are used */ tcg_gen_andi_i32(t0, arg2, 0x3F); @@ -XXX,XX +XXX,XX @@ static inline void gen_op_evslw(TCGv_i32 ret, TCGv_i32 arg1, TCGv_i32 arg2) { TCGLabel *l1 = gen_new_label(); TCGLabel *l2 = gen_new_label(); - TCGv_i32 t0 = tcg_temp_local_new_i32(); + TCGv_i32 t0 = tcg_temp_new_i32(); /* No error here: 6 bits are used */ tcg_gen_andi_i32(t0, arg2, 0x3F); @@ -XXX,XX +XXX,XX @@ static inline void gen_evsel(DisasContext *ctx) TCGLabel *l2 = gen_new_label(); TCGLabel *l3 = gen_new_label(); TCGLabel *l4 = gen_new_label(); - TCGv_i32 t0 = tcg_temp_local_new_i32(); + TCGv_i32 t0 = tcg_temp_new_i32(); tcg_gen_andi_i32(t0, cpu_crf[ctx->opcode & 0x07], 1 << 3); tcg_gen_brcondi_i32(TCG_COND_EQ, t0, 0, l1); diff --git a/target/ppc/translate/vmx-impl.c.inc b/target/ppc/translate/vmx-impl.c.inc index XXXXXXX..XXXXXXX 100644 --- a/target/ppc/translate/vmx-impl.c.inc +++ b/target/ppc/translate/vmx-impl.c.inc @@ -XXX,XX +XXX,XX @@ static bool do_vcmpq(DisasContext *ctx, arg_VX_bf *a, bool sign) REQUIRE_INSNS_FLAGS2(ctx, ISA310); REQUIRE_VECTOR(ctx); - vra = tcg_temp_local_new_i64(); - vrb = tcg_temp_local_new_i64(); + vra = tcg_temp_new_i64(); + vrb = tcg_temp_new_i64(); gt = gen_new_label(); lt = gen_new_label(); done = gen_new_label(); -- 2.34.1 Since tcg_temp_new_* is now identical, use those. Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/xtensa/translate.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/target/xtensa/translate.c b/target/xtensa/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/xtensa/translate.c +++ b/target/xtensa/translate.c @@ -XXX,XX +XXX,XX @@ static void gen_right_shift_sar(DisasContext *dc, TCGv_i32 sa) static void gen_left_shift_sar(DisasContext *dc, TCGv_i32 sa) { if (!dc->sar_m32_allocated) { - dc->sar_m32 = tcg_temp_local_new_i32(); + dc->sar_m32 = tcg_temp_new_i32(); dc->sar_m32_allocated = true; } tcg_gen_andi_i32(dc->sar_m32, sa, 0x1f); @@ -XXX,XX +XXX,XX @@ static void disas_xtensa_insn(CPUXtensaState *env, DisasContext *dc) if (i == 0 || arg_copy[i].resource != resource) { resource = arg_copy[i].resource; if (arg_copy[i].arg->num_bits <= 32) { - temp = tcg_temp_local_new_i32(); + temp = tcg_temp_new_i32(); tcg_gen_mov_i32(temp, arg_copy[i].arg->in); } else if (arg_copy[i].arg->num_bits <= 64) { - temp = tcg_temp_local_new_i64(); + temp = tcg_temp_new_i64(); tcg_gen_mov_i64(temp, arg_copy[i].arg->in); } else { g_assert_not_reached(); @@ -XXX,XX +XXX,XX @@ static void xtensa_tr_tb_start(DisasContextBase *dcbase, CPUState *cpu) DisasContext *dc = container_of(dcbase, DisasContext, base); if (dc->icount) { - dc->next_icount = tcg_temp_local_new_i32(); + dc->next_icount = tcg_temp_new_i32(); } } @@ -XXX,XX +XXX,XX @@ static void gen_check_atomctl(DisasContext *dc, TCGv_i32 addr) static void translate_s32c1i(DisasContext *dc, const OpcodeArg arg[], const uint32_t par[]) { - TCGv_i32 tmp = tcg_temp_local_new_i32(); - TCGv_i32 addr = tcg_temp_local_new_i32(); + TCGv_i32 tmp = tcg_temp_new_i32(); + TCGv_i32 addr = tcg_temp_new_i32(); MemOp mop; tcg_gen_mov_i32(tmp, arg[0].in); @@ -XXX,XX +XXX,XX @@ static void translate_s32ex(DisasContext *dc, const OpcodeArg arg[], const uint32_t par[]) { TCGv_i32 prev = tcg_temp_new_i32(); - TCGv_i32 addr = tcg_temp_local_new_i32(); - TCGv_i32 res = tcg_temp_local_new_i32(); + TCGv_i32 addr = tcg_temp_new_i32(); + TCGv_i32 res = tcg_temp_new_i32(); TCGLabel *label = gen_new_label(); MemOp mop; -- 2.34.1 Since tcg_temp_new_i32 is now identical, use that. Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- include/exec/gen-icount.h | 8 +------- 1 file changed, 1 insertion(+), 7 deletions(-) diff --git a/include/exec/gen-icount.h b/include/exec/gen-icount.h index XXXXXXX..XXXXXXX 100644 --- a/include/exec/gen-icount.h +++ b/include/exec/gen-icount.h @@ -XXX,XX +XXX,XX @@ static inline void gen_io_start(void) static inline void gen_tb_start(const TranslationBlock *tb) { - TCGv_i32 count; - - if (tb_cflags(tb) & CF_USE_ICOUNT) { - count = tcg_temp_local_new_i32(); - } else { - count = tcg_temp_new_i32(); - } + TCGv_i32 count = tcg_temp_new_i32(); tcg_gen_ld_i32(count, cpu_env, offsetof(ArchCPU, neg.icount_decr.u32) - -- 2.34.1 These symbols are now unused. Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- include/tcg/tcg-op.h | 2 -- include/tcg/tcg.h | 28 ---------------------------- tcg/tcg.c | 16 ---------------- 3 files changed, 46 deletions(-) diff --git a/include/tcg/tcg-op.h b/include/tcg/tcg-op.h index XXXXXXX..XXXXXXX 100644 --- a/include/tcg/tcg-op.h +++ b/include/tcg/tcg-op.h @@ -XXX,XX +XXX,XX @@ static inline void tcg_gen_plugin_cb_end(void) #if TARGET_LONG_BITS == 32 #define tcg_temp_new() tcg_temp_new_i32() #define tcg_global_mem_new tcg_global_mem_new_i32 -#define tcg_temp_local_new() tcg_temp_local_new_i32() #define tcg_temp_free tcg_temp_free_i32 #define tcg_gen_qemu_ld_tl tcg_gen_qemu_ld_i32 #define tcg_gen_qemu_st_tl tcg_gen_qemu_st_i32 #else #define tcg_temp_new() tcg_temp_new_i64() #define tcg_global_mem_new tcg_global_mem_new_i64 -#define tcg_temp_local_new() tcg_temp_local_new_i64() #define tcg_temp_free tcg_temp_free_i64 #define tcg_gen_qemu_ld_tl tcg_gen_qemu_ld_i64 #define tcg_gen_qemu_st_tl tcg_gen_qemu_st_i64 diff --git a/include/tcg/tcg.h b/include/tcg/tcg.h index XXXXXXX..XXXXXXX 100644 --- a/include/tcg/tcg.h +++ b/include/tcg/tcg.h @@ -XXX,XX +XXX,XX @@ static inline TCGv_i32 tcg_temp_new_i32(void) return temp_tcgv_i32(t); } -static inline TCGv_i32 tcg_temp_local_new_i32(void) -{ - TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_I32, TEMP_TB); - return temp_tcgv_i32(t); -} - static inline TCGv_i64 tcg_global_mem_new_i64(TCGv_ptr reg, intptr_t offset, const char *name) { @@ -XXX,XX +XXX,XX @@ static inline TCGv_i64 tcg_temp_new_i64(void) return temp_tcgv_i64(t); } -static inline TCGv_i64 tcg_temp_local_new_i64(void) -{ - TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_I64, TEMP_TB); - return temp_tcgv_i64(t); -} - /* Used only by tcg infrastructure: tcg-op.c or plugin-gen.c */ static inline TCGv_i128 tcg_temp_ebb_new_i128(void) { @@ -XXX,XX +XXX,XX @@ static inline TCGv_i128 tcg_temp_new_i128(void) return temp_tcgv_i128(t); } -static inline TCGv_i128 tcg_temp_local_new_i128(void) -{ - TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_I128, TEMP_TB); - return temp_tcgv_i128(t); -} - static inline TCGv_ptr tcg_global_mem_new_ptr(TCGv_ptr reg, intptr_t offset, const char *name) { @@ -XXX,XX +XXX,XX @@ static inline TCGv_ptr tcg_temp_new_ptr(void) return temp_tcgv_ptr(t); } -static inline TCGv_ptr tcg_temp_local_new_ptr(void) -{ - TCGTemp *t = tcg_temp_new_internal(TCG_TYPE_PTR, TEMP_TB); - return temp_tcgv_ptr(t); -} - #if defined(CONFIG_DEBUG_TCG) /* If you call tcg_clear_temp_count() at the start of a section of * code which is not supposed to leak any TCG temporaries, then @@ -XXX,XX +XXX,XX @@ void tcg_optimize(TCGContext *s); /* Allocate a new temporary and initialize it with a constant. */ TCGv_i32 tcg_const_i32(int32_t val); TCGv_i64 tcg_const_i64(int64_t val); -TCGv_i32 tcg_const_local_i32(int32_t val); -TCGv_i64 tcg_const_local_i64(int64_t val); TCGv_vec tcg_const_zeros_vec(TCGType); TCGv_vec tcg_const_ones_vec(TCGType); TCGv_vec tcg_const_zeros_vec_matching(TCGv_vec); @@ -XXX,XX +XXX,XX @@ TCGv_vec tcg_constant_vec_matching(TCGv_vec match, unsigned vece, int64_t val); #if UINTPTR_MAX == UINT32_MAX # define tcg_const_ptr(x) ((TCGv_ptr)tcg_const_i32((intptr_t)(x))) -# define tcg_const_local_ptr(x) ((TCGv_ptr)tcg_const_local_i32((intptr_t)(x))) # define tcg_constant_ptr(x) ((TCGv_ptr)tcg_constant_i32((intptr_t)(x))) #else # define tcg_const_ptr(x) ((TCGv_ptr)tcg_const_i64((intptr_t)(x))) -# define tcg_const_local_ptr(x) ((TCGv_ptr)tcg_const_local_i64((intptr_t)(x))) # define tcg_constant_ptr(x) ((TCGv_ptr)tcg_constant_i64((intptr_t)(x))) #endif diff --git a/tcg/tcg.c b/tcg/tcg.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/tcg.c +++ b/tcg/tcg.c @@ -XXX,XX +XXX,XX @@ TCGv_i64 tcg_const_i64(int64_t val) return t0; } -TCGv_i32 tcg_const_local_i32(int32_t val) -{ - TCGv_i32 t0; - t0 = tcg_temp_local_new_i32(); - tcg_gen_movi_i32(t0, val); - return t0; -} - -TCGv_i64 tcg_const_local_i64(int64_t val) -{ - TCGv_i64 t0; - t0 = tcg_temp_local_new_i64(); - tcg_gen_movi_i64(t0, val); - return t0; -} - #if defined(CONFIG_DEBUG_TCG) void tcg_clear_temp_count(void) { -- 2.34.1 Rewrite the sections which talked about 'local temporaries'. Remove some assumptions which no longer hold. Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- docs/devel/tcg-ops.rst | 230 +++++++++++++++++++++++------------------ 1 file changed, 129 insertions(+), 101 deletions(-) diff --git a/docs/devel/tcg-ops.rst b/docs/devel/tcg-ops.rst index XXXXXXX..XXXXXXX 100644 --- a/docs/devel/tcg-ops.rst +++ b/docs/devel/tcg-ops.rst @@ -XXX,XX +XXX,XX @@ TCG Intermediate Representation Introduction ============ -TCG (Tiny Code Generator) began as a generic backend for a C -compiler. It was simplified to be used in QEMU. It also has its roots -in the QOP code generator written by Paul Brook. +TCG (Tiny Code Generator) began as a generic backend for a C compiler. +It was simplified to be used in QEMU. It also has its roots in the +QOP code generator written by Paul Brook. Definitions =========== -TCG receives RISC-like *TCG ops* and performs some optimizations on them, -including liveness analysis and trivial constant expression -evaluation. TCG ops are then implemented in the host CPU back end, -also known as the TCG target. - -The TCG *target* is the architecture for which we generate the -code. It is of course not the same as the "target" of QEMU which is -the emulated architecture. As TCG started as a generic C backend used -for cross compiling, it is assumed that the TCG target is different -from the host, although it is never the case for QEMU. +The TCG *target* is the architecture for which we generate the code. +It is of course not the same as the "target" of QEMU which is the +emulated architecture. As TCG started as a generic C backend used +for cross compiling, the assumption was that TCG target might be +different from the host, although this is never the case for QEMU. In this document, we use *guest* to specify what architecture we are emulating; *target* always means the TCG target, the machine on which we are running QEMU. -A TCG *function* corresponds to a QEMU Translated Block (TB). - -A TCG *temporary* is a variable only live in a basic block. Temporaries are allocated explicitly in each function. - -A TCG *local temporary* is a variable only live in a function. Local temporaries are allocated explicitly in each function. - -A TCG *global* is a variable which is live in all the functions -(equivalent of a C global variable). They are defined before the -functions defined. A TCG global can be a memory location (e.g. a QEMU -CPU register), a fixed host register (e.g. the QEMU CPU state pointer) -or a memory location which is stored in a register outside QEMU TBs -(not implemented yet). - -A TCG *basic block* corresponds to a list of instructions terminated -by a branch instruction. - An operation with *undefined behavior* may result in a crash. An operation with *unspecified behavior* shall not crash. However, the result may be one of several possibilities so may be considered an *undefined result*. -Intermediate representation -=========================== +Basic Blocks +============ -Introduction ------------- +A TCG *basic block* is a single entry, multiple exit region which +corresponds to a list of instructions terminated by a label, or +any branch instruction. -TCG instructions operate on variables which are temporaries, local -temporaries or globals. TCG instructions and variables are strongly -typed. Two types are supported: 32 bit integers and 64 bit -integers. Pointers are defined as an alias to 32 bit or 64 bit -integers depending on the TCG target word size. +A TCG *extended basic block* is a single entry, multiple exit region +which corresponds to a list of instructions terminated by a label or +an unconditional branch. Specifically, an extended basic block is +a sequence of basic blocks connected by the fall-through paths of +zero or more conditional branch instructions. -Each instruction has a fixed number of output variable operands, input -variable operands and always constant operands. +Operations +========== -The notable exception is the call instruction which has a variable -number of outputs and inputs. +TCG instructions or *ops* operate on TCG *variables*, both of which +are strongly typed. Each instruction has a fixed number of output +variable operands, input variable operands and constant operands. +Vector instructions have a field specifying the element size within +the vector. The notable exception is the call instruction which has +a variable number of outputs and inputs. In the textual form, output operands usually come first, followed by input operands, followed by constant operands. The output type is @@ -XXX,XX +XXX,XX @@ included in the instruction name. Constants are prefixed with a '$'. add_i32 t0, t1, t2 /* (t0 <- t1 + t2) */ +Variables +========= -Assumptions ------------ +* ``TEMP_FIXED`` -Basic blocks -^^^^^^^^^^^^ + There is one TCG *fixed global* variable, ``cpu_env``, which is + live in all translation blocks, and holds a pointer to ``CPUArchState``. + This variable is held in a host cpu register at all times in all + translation blocks. -* Basic blocks end after branches (e.g. brcond_i32 instruction), - goto_tb and exit_tb instructions. +* ``TEMP_GLOBAL`` -* Basic blocks start after the end of a previous basic block, or at a - set_label instruction. + A TCG *global* is a variable which is live in all translation blocks, + and corresponds to memory location that is within ``CPUArchState``. + These may be specified as an offset from ``cpu_env``, in which case + they are called *direct globals*, or may be specified as an offset + from a direct global, in which case they are called *indirect globals*. + Even indirect globals should still reference memory within + ``CPUArchState``. All TCG globals are defined during + ``TCGCPUOps.initialize``, before any translation blocks are generated. -After the end of a basic block, the content of temporaries is -destroyed, but local temporaries and globals are preserved. +* ``TEMP_CONST`` -Floating point types -^^^^^^^^^^^^^^^^^^^^ + A TCG *constant* is a variable which is live throughout the entire + translation block, and contains a constant value. These variables + are allocated on demand during translation and are hashed so that + there is exactly one variable holding a given value. -* Floating point types are not supported yet +* ``TEMP_TB`` -Pointers -^^^^^^^^ + A TCG *translation block temporary* is a variable which is live + throughout the entire translation block, but dies on any exit. + These temporaries are allocated explicitly during translation. -* Depending on the TCG target, pointer size is 32 bit or 64 - bit. The type ``TCG_TYPE_PTR`` is an alias to ``TCG_TYPE_I32`` or - ``TCG_TYPE_I64``. +* ``TEMP_EBB`` + + A TCG *extended basic block temporary* is a variable which is live + throughout an extended basic block, but dies on any exit. + These temporaries are allocated explicitly during translation. + +Types +===== + +* ``TCG_TYPE_I32`` + + A 32-bit integer. + +* ``TCG_TYPE_I64`` + + A 64-bit integer. For 32-bit hosts, such variables are split into a pair + of variables with ``type=TCG_TYPE_I32`` and ``base_type=TCG_TYPE_I64``. + The ``temp_subindex`` for each indicates where it falls within the + host-endian representation. + +* ``TCG_TYPE_PTR`` + + An alias for ``TCG_TYPE_I32`` or ``TCG_TYPE_I64``, depending on the size + of a pointer for the host. + +* ``TCG_TYPE_REG`` + + An alias for ``TCG_TYPE_I32`` or ``TCG_TYPE_I64``, depending on the size + of the integer registers for the host. This may be larger + than ``TCG_TYPE_PTR`` depending on the host ABI. + +* ``TCG_TYPE_I128`` + + A 128-bit integer. For all hosts, such variables are split into a number + of variables with ``type=TCG_TYPE_REG`` and ``base_type=TCG_TYPE_I128``. + The ``temp_subindex`` for each indicates where it falls within the + host-endian representation. + +* ``TCG_TYPE_V64`` + + A 64-bit vector. This type is valid only if the TCG target + sets ``TCG_TARGET_HAS_v64``. + +* ``TCG_TYPE_V128`` + + A 128-bit vector. This type is valid only if the TCG target + sets ``TCG_TARGET_HAS_v128``. + +* ``TCG_TYPE_V256`` + + A 256-bit vector. This type is valid only if the TCG target + sets ``TCG_TARGET_HAS_v256``. Helpers -^^^^^^^ +======= -* Using the tcg_gen_helper_x_y it is possible to call any function - taking i32, i64 or pointer types. By default, before calling a helper, - all globals are stored at their canonical location and it is assumed - that the function can modify them. By default, the helper is allowed to - modify the CPU state or raise an exception. +Helpers are registered in a guest-specific ``helper.h``, +which is processed to generate ``tcg_gen_helper_*`` functions. +With these functions it is possible to call a function taking +i32, i64, i128 or pointer types. - This can be overridden using the following function modifiers: +By default, before calling a helper, all globals are stored at their +canonical location. By default, the helper is allowed to modify the +CPU state (including the state represented by tcg globals) +or may raise an exception. This default can be overridden using the +following function modifiers: - - ``TCG_CALL_NO_READ_GLOBALS`` means that the helper does not read globals, - either directly or via an exception. They will not be saved to their - canonical locations before calling the helper. +* ``TCG_CALL_NO_WRITE_GLOBALS`` - - ``TCG_CALL_NO_WRITE_GLOBALS`` means that the helper does not modify any globals. - They will only be saved to their canonical location before calling helpers, - but they won't be reloaded afterwards. + The helper does not modify any globals, but may read them. + Globals will be saved to their canonical location before calling helpers, + but need not be reloaded afterwards. - - ``TCG_CALL_NO_SIDE_EFFECTS`` means that the call to the function is removed if - the return value is not used. +* ``TCG_CALL_NO_READ_GLOBALS`` - Note that ``TCG_CALL_NO_READ_GLOBALS`` implies ``TCG_CALL_NO_WRITE_GLOBALS``. + The helper does not read globals, either directly or via an exception. + They will not be saved to their canonical locations before calling + the helper. This implies ``TCG_CALL_NO_WRITE_GLOBALS``. - On some TCG targets (e.g. x86), several calling conventions are - supported. +* ``TCG_CALL_NO_SIDE_EFFECTS`` -Branches -^^^^^^^^ - -* Use the instruction 'br' to jump to a label. + The call to the helper function may be removed if the return value is + not used. This means that it may not modify any CPU state nor may it + raise an exception. Code Optimizations ------------------- +================== When generating instructions, you can count on at least the following optimizations: @@ -XXX,XX +XXX,XX @@ Recommended coding rules for best performance often modified, e.g. the integer registers and the condition codes. TCG will be able to use host registers to store them. -- Avoid globals stored in fixed registers. They must be used only to - store the pointer to the CPU state and possibly to store a pointer - to a register window. - -- Use temporaries. Use local temporaries only when really needed, - e.g. when you need to use a value after a jump. Local temporaries - introduce a performance hit in the current TCG implementation: their - content is saved to memory at end of each basic block. - -- Free temporaries and local temporaries when they are no longer used - (tcg_temp_free). Since tcg_const_x() also creates a temporary, you - should free it after it is used. Freeing temporaries does not yield - a better generated code, but it reduces the memory usage of TCG and - the speed of the translation. +- Free temporaries when they are no longer used (``tcg_temp_free``). + Since ``tcg_const_x`` also creates a temporary, you should free it + after it is used. - Don't hesitate to use helpers for complicated or seldom used guest instructions. There is little performance advantage in using TCG to @@ -XXX,XX +XXX,XX @@ Recommended coding rules for best performance the instruction is mostly doing loads and stores, and in those cases inline TCG may still be faster for longer sequences. -- The hard limit on the number of TCG instructions you can generate - per guest instruction is set by ``MAX_OP_PER_INSTR`` in ``exec-all.h`` -- - you cannot exceed this without risking a buffer overrun. - - Use the 'discard' instruction if you know that TCG won't be able to prove that a given global is "dead" at a given program point. The x86 guest uses it to improve the condition codes optimisation. -- 2.34.1 The following changes since commit aa3a285b5bc56a4208b3b57d4a55291e9c260107: Merge tag 'mem-2024-12-21' of https://github.com/davidhildenbrand/qemu into staging (2024-12-22 14:33:27 -0500) are available in the Git repository at: https://gitlab.com/rth7680/qemu.git tags/pull-tcg-20241224 for you to fetch changes up to e4a8e093dc74be049f4829831dce76e5edab0003: accel/tcg: Move gen_intermediate_code to TCGCPUOps.translate_core (2024-12-24 08:32:15 -0800) ---------------------------------------------------------------- tcg/optimize: Remove in-flight mask data from OptContext fpu: Add float*_muladd_scalbn fpu: Remove float_muladd_halve_result fpu: Add float_round_nearest_even_max fpu: Add float_muladd_suppress_add_product_zero target/hexagon: Use float32_muladd accel/tcg: Move gen_intermediate_code to TCGCPUOps.translate_core ---------------------------------------------------------------- Ilya Leoshkevich (1): tests/tcg: Do not use inttypes.h in multiarch/system/memory.c Pierrick Bouvier (1): plugins: optimize cpu_index code generation Richard Henderson (70): tcg/optimize: Split out finish_bb, finish_ebb tcg/optimize: Split out fold_affected_mask tcg/optimize: Copy mask writeback to fold_masks tcg/optimize: Split out fold_masks_zs tcg/optimize: Augment s_mask from z_mask in fold_masks_zs tcg/optimize: Change representation of s_mask tcg/optimize: Use finish_folding in fold_add, fold_add_vec, fold_addsub2 tcg/optimize: Introduce const value accessors for TempOptInfo tcg/optimize: Use fold_masks_zs in fold_and tcg/optimize: Use fold_masks_zs in fold_andc tcg/optimize: Use fold_masks_zs in fold_bswap tcg/optimize: Use fold_masks_zs in fold_count_zeros tcg/optimize: Use fold_masks_z in fold_ctpop tcg/optimize: Use fold_and and fold_masks_z in fold_deposit tcg/optimize: Compute sign mask in fold_deposit tcg/optimize: Use finish_folding in fold_divide tcg/optimize: Use finish_folding in fold_dup, fold_dup2 tcg/optimize: Use fold_masks_s in fold_eqv tcg/optimize: Use fold_masks_z in fold_extract tcg/optimize: Use finish_folding in fold_extract2 tcg/optimize: Use fold_masks_zs in fold_exts tcg/optimize: Use fold_masks_z in fold_extu tcg/optimize: Use fold_masks_zs in fold_movcond tcg/optimize: Use finish_folding in fold_mul* tcg/optimize: Use fold_masks_s in fold_nand tcg/optimize: Use fold_masks_z in fold_neg_no_const tcg/optimize: Use fold_masks_s in fold_nor tcg/optimize: Use fold_masks_s in fold_not tcg/optimize: Use fold_masks_zs in fold_or tcg/optimize: Use fold_masks_zs in fold_orc tcg/optimize: Use fold_masks_zs in fold_qemu_ld tcg/optimize: Return true from fold_qemu_st, fold_tcg_st tcg/optimize: Use finish_folding in fold_remainder tcg/optimize: Distinguish simplification in fold_setcond_zmask tcg/optimize: Use fold_masks_z in fold_setcond tcg/optimize: Use fold_masks_s in fold_negsetcond tcg/optimize: Use fold_masks_z in fold_setcond2 tcg/optimize: Use finish_folding in fold_cmp_vec tcg/optimize: Use finish_folding in fold_cmpsel_vec tcg/optimize: Use fold_masks_zs in fold_sextract tcg/optimize: Use fold_masks_zs, fold_masks_s in fold_shift tcg/optimize: Simplify sign bit test in fold_shift tcg/optimize: Use finish_folding in fold_sub, fold_sub_vec tcg/optimize: Use fold_masks_zs in fold_tcg_ld tcg/optimize: Use finish_folding in fold_tcg_ld_memcopy tcg/optimize: Use fold_masks_zs in fold_xor tcg/optimize: Use finish_folding in fold_bitsel_vec tcg/optimize: Use finish_folding as default in tcg_optimize tcg/optimize: Remove z_mask, s_mask from OptContext tcg/optimize: Re-enable sign-mask optimizations tcg/optimize: Move fold_bitsel_vec into alphabetic sort tcg/optimize: Move fold_cmp_vec, fold_cmpsel_vec into alphabetic sort softfloat: Add float{16,32,64}_muladd_scalbn target/arm: Use float*_muladd_scalbn target/sparc: Use float*_muladd_scalbn softfloat: Remove float_muladd_halve_result softfloat: Add float_round_nearest_even_max softfloat: Add float_muladd_suppress_add_product_zero target/hexagon: Use float32_mul in helper_sfmpy target/hexagon: Use float32_muladd for helper_sffma target/hexagon: Use float32_muladd for helper_sffms target/hexagon: Use float32_muladd_scalbn for helper_sffma_sc target/hexagon: Use float32_muladd for helper_sffm[as]_lib target/hexagon: Remove internal_fmafx target/hexagon: Expand GEN_XF_ROUND target/hexagon: Remove Float target/hexagon: Remove Double target/hexagon: Use mulu64 for int128_mul_6464 target/hexagon: Simplify internal_mpyhh setup accel/tcg: Move gen_intermediate_code to TCGCPUOps.translate_core include/exec/translator.h | 14 - include/fpu/softfloat-types.h | 2 + include/fpu/softfloat.h | 14 +- include/hw/core/tcg-cpu-ops.h | 13 + target/alpha/cpu.h | 2 + target/arm/internals.h | 2 + target/avr/cpu.h | 2 + target/hexagon/cpu.h | 2 + target/hexagon/fma_emu.h | 3 - target/hppa/cpu.h | 2 + target/i386/tcg/helper-tcg.h | 2 + target/loongarch/internals.h | 2 + target/m68k/cpu.h | 2 + target/microblaze/cpu.h | 2 + target/mips/tcg/tcg-internal.h | 2 + target/openrisc/cpu.h | 2 + target/ppc/cpu.h | 2 + target/riscv/cpu.h | 3 + target/rx/cpu.h | 2 + target/s390x/s390x-internal.h | 2 + target/sh4/cpu.h | 2 + target/sparc/cpu.h | 2 + target/sparc/helper.h | 4 +- target/tricore/cpu.h | 2 + target/xtensa/cpu.h | 2 + accel/tcg/cpu-exec.c | 8 +- accel/tcg/plugin-gen.c | 9 + accel/tcg/translate-all.c | 8 +- fpu/softfloat.c | 63 +-- target/alpha/cpu.c | 1 + target/alpha/translate.c | 4 +- target/arm/cpu.c | 1 + target/arm/tcg/cpu-v7m.c | 1 + target/arm/tcg/helper-a64.c | 6 +- target/arm/tcg/translate.c | 5 +- target/avr/cpu.c | 1 + target/avr/translate.c | 6 +- target/hexagon/cpu.c | 1 + target/hexagon/fma_emu.c | 496 ++++++--------------- target/hexagon/op_helper.c | 125 ++---- target/hexagon/translate.c | 4 +- target/hppa/cpu.c | 1 + target/hppa/translate.c | 4 +- target/i386/tcg/tcg-cpu.c | 1 + target/i386/tcg/translate.c | 5 +- target/loongarch/cpu.c | 1 + target/loongarch/tcg/translate.c | 4 +- target/m68k/cpu.c | 1 + target/m68k/translate.c | 4 +- target/microblaze/cpu.c | 1 + target/microblaze/translate.c | 4 +- target/mips/cpu.c | 1 + target/mips/tcg/translate.c | 4 +- target/openrisc/cpu.c | 1 + target/openrisc/translate.c | 4 +- target/ppc/cpu_init.c | 1 + target/ppc/translate.c | 4 +- target/riscv/tcg/tcg-cpu.c | 1 + target/riscv/translate.c | 4 +- target/rx/cpu.c | 1 + target/rx/translate.c | 4 +- target/s390x/cpu.c | 1 + target/s390x/tcg/translate.c | 4 +- target/sh4/cpu.c | 1 + target/sh4/translate.c | 4 +- target/sparc/cpu.c | 1 + target/sparc/fop_helper.c | 8 +- target/sparc/translate.c | 84 ++-- target/tricore/cpu.c | 1 + target/tricore/translate.c | 5 +- target/xtensa/cpu.c | 1 + target/xtensa/translate.c | 4 +- tcg/optimize.c | 857 +++++++++++++++++++----------------- tests/tcg/multiarch/system/memory.c | 9 +- fpu/softfloat-parts.c.inc | 16 +- 75 files changed, 866 insertions(+), 1009 deletions(-) From: Ilya Leoshkevich <iii@linux.ibm.com> make check-tcg fails on Fedora with the following error message: alpha-linux-gnu-gcc [...] qemu/tests/tcg/multiarch/system/memory.c -o memory [...] qemu/tests/tcg/multiarch/system/memory.c:17:10: fatal error: inttypes.h: No such file or directory 17 | #include <inttypes.h> | ^~~~~~~~~~~~ compilation terminated. The reason is that Fedora has cross-compilers, but no cross-glibc headers. Fix by hardcoding the format specifiers and dropping the include. An alternative fix would be to introduce a configure check for inttypes.h. But this would make it impossible to use Fedora cross-compilers for softmmu tests, which used to work so far. Fixes: ecbcc9ead2f8 ("tests/tcg: add a system test to check memory instrumentation") Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com> Reviewed-by: Paolo Bonzini <pbonzini@redhat.com> Message-ID: <20241010085906.226249-1-iii@linux.ibm.com> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tests/tcg/multiarch/system/memory.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) diff --git a/tests/tcg/multiarch/system/memory.c b/tests/tcg/multiarch/system/memory.c index XXXXXXX..XXXXXXX 100644 --- a/tests/tcg/multiarch/system/memory.c +++ b/tests/tcg/multiarch/system/memory.c @@ -XXX,XX +XXX,XX @@ #include <stdint.h> #include <stdbool.h> -#include <inttypes.h> #include <minilib.h> #ifndef CHECK_UNALIGNED @@ -XXX,XX +XXX,XX @@ int main(void) int i; bool ok = true; - ml_printf("Test data start: 0x%"PRIxPTR"\n", &test_data[0]); - ml_printf("Test data end: 0x%"PRIxPTR"\n", &test_data[TEST_SIZE]); + ml_printf("Test data start: 0x%lx\n", (unsigned long)&test_data[0]); + ml_printf("Test data end: 0x%lx\n", (unsigned long)&test_data[TEST_SIZE]); /* Run through the unsigned tests first */ for (i = 0; i < ARRAY_SIZE(init_ufns) && ok; i++) { @@ -XXX,XX +XXX,XX @@ int main(void) ok = do_signed_reads(true); } - ml_printf("Test data read: %"PRId32"\n", test_read_count); - ml_printf("Test data write: %"PRId32"\n", test_write_count); + ml_printf("Test data read: %lu\n", (unsigned long)test_read_count); + ml_printf("Test data write: %lu\n", (unsigned long)test_write_count); ml_printf("Test complete: %s\n", ok ? "PASSED" : "FAILED"); return ok ? 0 : -1; } -- 2.43.0 From: Pierrick Bouvier <pierrick.bouvier@linaro.org> When running with a single vcpu, we can return a constant instead of a load when accessing cpu_index. A side effect is that all tcg operations using it are optimized, most notably scoreboard access. When running a simple loop in user-mode, the speedup is around 20%. Signed-off-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-ID: <20241128213843.1023080-1-pierrick.bouvier@linaro.org> --- accel/tcg/plugin-gen.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/accel/tcg/plugin-gen.c b/accel/tcg/plugin-gen.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/plugin-gen.c +++ b/accel/tcg/plugin-gen.c @@ -XXX,XX +XXX,XX @@ static void gen_disable_mem_helper(void) static TCGv_i32 gen_cpu_index(void) { + /* + * Optimize when we run with a single vcpu. All values using cpu_index, + * including scoreboard index, will be optimized out. + * User-mode calls tb_flush when setting this flag. In system-mode, all + * vcpus are created before generating code. + */ + if (!tcg_cflags_has(current_cpu, CF_PARALLEL)) { + return tcg_constant_i32(current_cpu->cpu_index); + } TCGv_i32 cpu_index = tcg_temp_ebb_new_i32(); tcg_gen_ld_i32(cpu_index, tcg_env, -offsetof(ArchCPU, env) + offsetof(CPUState, cpu_index)); -- 2.43.0 Call them directly from the opcode switch statement in tcg_optimize, rather than in finish_folding based on opcode flags. Adjust folding of conditional branches to match. Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 47 +++++++++++++++++++++++++++++++---------------- 1 file changed, 31 insertions(+), 16 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static void copy_propagate(OptContext *ctx, TCGOp *op, } } +static void finish_bb(OptContext *ctx) +{ + /* We only optimize memory barriers across basic blocks. */ + ctx->prev_mb = NULL; +} + +static void finish_ebb(OptContext *ctx) +{ + finish_bb(ctx); + /* We only optimize across extended basic blocks. */ + memset(&ctx->temps_used, 0, sizeof(ctx->temps_used)); + remove_mem_copy_all(ctx); +} + static void finish_folding(OptContext *ctx, TCGOp *op) { const TCGOpDef *def = &tcg_op_defs[op->opc]; int i, nb_oargs; - /* - * We only optimize extended basic blocks. If the opcode ends a BB - * and is not a conditional branch, reset all temp data. - */ - if (def->flags & TCG_OPF_BB_END) { - ctx->prev_mb = NULL; - if (!(def->flags & TCG_OPF_COND_BRANCH)) { - memset(&ctx->temps_used, 0, sizeof(ctx->temps_used)); - remove_mem_copy_all(ctx); - } - return; - } - nb_oargs = def->nb_oargs; for (i = 0; i < nb_oargs; i++) { TCGTemp *ts = arg_temp(op->args[i]); @@ -XXX,XX +XXX,XX @@ static bool fold_brcond(OptContext *ctx, TCGOp *op) if (i > 0) { op->opc = INDEX_op_br; op->args[0] = op->args[3]; + finish_ebb(ctx); + } else { + finish_bb(ctx); } - return false; + return true; } static bool fold_brcond2(OptContext *ctx, TCGOp *op) @@ -XXX,XX +XXX,XX @@ static bool fold_brcond2(OptContext *ctx, TCGOp *op) } op->opc = INDEX_op_br; op->args[0] = label; - break; + finish_ebb(ctx); + return true; } - return false; + + finish_bb(ctx); + return true; } static bool fold_bswap(OptContext *ctx, TCGOp *op) @@ -XXX,XX +XXX,XX @@ void tcg_optimize(TCGContext *s) CASE_OP_32_64_VEC(xor): done = fold_xor(&ctx, op); break; + case INDEX_op_set_label: + case INDEX_op_br: + case INDEX_op_exit_tb: + case INDEX_op_goto_tb: + case INDEX_op_goto_ptr: + finish_ebb(&ctx); + done = true; + break; default: break; } -- 2.43.0 There are only a few logical operations which can compute an "affected" mask. Split out handling of this optimization to a separate function, only to be called when applicable. Remove the a_mask field from OptContext, as the mask is no longer stored anywhere. Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 42 +++++++++++++++++++++++++++--------------- 1 file changed, 27 insertions(+), 15 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ typedef struct OptContext { QSIMPLEQ_HEAD(, MemCopyInfo) mem_free; /* In flight values from optimization. */ - uint64_t a_mask; /* mask bit is 0 iff value identical to first input */ uint64_t z_mask; /* mask bit is 0 iff value bit is 0 */ uint64_t s_mask; /* mask of clrsb(value) bits */ TCGType type; @@ -XXX,XX +XXX,XX @@ static bool fold_const2_commutative(OptContext *ctx, TCGOp *op) static bool fold_masks(OptContext *ctx, TCGOp *op) { - uint64_t a_mask = ctx->a_mask; uint64_t z_mask = ctx->z_mask; uint64_t s_mask = ctx->s_mask; @@ -XXX,XX +XXX,XX @@ static bool fold_masks(OptContext *ctx, TCGOp *op) * type changing opcodes. */ if (ctx->type == TCG_TYPE_I32) { - a_mask = (int32_t)a_mask; z_mask = (int32_t)z_mask; s_mask |= MAKE_64BIT_MASK(32, 32); ctx->z_mask = z_mask; @@ -XXX,XX +XXX,XX @@ static bool fold_masks(OptContext *ctx, TCGOp *op) if (z_mask == 0) { return tcg_opt_gen_movi(ctx, op, op->args[0], 0); } + return false; +} + +/* + * An "affected" mask bit is 0 if and only if the result is identical + * to the first input. Thus if the entire mask is 0, the operation + * is equivalent to a copy. + */ +static bool fold_affected_mask(OptContext *ctx, TCGOp *op, uint64_t a_mask) +{ + if (ctx->type == TCG_TYPE_I32) { + a_mask = (uint32_t)a_mask; + } if (a_mask == 0) { return tcg_opt_gen_mov(ctx, op, op->args[0], op->args[1]); } @@ -XXX,XX +XXX,XX @@ static bool fold_and(OptContext *ctx, TCGOp *op) * Known-zeros does not imply known-ones. Therefore unless * arg2 is constant, we can't infer affected bits from it. */ - if (arg_is_const(op->args[2])) { - ctx->a_mask = z1 & ~z2; + if (arg_is_const(op->args[2]) && + fold_affected_mask(ctx, op, z1 & ~z2)) { + return true; } return fold_masks(ctx, op); @@ -XXX,XX +XXX,XX @@ static bool fold_andc(OptContext *ctx, TCGOp *op) */ if (arg_is_const(op->args[2])) { uint64_t z2 = ~arg_info(op->args[2])->z_mask; - ctx->a_mask = z1 & ~z2; + if (fold_affected_mask(ctx, op, z1 & ~z2)) { + return true; + } z1 &= z2; } ctx->z_mask = z1; @@ -XXX,XX +XXX,XX @@ static bool fold_extract(OptContext *ctx, TCGOp *op) z_mask_old = arg_info(op->args[1])->z_mask; z_mask = extract64(z_mask_old, pos, len); - if (pos == 0) { - ctx->a_mask = z_mask_old ^ z_mask; + if (pos == 0 && fold_affected_mask(ctx, op, z_mask_old ^ z_mask)) { + return true; } ctx->z_mask = z_mask; ctx->s_mask = smask_from_zmask(z_mask); @@ -XXX,XX +XXX,XX @@ static bool fold_exts(OptContext *ctx, TCGOp *op) ctx->z_mask = z_mask; ctx->s_mask = s_mask; - if (!type_change) { - ctx->a_mask = s_mask & ~s_mask_old; + if (!type_change && fold_affected_mask(ctx, op, s_mask & ~s_mask_old)) { + return true; } return fold_masks(ctx, op); @@ -XXX,XX +XXX,XX @@ static bool fold_extu(OptContext *ctx, TCGOp *op) ctx->z_mask = z_mask; ctx->s_mask = smask_from_zmask(z_mask); - if (!type_change) { - ctx->a_mask = z_mask_old ^ z_mask; + if (!type_change && fold_affected_mask(ctx, op, z_mask_old ^ z_mask)) { + return true; } return fold_masks(ctx, op); } @@ -XXX,XX +XXX,XX @@ static bool fold_sextract(OptContext *ctx, TCGOp *op) s_mask |= MAKE_64BIT_MASK(len, 64 - len); ctx->s_mask = s_mask; - if (pos == 0) { - ctx->a_mask = s_mask & ~s_mask_old; + if (pos == 0 && fold_affected_mask(ctx, op, s_mask & ~s_mask_old)) { + return true; } return fold_masks(ctx, op); @@ -XXX,XX +XXX,XX @@ void tcg_optimize(TCGContext *s) } /* Assume all bits affected, no bits known zero, no sign reps. */ - ctx.a_mask = -1; ctx.z_mask = -1; ctx.s_mask = 0; -- 2.43.0 Use of fold_masks should be restricted to those opcodes that can reliably make use of it -- those with a single output, and from higher-level folders that set up the masks. Prepare for conversion of each folder in turn. Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 17 ++++++++++++++--- 1 file changed, 14 insertions(+), 3 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_masks(OptContext *ctx, TCGOp *op) { uint64_t z_mask = ctx->z_mask; uint64_t s_mask = ctx->s_mask; + const TCGOpDef *def = &tcg_op_defs[op->opc]; + TCGTemp *ts; + TempOptInfo *ti; + + /* Only single-output opcodes are supported here. */ + tcg_debug_assert(def->nb_oargs == 1); /* * 32-bit ops generate 32-bit results, which for the purpose of @@ -XXX,XX +XXX,XX @@ static bool fold_masks(OptContext *ctx, TCGOp *op) if (ctx->type == TCG_TYPE_I32) { z_mask = (int32_t)z_mask; s_mask |= MAKE_64BIT_MASK(32, 32); - ctx->z_mask = z_mask; - ctx->s_mask = s_mask; } if (z_mask == 0) { return tcg_opt_gen_movi(ctx, op, op->args[0], 0); } - return false; + + ts = arg_temp(op->args[0]); + reset_ts(ctx, ts); + + ti = ts_info(ts); + ti->z_mask = z_mask; + ti->s_mask = s_mask; + return true; } /* -- 2.43.0 Add a routine to which masks can be passed directly, rather than storing them into OptContext. To be used in upcoming patches. Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 15 ++++++++++++--- 1 file changed, 12 insertions(+), 3 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_const2_commutative(OptContext *ctx, TCGOp *op) return fold_const2(ctx, op); } -static bool fold_masks(OptContext *ctx, TCGOp *op) +/* + * Record "zero" and "sign" masks for the single output of @op. + * See TempOptInfo definition of z_mask and s_mask. + * If z_mask allows, fold the output to constant zero. + */ +static bool fold_masks_zs(OptContext *ctx, TCGOp *op, + uint64_t z_mask, uint64_t s_mask) { - uint64_t z_mask = ctx->z_mask; - uint64_t s_mask = ctx->s_mask; const TCGOpDef *def = &tcg_op_defs[op->opc]; TCGTemp *ts; TempOptInfo *ti; @@ -XXX,XX +XXX,XX @@ static bool fold_masks(OptContext *ctx, TCGOp *op) return true; } +static bool fold_masks(OptContext *ctx, TCGOp *op) +{ + return fold_masks_zs(ctx, op, ctx->z_mask, ctx->s_mask); +} + /* * An "affected" mask bit is 0 if and only if the result is identical * to the first input. Thus if the entire mask is 0, the operation -- 2.43.0 Consider the passed s_mask to be a minimum deduced from either existing s_mask or from a sign-extension operation. We may be able to deduce more from the set of known zeros. Remove identical logic from several opcode folders. Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 21 ++++++--------------- 1 file changed, 6 insertions(+), 15 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_const2_commutative(OptContext *ctx, TCGOp *op) * Record "zero" and "sign" masks for the single output of @op. * See TempOptInfo definition of z_mask and s_mask. * If z_mask allows, fold the output to constant zero. + * The passed s_mask may be augmented by z_mask. */ static bool fold_masks_zs(OptContext *ctx, TCGOp *op, uint64_t z_mask, uint64_t s_mask) @@ -XXX,XX +XXX,XX @@ static bool fold_masks_zs(OptContext *ctx, TCGOp *op, ti = ts_info(ts); ti->z_mask = z_mask; - ti->s_mask = s_mask; + ti->s_mask = s_mask | smask_from_zmask(z_mask); return true; } @@ -XXX,XX +XXX,XX @@ static bool fold_bswap(OptContext *ctx, TCGOp *op) default: g_assert_not_reached(); } - s_mask = smask_from_zmask(z_mask); + s_mask = 0; switch (op->args[2] & (TCG_BSWAP_OZ | TCG_BSWAP_OS)) { case TCG_BSWAP_OZ: break; @@ -XXX,XX +XXX,XX @@ static bool fold_bswap(OptContext *ctx, TCGOp *op) default: /* The high bits are undefined: force all bits above the sign to 1. */ z_mask |= sign << 1; - s_mask = 0; break; } ctx->z_mask = z_mask; @@ -XXX,XX +XXX,XX @@ static bool fold_count_zeros(OptContext *ctx, TCGOp *op) g_assert_not_reached(); } ctx->z_mask = arg_info(op->args[2])->z_mask | z_mask; - ctx->s_mask = smask_from_zmask(ctx->z_mask); return false; } @@ -XXX,XX +XXX,XX @@ static bool fold_ctpop(OptContext *ctx, TCGOp *op) default: g_assert_not_reached(); } - ctx->s_mask = smask_from_zmask(ctx->z_mask); return false; } @@ -XXX,XX +XXX,XX @@ static bool fold_extract(OptContext *ctx, TCGOp *op) return true; } ctx->z_mask = z_mask; - ctx->s_mask = smask_from_zmask(z_mask); return fold_masks(ctx, op); } @@ -XXX,XX +XXX,XX @@ static bool fold_extu(OptContext *ctx, TCGOp *op) } ctx->z_mask = z_mask; - ctx->s_mask = smask_from_zmask(z_mask); if (!type_change && fold_affected_mask(ctx, op, z_mask_old ^ z_mask)) { return true; } @@ -XXX,XX +XXX,XX @@ static bool fold_qemu_ld(OptContext *ctx, TCGOp *op) int width = 8 * memop_size(mop); if (width < 64) { - ctx->s_mask = MAKE_64BIT_MASK(width, 64 - width); - if (!(mop & MO_SIGN)) { + if (mop & MO_SIGN) { + ctx->s_mask = MAKE_64BIT_MASK(width, 64 - width); + } else { ctx->z_mask = MAKE_64BIT_MASK(0, width); - ctx->s_mask <<= 1; } } @@ -XXX,XX +XXX,XX @@ static bool fold_setcond(OptContext *ctx, TCGOp *op) fold_setcond_tst_pow2(ctx, op, false); ctx->z_mask = 1; - ctx->s_mask = smask_from_zmask(1); return false; } @@ -XXX,XX +XXX,XX @@ static bool fold_setcond2(OptContext *ctx, TCGOp *op) } ctx->z_mask = 1; - ctx->s_mask = smask_from_zmask(1); return false; do_setcond_const: @@ -XXX,XX +XXX,XX @@ static bool fold_tcg_ld(OptContext *ctx, TCGOp *op) break; CASE_OP_32_64(ld8u): ctx->z_mask = MAKE_64BIT_MASK(0, 8); - ctx->s_mask = MAKE_64BIT_MASK(9, 55); break; CASE_OP_32_64(ld16s): ctx->s_mask = MAKE_64BIT_MASK(16, 48); break; CASE_OP_32_64(ld16u): ctx->z_mask = MAKE_64BIT_MASK(0, 16); - ctx->s_mask = MAKE_64BIT_MASK(17, 47); break; case INDEX_op_ld32s_i64: ctx->s_mask = MAKE_64BIT_MASK(32, 32); break; case INDEX_op_ld32u_i64: ctx->z_mask = MAKE_64BIT_MASK(0, 32); - ctx->s_mask = MAKE_64BIT_MASK(33, 31); break; default: g_assert_not_reached(); -- 2.43.0 Change the representation from sign bit repetitions to all bits equal to the sign bit, including the sign bit itself. The previous format has a problem in that it is difficult to recreate a valid sign mask after a shift operation: the "repetitions" part of the previous format meant that applying the same shift as for the value lead to an off-by-one value. The new format, including the sign bit itself, means that the sign mask can be manipulated in exactly the same way as the value, canonicalization is easier. Canonicalize the s_mask in fold_masks_zs, rather than requiring callers to do so. Treat 0 as a non-canonical but typeless input for no sign information, which will be reset as appropriate for the data type. We can easily fold in the data from z_mask while canonicalizing. Temporarily disable optimizations using s_mask while each operation is converted to use fold_masks_zs and to the new form. Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 64 ++++++++++++-------------------------------------- 1 file changed, 15 insertions(+), 49 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ typedef struct TempOptInfo { QSIMPLEQ_HEAD(, MemCopyInfo) mem_copy; uint64_t val; uint64_t z_mask; /* mask bit is 0 if and only if value bit is 0 */ - uint64_t s_mask; /* a left-aligned mask of clrsb(value) bits. */ + uint64_t s_mask; /* mask bit is 1 if value bit matches msb */ } TempOptInfo; typedef struct OptContext { @@ -XXX,XX +XXX,XX @@ typedef struct OptContext { /* In flight values from optimization. */ uint64_t z_mask; /* mask bit is 0 iff value bit is 0 */ - uint64_t s_mask; /* mask of clrsb(value) bits */ + uint64_t s_mask; /* mask bit is 1 if value bit matches msb */ TCGType type; } OptContext; -/* Calculate the smask for a specific value. */ -static uint64_t smask_from_value(uint64_t value) -{ - int rep = clrsb64(value); - return ~(~0ull >> rep); -} - -/* - * Calculate the smask for a given set of known-zeros. - * If there are lots of zeros on the left, we can consider the remainder - * an unsigned field, and thus the corresponding signed field is one bit - * larger. - */ -static uint64_t smask_from_zmask(uint64_t zmask) -{ - /* - * Only the 0 bits are significant for zmask, thus the msb itself - * must be zero, else we have no sign information. - */ - int rep = clz64(zmask); - if (rep == 0) { - return 0; - } - rep -= 1; - return ~(~0ull >> rep); -} - -/* - * Recreate a properly left-aligned smask after manipulation. - * Some bit-shuffling, particularly shifts and rotates, may - * retain sign bits on the left, but may scatter disconnected - * sign bits on the right. Retain only what remains to the left. - */ -static uint64_t smask_from_smask(int64_t smask) -{ - /* Only the 1 bits are significant for smask */ - return smask_from_zmask(~smask); -} - static inline TempOptInfo *ts_info(TCGTemp *ts) { return ts->state_ptr; @@ -XXX,XX +XXX,XX @@ static void init_ts_info(OptContext *ctx, TCGTemp *ts) ti->is_const = true; ti->val = ts->val; ti->z_mask = ts->val; - ti->s_mask = smask_from_value(ts->val); + ti->s_mask = INT64_MIN >> clrsb64(ts->val); } else { ti->is_const = false; ti->z_mask = -1; @@ -XXX,XX +XXX,XX @@ static void finish_folding(OptContext *ctx, TCGOp *op) */ if (i == 0) { ts_info(ts)->z_mask = ctx->z_mask; - ts_info(ts)->s_mask = ctx->s_mask; } } } @@ -XXX,XX +XXX,XX @@ static bool fold_const2_commutative(OptContext *ctx, TCGOp *op) * The passed s_mask may be augmented by z_mask. */ static bool fold_masks_zs(OptContext *ctx, TCGOp *op, - uint64_t z_mask, uint64_t s_mask) + uint64_t z_mask, int64_t s_mask) { const TCGOpDef *def = &tcg_op_defs[op->opc]; TCGTemp *ts; TempOptInfo *ti; + int rep; /* Only single-output opcodes are supported here. */ tcg_debug_assert(def->nb_oargs == 1); @@ -XXX,XX +XXX,XX @@ static bool fold_masks_zs(OptContext *ctx, TCGOp *op, */ if (ctx->type == TCG_TYPE_I32) { z_mask = (int32_t)z_mask; - s_mask |= MAKE_64BIT_MASK(32, 32); + s_mask |= INT32_MIN; } if (z_mask == 0) { @@ -XXX,XX +XXX,XX @@ static bool fold_masks_zs(OptContext *ctx, TCGOp *op, ti = ts_info(ts); ti->z_mask = z_mask; - ti->s_mask = s_mask | smask_from_zmask(z_mask); + + /* Canonicalize s_mask and incorporate data from z_mask. */ + rep = clz64(~s_mask); + rep = MAX(rep, clz64(z_mask)); + rep = MAX(rep - 1, 0); + ti->s_mask = INT64_MIN >> rep; + return true; } @@ -XXX,XX +XXX,XX @@ static bool fold_exts(OptContext *ctx, TCGOp *op) ctx->z_mask = z_mask; ctx->s_mask = s_mask; - if (!type_change && fold_affected_mask(ctx, op, s_mask & ~s_mask_old)) { + if (0 && !type_change && fold_affected_mask(ctx, op, s_mask & ~s_mask_old)) { return true; } @@ -XXX,XX +XXX,XX @@ static bool fold_sextract(OptContext *ctx, TCGOp *op) s_mask |= MAKE_64BIT_MASK(len, 64 - len); ctx->s_mask = s_mask; - if (pos == 0 && fold_affected_mask(ctx, op, s_mask & ~s_mask_old)) { + if (0 && pos == 0 && fold_affected_mask(ctx, op, s_mask & ~s_mask_old)) { return true; } @@ -XXX,XX +XXX,XX @@ static bool fold_shift(OptContext *ctx, TCGOp *op) ctx->z_mask = do_constant_folding(op->opc, ctx->type, z_mask, sh); s_mask = do_constant_folding(op->opc, ctx->type, s_mask, sh); - ctx->s_mask = smask_from_smask(s_mask); return fold_masks(ctx, op); } -- 2.43.0 Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static void finish_ebb(OptContext *ctx) remove_mem_copy_all(ctx); } -static void finish_folding(OptContext *ctx, TCGOp *op) +static bool finish_folding(OptContext *ctx, TCGOp *op) { const TCGOpDef *def = &tcg_op_defs[op->opc]; int i, nb_oargs; @@ -XXX,XX +XXX,XX @@ static void finish_folding(OptContext *ctx, TCGOp *op) ts_info(ts)->z_mask = ctx->z_mask; } } + return true; } /* @@ -XXX,XX +XXX,XX @@ static bool fold_add(OptContext *ctx, TCGOp *op) fold_xi_to_x(ctx, op, 0)) { return true; } - return false; + return finish_folding(ctx, op); } /* We cannot as yet do_constant_folding with vectors. */ @@ -XXX,XX +XXX,XX @@ static bool fold_add_vec(OptContext *ctx, TCGOp *op) fold_xi_to_x(ctx, op, 0)) { return true; } - return false; + return finish_folding(ctx, op); } static bool fold_addsub2(OptContext *ctx, TCGOp *op, bool add) @@ -XXX,XX +XXX,XX @@ static bool fold_addsub2(OptContext *ctx, TCGOp *op, bool add) op->args[4] = arg_new_constant(ctx, bl); op->args[5] = arg_new_constant(ctx, bh); } - return false; + return finish_folding(ctx, op); } static bool fold_add2(OptContext *ctx, TCGOp *op) -- 2.43.0 Introduce ti_is_const, ti_const_val, ti_is_const_val. Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 20 +++++++++++++++++--- 1 file changed, 17 insertions(+), 3 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static inline TempOptInfo *arg_info(TCGArg arg) return ts_info(arg_temp(arg)); } +static inline bool ti_is_const(TempOptInfo *ti) +{ + return ti->is_const; +} + +static inline uint64_t ti_const_val(TempOptInfo *ti) +{ + return ti->val; +} + +static inline bool ti_is_const_val(TempOptInfo *ti, uint64_t val) +{ + return ti_is_const(ti) && ti_const_val(ti) == val; +} + static inline bool ts_is_const(TCGTemp *ts) { - return ts_info(ts)->is_const; + return ti_is_const(ts_info(ts)); } static inline bool ts_is_const_val(TCGTemp *ts, uint64_t val) { - TempOptInfo *ti = ts_info(ts); - return ti->is_const && ti->val == val; + return ti_is_const_val(ts_info(ts), val); } static inline bool arg_is_const(TCGArg arg) -- 2.43.0 Avoid the use of the OptContext slots. Find TempOptInfo once. Sink mask computation below fold_affected_mask early exit. Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 30 ++++++++++++++++-------------- 1 file changed, 16 insertions(+), 14 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_add2(OptContext *ctx, TCGOp *op) static bool fold_and(OptContext *ctx, TCGOp *op) { - uint64_t z1, z2; + uint64_t z1, z2, z_mask, s_mask; + TempOptInfo *t1, *t2; if (fold_const2_commutative(ctx, op) || fold_xi_to_i(ctx, op, 0) || @@ -XXX,XX +XXX,XX @@ static bool fold_and(OptContext *ctx, TCGOp *op) return true; } - z1 = arg_info(op->args[1])->z_mask; - z2 = arg_info(op->args[2])->z_mask; - ctx->z_mask = z1 & z2; - - /* - * Sign repetitions are perforce all identical, whether they are 1 or 0. - * Bitwise operations preserve the relative quantity of the repetitions. - */ - ctx->s_mask = arg_info(op->args[1])->s_mask - & arg_info(op->args[2])->s_mask; + t1 = arg_info(op->args[1]); + t2 = arg_info(op->args[2]); + z1 = t1->z_mask; + z2 = t2->z_mask; /* * Known-zeros does not imply known-ones. Therefore unless * arg2 is constant, we can't infer affected bits from it. */ - if (arg_is_const(op->args[2]) && - fold_affected_mask(ctx, op, z1 & ~z2)) { + if (ti_is_const(t2) && fold_affected_mask(ctx, op, z1 & ~z2)) { return true; } - return fold_masks(ctx, op); + z_mask = z1 & z2; + + /* + * Sign repetitions are perforce all identical, whether they are 1 or 0. + * Bitwise operations preserve the relative quantity of the repetitions. + */ + s_mask = t1->s_mask & t2->s_mask; + + return fold_masks_zs(ctx, op, z_mask, s_mask); } static bool fold_andc(OptContext *ctx, TCGOp *op) -- 2.43.0 Avoid the use of the OptContext slots. Find TempOptInfo once. Avoid double inversion of the value of second const operand. Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 21 +++++++++++---------- 1 file changed, 11 insertions(+), 10 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_and(OptContext *ctx, TCGOp *op) static bool fold_andc(OptContext *ctx, TCGOp *op) { - uint64_t z1; + uint64_t z_mask, s_mask; + TempOptInfo *t1, *t2; if (fold_const2(ctx, op) || fold_xx_to_i(ctx, op, 0) || @@ -XXX,XX +XXX,XX @@ static bool fold_andc(OptContext *ctx, TCGOp *op) return true; } - z1 = arg_info(op->args[1])->z_mask; + t1 = arg_info(op->args[1]); + t2 = arg_info(op->args[2]); + z_mask = t1->z_mask; /* * Known-zeros does not imply known-ones. Therefore unless * arg2 is constant, we can't infer anything from it. */ - if (arg_is_const(op->args[2])) { - uint64_t z2 = ~arg_info(op->args[2])->z_mask; - if (fold_affected_mask(ctx, op, z1 & ~z2)) { + if (ti_is_const(t2)) { + uint64_t v2 = ti_const_val(t2); + if (fold_affected_mask(ctx, op, z_mask & v2)) { return true; } - z1 &= z2; + z_mask &= ~v2; } - ctx->z_mask = z1; - ctx->s_mask = arg_info(op->args[1])->s_mask - & arg_info(op->args[2])->s_mask; - return fold_masks(ctx, op); + s_mask = t1->s_mask & t2->s_mask; + return fold_masks_zs(ctx, op, z_mask, s_mask); } static bool fold_brcond(OptContext *ctx, TCGOp *op) -- 2.43.0 Avoid the use of the OptContext slots. Find TempOptInfo once. Always set s_mask along the BSWAP_OS path, since the result is being explicitly sign-extended. Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 21 ++++++++++----------- 1 file changed, 10 insertions(+), 11 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_brcond2(OptContext *ctx, TCGOp *op) static bool fold_bswap(OptContext *ctx, TCGOp *op) { uint64_t z_mask, s_mask, sign; + TempOptInfo *t1 = arg_info(op->args[1]); - if (arg_is_const(op->args[1])) { - uint64_t t = arg_info(op->args[1])->val; - - t = do_constant_folding(op->opc, ctx->type, t, op->args[2]); - return tcg_opt_gen_movi(ctx, op, op->args[0], t); + if (ti_is_const(t1)) { + return tcg_opt_gen_movi(ctx, op, op->args[0], + do_constant_folding(op->opc, ctx->type, + ti_const_val(t1), + op->args[2])); } - z_mask = arg_info(op->args[1])->z_mask; - + z_mask = t1->z_mask; switch (op->opc) { case INDEX_op_bswap16_i32: case INDEX_op_bswap16_i64: @@ -XXX,XX +XXX,XX @@ static bool fold_bswap(OptContext *ctx, TCGOp *op) /* If the sign bit may be 1, force all the bits above to 1. */ if (z_mask & sign) { z_mask |= sign; - s_mask = sign << 1; } + /* The value and therefore s_mask is explicitly sign-extended. */ + s_mask = sign; break; default: /* The high bits are undefined: force all bits above the sign to 1. */ z_mask |= sign << 1; break; } - ctx->z_mask = z_mask; - ctx->s_mask = s_mask; - return fold_masks(ctx, op); + return fold_masks_zs(ctx, op, z_mask, s_mask); } static bool fold_call(OptContext *ctx, TCGOp *op) -- 2.43.0 Avoid the use of the OptContext slots. Find TempOptInfo once. Compute s_mask from the union of the maximum count and the op2 fallback for op1 being zero. Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_call(OptContext *ctx, TCGOp *op) static bool fold_count_zeros(OptContext *ctx, TCGOp *op) { - uint64_t z_mask; + uint64_t z_mask, s_mask; + TempOptInfo *t1 = arg_info(op->args[1]); + TempOptInfo *t2 = arg_info(op->args[2]); - if (arg_is_const(op->args[1])) { - uint64_t t = arg_info(op->args[1])->val; + if (ti_is_const(t1)) { + uint64_t t = ti_const_val(t1); if (t != 0) { t = do_constant_folding(op->opc, ctx->type, t, 0); @@ -XXX,XX +XXX,XX @@ static bool fold_count_zeros(OptContext *ctx, TCGOp *op) default: g_assert_not_reached(); } - ctx->z_mask = arg_info(op->args[2])->z_mask | z_mask; - return false; + s_mask = ~z_mask; + z_mask |= t2->z_mask; + s_mask &= t2->s_mask; + + return fold_masks_zs(ctx, op, z_mask, s_mask); } static bool fold_ctpop(OptContext *ctx, TCGOp *op) -- 2.43.0 Add fold_masks_z as a trivial wrapper around fold_masks_zs. Avoid the use of the OptContext slots. Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_masks_zs(OptContext *ctx, TCGOp *op, return true; } +static bool fold_masks_z(OptContext *ctx, TCGOp *op, uint64_t z_mask) +{ + return fold_masks_zs(ctx, op, z_mask, 0); +} + static bool fold_masks(OptContext *ctx, TCGOp *op) { return fold_masks_zs(ctx, op, ctx->z_mask, ctx->s_mask); @@ -XXX,XX +XXX,XX @@ static bool fold_count_zeros(OptContext *ctx, TCGOp *op) static bool fold_ctpop(OptContext *ctx, TCGOp *op) { + uint64_t z_mask; + if (fold_const1(ctx, op)) { return true; } switch (ctx->type) { case TCG_TYPE_I32: - ctx->z_mask = 32 | 31; + z_mask = 32 | 31; break; case TCG_TYPE_I64: - ctx->z_mask = 64 | 63; + z_mask = 64 | 63; break; default: g_assert_not_reached(); } - return false; + return fold_masks_z(ctx, op, z_mask); } static bool fold_deposit(OptContext *ctx, TCGOp *op) -- 2.43.0 Avoid the use of the OptContext slots. Find TempOptInfo once. When we fold to and, use fold_and. Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 35 +++++++++++++++++------------------ 1 file changed, 17 insertions(+), 18 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_ctpop(OptContext *ctx, TCGOp *op) static bool fold_deposit(OptContext *ctx, TCGOp *op) { + TempOptInfo *t1 = arg_info(op->args[1]); + TempOptInfo *t2 = arg_info(op->args[2]); + int ofs = op->args[3]; + int len = op->args[4]; TCGOpcode and_opc; + uint64_t z_mask; - if (arg_is_const(op->args[1]) && arg_is_const(op->args[2])) { - uint64_t t1 = arg_info(op->args[1])->val; - uint64_t t2 = arg_info(op->args[2])->val; - - t1 = deposit64(t1, op->args[3], op->args[4], t2); - return tcg_opt_gen_movi(ctx, op, op->args[0], t1); + if (ti_is_const(t1) && ti_is_const(t2)) { + return tcg_opt_gen_movi(ctx, op, op->args[0], + deposit64(ti_const_val(t1), ofs, len, + ti_const_val(t2))); } switch (ctx->type) { @@ -XXX,XX +XXX,XX @@ static bool fold_deposit(OptContext *ctx, TCGOp *op) } /* Inserting a value into zero at offset 0. */ - if (arg_is_const_val(op->args[1], 0) && op->args[3] == 0) { - uint64_t mask = MAKE_64BIT_MASK(0, op->args[4]); + if (ti_is_const_val(t1, 0) && ofs == 0) { + uint64_t mask = MAKE_64BIT_MASK(0, len); op->opc = and_opc; op->args[1] = op->args[2]; op->args[2] = arg_new_constant(ctx, mask); - ctx->z_mask = mask & arg_info(op->args[1])->z_mask; - return false; + return fold_and(ctx, op); } /* Inserting zero into a value. */ - if (arg_is_const_val(op->args[2], 0)) { - uint64_t mask = deposit64(-1, op->args[3], op->args[4], 0); + if (ti_is_const_val(t2, 0)) { + uint64_t mask = deposit64(-1, ofs, len, 0); op->opc = and_opc; op->args[2] = arg_new_constant(ctx, mask); - ctx->z_mask = mask & arg_info(op->args[1])->z_mask; - return false; + return fold_and(ctx, op); } - ctx->z_mask = deposit64(arg_info(op->args[1])->z_mask, - op->args[3], op->args[4], - arg_info(op->args[2])->z_mask); - return false; + z_mask = deposit64(t1->z_mask, ofs, len, t2->z_mask); + return fold_masks_z(ctx, op, z_mask); } static bool fold_divide(OptContext *ctx, TCGOp *op) -- 2.43.0 The input which overlaps the sign bit of the output can have its input s_mask propagated to the output s_mask. Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_deposit(OptContext *ctx, TCGOp *op) TempOptInfo *t2 = arg_info(op->args[2]); int ofs = op->args[3]; int len = op->args[4]; + int width; TCGOpcode and_opc; - uint64_t z_mask; + uint64_t z_mask, s_mask; if (ti_is_const(t1) && ti_is_const(t2)) { return tcg_opt_gen_movi(ctx, op, op->args[0], @@ -XXX,XX +XXX,XX @@ static bool fold_deposit(OptContext *ctx, TCGOp *op) switch (ctx->type) { case TCG_TYPE_I32: and_opc = INDEX_op_and_i32; + width = 32; break; case TCG_TYPE_I64: and_opc = INDEX_op_and_i64; + width = 64; break; default: g_assert_not_reached(); @@ -XXX,XX +XXX,XX @@ static bool fold_deposit(OptContext *ctx, TCGOp *op) return fold_and(ctx, op); } + /* The s_mask from the top portion of the deposit is still valid. */ + if (ofs + len == width) { + s_mask = t2->s_mask << ofs; + } else { + s_mask = t1->s_mask & ~MAKE_64BIT_MASK(0, ofs + len); + } + z_mask = deposit64(t1->z_mask, ofs, len, t2->z_mask); - return fold_masks_z(ctx, op, z_mask); + return fold_masks_zs(ctx, op, z_mask, s_mask); } static bool fold_divide(OptContext *ctx, TCGOp *op) -- 2.43.0 Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_divide(OptContext *ctx, TCGOp *op) fold_xi_to_x(ctx, op, 1)) { return true; } - return false; + return finish_folding(ctx, op); } static bool fold_dup(OptContext *ctx, TCGOp *op) -- 2.43.0 Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_dup(OptContext *ctx, TCGOp *op) t = dup_const(TCGOP_VECE(op), t); return tcg_opt_gen_movi(ctx, op, op->args[0], t); } - return false; + return finish_folding(ctx, op); } static bool fold_dup2(OptContext *ctx, TCGOp *op) @@ -XXX,XX +XXX,XX @@ static bool fold_dup2(OptContext *ctx, TCGOp *op) op->opc = INDEX_op_dup_vec; TCGOP_VECE(op) = MO_32; } - return false; + return finish_folding(ctx, op); } static bool fold_eqv(OptContext *ctx, TCGOp *op) -- 2.43.0 Add fold_masks_s as a trivial wrapper around fold_masks_zs. Avoid the use of the OptContext slots. Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_masks_z(OptContext *ctx, TCGOp *op, uint64_t z_mask) return fold_masks_zs(ctx, op, z_mask, 0); } +static bool fold_masks_s(OptContext *ctx, TCGOp *op, uint64_t s_mask) +{ + return fold_masks_zs(ctx, op, -1, s_mask); +} + static bool fold_masks(OptContext *ctx, TCGOp *op) { return fold_masks_zs(ctx, op, ctx->z_mask, ctx->s_mask); @@ -XXX,XX +XXX,XX @@ static bool fold_dup2(OptContext *ctx, TCGOp *op) static bool fold_eqv(OptContext *ctx, TCGOp *op) { + uint64_t s_mask; + if (fold_const2_commutative(ctx, op) || fold_xi_to_x(ctx, op, -1) || fold_xi_to_not(ctx, op, 0)) { return true; } - ctx->s_mask = arg_info(op->args[1])->s_mask - & arg_info(op->args[2])->s_mask; - return false; + s_mask = arg_info(op->args[1])->s_mask + & arg_info(op->args[2])->s_mask; + return fold_masks_s(ctx, op, s_mask); } static bool fold_extract(OptContext *ctx, TCGOp *op) -- 2.43.0 Avoid the use of the OptContext slots. Find TempOptInfo once. Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 15 ++++++--------- 1 file changed, 6 insertions(+), 9 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_eqv(OptContext *ctx, TCGOp *op) static bool fold_extract(OptContext *ctx, TCGOp *op) { uint64_t z_mask_old, z_mask; + TempOptInfo *t1 = arg_info(op->args[1]); int pos = op->args[2]; int len = op->args[3]; - if (arg_is_const(op->args[1])) { - uint64_t t; - - t = arg_info(op->args[1])->val; - t = extract64(t, pos, len); - return tcg_opt_gen_movi(ctx, op, op->args[0], t); + if (ti_is_const(t1)) { + return tcg_opt_gen_movi(ctx, op, op->args[0], + extract64(ti_const_val(t1), pos, len)); } - z_mask_old = arg_info(op->args[1])->z_mask; + z_mask_old = t1->z_mask; z_mask = extract64(z_mask_old, pos, len); if (pos == 0 && fold_affected_mask(ctx, op, z_mask_old ^ z_mask)) { return true; } - ctx->z_mask = z_mask; - return fold_masks(ctx, op); + return fold_masks_z(ctx, op, z_mask); } static bool fold_extract2(OptContext *ctx, TCGOp *op) -- 2.43.0 Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_extract2(OptContext *ctx, TCGOp *op) } return tcg_opt_gen_movi(ctx, op, op->args[0], v1 | v2); } - return false; + return finish_folding(ctx, op); } static bool fold_exts(OptContext *ctx, TCGOp *op) -- 2.43.0 Avoid the use of the OptContext slots. Find TempOptInfo once. Explicitly sign-extend z_mask instead of doing that manually. Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 29 ++++++++++++----------------- 1 file changed, 12 insertions(+), 17 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_extract2(OptContext *ctx, TCGOp *op) static bool fold_exts(OptContext *ctx, TCGOp *op) { - uint64_t s_mask_old, s_mask, z_mask, sign; + uint64_t s_mask_old, s_mask, z_mask; bool type_change = false; + TempOptInfo *t1; if (fold_const1(ctx, op)) { return true; } - z_mask = arg_info(op->args[1])->z_mask; - s_mask = arg_info(op->args[1])->s_mask; + t1 = arg_info(op->args[1]); + z_mask = t1->z_mask; + s_mask = t1->s_mask; s_mask_old = s_mask; switch (op->opc) { CASE_OP_32_64(ext8s): - sign = INT8_MIN; - z_mask = (uint8_t)z_mask; + s_mask |= INT8_MIN; + z_mask = (int8_t)z_mask; break; CASE_OP_32_64(ext16s): - sign = INT16_MIN; - z_mask = (uint16_t)z_mask; + s_mask |= INT16_MIN; + z_mask = (int16_t)z_mask; break; case INDEX_op_ext_i32_i64: type_change = true; QEMU_FALLTHROUGH; case INDEX_op_ext32s_i64: - sign = INT32_MIN; - z_mask = (uint32_t)z_mask; + s_mask |= INT32_MIN; + z_mask = (int32_t)z_mask; break; default: g_assert_not_reached(); } - if (z_mask & sign) { - z_mask |= sign; - } - s_mask |= sign << 1; - - ctx->z_mask = z_mask; - ctx->s_mask = s_mask; if (0 && !type_change && fold_affected_mask(ctx, op, s_mask & ~s_mask_old)) { return true; } - return fold_masks(ctx, op); + return fold_masks_zs(ctx, op, z_mask, s_mask); } static bool fold_extu(OptContext *ctx, TCGOp *op) -- 2.43.0 Avoid the use of the OptContext slots. Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_extu(OptContext *ctx, TCGOp *op) g_assert_not_reached(); } - ctx->z_mask = z_mask; if (!type_change && fold_affected_mask(ctx, op, z_mask_old ^ z_mask)) { return true; } - return fold_masks(ctx, op); + + return fold_masks_z(ctx, op, z_mask); } static bool fold_mb(OptContext *ctx, TCGOp *op) -- 2.43.0 Avoid the use of the OptContext slots. Find TempOptInfo once. Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 19 +++++++++++-------- 1 file changed, 11 insertions(+), 8 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_mov(OptContext *ctx, TCGOp *op) static bool fold_movcond(OptContext *ctx, TCGOp *op) { + uint64_t z_mask, s_mask; + TempOptInfo *tt, *ft; int i; /* If true and false values are the same, eliminate the cmp. */ @@ -XXX,XX +XXX,XX @@ static bool fold_movcond(OptContext *ctx, TCGOp *op) return tcg_opt_gen_mov(ctx, op, op->args[0], op->args[4 - i]); } - ctx->z_mask = arg_info(op->args[3])->z_mask - | arg_info(op->args[4])->z_mask; - ctx->s_mask = arg_info(op->args[3])->s_mask - & arg_info(op->args[4])->s_mask; + tt = arg_info(op->args[3]); + ft = arg_info(op->args[4]); + z_mask = tt->z_mask | ft->z_mask; + s_mask = tt->s_mask & ft->s_mask; - if (arg_is_const(op->args[3]) && arg_is_const(op->args[4])) { - uint64_t tv = arg_info(op->args[3])->val; - uint64_t fv = arg_info(op->args[4])->val; + if (ti_is_const(tt) && ti_is_const(ft)) { + uint64_t tv = ti_const_val(tt); + uint64_t fv = ti_const_val(ft); TCGOpcode opc, negopc = 0; TCGCond cond = op->args[5]; @@ -XXX,XX +XXX,XX @@ static bool fold_movcond(OptContext *ctx, TCGOp *op) } } } - return false; + + return fold_masks_zs(ctx, op, z_mask, s_mask); } static bool fold_mul(OptContext *ctx, TCGOp *op) -- 2.43.0 Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_mul(OptContext *ctx, TCGOp *op) fold_xi_to_x(ctx, op, 1)) { return true; } - return false; + return finish_folding(ctx, op); } static bool fold_mul_highpart(OptContext *ctx, TCGOp *op) @@ -XXX,XX +XXX,XX @@ static bool fold_mul_highpart(OptContext *ctx, TCGOp *op) fold_xi_to_i(ctx, op, 0)) { return true; } - return false; + return finish_folding(ctx, op); } static bool fold_multiply2(OptContext *ctx, TCGOp *op) @@ -XXX,XX +XXX,XX @@ static bool fold_multiply2(OptContext *ctx, TCGOp *op) tcg_opt_gen_movi(ctx, op2, rh, h); return true; } - return false; + return finish_folding(ctx, op); } static bool fold_nand(OptContext *ctx, TCGOp *op) -- 2.43.0 Avoid the use of the OptContext slots. Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_multiply2(OptContext *ctx, TCGOp *op) static bool fold_nand(OptContext *ctx, TCGOp *op) { + uint64_t s_mask; + if (fold_const2_commutative(ctx, op) || fold_xi_to_not(ctx, op, -1)) { return true; } - ctx->s_mask = arg_info(op->args[1])->s_mask - & arg_info(op->args[2])->s_mask; - return false; + s_mask = arg_info(op->args[1])->s_mask + & arg_info(op->args[2])->s_mask; + return fold_masks_s(ctx, op, s_mask); } static bool fold_neg_no_const(OptContext *ctx, TCGOp *op) -- 2.43.0 Avoid the use of the OptContext slots. Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 9 ++------- 1 file changed, 2 insertions(+), 7 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_neg_no_const(OptContext *ctx, TCGOp *op) { /* Set to 1 all bits to the left of the rightmost. */ uint64_t z_mask = arg_info(op->args[1])->z_mask; - ctx->z_mask = -(z_mask & -z_mask); + z_mask = -(z_mask & -z_mask); - /* - * Because of fold_sub_to_neg, we want to always return true, - * via finish_folding. - */ - finish_folding(ctx, op); - return true; + return fold_masks_z(ctx, op, z_mask); } static bool fold_neg(OptContext *ctx, TCGOp *op) -- 2.43.0 Avoid the use of the OptContext slots. Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_neg(OptContext *ctx, TCGOp *op) static bool fold_nor(OptContext *ctx, TCGOp *op) { + uint64_t s_mask; + if (fold_const2_commutative(ctx, op) || fold_xi_to_not(ctx, op, 0)) { return true; } - ctx->s_mask = arg_info(op->args[1])->s_mask - & arg_info(op->args[2])->s_mask; - return false; + s_mask = arg_info(op->args[1])->s_mask + & arg_info(op->args[2])->s_mask; + return fold_masks_s(ctx, op, s_mask); } static bool fold_not(OptContext *ctx, TCGOp *op) -- 2.43.0 Avoid the use of the OptContext slots. Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 7 +------ 1 file changed, 1 insertion(+), 6 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_not(OptContext *ctx, TCGOp *op) if (fold_const1(ctx, op)) { return true; } - - ctx->s_mask = arg_info(op->args[1])->s_mask; - - /* Because of fold_to_not, we want to always return true, via finish. */ - finish_folding(ctx, op); - return true; + return fold_masks_s(ctx, op, arg_info(op->args[1])->s_mask); } static bool fold_or(OptContext *ctx, TCGOp *op) -- 2.43.0 Avoid the use of the OptContext slots. Find TempOptInfo once. Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_not(OptContext *ctx, TCGOp *op) static bool fold_or(OptContext *ctx, TCGOp *op) { + uint64_t z_mask, s_mask; + TempOptInfo *t1, *t2; + if (fold_const2_commutative(ctx, op) || fold_xi_to_x(ctx, op, 0) || fold_xx_to_x(ctx, op)) { return true; } - ctx->z_mask = arg_info(op->args[1])->z_mask - | arg_info(op->args[2])->z_mask; - ctx->s_mask = arg_info(op->args[1])->s_mask - & arg_info(op->args[2])->s_mask; - return fold_masks(ctx, op); + t1 = arg_info(op->args[1]); + t2 = arg_info(op->args[2]); + z_mask = t1->z_mask | t2->z_mask; + s_mask = t1->s_mask & t2->s_mask; + return fold_masks_zs(ctx, op, z_mask, s_mask); } static bool fold_orc(OptContext *ctx, TCGOp *op) -- 2.43.0 Avoid the use of the OptContext slots. Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_or(OptContext *ctx, TCGOp *op) static bool fold_orc(OptContext *ctx, TCGOp *op) { + uint64_t s_mask; + if (fold_const2(ctx, op) || fold_xx_to_i(ctx, op, -1) || fold_xi_to_x(ctx, op, -1) || @@ -XXX,XX +XXX,XX @@ static bool fold_orc(OptContext *ctx, TCGOp *op) return true; } - ctx->s_mask = arg_info(op->args[1])->s_mask - & arg_info(op->args[2])->s_mask; - return false; + s_mask = arg_info(op->args[1])->s_mask + & arg_info(op->args[2])->s_mask; + return fold_masks_s(ctx, op, s_mask); } static bool fold_qemu_ld(OptContext *ctx, TCGOp *op) -- 2.43.0 Avoid the use of the OptContext slots. Be careful not to call fold_masks_zs when the memory operation is wide enough to require multiple outputs, so split into two functions: fold_qemu_ld_1reg and fold_qemu_ld_2reg. Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 26 +++++++++++++++++++++----- 1 file changed, 21 insertions(+), 5 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_orc(OptContext *ctx, TCGOp *op) return fold_masks_s(ctx, op, s_mask); } -static bool fold_qemu_ld(OptContext *ctx, TCGOp *op) +static bool fold_qemu_ld_1reg(OptContext *ctx, TCGOp *op) { const TCGOpDef *def = &tcg_op_defs[op->opc]; MemOpIdx oi = op->args[def->nb_oargs + def->nb_iargs]; MemOp mop = get_memop(oi); int width = 8 * memop_size(mop); + uint64_t z_mask = -1, s_mask = 0; if (width < 64) { if (mop & MO_SIGN) { - ctx->s_mask = MAKE_64BIT_MASK(width, 64 - width); + s_mask = MAKE_64BIT_MASK(width - 1, 64 - (width - 1)); } else { - ctx->z_mask = MAKE_64BIT_MASK(0, width); + z_mask = MAKE_64BIT_MASK(0, width); } } /* Opcodes that touch guest memory stop the mb optimization. */ ctx->prev_mb = NULL; - return false; + + return fold_masks_zs(ctx, op, z_mask, s_mask); +} + +static bool fold_qemu_ld_2reg(OptContext *ctx, TCGOp *op) +{ + /* Opcodes that touch guest memory stop the mb optimization. */ + ctx->prev_mb = NULL; + return finish_folding(ctx, op); } static bool fold_qemu_st(OptContext *ctx, TCGOp *op) @@ -XXX,XX +XXX,XX @@ void tcg_optimize(TCGContext *s) break; case INDEX_op_qemu_ld_a32_i32: case INDEX_op_qemu_ld_a64_i32: + done = fold_qemu_ld_1reg(&ctx, op); + break; case INDEX_op_qemu_ld_a32_i64: case INDEX_op_qemu_ld_a64_i64: + if (TCG_TARGET_REG_BITS == 64) { + done = fold_qemu_ld_1reg(&ctx, op); + break; + } + QEMU_FALLTHROUGH; case INDEX_op_qemu_ld_a32_i128: case INDEX_op_qemu_ld_a64_i128: - done = fold_qemu_ld(&ctx, op); + done = fold_qemu_ld_2reg(&ctx, op); break; case INDEX_op_qemu_st8_a32_i32: case INDEX_op_qemu_st8_a64_i32: -- 2.43.0 Stores have no output operands, and so need no further work. Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_qemu_st(OptContext *ctx, TCGOp *op) { /* Opcodes that touch guest memory stop the mb optimization. */ ctx->prev_mb = NULL; - return false; + return true; } static bool fold_remainder(OptContext *ctx, TCGOp *op) @@ -XXX,XX +XXX,XX @@ static bool fold_tcg_st(OptContext *ctx, TCGOp *op) if (op->args[1] != tcgv_ptr_arg(tcg_env)) { remove_mem_copy_all(ctx); - return false; + return true; } switch (op->opc) { @@ -XXX,XX +XXX,XX @@ static bool fold_tcg_st(OptContext *ctx, TCGOp *op) g_assert_not_reached(); } remove_mem_copy_in(ctx, ofs, ofs + lm1); - return false; + return true; } static bool fold_tcg_st_memcopy(OptContext *ctx, TCGOp *op) @@ -XXX,XX +XXX,XX @@ static bool fold_tcg_st_memcopy(OptContext *ctx, TCGOp *op) TCGType type; if (op->args[1] != tcgv_ptr_arg(tcg_env)) { - fold_tcg_st(ctx, op); - return false; + return fold_tcg_st(ctx, op); } src = arg_temp(op->args[0]); @@ -XXX,XX +XXX,XX @@ static bool fold_tcg_st_memcopy(OptContext *ctx, TCGOp *op) last = ofs + tcg_type_size(type) - 1; remove_mem_copy_in(ctx, ofs, last); record_mem_copy(ctx, type, src, ofs, last); - return false; + return true; } static bool fold_xor(OptContext *ctx, TCGOp *op) -- 2.43.0 Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_remainder(OptContext *ctx, TCGOp *op) fold_xx_to_i(ctx, op, 0)) { return true; } - return false; + return finish_folding(ctx, op); } static bool fold_setcond_zmask(OptContext *ctx, TCGOp *op, bool neg) -- 2.43.0 Change return from bool to int; distinguish between complete folding, simplification, and no change. Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 22 ++++++++++++++-------- 1 file changed, 14 insertions(+), 8 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_remainder(OptContext *ctx, TCGOp *op) return finish_folding(ctx, op); } -static bool fold_setcond_zmask(OptContext *ctx, TCGOp *op, bool neg) +/* Return 1 if finished, -1 if simplified, 0 if unchanged. */ +static int fold_setcond_zmask(OptContext *ctx, TCGOp *op, bool neg) { uint64_t a_zmask, b_val; TCGCond cond; @@ -XXX,XX +XXX,XX @@ static bool fold_setcond_zmask(OptContext *ctx, TCGOp *op, bool neg) op->opc = xor_opc; op->args[2] = arg_new_constant(ctx, 1); } - return false; + return -1; } } - - return false; + return 0; } static void fold_setcond_tst_pow2(OptContext *ctx, TCGOp *op, bool neg) @@ -XXX,XX +XXX,XX @@ static bool fold_setcond(OptContext *ctx, TCGOp *op) return tcg_opt_gen_movi(ctx, op, op->args[0], i); } - if (fold_setcond_zmask(ctx, op, false)) { + i = fold_setcond_zmask(ctx, op, false); + if (i > 0) { return true; } - fold_setcond_tst_pow2(ctx, op, false); + if (i == 0) { + fold_setcond_tst_pow2(ctx, op, false); + } ctx->z_mask = 1; return false; @@ -XXX,XX +XXX,XX @@ static bool fold_negsetcond(OptContext *ctx, TCGOp *op) return tcg_opt_gen_movi(ctx, op, op->args[0], -i); } - if (fold_setcond_zmask(ctx, op, true)) { + i = fold_setcond_zmask(ctx, op, true); + if (i > 0) { return true; } - fold_setcond_tst_pow2(ctx, op, true); + if (i == 0) { + fold_setcond_tst_pow2(ctx, op, true); + } /* Value is {0,-1} so all bits are repetitions of the sign. */ ctx->s_mask = -1; -- 2.43.0 Avoid the use of the OptContext slots. Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_setcond(OptContext *ctx, TCGOp *op) fold_setcond_tst_pow2(ctx, op, false); } - ctx->z_mask = 1; - return false; + return fold_masks_z(ctx, op, 1); } static bool fold_negsetcond(OptContext *ctx, TCGOp *op) -- 2.43.0 Avoid the use of the OptContext slots. Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_negsetcond(OptContext *ctx, TCGOp *op) } /* Value is {0,-1} so all bits are repetitions of the sign. */ - ctx->s_mask = -1; - return false; + return fold_masks_s(ctx, op, -1); } static bool fold_setcond2(OptContext *ctx, TCGOp *op) -- 2.43.0 Avoid the use of the OptContext slots. Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_setcond2(OptContext *ctx, TCGOp *op) return fold_setcond(ctx, op); } - ctx->z_mask = 1; - return false; + return fold_masks_z(ctx, op, 1); do_setcond_const: return tcg_opt_gen_movi(ctx, op, op->args[0], i); -- 2.43.0 Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_cmp_vec(OptContext *ctx, TCGOp *op) if (swap_commutative(NO_DEST, &op->args[1], &op->args[2])) { op->args[3] = tcg_swap_cond(op->args[3]); } - return false; + return finish_folding(ctx, op); } static bool fold_cmpsel_vec(OptContext *ctx, TCGOp *op) -- 2.43.0 Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_cmpsel_vec(OptContext *ctx, TCGOp *op) if (swap_commutative(op->args[0], &op->args[4], &op->args[3])) { op->args[5] = tcg_invert_cond(op->args[5]); } - return false; + return finish_folding(ctx, op); } static bool fold_sextract(OptContext *ctx, TCGOp *op) -- 2.43.0 Avoid the use of the OptContext slots. Find TempOptInfo once. Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 24 +++++++++--------------- 1 file changed, 9 insertions(+), 15 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_cmpsel_vec(OptContext *ctx, TCGOp *op) static bool fold_sextract(OptContext *ctx, TCGOp *op) { uint64_t z_mask, s_mask, s_mask_old; + TempOptInfo *t1 = arg_info(op->args[1]); int pos = op->args[2]; int len = op->args[3]; - if (arg_is_const(op->args[1])) { - uint64_t t; - - t = arg_info(op->args[1])->val; - t = sextract64(t, pos, len); - return tcg_opt_gen_movi(ctx, op, op->args[0], t); + if (ti_is_const(t1)) { + return tcg_opt_gen_movi(ctx, op, op->args[0], + sextract64(ti_const_val(t1), pos, len)); } - z_mask = arg_info(op->args[1])->z_mask; - z_mask = sextract64(z_mask, pos, len); - ctx->z_mask = z_mask; - - s_mask_old = arg_info(op->args[1])->s_mask; - s_mask = sextract64(s_mask_old, pos, len); - s_mask |= MAKE_64BIT_MASK(len, 64 - len); - ctx->s_mask = s_mask; + s_mask_old = t1->s_mask; + s_mask = s_mask_old >> pos; + s_mask |= -1ull << (len - 1); if (0 && pos == 0 && fold_affected_mask(ctx, op, s_mask & ~s_mask_old)) { return true; } - return fold_masks(ctx, op); + z_mask = sextract64(t1->z_mask, pos, len); + return fold_masks_zs(ctx, op, z_mask, s_mask); } static bool fold_shift(OptContext *ctx, TCGOp *op) -- 2.43.0 Avoid the use of the OptContext slots. Find TempOptInfo once. Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 27 ++++++++++++++------------- 1 file changed, 14 insertions(+), 13 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_sextract(OptContext *ctx, TCGOp *op) static bool fold_shift(OptContext *ctx, TCGOp *op) { uint64_t s_mask, z_mask, sign; + TempOptInfo *t1, *t2; if (fold_const2(ctx, op) || fold_ix_to_i(ctx, op, 0) || @@ -XXX,XX +XXX,XX @@ static bool fold_shift(OptContext *ctx, TCGOp *op) return true; } - s_mask = arg_info(op->args[1])->s_mask; - z_mask = arg_info(op->args[1])->z_mask; + t1 = arg_info(op->args[1]); + t2 = arg_info(op->args[2]); + s_mask = t1->s_mask; + z_mask = t1->z_mask; - if (arg_is_const(op->args[2])) { - int sh = arg_info(op->args[2])->val; - - ctx->z_mask = do_constant_folding(op->opc, ctx->type, z_mask, sh); + if (ti_is_const(t2)) { + int sh = ti_const_val(t2); + z_mask = do_constant_folding(op->opc, ctx->type, z_mask, sh); s_mask = do_constant_folding(op->opc, ctx->type, s_mask, sh); - return fold_masks(ctx, op); + return fold_masks_zs(ctx, op, z_mask, s_mask); } switch (op->opc) { @@ -XXX,XX +XXX,XX @@ static bool fold_shift(OptContext *ctx, TCGOp *op) * Arithmetic right shift will not reduce the number of * input sign repetitions. */ - ctx->s_mask = s_mask; - break; + return fold_masks_s(ctx, op, s_mask); CASE_OP_32_64(shr): /* * If the sign bit is known zero, then logical right shift - * will not reduced the number of input sign repetitions. + * will not reduce the number of input sign repetitions. */ - sign = (s_mask & -s_mask) >> 1; + sign = -s_mask; if (sign && !(z_mask & sign)) { - ctx->s_mask = s_mask; + return fold_masks_s(ctx, op, s_mask); } break; default: break; } - return false; + return finish_folding(ctx, op); } static bool fold_sub_to_neg(OptContext *ctx, TCGOp *op) -- 2.43.0 Merge the two conditions, sign != 0 && !(z_mask & sign), by testing ~z_mask & sign. If sign == 0, the logical and will produce false. Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_sextract(OptContext *ctx, TCGOp *op) static bool fold_shift(OptContext *ctx, TCGOp *op) { - uint64_t s_mask, z_mask, sign; + uint64_t s_mask, z_mask; TempOptInfo *t1, *t2; if (fold_const2(ctx, op) || @@ -XXX,XX +XXX,XX @@ static bool fold_shift(OptContext *ctx, TCGOp *op) * If the sign bit is known zero, then logical right shift * will not reduce the number of input sign repetitions. */ - sign = -s_mask; - if (sign && !(z_mask & sign)) { + if (~z_mask & -s_mask) { return fold_masks_s(ctx, op, s_mask); } break; -- 2.43.0 Duplicate fold_sub_vec into fold_sub instead of calling it, now that fold_sub_vec always returns true. Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_sub_vec(OptContext *ctx, TCGOp *op) fold_sub_to_neg(ctx, op)) { return true; } - return false; + return finish_folding(ctx, op); } static bool fold_sub(OptContext *ctx, TCGOp *op) { - if (fold_const2(ctx, op) || fold_sub_vec(ctx, op)) { + if (fold_const2(ctx, op) || + fold_xx_to_i(ctx, op, 0) || + fold_xi_to_x(ctx, op, 0) || + fold_sub_to_neg(ctx, op)) { return true; } @@ -XXX,XX +XXX,XX @@ static bool fold_sub(OptContext *ctx, TCGOp *op) ? INDEX_op_add_i32 : INDEX_op_add_i64); op->args[2] = arg_new_constant(ctx, -val); } - return false; + return finish_folding(ctx, op); } static bool fold_sub2(OptContext *ctx, TCGOp *op) -- 2.43.0 Avoid the use of the OptContext slots. Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 16 +++++++++------- 1 file changed, 9 insertions(+), 7 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_sub2(OptContext *ctx, TCGOp *op) static bool fold_tcg_ld(OptContext *ctx, TCGOp *op) { + uint64_t z_mask = -1, s_mask = 0; + /* We can't do any folding with a load, but we can record bits. */ switch (op->opc) { CASE_OP_32_64(ld8s): - ctx->s_mask = MAKE_64BIT_MASK(8, 56); + s_mask = INT8_MIN; break; CASE_OP_32_64(ld8u): - ctx->z_mask = MAKE_64BIT_MASK(0, 8); + z_mask = MAKE_64BIT_MASK(0, 8); break; CASE_OP_32_64(ld16s): - ctx->s_mask = MAKE_64BIT_MASK(16, 48); + s_mask = INT16_MIN; break; CASE_OP_32_64(ld16u): - ctx->z_mask = MAKE_64BIT_MASK(0, 16); + z_mask = MAKE_64BIT_MASK(0, 16); break; case INDEX_op_ld32s_i64: - ctx->s_mask = MAKE_64BIT_MASK(32, 32); + s_mask = INT32_MIN; break; case INDEX_op_ld32u_i64: - ctx->z_mask = MAKE_64BIT_MASK(0, 32); + z_mask = MAKE_64BIT_MASK(0, 32); break; default: g_assert_not_reached(); } - return false; + return fold_masks_zs(ctx, op, z_mask, s_mask); } static bool fold_tcg_ld_memcopy(OptContext *ctx, TCGOp *op) -- 2.43.0 Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_tcg_ld_memcopy(OptContext *ctx, TCGOp *op) TCGType type; if (op->args[1] != tcgv_ptr_arg(tcg_env)) { - return false; + return finish_folding(ctx, op); } type = ctx->type; -- 2.43.0 Avoid the use of the OptContext slots. Find TempOptInfo once. Remove fold_masks as the function becomes unused. Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 18 ++++++++---------- 1 file changed, 8 insertions(+), 10 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_masks_s(OptContext *ctx, TCGOp *op, uint64_t s_mask) return fold_masks_zs(ctx, op, -1, s_mask); } -static bool fold_masks(OptContext *ctx, TCGOp *op) -{ - return fold_masks_zs(ctx, op, ctx->z_mask, ctx->s_mask); -} - /* * An "affected" mask bit is 0 if and only if the result is identical * to the first input. Thus if the entire mask is 0, the operation @@ -XXX,XX +XXX,XX @@ static bool fold_tcg_st_memcopy(OptContext *ctx, TCGOp *op) static bool fold_xor(OptContext *ctx, TCGOp *op) { + uint64_t z_mask, s_mask; + TempOptInfo *t1, *t2; + if (fold_const2_commutative(ctx, op) || fold_xx_to_i(ctx, op, 0) || fold_xi_to_x(ctx, op, 0) || @@ -XXX,XX +XXX,XX @@ static bool fold_xor(OptContext *ctx, TCGOp *op) return true; } - ctx->z_mask = arg_info(op->args[1])->z_mask - | arg_info(op->args[2])->z_mask; - ctx->s_mask = arg_info(op->args[1])->s_mask - & arg_info(op->args[2])->s_mask; - return fold_masks(ctx, op); + t1 = arg_info(op->args[1]); + t2 = arg_info(op->args[2]); + z_mask = t1->z_mask | t2->z_mask; + s_mask = t1->s_mask & t2->s_mask; + return fold_masks_zs(ctx, op, z_mask, s_mask); } static bool fold_bitsel_vec(OptContext *ctx, TCGOp *op) -- 2.43.0 Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_bitsel_vec(OptContext *ctx, TCGOp *op) return fold_orc(ctx, op); } } - return false; + return finish_folding(ctx, op); } /* Propagate constants and copies, fold constant expressions. */ -- 2.43.0 All non-default cases now finish folding within each function. Do the same with the default case and assert it is done after. Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ void tcg_optimize(TCGContext *s) done = true; break; default: + done = finish_folding(&ctx, op); break; } - - if (!done) { - finish_folding(&ctx, op); - } + tcg_debug_assert(done); } } -- 2.43.0 All mask setting is now done with parameters via fold_masks_*. Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 13 ------------- 1 file changed, 13 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ typedef struct OptContext { QSIMPLEQ_HEAD(, MemCopyInfo) mem_free; /* In flight values from optimization. */ - uint64_t z_mask; /* mask bit is 0 iff value bit is 0 */ - uint64_t s_mask; /* mask bit is 1 if value bit matches msb */ TCGType type; } OptContext; @@ -XXX,XX +XXX,XX @@ static bool finish_folding(OptContext *ctx, TCGOp *op) for (i = 0; i < nb_oargs; i++) { TCGTemp *ts = arg_temp(op->args[i]); reset_ts(ctx, ts); - /* - * Save the corresponding known-zero/sign bits mask for the - * first output argument (only one supported so far). - */ - if (i == 0) { - ts_info(ts)->z_mask = ctx->z_mask; - } } return true; } @@ -XXX,XX +XXX,XX @@ void tcg_optimize(TCGContext *s) ctx.type = TCG_TYPE_I32; } - /* Assume all bits affected, no bits known zero, no sign reps. */ - ctx.z_mask = -1; - ctx.s_mask = 0; - /* * Process each opcode. * Sorted alphabetically by opcode as much as possible. -- 2.43.0 All instances of s_mask have been converted to the new representation. We can now re-enable usage. Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_exts(OptContext *ctx, TCGOp *op) g_assert_not_reached(); } - if (0 && !type_change && fold_affected_mask(ctx, op, s_mask & ~s_mask_old)) { + if (!type_change && fold_affected_mask(ctx, op, s_mask & ~s_mask_old)) { return true; } @@ -XXX,XX +XXX,XX @@ static bool fold_sextract(OptContext *ctx, TCGOp *op) s_mask = s_mask_old >> pos; s_mask |= -1ull << (len - 1); - if (0 && pos == 0 && fold_affected_mask(ctx, op, s_mask & ~s_mask_old)) { + if (pos == 0 && fold_affected_mask(ctx, op, s_mask & ~s_mask_old)) { return true; } -- 2.43.0 The big comment just above says functions should be sorted. Add forward declarations as needed. Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 114 +++++++++++++++++++++++++------------------------ 1 file changed, 59 insertions(+), 55 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_xx_to_x(OptContext *ctx, TCGOp *op) * 3) those that produce information about the result value. */ +static bool fold_or(OptContext *ctx, TCGOp *op); +static bool fold_orc(OptContext *ctx, TCGOp *op); +static bool fold_xor(OptContext *ctx, TCGOp *op); + static bool fold_add(OptContext *ctx, TCGOp *op) { if (fold_const2_commutative(ctx, op) || @@ -XXX,XX +XXX,XX @@ static bool fold_andc(OptContext *ctx, TCGOp *op) return fold_masks_zs(ctx, op, z_mask, s_mask); } +static bool fold_bitsel_vec(OptContext *ctx, TCGOp *op) +{ + /* If true and false values are the same, eliminate the cmp. */ + if (args_are_copies(op->args[2], op->args[3])) { + return tcg_opt_gen_mov(ctx, op, op->args[0], op->args[2]); + } + + if (arg_is_const(op->args[2]) && arg_is_const(op->args[3])) { + uint64_t tv = arg_info(op->args[2])->val; + uint64_t fv = arg_info(op->args[3])->val; + + if (tv == -1 && fv == 0) { + return tcg_opt_gen_mov(ctx, op, op->args[0], op->args[1]); + } + if (tv == 0 && fv == -1) { + if (TCG_TARGET_HAS_not_vec) { + op->opc = INDEX_op_not_vec; + return fold_not(ctx, op); + } else { + op->opc = INDEX_op_xor_vec; + op->args[2] = arg_new_constant(ctx, -1); + return fold_xor(ctx, op); + } + } + } + if (arg_is_const(op->args[2])) { + uint64_t tv = arg_info(op->args[2])->val; + if (tv == -1) { + op->opc = INDEX_op_or_vec; + op->args[2] = op->args[3]; + return fold_or(ctx, op); + } + if (tv == 0 && TCG_TARGET_HAS_andc_vec) { + op->opc = INDEX_op_andc_vec; + op->args[2] = op->args[1]; + op->args[1] = op->args[3]; + return fold_andc(ctx, op); + } + } + if (arg_is_const(op->args[3])) { + uint64_t fv = arg_info(op->args[3])->val; + if (fv == 0) { + op->opc = INDEX_op_and_vec; + return fold_and(ctx, op); + } + if (fv == -1 && TCG_TARGET_HAS_orc_vec) { + op->opc = INDEX_op_orc_vec; + op->args[2] = op->args[1]; + op->args[1] = op->args[3]; + return fold_orc(ctx, op); + } + } + return finish_folding(ctx, op); +} + static bool fold_brcond(OptContext *ctx, TCGOp *op) { int i = do_constant_folding_cond1(ctx, op, NO_DEST, &op->args[0], @@ -XXX,XX +XXX,XX @@ static bool fold_xor(OptContext *ctx, TCGOp *op) return fold_masks_zs(ctx, op, z_mask, s_mask); } -static bool fold_bitsel_vec(OptContext *ctx, TCGOp *op) -{ - /* If true and false values are the same, eliminate the cmp. */ - if (args_are_copies(op->args[2], op->args[3])) { - return tcg_opt_gen_mov(ctx, op, op->args[0], op->args[2]); - } - - if (arg_is_const(op->args[2]) && arg_is_const(op->args[3])) { - uint64_t tv = arg_info(op->args[2])->val; - uint64_t fv = arg_info(op->args[3])->val; - - if (tv == -1 && fv == 0) { - return tcg_opt_gen_mov(ctx, op, op->args[0], op->args[1]); - } - if (tv == 0 && fv == -1) { - if (TCG_TARGET_HAS_not_vec) { - op->opc = INDEX_op_not_vec; - return fold_not(ctx, op); - } else { - op->opc = INDEX_op_xor_vec; - op->args[2] = arg_new_constant(ctx, -1); - return fold_xor(ctx, op); - } - } - } - if (arg_is_const(op->args[2])) { - uint64_t tv = arg_info(op->args[2])->val; - if (tv == -1) { - op->opc = INDEX_op_or_vec; - op->args[2] = op->args[3]; - return fold_or(ctx, op); - } - if (tv == 0 && TCG_TARGET_HAS_andc_vec) { - op->opc = INDEX_op_andc_vec; - op->args[2] = op->args[1]; - op->args[1] = op->args[3]; - return fold_andc(ctx, op); - } - } - if (arg_is_const(op->args[3])) { - uint64_t fv = arg_info(op->args[3])->val; - if (fv == 0) { - op->opc = INDEX_op_and_vec; - return fold_and(ctx, op); - } - if (fv == -1 && TCG_TARGET_HAS_orc_vec) { - op->opc = INDEX_op_orc_vec; - op->args[2] = op->args[1]; - op->args[1] = op->args[3]; - return fold_orc(ctx, op); - } - } - return finish_folding(ctx, op); -} - /* Propagate constants and copies, fold constant expressions. */ void tcg_optimize(TCGContext *s) { -- 2.43.0 The big comment just above says functions should be sorted. Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tcg/optimize.c | 60 +++++++++++++++++++++++++------------------------- 1 file changed, 30 insertions(+), 30 deletions(-) diff --git a/tcg/optimize.c b/tcg/optimize.c index XXXXXXX..XXXXXXX 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -XXX,XX +XXX,XX @@ static bool fold_call(OptContext *ctx, TCGOp *op) return true; } +static bool fold_cmp_vec(OptContext *ctx, TCGOp *op) +{ + /* Canonicalize the comparison to put immediate second. */ + if (swap_commutative(NO_DEST, &op->args[1], &op->args[2])) { + op->args[3] = tcg_swap_cond(op->args[3]); + } + return finish_folding(ctx, op); +} + +static bool fold_cmpsel_vec(OptContext *ctx, TCGOp *op) +{ + /* If true and false values are the same, eliminate the cmp. */ + if (args_are_copies(op->args[3], op->args[4])) { + return tcg_opt_gen_mov(ctx, op, op->args[0], op->args[3]); + } + + /* Canonicalize the comparison to put immediate second. */ + if (swap_commutative(NO_DEST, &op->args[1], &op->args[2])) { + op->args[5] = tcg_swap_cond(op->args[5]); + } + /* + * Canonicalize the "false" input reg to match the destination, + * so that the tcg backend can implement "move if true". + */ + if (swap_commutative(op->args[0], &op->args[4], &op->args[3])) { + op->args[5] = tcg_invert_cond(op->args[5]); + } + return finish_folding(ctx, op); +} + static bool fold_count_zeros(OptContext *ctx, TCGOp *op) { uint64_t z_mask, s_mask; @@ -XXX,XX +XXX,XX @@ static bool fold_setcond2(OptContext *ctx, TCGOp *op) return tcg_opt_gen_movi(ctx, op, op->args[0], i); } -static bool fold_cmp_vec(OptContext *ctx, TCGOp *op) -{ - /* Canonicalize the comparison to put immediate second. */ - if (swap_commutative(NO_DEST, &op->args[1], &op->args[2])) { - op->args[3] = tcg_swap_cond(op->args[3]); - } - return finish_folding(ctx, op); -} - -static bool fold_cmpsel_vec(OptContext *ctx, TCGOp *op) -{ - /* If true and false values are the same, eliminate the cmp. */ - if (args_are_copies(op->args[3], op->args[4])) { - return tcg_opt_gen_mov(ctx, op, op->args[0], op->args[3]); - } - - /* Canonicalize the comparison to put immediate second. */ - if (swap_commutative(NO_DEST, &op->args[1], &op->args[2])) { - op->args[5] = tcg_swap_cond(op->args[5]); - } - /* - * Canonicalize the "false" input reg to match the destination, - * so that the tcg backend can implement "move if true". - */ - if (swap_commutative(op->args[0], &op->args[4], &op->args[3])) { - op->args[5] = tcg_invert_cond(op->args[5]); - } - return finish_folding(ctx, op); -} - static bool fold_sextract(OptContext *ctx, TCGOp *op) { uint64_t z_mask, s_mask, s_mask_old; -- 2.43.0 We currently have a flag, float_muladd_halve_result, to scale the result by 2**-1. Extend this to handle arbitrary scaling. Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- include/fpu/softfloat.h | 6 ++++ fpu/softfloat.c | 58 ++++++++++++++++++++++----------------- fpu/softfloat-parts.c.inc | 7 +++-- 3 files changed, 44 insertions(+), 27 deletions(-) diff --git a/include/fpu/softfloat.h b/include/fpu/softfloat.h index XXXXXXX..XXXXXXX 100644 --- a/include/fpu/softfloat.h +++ b/include/fpu/softfloat.h @@ -XXX,XX +XXX,XX @@ float16 float16_add(float16, float16, float_status *status); float16 float16_sub(float16, float16, float_status *status); float16 float16_mul(float16, float16, float_status *status); float16 float16_muladd(float16, float16, float16, int, float_status *status); +float16 float16_muladd_scalbn(float16, float16, float16, + int, int, float_status *status); float16 float16_div(float16, float16, float_status *status); float16 float16_scalbn(float16, int, float_status *status); float16 float16_min(float16, float16, float_status *status); @@ -XXX,XX +XXX,XX @@ float32 float32_mul(float32, float32, float_status *status); float32 float32_div(float32, float32, float_status *status); float32 float32_rem(float32, float32, float_status *status); float32 float32_muladd(float32, float32, float32, int, float_status *status); +float32 float32_muladd_scalbn(float32, float32, float32, + int, int, float_status *status); float32 float32_sqrt(float32, float_status *status); float32 float32_exp2(float32, float_status *status); float32 float32_log2(float32, float_status *status); @@ -XXX,XX +XXX,XX @@ float64 float64_mul(float64, float64, float_status *status); float64 float64_div(float64, float64, float_status *status); float64 float64_rem(float64, float64, float_status *status); float64 float64_muladd(float64, float64, float64, int, float_status *status); +float64 float64_muladd_scalbn(float64, float64, float64, + int, int, float_status *status); float64 float64_sqrt(float64, float_status *status); float64 float64_log2(float64, float_status *status); FloatRelation float64_compare(float64, float64, float_status *status); diff --git a/fpu/softfloat.c b/fpu/softfloat.c index XXXXXXX..XXXXXXX 100644 --- a/fpu/softfloat.c +++ b/fpu/softfloat.c @@ -XXX,XX +XXX,XX @@ static FloatParts128 *parts128_mul(FloatParts128 *a, FloatParts128 *b, #define parts_mul(A, B, S) \ PARTS_GENERIC_64_128(mul, A)(A, B, S) -static FloatParts64 *parts64_muladd(FloatParts64 *a, FloatParts64 *b, - FloatParts64 *c, int flags, - float_status *s); -static FloatParts128 *parts128_muladd(FloatParts128 *a, FloatParts128 *b, - FloatParts128 *c, int flags, - float_status *s); +static FloatParts64 *parts64_muladd_scalbn(FloatParts64 *a, FloatParts64 *b, + FloatParts64 *c, int scale, + int flags, float_status *s); +static FloatParts128 *parts128_muladd_scalbn(FloatParts128 *a, FloatParts128 *b, + FloatParts128 *c, int scale, + int flags, float_status *s); -#define parts_muladd(A, B, C, Z, S) \ - PARTS_GENERIC_64_128(muladd, A)(A, B, C, Z, S) +#define parts_muladd_scalbn(A, B, C, Z, Y, S) \ + PARTS_GENERIC_64_128(muladd_scalbn, A)(A, B, C, Z, Y, S) static FloatParts64 *parts64_div(FloatParts64 *a, FloatParts64 *b, float_status *s); @@ -XXX,XX +XXX,XX @@ floatx80_mul(floatx80 a, floatx80 b, float_status *status) * Fused multiply-add */ -float16 QEMU_FLATTEN float16_muladd(float16 a, float16 b, float16 c, - int flags, float_status *status) +float16 QEMU_FLATTEN +float16_muladd_scalbn(float16 a, float16 b, float16 c, + int scale, int flags, float_status *status) { FloatParts64 pa, pb, pc, *pr; float16_unpack_canonical(&pa, a, status); float16_unpack_canonical(&pb, b, status); float16_unpack_canonical(&pc, c, status); - pr = parts_muladd(&pa, &pb, &pc, flags, status); + pr = parts_muladd_scalbn(&pa, &pb, &pc, scale, flags, status); return float16_round_pack_canonical(pr, status); } -static float32 QEMU_SOFTFLOAT_ATTR -soft_f32_muladd(float32 a, float32 b, float32 c, int flags, - float_status *status) +float16 float16_muladd(float16 a, float16 b, float16 c, + int flags, float_status *status) +{ + return float16_muladd_scalbn(a, b, c, 0, flags, status); +} + +float32 QEMU_SOFTFLOAT_ATTR +float32_muladd_scalbn(float32 a, float32 b, float32 c, + int scale, int flags, float_status *status) { FloatParts64 pa, pb, pc, *pr; float32_unpack_canonical(&pa, a, status); float32_unpack_canonical(&pb, b, status); float32_unpack_canonical(&pc, c, status); - pr = parts_muladd(&pa, &pb, &pc, flags, status); + pr = parts_muladd_scalbn(&pa, &pb, &pc, scale, flags, status); return float32_round_pack_canonical(pr, status); } -static float64 QEMU_SOFTFLOAT_ATTR -soft_f64_muladd(float64 a, float64 b, float64 c, int flags, - float_status *status) +float64 QEMU_SOFTFLOAT_ATTR +float64_muladd_scalbn(float64 a, float64 b, float64 c, + int scale, int flags, float_status *status) { FloatParts64 pa, pb, pc, *pr; float64_unpack_canonical(&pa, a, status); float64_unpack_canonical(&pb, b, status); float64_unpack_canonical(&pc, c, status); - pr = parts_muladd(&pa, &pb, &pc, flags, status); + pr = parts_muladd_scalbn(&pa, &pb, &pc, scale, flags, status); return float64_round_pack_canonical(pr, status); } @@ -XXX,XX +XXX,XX @@ float32_muladd(float32 xa, float32 xb, float32 xc, int flags, float_status *s) return ur.s; soft: - return soft_f32_muladd(ua.s, ub.s, uc.s, flags, s); + return float32_muladd_scalbn(ua.s, ub.s, uc.s, 0, flags, s); } float64 QEMU_FLATTEN @@ -XXX,XX +XXX,XX @@ float64_muladd(float64 xa, float64 xb, float64 xc, int flags, float_status *s) return ur.s; soft: - return soft_f64_muladd(ua.s, ub.s, uc.s, flags, s); + return float64_muladd_scalbn(ua.s, ub.s, uc.s, 0, flags, s); } float64 float64r32_muladd(float64 a, float64 b, float64 c, @@ -XXX,XX +XXX,XX @@ float64 float64r32_muladd(float64 a, float64 b, float64 c, float64_unpack_canonical(&pa, a, status); float64_unpack_canonical(&pb, b, status); float64_unpack_canonical(&pc, c, status); - pr = parts_muladd(&pa, &pb, &pc, flags, status); + pr = parts_muladd_scalbn(&pa, &pb, &pc, 0, flags, status); return float64r32_round_pack_canonical(pr, status); } @@ -XXX,XX +XXX,XX @@ bfloat16 QEMU_FLATTEN bfloat16_muladd(bfloat16 a, bfloat16 b, bfloat16 c, bfloat16_unpack_canonical(&pa, a, status); bfloat16_unpack_canonical(&pb, b, status); bfloat16_unpack_canonical(&pc, c, status); - pr = parts_muladd(&pa, &pb, &pc, flags, status); + pr = parts_muladd_scalbn(&pa, &pb, &pc, 0, flags, status); return bfloat16_round_pack_canonical(pr, status); } @@ -XXX,XX +XXX,XX @@ float128 QEMU_FLATTEN float128_muladd(float128 a, float128 b, float128 c, float128_unpack_canonical(&pa, a, status); float128_unpack_canonical(&pb, b, status); float128_unpack_canonical(&pc, c, status); - pr = parts_muladd(&pa, &pb, &pc, flags, status); + pr = parts_muladd_scalbn(&pa, &pb, &pc, 0, flags, status); return float128_round_pack_canonical(pr, status); } @@ -XXX,XX +XXX,XX @@ float32 float32_exp2(float32 a, float_status *status) float64_unpack_canonical(&rp, float64_one, status); for (i = 0 ; i < 15 ; i++) { + float64_unpack_canonical(&tp, float32_exp2_coefficients[i], status); - rp = *parts_muladd(&tp, &xnp, &rp, 0, status); + rp = *parts_muladd_scalbn(&tp, &xnp, &rp, 0, 0, status); xnp = *parts_mul(&xnp, &xp, status); } diff --git a/fpu/softfloat-parts.c.inc b/fpu/softfloat-parts.c.inc index XXXXXXX..XXXXXXX 100644 --- a/fpu/softfloat-parts.c.inc +++ b/fpu/softfloat-parts.c.inc @@ -XXX,XX +XXX,XX @@ static FloatPartsN *partsN(mul)(FloatPartsN *a, FloatPartsN *b, * Requires A and C extracted into a double-sized structure to provide the * extra space for the widening multiply. */ -static FloatPartsN *partsN(muladd)(FloatPartsN *a, FloatPartsN *b, - FloatPartsN *c, int flags, float_status *s) +static FloatPartsN *partsN(muladd_scalbn)(FloatPartsN *a, FloatPartsN *b, + FloatPartsN *c, int scale, + int flags, float_status *s) { int ab_mask, abc_mask; FloatPartsW p_widen, c_widen; @@ -XXX,XX +XXX,XX @@ static FloatPartsN *partsN(muladd)(FloatPartsN *a, FloatPartsN *b, a->exp = p_widen.exp; return_normal: + /* TODO: Replace all use of float_muladd_halve_result with scale. */ if (flags & float_muladd_halve_result) { a->exp -= 1; } + a->exp += scale; finish_sign: if (flags & float_muladd_negate_result) { a->sign ^= 1; -- 2.43.0 Use the scalbn interface instead of float_muladd_halve_result. Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/arm/tcg/helper-a64.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/target/arm/tcg/helper-a64.c b/target/arm/tcg/helper-a64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/helper-a64.c +++ b/target/arm/tcg/helper-a64.c @@ -XXX,XX +XXX,XX @@ uint32_t HELPER(rsqrtsf_f16)(uint32_t a, uint32_t b, float_status *fpst) (float16_is_infinity(b) && float16_is_zero(a))) { return float16_one_point_five; } - return float16_muladd(a, b, float16_three, float_muladd_halve_result, fpst); + return float16_muladd_scalbn(a, b, float16_three, -1, 0, fpst); } float32 HELPER(rsqrtsf_f32)(float32 a, float32 b, float_status *fpst) @@ -XXX,XX +XXX,XX @@ float32 HELPER(rsqrtsf_f32)(float32 a, float32 b, float_status *fpst) (float32_is_infinity(b) && float32_is_zero(a))) { return float32_one_point_five; } - return float32_muladd(a, b, float32_three, float_muladd_halve_result, fpst); + return float32_muladd_scalbn(a, b, float32_three, -1, 0, fpst); } float64 HELPER(rsqrtsf_f64)(float64 a, float64 b, float_status *fpst) @@ -XXX,XX +XXX,XX @@ float64 HELPER(rsqrtsf_f64)(float64 a, float64 b, float_status *fpst) (float64_is_infinity(b) && float64_is_zero(a))) { return float64_one_point_five; } - return float64_muladd(a, b, float64_three, float_muladd_halve_result, fpst); + return float64_muladd_scalbn(a, b, float64_three, -1, 0, fpst); } /* Floating-point reciprocal exponent - see FPRecpX in ARM ARM */ -- 2.43.0 Use the scalbn interface instead of float_muladd_halve_result. Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/sparc/helper.h | 4 +- target/sparc/fop_helper.c | 8 ++-- target/sparc/translate.c | 80 +++++++++++++++++++++++---------------- 3 files changed, 54 insertions(+), 38 deletions(-) diff --git a/target/sparc/helper.h b/target/sparc/helper.h index XXXXXXX..XXXXXXX 100644 --- a/target/sparc/helper.h +++ b/target/sparc/helper.h @@ -XXX,XX +XXX,XX @@ DEF_HELPER_FLAGS_3(faddd, TCG_CALL_NO_WG, f64, env, f64, f64) DEF_HELPER_FLAGS_3(fsubd, TCG_CALL_NO_WG, f64, env, f64, f64) DEF_HELPER_FLAGS_3(fmuld, TCG_CALL_NO_WG, f64, env, f64, f64) DEF_HELPER_FLAGS_3(fdivd, TCG_CALL_NO_WG, f64, env, f64, f64) -DEF_HELPER_FLAGS_5(fmaddd, TCG_CALL_NO_WG, f64, env, f64, f64, f64, i32) +DEF_HELPER_FLAGS_6(fmaddd, TCG_CALL_NO_WG, f64, env, f64, f64, f64, s32, i32) DEF_HELPER_FLAGS_3(fnaddd, TCG_CALL_NO_WG, f64, env, f64, f64) DEF_HELPER_FLAGS_3(fnmuld, TCG_CALL_NO_WG, f64, env, f64, f64) @@ -XXX,XX +XXX,XX @@ DEF_HELPER_FLAGS_3(fadds, TCG_CALL_NO_WG, f32, env, f32, f32) DEF_HELPER_FLAGS_3(fsubs, TCG_CALL_NO_WG, f32, env, f32, f32) DEF_HELPER_FLAGS_3(fmuls, TCG_CALL_NO_WG, f32, env, f32, f32) DEF_HELPER_FLAGS_3(fdivs, TCG_CALL_NO_WG, f32, env, f32, f32) -DEF_HELPER_FLAGS_5(fmadds, TCG_CALL_NO_WG, f32, env, f32, f32, f32, i32) +DEF_HELPER_FLAGS_6(fmadds, TCG_CALL_NO_WG, f32, env, f32, f32, f32, s32, i32) DEF_HELPER_FLAGS_3(fnadds, TCG_CALL_NO_WG, f32, env, f32, f32) DEF_HELPER_FLAGS_3(fnmuls, TCG_CALL_NO_WG, f32, env, f32, f32) diff --git a/target/sparc/fop_helper.c b/target/sparc/fop_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/sparc/fop_helper.c +++ b/target/sparc/fop_helper.c @@ -XXX,XX +XXX,XX @@ Int128 helper_fsqrtq(CPUSPARCState *env, Int128 src) } float32 helper_fmadds(CPUSPARCState *env, float32 s1, - float32 s2, float32 s3, uint32_t op) + float32 s2, float32 s3, int32_t sc, uint32_t op) { - float32 ret = float32_muladd(s1, s2, s3, op, &env->fp_status); + float32 ret = float32_muladd_scalbn(s1, s2, s3, sc, op, &env->fp_status); check_ieee_exceptions(env, GETPC()); return ret; } float64 helper_fmaddd(CPUSPARCState *env, float64 s1, - float64 s2, float64 s3, uint32_t op) + float64 s2, float64 s3, int32_t sc, uint32_t op) { - float64 ret = float64_muladd(s1, s2, s3, op, &env->fp_status); + float64 ret = float64_muladd_scalbn(s1, s2, s3, sc, op, &env->fp_status); check_ieee_exceptions(env, GETPC()); return ret; } diff --git a/target/sparc/translate.c b/target/sparc/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/sparc/translate.c +++ b/target/sparc/translate.c @@ -XXX,XX +XXX,XX @@ static void gen_op_fabsq(TCGv_i128 dst, TCGv_i128 src) static void gen_op_fmadds(TCGv_i32 d, TCGv_i32 s1, TCGv_i32 s2, TCGv_i32 s3) { - gen_helper_fmadds(d, tcg_env, s1, s2, s3, tcg_constant_i32(0)); + TCGv_i32 z = tcg_constant_i32(0); + gen_helper_fmadds(d, tcg_env, s1, s2, s3, z, z); } static void gen_op_fmaddd(TCGv_i64 d, TCGv_i64 s1, TCGv_i64 s2, TCGv_i64 s3) { - gen_helper_fmaddd(d, tcg_env, s1, s2, s3, tcg_constant_i32(0)); + TCGv_i32 z = tcg_constant_i32(0); + gen_helper_fmaddd(d, tcg_env, s1, s2, s3, z, z); } static void gen_op_fmsubs(TCGv_i32 d, TCGv_i32 s1, TCGv_i32 s2, TCGv_i32 s3) { - int op = float_muladd_negate_c; - gen_helper_fmadds(d, tcg_env, s1, s2, s3, tcg_constant_i32(op)); + TCGv_i32 z = tcg_constant_i32(0); + TCGv_i32 op = tcg_constant_i32(float_muladd_negate_c); + gen_helper_fmadds(d, tcg_env, s1, s2, s3, z, op); } static void gen_op_fmsubd(TCGv_i64 d, TCGv_i64 s1, TCGv_i64 s2, TCGv_i64 s3) { - int op = float_muladd_negate_c; - gen_helper_fmaddd(d, tcg_env, s1, s2, s3, tcg_constant_i32(op)); + TCGv_i32 z = tcg_constant_i32(0); + TCGv_i32 op = tcg_constant_i32(float_muladd_negate_c); + gen_helper_fmaddd(d, tcg_env, s1, s2, s3, z, op); } static void gen_op_fnmsubs(TCGv_i32 d, TCGv_i32 s1, TCGv_i32 s2, TCGv_i32 s3) { - int op = float_muladd_negate_c | float_muladd_negate_result; - gen_helper_fmadds(d, tcg_env, s1, s2, s3, tcg_constant_i32(op)); + TCGv_i32 z = tcg_constant_i32(0); + TCGv_i32 op = tcg_constant_i32(float_muladd_negate_c | + float_muladd_negate_result); + gen_helper_fmadds(d, tcg_env, s1, s2, s3, z, op); } static void gen_op_fnmsubd(TCGv_i64 d, TCGv_i64 s1, TCGv_i64 s2, TCGv_i64 s3) { - int op = float_muladd_negate_c | float_muladd_negate_result; - gen_helper_fmaddd(d, tcg_env, s1, s2, s3, tcg_constant_i32(op)); + TCGv_i32 z = tcg_constant_i32(0); + TCGv_i32 op = tcg_constant_i32(float_muladd_negate_c | + float_muladd_negate_result); + gen_helper_fmaddd(d, tcg_env, s1, s2, s3, z, op); } static void gen_op_fnmadds(TCGv_i32 d, TCGv_i32 s1, TCGv_i32 s2, TCGv_i32 s3) { - int op = float_muladd_negate_result; - gen_helper_fmadds(d, tcg_env, s1, s2, s3, tcg_constant_i32(op)); + TCGv_i32 z = tcg_constant_i32(0); + TCGv_i32 op = tcg_constant_i32(float_muladd_negate_result); + gen_helper_fmadds(d, tcg_env, s1, s2, s3, z, op); } static void gen_op_fnmaddd(TCGv_i64 d, TCGv_i64 s1, TCGv_i64 s2, TCGv_i64 s3) { - int op = float_muladd_negate_result; - gen_helper_fmaddd(d, tcg_env, s1, s2, s3, tcg_constant_i32(op)); + TCGv_i32 z = tcg_constant_i32(0); + TCGv_i32 op = tcg_constant_i32(float_muladd_negate_result); + gen_helper_fmaddd(d, tcg_env, s1, s2, s3, z, op); } /* Use muladd to compute (1 * src1) + src2 / 2 with one rounding. */ static void gen_op_fhadds(TCGv_i32 d, TCGv_i32 s1, TCGv_i32 s2) { - TCGv_i32 one = tcg_constant_i32(float32_one); - int op = float_muladd_halve_result; - gen_helper_fmadds(d, tcg_env, one, s1, s2, tcg_constant_i32(op)); + TCGv_i32 fone = tcg_constant_i32(float32_one); + TCGv_i32 mone = tcg_constant_i32(-1); + TCGv_i32 op = tcg_constant_i32(0); + gen_helper_fmadds(d, tcg_env, fone, s1, s2, mone, op); } static void gen_op_fhaddd(TCGv_i64 d, TCGv_i64 s1, TCGv_i64 s2) { - TCGv_i64 one = tcg_constant_i64(float64_one); - int op = float_muladd_halve_result; - gen_helper_fmaddd(d, tcg_env, one, s1, s2, tcg_constant_i32(op)); + TCGv_i64 fone = tcg_constant_i64(float64_one); + TCGv_i32 mone = tcg_constant_i32(-1); + TCGv_i32 op = tcg_constant_i32(0); + gen_helper_fmaddd(d, tcg_env, fone, s1, s2, mone, op); } /* Use muladd to compute (1 * src1) - src2 / 2 with one rounding. */ static void gen_op_fhsubs(TCGv_i32 d, TCGv_i32 s1, TCGv_i32 s2) { - TCGv_i32 one = tcg_constant_i32(float32_one); - int op = float_muladd_negate_c | float_muladd_halve_result; - gen_helper_fmadds(d, tcg_env, one, s1, s2, tcg_constant_i32(op)); + TCGv_i32 fone = tcg_constant_i32(float32_one); + TCGv_i32 mone = tcg_constant_i32(-1); + TCGv_i32 op = tcg_constant_i32(float_muladd_negate_c); + gen_helper_fmadds(d, tcg_env, fone, s1, s2, mone, op); } static void gen_op_fhsubd(TCGv_i64 d, TCGv_i64 s1, TCGv_i64 s2) { - TCGv_i64 one = tcg_constant_i64(float64_one); - int op = float_muladd_negate_c | float_muladd_halve_result; - gen_helper_fmaddd(d, tcg_env, one, s1, s2, tcg_constant_i32(op)); + TCGv_i64 fone = tcg_constant_i64(float64_one); + TCGv_i32 mone = tcg_constant_i32(-1); + TCGv_i32 op = tcg_constant_i32(float_muladd_negate_c); + gen_helper_fmaddd(d, tcg_env, fone, s1, s2, mone, op); } /* Use muladd to compute -((1 * src1) + src2 / 2) with one rounding. */ static void gen_op_fnhadds(TCGv_i32 d, TCGv_i32 s1, TCGv_i32 s2) { - TCGv_i32 one = tcg_constant_i32(float32_one); - int op = float_muladd_negate_result | float_muladd_halve_result; - gen_helper_fmadds(d, tcg_env, one, s1, s2, tcg_constant_i32(op)); + TCGv_i32 fone = tcg_constant_i32(float32_one); + TCGv_i32 mone = tcg_constant_i32(-1); + TCGv_i32 op = tcg_constant_i32(float_muladd_negate_result); + gen_helper_fmadds(d, tcg_env, fone, s1, s2, mone, op); } static void gen_op_fnhaddd(TCGv_i64 d, TCGv_i64 s1, TCGv_i64 s2) { - TCGv_i64 one = tcg_constant_i64(float64_one); - int op = float_muladd_negate_result | float_muladd_halve_result; - gen_helper_fmaddd(d, tcg_env, one, s1, s2, tcg_constant_i32(op)); + TCGv_i64 fone = tcg_constant_i64(float64_one); + TCGv_i32 mone = tcg_constant_i32(-1); + TCGv_i32 op = tcg_constant_i32(float_muladd_negate_result); + gen_helper_fmaddd(d, tcg_env, fone, s1, s2, mone, op); } static void gen_op_fpexception_im(DisasContext *dc, int ftt) -- 2.43.0 All uses have been convered to float*_muladd_scalbn. Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- include/fpu/softfloat.h | 3 --- fpu/softfloat.c | 6 ------ fpu/softfloat-parts.c.inc | 4 ---- 3 files changed, 13 deletions(-) diff --git a/include/fpu/softfloat.h b/include/fpu/softfloat.h index XXXXXXX..XXXXXXX 100644 --- a/include/fpu/softfloat.h +++ b/include/fpu/softfloat.h @@ -XXX,XX +XXX,XX @@ bfloat16 bfloat16_squash_input_denormal(bfloat16 a, float_status *status); | Using these differs from negating an input or output before calling | the muladd function in that this means that a NaN doesn't have its | sign bit inverted before it is propagated. -| We also support halving the result before rounding, as a special -| case to support the ARM fused-sqrt-step instruction FRSQRTS. *----------------------------------------------------------------------------*/ enum { float_muladd_negate_c = 1, float_muladd_negate_product = 2, float_muladd_negate_result = 4, - float_muladd_halve_result = 8, }; /*---------------------------------------------------------------------------- diff --git a/fpu/softfloat.c b/fpu/softfloat.c index XXXXXXX..XXXXXXX 100644 --- a/fpu/softfloat.c +++ b/fpu/softfloat.c @@ -XXX,XX +XXX,XX @@ float32_muladd(float32 xa, float32 xb, float32 xc, int flags, float_status *s) if (unlikely(!can_use_fpu(s))) { goto soft; } - if (unlikely(flags & float_muladd_halve_result)) { - goto soft; - } float32_input_flush3(&ua.s, &ub.s, &uc.s, s); if (unlikely(!f32_is_zon3(ua, ub, uc))) { @@ -XXX,XX +XXX,XX @@ float64_muladd(float64 xa, float64 xb, float64 xc, int flags, float_status *s) if (unlikely(!can_use_fpu(s))) { goto soft; } - if (unlikely(flags & float_muladd_halve_result)) { - goto soft; - } float64_input_flush3(&ua.s, &ub.s, &uc.s, s); if (unlikely(!f64_is_zon3(ua, ub, uc))) { diff --git a/fpu/softfloat-parts.c.inc b/fpu/softfloat-parts.c.inc index XXXXXXX..XXXXXXX 100644 --- a/fpu/softfloat-parts.c.inc +++ b/fpu/softfloat-parts.c.inc @@ -XXX,XX +XXX,XX @@ static FloatPartsN *partsN(muladd_scalbn)(FloatPartsN *a, FloatPartsN *b, a->exp = p_widen.exp; return_normal: - /* TODO: Replace all use of float_muladd_halve_result with scale. */ - if (flags & float_muladd_halve_result) { - a->exp -= 1; - } a->exp += scale; finish_sign: if (flags & float_muladd_negate_result) { -- 2.43.0 This rounding mode is used by Hexagon. Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- include/fpu/softfloat-types.h | 2 ++ fpu/softfloat-parts.c.inc | 3 +++ 2 files changed, 5 insertions(+) diff --git a/include/fpu/softfloat-types.h b/include/fpu/softfloat-types.h index XXXXXXX..XXXXXXX 100644 --- a/include/fpu/softfloat-types.h +++ b/include/fpu/softfloat-types.h @@ -XXX,XX +XXX,XX @@ typedef enum __attribute__((__packed__)) { float_round_to_odd = 5, /* Not an IEEE rounding mode: round to closest odd, overflow to inf */ float_round_to_odd_inf = 6, + /* Not an IEEE rounding mode: round to nearest even, overflow to max */ + float_round_nearest_even_max = 7, } FloatRoundMode; /* diff --git a/fpu/softfloat-parts.c.inc b/fpu/softfloat-parts.c.inc index XXXXXXX..XXXXXXX 100644 --- a/fpu/softfloat-parts.c.inc +++ b/fpu/softfloat-parts.c.inc @@ -XXX,XX +XXX,XX @@ static void partsN(uncanon_normal)(FloatPartsN *p, float_status *s, int exp, flags = 0; switch (s->float_rounding_mode) { + case float_round_nearest_even_max: + overflow_norm = true; + /* fall through */ case float_round_nearest_even: if (N > 64 && frac_lsb == 0) { inc = ((p->frac_hi & 1) || (p->frac_lo & round_mask) != frac_lsbm1 -- 2.43.0 Certain Hexagon instructions suppress changes to the result when the product of fma() is a true zero. Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- include/fpu/softfloat.h | 5 +++++ fpu/softfloat.c | 3 +++ fpu/softfloat-parts.c.inc | 4 +++- 3 files changed, 11 insertions(+), 1 deletion(-) diff --git a/include/fpu/softfloat.h b/include/fpu/softfloat.h index XXXXXXX..XXXXXXX 100644 --- a/include/fpu/softfloat.h +++ b/include/fpu/softfloat.h @@ -XXX,XX +XXX,XX @@ bfloat16 bfloat16_squash_input_denormal(bfloat16 a, float_status *status); | Using these differs from negating an input or output before calling | the muladd function in that this means that a NaN doesn't have its | sign bit inverted before it is propagated. +| +| With float_muladd_suppress_add_product_zero, if A or B is zero +| such that the product is a true zero, then return C without addition. +| This preserves the sign of C when C is +/- 0. Used for Hexagon. *----------------------------------------------------------------------------*/ enum { float_muladd_negate_c = 1, float_muladd_negate_product = 2, float_muladd_negate_result = 4, + float_muladd_suppress_add_product_zero = 8, }; /*---------------------------------------------------------------------------- diff --git a/fpu/softfloat.c b/fpu/softfloat.c index XXXXXXX..XXXXXXX 100644 --- a/fpu/softfloat.c +++ b/fpu/softfloat.c @@ -XXX,XX +XXX,XX @@ float32_muladd(float32 xa, float32 xb, float32 xc, int flags, float_status *s) if (unlikely(!can_use_fpu(s))) { goto soft; } + if (unlikely(flags & float_muladd_suppress_add_product_zero)) { + goto soft; + } float32_input_flush3(&ua.s, &ub.s, &uc.s, s); if (unlikely(!f32_is_zon3(ua, ub, uc))) { diff --git a/fpu/softfloat-parts.c.inc b/fpu/softfloat-parts.c.inc index XXXXXXX..XXXXXXX 100644 --- a/fpu/softfloat-parts.c.inc +++ b/fpu/softfloat-parts.c.inc @@ -XXX,XX +XXX,XX @@ static FloatPartsN *partsN(muladd_scalbn)(FloatPartsN *a, FloatPartsN *b, goto return_normal; } if (c->cls == float_class_zero) { - if (a->sign != c->sign) { + if (flags & float_muladd_suppress_add_product_zero) { + a->sign = c->sign; + } else if (a->sign != c->sign) { goto return_sub_zero; } goto return_zero; -- 2.43.0 There are no special cases for this instruction. Remove internal_mpyf as unused. Reviewed-by: Brian Cain <brian.cain@oss.qualcomm.com> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/hexagon/fma_emu.h | 1 - target/hexagon/fma_emu.c | 8 -------- target/hexagon/op_helper.c | 2 +- 3 files changed, 1 insertion(+), 10 deletions(-) diff --git a/target/hexagon/fma_emu.h b/target/hexagon/fma_emu.h index XXXXXXX..XXXXXXX 100644 --- a/target/hexagon/fma_emu.h +++ b/target/hexagon/fma_emu.h @@ -XXX,XX +XXX,XX @@ int32_t float32_getexp(float32 f32); float32 infinite_float32(uint8_t sign); float32 internal_fmafx(float32 a, float32 b, float32 c, int scale, float_status *fp_status); -float32 internal_mpyf(float32 a, float32 b, float_status *fp_status); float64 internal_mpyhh(float64 a, float64 b, unsigned long long int accumulated, float_status *fp_status); diff --git a/target/hexagon/fma_emu.c b/target/hexagon/fma_emu.c index XXXXXXX..XXXXXXX 100644 --- a/target/hexagon/fma_emu.c +++ b/target/hexagon/fma_emu.c @@ -XXX,XX +XXX,XX @@ float32 internal_fmafx(float32 a, float32 b, float32 c, int scale, return accum_round_float32(result, fp_status); } -float32 internal_mpyf(float32 a, float32 b, float_status *fp_status) -{ - if (float32_is_zero(a) || float32_is_zero(b)) { - return float32_mul(a, b, fp_status); - } - return internal_fmafx(a, b, float32_zero, 0, fp_status); -} - float64 internal_mpyhh(float64 a, float64 b, unsigned long long int accumulated, float_status *fp_status) diff --git a/target/hexagon/op_helper.c b/target/hexagon/op_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/hexagon/op_helper.c +++ b/target/hexagon/op_helper.c @@ -XXX,XX +XXX,XX @@ float32 HELPER(sfmpy)(CPUHexagonState *env, float32 RsV, float32 RtV) { float32 RdV; arch_fpop_start(env); - RdV = internal_mpyf(RsV, RtV, &env->fp_status); + RdV = float32_mul(RsV, RtV, &env->fp_status); arch_fpop_end(env); return RdV; } -- 2.43.0 There are no special cases for this instruction. Reviewed-by: Brian Cain <brian.cain@oss.qualcomm.com> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/hexagon/op_helper.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/target/hexagon/op_helper.c b/target/hexagon/op_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/hexagon/op_helper.c +++ b/target/hexagon/op_helper.c @@ -XXX,XX +XXX,XX @@ float32 HELPER(sffma)(CPUHexagonState *env, float32 RxV, float32 RsV, float32 RtV) { arch_fpop_start(env); - RxV = internal_fmafx(RsV, RtV, RxV, 0, &env->fp_status); + RxV = float32_muladd(RsV, RtV, RxV, 0, &env->fp_status); arch_fpop_end(env); return RxV; } -- 2.43.0 There are no special cases for this instruction. Since hexagon always uses default-nan mode, explicitly negating the first input is unnecessary. Use float_muladd_negate_product instead. Reviewed-by: Brian Cain <brian.cain@oss.qualcomm.com> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/hexagon/op_helper.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/target/hexagon/op_helper.c b/target/hexagon/op_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/hexagon/op_helper.c +++ b/target/hexagon/op_helper.c @@ -XXX,XX +XXX,XX @@ float32 HELPER(sffma_sc)(CPUHexagonState *env, float32 RxV, float32 HELPER(sffms)(CPUHexagonState *env, float32 RxV, float32 RsV, float32 RtV) { - float32 neg_RsV; arch_fpop_start(env); - neg_RsV = float32_set_sign(RsV, float32_is_neg(RsV) ? 0 : 1); - RxV = internal_fmafx(neg_RsV, RtV, RxV, 0, &env->fp_status); + RxV = float32_muladd(RsV, RtV, RxV, float_muladd_negate_product, + &env->fp_status); arch_fpop_end(env); return RxV; } -- 2.43.0 This instruction has a special case that 0 * x + c returns c without the normal sign folding that comes with 0 + -0. Use the new float_muladd_suppress_add_product_zero to describe this. Reviewed-by: Brian Cain <brian.cain@oss.qualcomm.com> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/hexagon/op_helper.c | 11 +++-------- 1 file changed, 3 insertions(+), 8 deletions(-) diff --git a/target/hexagon/op_helper.c b/target/hexagon/op_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/hexagon/op_helper.c +++ b/target/hexagon/op_helper.c @@ -XXX,XX +XXX,XX @@ static float32 check_nan(float32 dst, float32 x, float_status *fp_status) float32 HELPER(sffma_sc)(CPUHexagonState *env, float32 RxV, float32 RsV, float32 RtV, float32 PuV) { - size4s_t tmp; arch_fpop_start(env); - RxV = check_nan(RxV, RxV, &env->fp_status); - RxV = check_nan(RxV, RsV, &env->fp_status); - RxV = check_nan(RxV, RtV, &env->fp_status); - tmp = internal_fmafx(RsV, RtV, RxV, fSXTN(8, 64, PuV), &env->fp_status); - if (!(float32_is_zero(RxV) && is_zero_prod(RsV, RtV))) { - RxV = tmp; - } + RxV = float32_muladd_scalbn(RsV, RtV, RxV, fSXTN(8, 64, PuV), + float_muladd_suppress_add_product_zero, + &env->fp_status); arch_fpop_end(env); return RxV; } -- 2.43.0 There are multiple special cases for this instruction. (1) The saturate to normal maximum instead of overflow to infinity is handled by the new float_round_nearest_even_max rounding mode. (2) The 0 * n + c special case is handled by the new float_muladd_suppress_add_product_zero flag. (3) The Inf - Inf -> 0 special case can be detected after the fact by examining float_flag_invalid_isi. Reviewed-by: Brian Cain <brian.cain@oss.qualcomm.com> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/hexagon/op_helper.c | 105 +++++++++---------------------------- 1 file changed, 26 insertions(+), 79 deletions(-) diff --git a/target/hexagon/op_helper.c b/target/hexagon/op_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/hexagon/op_helper.c +++ b/target/hexagon/op_helper.c @@ -XXX,XX +XXX,XX @@ float32 HELPER(sffma)(CPUHexagonState *env, float32 RxV, return RxV; } -static bool is_zero_prod(float32 a, float32 b) -{ - return ((float32_is_zero(a) && is_finite(b)) || - (float32_is_zero(b) && is_finite(a))); -} - -static float32 check_nan(float32 dst, float32 x, float_status *fp_status) -{ - float32 ret = dst; - if (float32_is_any_nan(x)) { - if (extract32(x, 22, 1) == 0) { - float_raise(float_flag_invalid, fp_status); - } - ret = make_float32(0xffffffff); /* nan */ - } - return ret; -} - float32 HELPER(sffma_sc)(CPUHexagonState *env, float32 RxV, float32 RsV, float32 RtV, float32 PuV) { @@ -XXX,XX +XXX,XX @@ float32 HELPER(sffms)(CPUHexagonState *env, float32 RxV, return RxV; } -static bool is_inf_prod(int32_t a, int32_t b) +static float32 do_sffma_lib(CPUHexagonState *env, float32 RxV, + float32 RsV, float32 RtV, int negate) { - return (float32_is_infinity(a) && float32_is_infinity(b)) || - (float32_is_infinity(a) && is_finite(b) && !float32_is_zero(b)) || - (float32_is_infinity(b) && is_finite(a) && !float32_is_zero(a)); + int flags; + + arch_fpop_start(env); + + set_float_rounding_mode(float_round_nearest_even_max, &env->fp_status); + RxV = float32_muladd(RsV, RtV, RxV, + negate | float_muladd_suppress_add_product_zero, + &env->fp_status); + + flags = get_float_exception_flags(&env->fp_status); + if (flags) { + /* Flags are suppressed by this instruction. */ + set_float_exception_flags(0, &env->fp_status); + + /* Return 0 for Inf - Inf. */ + if (flags & float_flag_invalid_isi) { + RxV = 0; + } + } + + arch_fpop_end(env); + return RxV; } float32 HELPER(sffma_lib)(CPUHexagonState *env, float32 RxV, float32 RsV, float32 RtV) { - bool infinp; - bool infminusinf; - float32 tmp; - - arch_fpop_start(env); - set_float_rounding_mode(float_round_nearest_even, &env->fp_status); - infminusinf = float32_is_infinity(RxV) && - is_inf_prod(RsV, RtV) && - (fGETBIT(31, RsV ^ RxV ^ RtV) != 0); - infinp = float32_is_infinity(RxV) || - float32_is_infinity(RtV) || - float32_is_infinity(RsV); - RxV = check_nan(RxV, RxV, &env->fp_status); - RxV = check_nan(RxV, RsV, &env->fp_status); - RxV = check_nan(RxV, RtV, &env->fp_status); - tmp = internal_fmafx(RsV, RtV, RxV, 0, &env->fp_status); - if (!(float32_is_zero(RxV) && is_zero_prod(RsV, RtV))) { - RxV = tmp; - } - set_float_exception_flags(0, &env->fp_status); - if (float32_is_infinity(RxV) && !infinp) { - RxV = RxV - 1; - } - if (infminusinf) { - RxV = 0; - } - arch_fpop_end(env); - return RxV; + return do_sffma_lib(env, RxV, RsV, RtV, 0); } float32 HELPER(sffms_lib)(CPUHexagonState *env, float32 RxV, float32 RsV, float32 RtV) { - bool infinp; - bool infminusinf; - float32 tmp; - - arch_fpop_start(env); - set_float_rounding_mode(float_round_nearest_even, &env->fp_status); - infminusinf = float32_is_infinity(RxV) && - is_inf_prod(RsV, RtV) && - (fGETBIT(31, RsV ^ RxV ^ RtV) == 0); - infinp = float32_is_infinity(RxV) || - float32_is_infinity(RtV) || - float32_is_infinity(RsV); - RxV = check_nan(RxV, RxV, &env->fp_status); - RxV = check_nan(RxV, RsV, &env->fp_status); - RxV = check_nan(RxV, RtV, &env->fp_status); - float32 minus_RsV = float32_sub(float32_zero, RsV, &env->fp_status); - tmp = internal_fmafx(minus_RsV, RtV, RxV, 0, &env->fp_status); - if (!(float32_is_zero(RxV) && is_zero_prod(RsV, RtV))) { - RxV = tmp; - } - set_float_exception_flags(0, &env->fp_status); - if (float32_is_infinity(RxV) && !infinp) { - RxV = RxV - 1; - } - if (infminusinf) { - RxV = 0; - } - arch_fpop_end(env); - return RxV; + return do_sffma_lib(env, RxV, RsV, RtV, float_muladd_negate_product); } float64 HELPER(dfmpyfix)(CPUHexagonState *env, float64 RssV, float64 RttV) -- 2.43.0 The function is now unused. Reviewed-by: Brian Cain <brian.cain@oss.qualcomm.com> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/hexagon/fma_emu.h | 2 - target/hexagon/fma_emu.c | 171 --------------------------------------- 2 files changed, 173 deletions(-) diff --git a/target/hexagon/fma_emu.h b/target/hexagon/fma_emu.h index XXXXXXX..XXXXXXX 100644 --- a/target/hexagon/fma_emu.h +++ b/target/hexagon/fma_emu.h @@ -XXX,XX +XXX,XX @@ static inline uint32_t float32_getexp_raw(float32 f32) } int32_t float32_getexp(float32 f32); float32 infinite_float32(uint8_t sign); -float32 internal_fmafx(float32 a, float32 b, float32 c, - int scale, float_status *fp_status); float64 internal_mpyhh(float64 a, float64 b, unsigned long long int accumulated, float_status *fp_status); diff --git a/target/hexagon/fma_emu.c b/target/hexagon/fma_emu.c index XXXXXXX..XXXXXXX 100644 --- a/target/hexagon/fma_emu.c +++ b/target/hexagon/fma_emu.c @@ -XXX,XX +XXX,XX @@ int32_t float64_getexp(float64 f64) return -1; } -static uint64_t float32_getmant(float32 f32) -{ - Float a = { .i = f32 }; - if (float32_is_normal(f32)) { - return a.mant | 1ULL << 23; - } - if (float32_is_zero(f32)) { - return 0; - } - if (float32_is_denormal(f32)) { - return a.mant; - } - return ~0ULL; -} - int32_t float32_getexp(float32 f32) { Float a = { .i = f32 }; @@ -XXX,XX +XXX,XX @@ float32 infinite_float32(uint8_t sign) } /* Return a maximum finite value with the requested sign */ -static float32 maxfinite_float32(uint8_t sign) -{ - if (sign) { - return make_float32(SF_MINUS_MAXF); - } else { - return make_float32(SF_MAXF); - } -} - -/* Return a zero value with requested sign */ -static float32 zero_float32(uint8_t sign) -{ - if (sign) { - return make_float32(0x80000000); - } else { - return float32_zero; - } -} - #define GEN_XF_ROUND(SUFFIX, MANTBITS, INF_EXP, INTERNAL_TYPE) \ static SUFFIX accum_round_##SUFFIX(Accum a, float_status * fp_status) \ { \ @@ -XXX,XX +XXX,XX @@ static SUFFIX accum_round_##SUFFIX(Accum a, float_status * fp_status) \ } GEN_XF_ROUND(float64, DF_MANTBITS, DF_INF_EXP, Double) -GEN_XF_ROUND(float32, SF_MANTBITS, SF_INF_EXP, Float) - -static bool is_inf_prod(float64 a, float64 b) -{ - return ((float64_is_infinity(a) && float64_is_infinity(b)) || - (float64_is_infinity(a) && is_finite(b) && (!float64_is_zero(b))) || - (float64_is_infinity(b) && is_finite(a) && (!float64_is_zero(a)))); -} - -static float64 special_fma(float64 a, float64 b, float64 c, - float_status *fp_status) -{ - float64 ret = make_float64(0); - - /* - * If A multiplied by B is an exact infinity and C is also an infinity - * but with the opposite sign, FMA returns NaN and raises invalid. - */ - uint8_t a_sign = float64_is_neg(a); - uint8_t b_sign = float64_is_neg(b); - uint8_t c_sign = float64_is_neg(c); - if (is_inf_prod(a, b) && float64_is_infinity(c)) { - if ((a_sign ^ b_sign) != c_sign) { - ret = make_float64(DF_NAN); - float_raise(float_flag_invalid, fp_status); - return ret; - } - } - if ((float64_is_infinity(a) && float64_is_zero(b)) || - (float64_is_zero(a) && float64_is_infinity(b))) { - ret = make_float64(DF_NAN); - float_raise(float_flag_invalid, fp_status); - return ret; - } - /* - * If none of the above checks are true and C is a NaN, - * a NaN shall be returned - * If A or B are NaN, a NAN shall be returned. - */ - if (float64_is_any_nan(a) || - float64_is_any_nan(b) || - float64_is_any_nan(c)) { - if (float64_is_any_nan(a) && (fGETBIT(51, a) == 0)) { - float_raise(float_flag_invalid, fp_status); - } - if (float64_is_any_nan(b) && (fGETBIT(51, b) == 0)) { - float_raise(float_flag_invalid, fp_status); - } - if (float64_is_any_nan(c) && (fGETBIT(51, c) == 0)) { - float_raise(float_flag_invalid, fp_status); - } - ret = make_float64(DF_NAN); - return ret; - } - /* - * We have checked for adding opposite-signed infinities. - * Other infinities return infinity with the correct sign - */ - if (float64_is_infinity(c)) { - ret = infinite_float64(c_sign); - return ret; - } - if (float64_is_infinity(a) || float64_is_infinity(b)) { - ret = infinite_float64(a_sign ^ b_sign); - return ret; - } - g_assert_not_reached(); -} - -static float32 special_fmaf(float32 a, float32 b, float32 c, - float_status *fp_status) -{ - float64 aa, bb, cc; - aa = float32_to_float64(a, fp_status); - bb = float32_to_float64(b, fp_status); - cc = float32_to_float64(c, fp_status); - return float64_to_float32(special_fma(aa, bb, cc, fp_status), fp_status); -} - -float32 internal_fmafx(float32 a, float32 b, float32 c, int scale, - float_status *fp_status) -{ - Accum prod; - Accum acc; - Accum result; - accum_init(&prod); - accum_init(&acc); - accum_init(&result); - - uint8_t a_sign = float32_is_neg(a); - uint8_t b_sign = float32_is_neg(b); - uint8_t c_sign = float32_is_neg(c); - if (float32_is_infinity(a) || - float32_is_infinity(b) || - float32_is_infinity(c)) { - return special_fmaf(a, b, c, fp_status); - } - if (float32_is_any_nan(a) || - float32_is_any_nan(b) || - float32_is_any_nan(c)) { - return special_fmaf(a, b, c, fp_status); - } - if ((scale == 0) && (float32_is_zero(a) || float32_is_zero(b))) { - float32 tmp = float32_mul(a, b, fp_status); - tmp = float32_add(tmp, c, fp_status); - return tmp; - } - - /* (a * 2**b) * (c * 2**d) == a*c * 2**(b+d) */ - prod.mant = int128_mul_6464(float32_getmant(a), float32_getmant(b)); - - /* - * Note: extracting the mantissa into an int is multiplying by - * 2**23, so adjust here - */ - prod.exp = float32_getexp(a) + float32_getexp(b) - SF_BIAS - 23; - prod.sign = a_sign ^ b_sign; - if (float32_is_zero(a) || float32_is_zero(b)) { - prod.exp = -2 * WAY_BIG_EXP; - } - if ((scale > 0) && float32_is_denormal(c)) { - acc.mant = int128_mul_6464(0, 0); - acc.exp = -WAY_BIG_EXP; - acc.sign = c_sign; - acc.sticky = 1; - result = accum_add(prod, acc); - } else if (!float32_is_zero(c)) { - acc.mant = int128_mul_6464(float32_getmant(c), 1); - acc.exp = float32_getexp(c); - acc.sign = c_sign; - result = accum_add(prod, acc); - } else { - result = prod; - } - result.exp += scale; - return accum_round_float32(result, fp_status); -} float64 internal_mpyhh(float64 a, float64 b, unsigned long long int accumulated, -- 2.43.0 This massive macro is now only used once. Expand it for use only by float64. Reviewed-by: Brian Cain <brian.cain@oss.qualcomm.com> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/hexagon/fma_emu.c | 255 +++++++++++++++++++-------------------- 1 file changed, 127 insertions(+), 128 deletions(-) diff --git a/target/hexagon/fma_emu.c b/target/hexagon/fma_emu.c index XXXXXXX..XXXXXXX 100644 --- a/target/hexagon/fma_emu.c +++ b/target/hexagon/fma_emu.c @@ -XXX,XX +XXX,XX @@ float32 infinite_float32(uint8_t sign) } /* Return a maximum finite value with the requested sign */ -#define GEN_XF_ROUND(SUFFIX, MANTBITS, INF_EXP, INTERNAL_TYPE) \ -static SUFFIX accum_round_##SUFFIX(Accum a, float_status * fp_status) \ -{ \ - if ((int128_gethi(a.mant) == 0) && (int128_getlo(a.mant) == 0) \ - && ((a.guard | a.round | a.sticky) == 0)) { \ - /* result zero */ \ - switch (fp_status->float_rounding_mode) { \ - case float_round_down: \ - return zero_##SUFFIX(1); \ - default: \ - return zero_##SUFFIX(0); \ - } \ - } \ - /* Normalize right */ \ - /* We want MANTBITS bits of mantissa plus the leading one. */ \ - /* That means that we want MANTBITS+1 bits, or 0x000000000000FF_FFFF */ \ - /* So we need to normalize right while the high word is non-zero and \ - * while the low word is nonzero when masked with 0xffe0_0000_0000_0000 */ \ - while ((int128_gethi(a.mant) != 0) || \ - ((int128_getlo(a.mant) >> (MANTBITS + 1)) != 0)) { \ - a = accum_norm_right(a, 1); \ - } \ - /* \ - * OK, now normalize left \ - * We want to normalize left until we have a leading one in bit 24 \ - * Theoretically, we only need to shift a maximum of one to the left if we \ - * shifted out lots of bits from B, or if we had no shift / 1 shift sticky \ - * should be 0 \ - */ \ - while ((int128_getlo(a.mant) & (1ULL << MANTBITS)) == 0) { \ - a = accum_norm_left(a); \ - } \ - /* \ - * OK, now we might need to denormalize because of potential underflow. \ - * We need to do this before rounding, and rounding might make us normal \ - * again \ - */ \ - while (a.exp <= 0) { \ - a = accum_norm_right(a, 1 - a.exp); \ - /* \ - * Do we have underflow? \ - * That's when we get an inexact answer because we ran out of bits \ - * in a denormal. \ - */ \ - if (a.guard || a.round || a.sticky) { \ - float_raise(float_flag_underflow, fp_status); \ - } \ - } \ - /* OK, we're relatively canonical... now we need to round */ \ - if (a.guard || a.round || a.sticky) { \ - float_raise(float_flag_inexact, fp_status); \ - switch (fp_status->float_rounding_mode) { \ - case float_round_to_zero: \ - /* Chop and we're done */ \ - break; \ - case float_round_up: \ - if (a.sign == 0) { \ - a.mant = int128_add(a.mant, int128_one()); \ - } \ - break; \ - case float_round_down: \ - if (a.sign != 0) { \ - a.mant = int128_add(a.mant, int128_one()); \ - } \ - break; \ - default: \ - if (a.round || a.sticky) { \ - /* round up if guard is 1, down if guard is zero */ \ - a.mant = int128_add(a.mant, int128_make64(a.guard)); \ - } else if (a.guard) { \ - /* exactly .5, round up if odd */ \ - a.mant = int128_add(a.mant, int128_and(a.mant, int128_one())); \ - } \ - break; \ - } \ - } \ - /* \ - * OK, now we might have carried all the way up. \ - * So we might need to shr once \ - * at least we know that the lsb should be zero if we rounded and \ - * got a carry out... \ - */ \ - if ((int128_getlo(a.mant) >> (MANTBITS + 1)) != 0) { \ - a = accum_norm_right(a, 1); \ - } \ - /* Overflow? */ \ - if (a.exp >= INF_EXP) { \ - /* Yep, inf result */ \ - float_raise(float_flag_overflow, fp_status); \ - float_raise(float_flag_inexact, fp_status); \ - switch (fp_status->float_rounding_mode) { \ - case float_round_to_zero: \ - return maxfinite_##SUFFIX(a.sign); \ - case float_round_up: \ - if (a.sign == 0) { \ - return infinite_##SUFFIX(a.sign); \ - } else { \ - return maxfinite_##SUFFIX(a.sign); \ - } \ - case float_round_down: \ - if (a.sign != 0) { \ - return infinite_##SUFFIX(a.sign); \ - } else { \ - return maxfinite_##SUFFIX(a.sign); \ - } \ - default: \ - return infinite_##SUFFIX(a.sign); \ - } \ - } \ - /* Underflow? */ \ - if (int128_getlo(a.mant) & (1ULL << MANTBITS)) { \ - /* Leading one means: No, we're normal. So, we should be done... */ \ - INTERNAL_TYPE ret; \ - ret.i = 0; \ - ret.sign = a.sign; \ - ret.exp = a.exp; \ - ret.mant = int128_getlo(a.mant); \ - return ret.i; \ - } \ - assert(a.exp == 1); \ - INTERNAL_TYPE ret; \ - ret.i = 0; \ - ret.sign = a.sign; \ - ret.exp = 0; \ - ret.mant = int128_getlo(a.mant); \ - return ret.i; \ +static float64 accum_round_float64(Accum a, float_status *fp_status) +{ + if ((int128_gethi(a.mant) == 0) && (int128_getlo(a.mant) == 0) + && ((a.guard | a.round | a.sticky) == 0)) { + /* result zero */ + switch (fp_status->float_rounding_mode) { + case float_round_down: + return zero_float64(1); + default: + return zero_float64(0); + } + } + /* + * Normalize right + * We want DF_MANTBITS bits of mantissa plus the leading one. + * That means that we want DF_MANTBITS+1 bits, or 0x000000000000FF_FFFF + * So we need to normalize right while the high word is non-zero and + * while the low word is nonzero when masked with 0xffe0_0000_0000_0000 + */ + while ((int128_gethi(a.mant) != 0) || + ((int128_getlo(a.mant) >> (DF_MANTBITS + 1)) != 0)) { + a = accum_norm_right(a, 1); + } + /* + * OK, now normalize left + * We want to normalize left until we have a leading one in bit 24 + * Theoretically, we only need to shift a maximum of one to the left if we + * shifted out lots of bits from B, or if we had no shift / 1 shift sticky + * should be 0 + */ + while ((int128_getlo(a.mant) & (1ULL << DF_MANTBITS)) == 0) { + a = accum_norm_left(a); + } + /* + * OK, now we might need to denormalize because of potential underflow. + * We need to do this before rounding, and rounding might make us normal + * again + */ + while (a.exp <= 0) { + a = accum_norm_right(a, 1 - a.exp); + /* + * Do we have underflow? + * That's when we get an inexact answer because we ran out of bits + * in a denormal. + */ + if (a.guard || a.round || a.sticky) { + float_raise(float_flag_underflow, fp_status); + } + } + /* OK, we're relatively canonical... now we need to round */ + if (a.guard || a.round || a.sticky) { + float_raise(float_flag_inexact, fp_status); + switch (fp_status->float_rounding_mode) { + case float_round_to_zero: + /* Chop and we're done */ + break; + case float_round_up: + if (a.sign == 0) { + a.mant = int128_add(a.mant, int128_one()); + } + break; + case float_round_down: + if (a.sign != 0) { + a.mant = int128_add(a.mant, int128_one()); + } + break; + default: + if (a.round || a.sticky) { + /* round up if guard is 1, down if guard is zero */ + a.mant = int128_add(a.mant, int128_make64(a.guard)); + } else if (a.guard) { + /* exactly .5, round up if odd */ + a.mant = int128_add(a.mant, int128_and(a.mant, int128_one())); + } + break; + } + } + /* + * OK, now we might have carried all the way up. + * So we might need to shr once + * at least we know that the lsb should be zero if we rounded and + * got a carry out... + */ + if ((int128_getlo(a.mant) >> (DF_MANTBITS + 1)) != 0) { + a = accum_norm_right(a, 1); + } + /* Overflow? */ + if (a.exp >= DF_INF_EXP) { + /* Yep, inf result */ + float_raise(float_flag_overflow, fp_status); + float_raise(float_flag_inexact, fp_status); + switch (fp_status->float_rounding_mode) { + case float_round_to_zero: + return maxfinite_float64(a.sign); + case float_round_up: + if (a.sign == 0) { + return infinite_float64(a.sign); + } else { + return maxfinite_float64(a.sign); + } + case float_round_down: + if (a.sign != 0) { + return infinite_float64(a.sign); + } else { + return maxfinite_float64(a.sign); + } + default: + return infinite_float64(a.sign); + } + } + /* Underflow? */ + if (int128_getlo(a.mant) & (1ULL << DF_MANTBITS)) { + /* Leading one means: No, we're normal. So, we should be done... */ + Double ret; + ret.i = 0; + ret.sign = a.sign; + ret.exp = a.exp; + ret.mant = int128_getlo(a.mant); + return ret.i; + } + assert(a.exp == 1); + Double ret; + ret.i = 0; + ret.sign = a.sign; + ret.exp = 0; + ret.mant = int128_getlo(a.mant); + return ret.i; } -GEN_XF_ROUND(float64, DF_MANTBITS, DF_INF_EXP, Double) - float64 internal_mpyhh(float64 a, float64 b, unsigned long long int accumulated, float_status *fp_status) -- 2.43.0 This structure, with bitfields, is incorrect for big-endian. Use the existing float32_getexp_raw which uses extract32. Reviewed-by: Brian Cain <brian.cain@oss.qualcomm.com> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/hexagon/fma_emu.c | 16 +++------------- 1 file changed, 3 insertions(+), 13 deletions(-) diff --git a/target/hexagon/fma_emu.c b/target/hexagon/fma_emu.c index XXXXXXX..XXXXXXX 100644 --- a/target/hexagon/fma_emu.c +++ b/target/hexagon/fma_emu.c @@ -XXX,XX +XXX,XX @@ typedef union { }; } Double; -typedef union { - float f; - uint32_t i; - struct { - uint32_t mant:23; - uint32_t exp:8; - uint32_t sign:1; - }; -} Float; - static uint64_t float64_getmant(float64 f64) { Double a = { .i = f64 }; @@ -XXX,XX +XXX,XX @@ int32_t float64_getexp(float64 f64) int32_t float32_getexp(float32 f32) { - Float a = { .i = f32 }; + int exp = float32_getexp_raw(f32); if (float32_is_normal(f32)) { - return a.exp; + return exp; } if (float32_is_denormal(f32)) { - return a.exp + 1; + return exp + 1; } return -1; } -- 2.43.0 This structure, with bitfields, is incorrect for big-endian. Use extract64 and deposit64 instead. Reviewed-by: Brian Cain <brian.cain@oss.qualcomm.com> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/hexagon/fma_emu.c | 46 ++++++++++++++-------------------------- 1 file changed, 16 insertions(+), 30 deletions(-) diff --git a/target/hexagon/fma_emu.c b/target/hexagon/fma_emu.c index XXXXXXX..XXXXXXX 100644 --- a/target/hexagon/fma_emu.c +++ b/target/hexagon/fma_emu.c @@ -XXX,XX +XXX,XX @@ #define WAY_BIG_EXP 4096 -typedef union { - double f; - uint64_t i; - struct { - uint64_t mant:52; - uint64_t exp:11; - uint64_t sign:1; - }; -} Double; - static uint64_t float64_getmant(float64 f64) { - Double a = { .i = f64 }; + uint64_t mant = extract64(f64, 0, 52); if (float64_is_normal(f64)) { - return a.mant | 1ULL << 52; + return mant | 1ULL << 52; } if (float64_is_zero(f64)) { return 0; } if (float64_is_denormal(f64)) { - return a.mant; + return mant; } return ~0ULL; } int32_t float64_getexp(float64 f64) { - Double a = { .i = f64 }; + int exp = extract64(f64, 52, 11); if (float64_is_normal(f64)) { - return a.exp; + return exp; } if (float64_is_denormal(f64)) { - return a.exp + 1; + return exp + 1; } return -1; } @@ -XXX,XX +XXX,XX @@ float32 infinite_float32(uint8_t sign) /* Return a maximum finite value with the requested sign */ static float64 accum_round_float64(Accum a, float_status *fp_status) { + uint64_t ret; + if ((int128_gethi(a.mant) == 0) && (int128_getlo(a.mant) == 0) && ((a.guard | a.round | a.sticky) == 0)) { /* result zero */ @@ -XXX,XX +XXX,XX @@ static float64 accum_round_float64(Accum a, float_status *fp_status) } } /* Underflow? */ - if (int128_getlo(a.mant) & (1ULL << DF_MANTBITS)) { + ret = int128_getlo(a.mant); + if (ret & (1ULL << DF_MANTBITS)) { /* Leading one means: No, we're normal. So, we should be done... */ - Double ret; - ret.i = 0; - ret.sign = a.sign; - ret.exp = a.exp; - ret.mant = int128_getlo(a.mant); - return ret.i; + ret = deposit64(ret, 52, 11, a.exp); + } else { + assert(a.exp == 1); + ret = deposit64(ret, 52, 11, 0); } - assert(a.exp == 1); - Double ret; - ret.i = 0; - ret.sign = a.sign; - ret.exp = 0; - ret.mant = int128_getlo(a.mant); - return ret.i; + ret = deposit64(ret, 63, 1, a.sign); + return ret; } float64 internal_mpyhh(float64 a, float64 b, -- 2.43.0 No need to open-code 64x64->128-bit multiplication. Reviewed-by: Brian Cain <brian.cain@oss.qualcomm.com> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/hexagon/fma_emu.c | 32 +++----------------------------- 1 file changed, 3 insertions(+), 29 deletions(-) diff --git a/target/hexagon/fma_emu.c b/target/hexagon/fma_emu.c index XXXXXXX..XXXXXXX 100644 --- a/target/hexagon/fma_emu.c +++ b/target/hexagon/fma_emu.c @@ -XXX,XX +XXX,XX @@ int32_t float32_getexp(float32 f32) return -1; } -static uint32_t int128_getw0(Int128 x) -{ - return int128_getlo(x); -} - -static uint32_t int128_getw1(Int128 x) -{ - return int128_getlo(x) >> 32; -} - static Int128 int128_mul_6464(uint64_t ai, uint64_t bi) { - Int128 a, b; - uint64_t pp0, pp1a, pp1b, pp1s, pp2; + uint64_t l, h; - a = int128_make64(ai); - b = int128_make64(bi); - pp0 = (uint64_t)int128_getw0(a) * (uint64_t)int128_getw0(b); - pp1a = (uint64_t)int128_getw1(a) * (uint64_t)int128_getw0(b); - pp1b = (uint64_t)int128_getw1(b) * (uint64_t)int128_getw0(a); - pp2 = (uint64_t)int128_getw1(a) * (uint64_t)int128_getw1(b); - - pp1s = pp1a + pp1b; - if ((pp1s < pp1a) || (pp1s < pp1b)) { - pp2 += (1ULL << 32); - } - uint64_t ret_low = pp0 + (pp1s << 32); - if ((ret_low < pp0) || (ret_low < (pp1s << 32))) { - pp2 += 1; - } - - return int128_make128(ret_low, pp2 + (pp1s >> 32)); + mulu64(&l, &h, ai, bi); + return int128_make128(l, h); } static Int128 int128_sub_borrow(Int128 a, Int128 b, int borrow) -- 2.43.0 Initialize x with accumulated via direct assignment, rather than multiplying by 1. Reviewed-by: Brian Cain <brian.cain@oss.qualcomm.com> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- target/hexagon/fma_emu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/target/hexagon/fma_emu.c b/target/hexagon/fma_emu.c index XXXXXXX..XXXXXXX 100644 --- a/target/hexagon/fma_emu.c +++ b/target/hexagon/fma_emu.c @@ -XXX,XX +XXX,XX @@ float64 internal_mpyhh(float64 a, float64 b, float64_is_infinity(b)) { return float64_mul(a, b, fp_status); } - x.mant = int128_mul_6464(accumulated, 1); + x.mant = int128_make64(accumulated); x.sticky = sticky; prod = fGETUWORD(1, float64_getmant(a)) * fGETUWORD(1, float64_getmant(b)); x.mant = int128_add(x.mant, int128_mul_6464(prod, 0x100000000ULL)); -- 2.43.0 Convert all targets simultaneously, as the gen_intermediate_code function disappears from the target. While there are possible workarounds, they're larger than simply performing the conversion. Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- include/exec/translator.h | 14 -------------- include/hw/core/tcg-cpu-ops.h | 13 +++++++++++++ target/alpha/cpu.h | 2 ++ target/arm/internals.h | 2 ++ target/avr/cpu.h | 2 ++ target/hexagon/cpu.h | 2 ++ target/hppa/cpu.h | 2 ++ target/i386/tcg/helper-tcg.h | 2 ++ target/loongarch/internals.h | 2 ++ target/m68k/cpu.h | 2 ++ target/microblaze/cpu.h | 2 ++ target/mips/tcg/tcg-internal.h | 2 ++ target/openrisc/cpu.h | 2 ++ target/ppc/cpu.h | 2 ++ target/riscv/cpu.h | 3 +++ target/rx/cpu.h | 2 ++ target/s390x/s390x-internal.h | 2 ++ target/sh4/cpu.h | 2 ++ target/sparc/cpu.h | 2 ++ target/tricore/cpu.h | 2 ++ target/xtensa/cpu.h | 2 ++ accel/tcg/cpu-exec.c | 8 +++++--- accel/tcg/translate-all.c | 8 +++++--- target/alpha/cpu.c | 1 + target/alpha/translate.c | 4 ++-- target/arm/cpu.c | 1 + target/arm/tcg/cpu-v7m.c | 1 + target/arm/tcg/translate.c | 5 ++--- target/avr/cpu.c | 1 + target/avr/translate.c | 6 +++--- target/hexagon/cpu.c | 1 + target/hexagon/translate.c | 4 ++-- target/hppa/cpu.c | 1 + target/hppa/translate.c | 4 ++-- target/i386/tcg/tcg-cpu.c | 1 + target/i386/tcg/translate.c | 5 ++--- target/loongarch/cpu.c | 1 + target/loongarch/tcg/translate.c | 4 ++-- target/m68k/cpu.c | 1 + target/m68k/translate.c | 4 ++-- target/microblaze/cpu.c | 1 + target/microblaze/translate.c | 4 ++-- target/mips/cpu.c | 1 + target/mips/tcg/translate.c | 4 ++-- target/openrisc/cpu.c | 1 + target/openrisc/translate.c | 4 ++-- target/ppc/cpu_init.c | 1 + target/ppc/translate.c | 4 ++-- target/riscv/tcg/tcg-cpu.c | 1 + target/riscv/translate.c | 4 ++-- target/rx/cpu.c | 1 + target/rx/translate.c | 4 ++-- target/s390x/cpu.c | 1 + target/s390x/tcg/translate.c | 4 ++-- target/sh4/cpu.c | 1 + target/sh4/translate.c | 4 ++-- target/sparc/cpu.c | 1 + target/sparc/translate.c | 4 ++-- target/tricore/cpu.c | 1 + target/tricore/translate.c | 5 ++--- target/xtensa/cpu.c | 1 + target/xtensa/translate.c | 4 ++-- 62 files changed, 121 insertions(+), 62 deletions(-) diff --git a/include/exec/translator.h b/include/exec/translator.h index XXXXXXX..XXXXXXX 100644 --- a/include/exec/translator.h +++ b/include/exec/translator.h @@ -XXX,XX +XXX,XX @@ #include "qemu/bswap.h" #include "exec/vaddr.h" -/** - * gen_intermediate_code - * @cpu: cpu context - * @tb: translation block - * @max_insns: max number of instructions to translate - * @pc: guest virtual program counter address - * @host_pc: host physical program counter address - * - * This function must be provided by the target, which should create - * the target-specific DisasContext, and then invoke translator_loop. - */ -void gen_intermediate_code(CPUState *cpu, TranslationBlock *tb, int *max_insns, - vaddr pc, void *host_pc); - /** * DisasJumpType: * @DISAS_NEXT: Next instruction in program order. diff --git a/include/hw/core/tcg-cpu-ops.h b/include/hw/core/tcg-cpu-ops.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/core/tcg-cpu-ops.h +++ b/include/hw/core/tcg-cpu-ops.h @@ -XXX,XX +XXX,XX @@ struct TCGCPUOps { * Called when the first CPU is realized. */ void (*initialize)(void); + /** + * @translate_code: Translate guest instructions to TCGOps + * @cpu: cpu context + * @tb: translation block + * @max_insns: max number of instructions to translate + * @pc: guest virtual program counter address + * @host_pc: host physical program counter address + * + * This function must be provided by the target, which should create + * the target-specific DisasContext, and then invoke translator_loop. + */ + void (*translate_code)(CPUState *cpu, TranslationBlock *tb, + int *max_insns, vaddr pc, void *host_pc); /** * @synchronize_from_tb: Synchronize state from a TCG #TranslationBlock * diff --git a/target/alpha/cpu.h b/target/alpha/cpu.h index XXXXXXX..XXXXXXX 100644 --- a/target/alpha/cpu.h +++ b/target/alpha/cpu.h @@ -XXX,XX +XXX,XX @@ enum { }; void alpha_translate_init(void); +void alpha_translate_code(CPUState *cs, TranslationBlock *tb, + int *max_insns, vaddr pc, void *host_pc); #define CPU_RESOLVING_TYPE TYPE_ALPHA_CPU diff --git a/target/arm/internals.h b/target/arm/internals.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/internals.h +++ b/target/arm/internals.h @@ -XXX,XX +XXX,XX @@ void init_cpreg_list(ARMCPU *cpu); void arm_cpu_register_gdb_regs_for_features(ARMCPU *cpu); void arm_translate_init(void); +void arm_translate_code(CPUState *cs, TranslationBlock *tb, + int *max_insns, vaddr pc, void *host_pc); void arm_cpu_register_gdb_commands(ARMCPU *cpu); void aarch64_cpu_register_gdb_commands(ARMCPU *cpu, GString *, diff --git a/target/avr/cpu.h b/target/avr/cpu.h index XXXXXXX..XXXXXXX 100644 --- a/target/avr/cpu.h +++ b/target/avr/cpu.h @@ -XXX,XX +XXX,XX @@ static inline void set_avr_feature(CPUAVRState *env, int feature) } void avr_cpu_tcg_init(void); +void avr_cpu_translate_code(CPUState *cs, TranslationBlock *tb, + int *max_insns, vaddr pc, void *host_pc); int cpu_avr_exec(CPUState *cpu); diff --git a/target/hexagon/cpu.h b/target/hexagon/cpu.h index XXXXXXX..XXXXXXX 100644 --- a/target/hexagon/cpu.h +++ b/target/hexagon/cpu.h @@ -XXX,XX +XXX,XX @@ static inline void cpu_get_tb_cpu_state(CPUHexagonState *env, vaddr *pc, typedef HexagonCPU ArchCPU; void hexagon_translate_init(void); +void hexagon_translate_code(CPUState *cs, TranslationBlock *tb, + int *max_insns, vaddr pc, void *host_pc); #include "exec/cpu-all.h" diff --git a/target/hppa/cpu.h b/target/hppa/cpu.h index XXXXXXX..XXXXXXX 100644 --- a/target/hppa/cpu.h +++ b/target/hppa/cpu.h @@ -XXX,XX +XXX,XX @@ static inline int HPPA_BTLB_ENTRIES(CPUHPPAState *env) } void hppa_translate_init(void); +void hppa_translate_code(CPUState *cs, TranslationBlock *tb, + int *max_insns, vaddr pc, void *host_pc); #define CPU_RESOLVING_TYPE TYPE_HPPA_CPU diff --git a/target/i386/tcg/helper-tcg.h b/target/i386/tcg/helper-tcg.h index XXXXXXX..XXXXXXX 100644 --- a/target/i386/tcg/helper-tcg.h +++ b/target/i386/tcg/helper-tcg.h @@ -XXX,XX +XXX,XX @@ static inline target_long lshift(target_long x, int n) /* translate.c */ void tcg_x86_init(void); +void x86_translate_code(CPUState *cs, TranslationBlock *tb, + int *max_insns, vaddr pc, void *host_pc); /* excp_helper.c */ G_NORETURN void raise_exception(CPUX86State *env, int exception_index); diff --git a/target/loongarch/internals.h b/target/loongarch/internals.h index XXXXXXX..XXXXXXX 100644 --- a/target/loongarch/internals.h +++ b/target/loongarch/internals.h @@ -XXX,XX +XXX,XX @@ #define TARGET_VIRT_MASK MAKE_64BIT_MASK(0, TARGET_VIRT_ADDR_SPACE_BITS) void loongarch_translate_init(void); +void loongarch_translate_code(CPUState *cs, TranslationBlock *tb, + int *max_insns, vaddr pc, void *host_pc); void G_NORETURN do_raise_exception(CPULoongArchState *env, uint32_t exception, diff --git a/target/m68k/cpu.h b/target/m68k/cpu.h index XXXXXXX..XXXXXXX 100644 --- a/target/m68k/cpu.h +++ b/target/m68k/cpu.h @@ -XXX,XX +XXX,XX @@ int m68k_cpu_gdb_read_register(CPUState *cpu, GByteArray *buf, int reg); int m68k_cpu_gdb_write_register(CPUState *cpu, uint8_t *buf, int reg); void m68k_tcg_init(void); +void m68k_translate_code(CPUState *cs, TranslationBlock *tb, + int *max_insns, vaddr pc, void *host_pc); void m68k_cpu_init_gdb(M68kCPU *cpu); uint32_t cpu_m68k_get_ccr(CPUM68KState *env); void cpu_m68k_set_ccr(CPUM68KState *env, uint32_t); diff --git a/target/microblaze/cpu.h b/target/microblaze/cpu.h index XXXXXXX..XXXXXXX 100644 --- a/target/microblaze/cpu.h +++ b/target/microblaze/cpu.h @@ -XXX,XX +XXX,XX @@ static inline void mb_cpu_write_msr(CPUMBState *env, uint32_t val) } void mb_tcg_init(void); +void mb_translate_code(CPUState *cs, TranslationBlock *tb, + int *max_insns, vaddr pc, void *host_pc); #define CPU_RESOLVING_TYPE TYPE_MICROBLAZE_CPU diff --git a/target/mips/tcg/tcg-internal.h b/target/mips/tcg/tcg-internal.h index XXXXXXX..XXXXXXX 100644 --- a/target/mips/tcg/tcg-internal.h +++ b/target/mips/tcg/tcg-internal.h @@ -XXX,XX +XXX,XX @@ #include "cpu.h" void mips_tcg_init(void); +void mips_translate_code(CPUState *cs, TranslationBlock *tb, + int *max_insns, vaddr pc, void *host_pc); void mips_cpu_synchronize_from_tb(CPUState *cs, const TranslationBlock *tb); G_NORETURN void mips_cpu_do_unaligned_access(CPUState *cpu, vaddr addr, diff --git a/target/openrisc/cpu.h b/target/openrisc/cpu.h index XXXXXXX..XXXXXXX 100644 --- a/target/openrisc/cpu.h +++ b/target/openrisc/cpu.h @@ -XXX,XX +XXX,XX @@ void openrisc_cpu_dump_state(CPUState *cpu, FILE *f, int flags); int openrisc_cpu_gdb_read_register(CPUState *cpu, GByteArray *buf, int reg); int openrisc_cpu_gdb_write_register(CPUState *cpu, uint8_t *buf, int reg); void openrisc_translate_init(void); +void openrisc_translate_code(CPUState *cs, TranslationBlock *tb, + int *max_insns, vaddr pc, void *host_pc); int print_insn_or1k(bfd_vma addr, disassemble_info *info); #ifndef CONFIG_USER_ONLY diff --git a/target/ppc/cpu.h b/target/ppc/cpu.h index XXXXXXX..XXXXXXX 100644 --- a/target/ppc/cpu.h +++ b/target/ppc/cpu.h @@ -XXX,XX +XXX,XX @@ extern const VMStateDescription vmstate_ppc_cpu; /*****************************************************************************/ void ppc_translate_init(void); +void ppc_translate_code(CPUState *cs, TranslationBlock *tb, + int *max_insns, vaddr pc, void *host_pc); #if !defined(CONFIG_USER_ONLY) void ppc_store_sdr1(CPUPPCState *env, target_ulong value); diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/cpu.h +++ b/target/riscv/cpu.h @@ -XXX,XX +XXX,XX @@ RISCVException smstateen_acc_ok(CPURISCVState *env, int index, uint64_t bit); void riscv_cpu_set_mode(CPURISCVState *env, target_ulong newpriv, bool virt_en); void riscv_translate_init(void); +void riscv_translate_code(CPUState *cs, TranslationBlock *tb, + int *max_insns, vaddr pc, void *host_pc); + G_NORETURN void riscv_raise_exception(CPURISCVState *env, uint32_t exception, uintptr_t pc); diff --git a/target/rx/cpu.h b/target/rx/cpu.h index XXXXXXX..XXXXXXX 100644 --- a/target/rx/cpu.h +++ b/target/rx/cpu.h @@ -XXX,XX +XXX,XX @@ int rx_cpu_gdb_read_register(CPUState *cpu, GByteArray *buf, int reg); int rx_cpu_gdb_write_register(CPUState *cpu, uint8_t *buf, int reg); void rx_translate_init(void); +void rx_translate_code(CPUState *cs, TranslationBlock *tb, + int *max_insns, vaddr pc, void *host_pc); void rx_cpu_unpack_psw(CPURXState *env, uint32_t psw, int rte); #include "exec/cpu-all.h" diff --git a/target/s390x/s390x-internal.h b/target/s390x/s390x-internal.h index XXXXXXX..XXXXXXX 100644 --- a/target/s390x/s390x-internal.h +++ b/target/s390x/s390x-internal.h @@ -XXX,XX +XXX,XX @@ void handle_diag_308(CPUS390XState *env, uint64_t r1, uint64_t r3, /* translate.c */ void s390x_translate_init(void); +void s390x_translate_code(CPUState *cs, TranslationBlock *tb, + int *max_insns, vaddr pc, void *host_pc); void s390x_restore_state_to_opc(CPUState *cs, const TranslationBlock *tb, const uint64_t *data); diff --git a/target/sh4/cpu.h b/target/sh4/cpu.h index XXXXXXX..XXXXXXX 100644 --- a/target/sh4/cpu.h +++ b/target/sh4/cpu.h @@ -XXX,XX +XXX,XX @@ G_NORETURN void superh_cpu_do_unaligned_access(CPUState *cpu, vaddr addr, uintptr_t retaddr); void sh4_translate_init(void); +void sh4_translate_code(CPUState *cs, TranslationBlock *tb, + int *max_insns, vaddr pc, void *host_pc); #if !defined(CONFIG_USER_ONLY) hwaddr superh_cpu_get_phys_page_debug(CPUState *cpu, vaddr addr); diff --git a/target/sparc/cpu.h b/target/sparc/cpu.h index XXXXXXX..XXXXXXX 100644 --- a/target/sparc/cpu.h +++ b/target/sparc/cpu.h @@ -XXX,XX +XXX,XX @@ int sparc_cpu_memory_rw_debug(CPUState *cpu, vaddr addr, /* translate.c */ void sparc_tcg_init(void); +void sparc_translate_code(CPUState *cs, TranslationBlock *tb, + int *max_insns, vaddr pc, void *host_pc); /* fop_helper.c */ target_ulong cpu_get_fsr(CPUSPARCState *); diff --git a/target/tricore/cpu.h b/target/tricore/cpu.h index XXXXXXX..XXXXXXX 100644 --- a/target/tricore/cpu.h +++ b/target/tricore/cpu.h @@ -XXX,XX +XXX,XX @@ FIELD(TB_FLAGS, PRIV, 0, 2) void cpu_state_reset(CPUTriCoreState *s); void tricore_tcg_init(void); +void tricore_translate_code(CPUState *cs, TranslationBlock *tb, + int *max_insns, vaddr pc, void *host_pc); static inline void cpu_get_tb_cpu_state(CPUTriCoreState *env, vaddr *pc, uint64_t *cs_base, uint32_t *flags) diff --git a/target/xtensa/cpu.h b/target/xtensa/cpu.h index XXXXXXX..XXXXXXX 100644 --- a/target/xtensa/cpu.h +++ b/target/xtensa/cpu.h @@ -XXX,XX +XXX,XX @@ G_NORETURN void xtensa_cpu_do_unaligned_access(CPUState *cpu, vaddr addr, void xtensa_collect_sr_names(const XtensaConfig *config); void xtensa_translate_init(void); +void xtensa_translate_code(CPUState *cs, TranslationBlock *tb, + int *max_insns, vaddr pc, void *host_pc); void **xtensa_get_regfile_by_name(const char *name, int entries, int bits); void xtensa_breakpoint_handler(CPUState *cs); void xtensa_register_core(XtensaConfigList *node); diff --git a/accel/tcg/cpu-exec.c b/accel/tcg/cpu-exec.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/cpu-exec.c +++ b/accel/tcg/cpu-exec.c @@ -XXX,XX +XXX,XX @@ bool tcg_exec_realizefn(CPUState *cpu, Error **errp) if (!tcg_target_initialized) { /* Check mandatory TCGCPUOps handlers */ + const TCGCPUOps *tcg_ops = cpu->cc->tcg_ops; #ifndef CONFIG_USER_ONLY - assert(cpu->cc->tcg_ops->cpu_exec_halt); - assert(cpu->cc->tcg_ops->cpu_exec_interrupt); + assert(tcg_ops->cpu_exec_halt); + assert(tcg_ops->cpu_exec_interrupt); #endif /* !CONFIG_USER_ONLY */ - cpu->cc->tcg_ops->initialize(); + assert(tcg_ops->translate_code); + tcg_ops->initialize(); tcg_target_initialized = true; } diff --git a/accel/tcg/translate-all.c b/accel/tcg/translate-all.c index XXXXXXX..XXXXXXX 100644 --- a/accel/tcg/translate-all.c +++ b/accel/tcg/translate-all.c @@ -XXX,XX +XXX,XX @@ static int setjmp_gen_code(CPUArchState *env, TranslationBlock *tb, tcg_func_start(tcg_ctx); - tcg_ctx->cpu = env_cpu(env); - gen_intermediate_code(env_cpu(env), tb, max_insns, pc, host_pc); + CPUState *cs = env_cpu(env); + tcg_ctx->cpu = cs; + cs->cc->tcg_ops->translate_code(cs, tb, max_insns, pc, host_pc); + assert(tb->size != 0); tcg_ctx->cpu = NULL; *max_insns = tb->icount; @@ -XXX,XX +XXX,XX @@ TranslationBlock *tb_gen_code(CPUState *cpu, /* * Overflow of code_gen_buffer, or the current slice of it. * - * TODO: We don't need to re-do gen_intermediate_code, nor + * TODO: We don't need to re-do tcg_ops->translate_code, nor * should we re-do the tcg optimization currently hidden * inside tcg_gen_code. All that should be required is to * flush the TBs, allocate a new TB, re-initialize it per diff --git a/target/alpha/cpu.c b/target/alpha/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/alpha/cpu.c +++ b/target/alpha/cpu.c @@ -XXX,XX +XXX,XX @@ static const struct SysemuCPUOps alpha_sysemu_ops = { static const TCGCPUOps alpha_tcg_ops = { .initialize = alpha_translate_init, + .translate_code = alpha_translate_code, .synchronize_from_tb = alpha_cpu_synchronize_from_tb, .restore_state_to_opc = alpha_restore_state_to_opc, diff --git a/target/alpha/translate.c b/target/alpha/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/alpha/translate.c +++ b/target/alpha/translate.c @@ -XXX,XX +XXX,XX @@ static const TranslatorOps alpha_tr_ops = { .tb_stop = alpha_tr_tb_stop, }; -void gen_intermediate_code(CPUState *cpu, TranslationBlock *tb, int *max_insns, - vaddr pc, void *host_pc) +void alpha_translate_code(CPUState *cpu, TranslationBlock *tb, + int *max_insns, vaddr pc, void *host_pc) { DisasContext dc; translator_loop(cpu, tb, max_insns, pc, host_pc, &alpha_tr_ops, &dc.base); diff --git a/target/arm/cpu.c b/target/arm/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu.c +++ b/target/arm/cpu.c @@ -XXX,XX +XXX,XX @@ static const struct SysemuCPUOps arm_sysemu_ops = { #ifdef CONFIG_TCG static const TCGCPUOps arm_tcg_ops = { .initialize = arm_translate_init, + .translate_code = arm_translate_code, .synchronize_from_tb = arm_cpu_synchronize_from_tb, .debug_excp_handler = arm_debug_excp_handler, .restore_state_to_opc = arm_restore_state_to_opc, diff --git a/target/arm/tcg/cpu-v7m.c b/target/arm/tcg/cpu-v7m.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/cpu-v7m.c +++ b/target/arm/tcg/cpu-v7m.c @@ -XXX,XX +XXX,XX @@ static void cortex_m55_initfn(Object *obj) static const TCGCPUOps arm_v7m_tcg_ops = { .initialize = arm_translate_init, + .translate_code = arm_translate_code, .synchronize_from_tb = arm_cpu_synchronize_from_tb, .debug_excp_handler = arm_debug_excp_handler, .restore_state_to_opc = arm_restore_state_to_opc, diff --git a/target/arm/tcg/translate.c b/target/arm/tcg/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/tcg/translate.c +++ b/target/arm/tcg/translate.c @@ -XXX,XX +XXX,XX @@ static const TranslatorOps thumb_translator_ops = { .tb_stop = arm_tr_tb_stop, }; -/* generate intermediate code for basic block 'tb'. */ -void gen_intermediate_code(CPUState *cpu, TranslationBlock *tb, int *max_insns, - vaddr pc, void *host_pc) +void arm_translate_code(CPUState *cpu, TranslationBlock *tb, + int *max_insns, vaddr pc, void *host_pc) { DisasContext dc = { }; const TranslatorOps *ops = &arm_translator_ops; diff --git a/target/avr/cpu.c b/target/avr/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/avr/cpu.c +++ b/target/avr/cpu.c @@ -XXX,XX +XXX,XX @@ static const struct SysemuCPUOps avr_sysemu_ops = { static const TCGCPUOps avr_tcg_ops = { .initialize = avr_cpu_tcg_init, + .translate_code = avr_cpu_translate_code, .synchronize_from_tb = avr_cpu_synchronize_from_tb, .restore_state_to_opc = avr_restore_state_to_opc, .cpu_exec_interrupt = avr_cpu_exec_interrupt, diff --git a/target/avr/translate.c b/target/avr/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/avr/translate.c +++ b/target/avr/translate.c @@ -XXX,XX +XXX,XX @@ static bool trans_WDR(DisasContext *ctx, arg_WDR *a) * * - translate() * - canonicalize_skip() - * - gen_intermediate_code() + * - translate_code() * - restore_state_to_opc() * */ @@ -XXX,XX +XXX,XX @@ static const TranslatorOps avr_tr_ops = { .tb_stop = avr_tr_tb_stop, }; -void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int *max_insns, - vaddr pc, void *host_pc) +void avr_cpu_translate_code(CPUState *cs, TranslationBlock *tb, + int *max_insns, vaddr pc, void *host_pc) { DisasContext dc = { }; translator_loop(cs, tb, max_insns, pc, host_pc, &avr_tr_ops, &dc.base); diff --git a/target/hexagon/cpu.c b/target/hexagon/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/hexagon/cpu.c +++ b/target/hexagon/cpu.c @@ -XXX,XX +XXX,XX @@ static void hexagon_cpu_init(Object *obj) static const TCGCPUOps hexagon_tcg_ops = { .initialize = hexagon_translate_init, + .translate_code = hexagon_translate_code, .synchronize_from_tb = hexagon_cpu_synchronize_from_tb, .restore_state_to_opc = hexagon_restore_state_to_opc, }; diff --git a/target/hexagon/translate.c b/target/hexagon/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/hexagon/translate.c +++ b/target/hexagon/translate.c @@ -XXX,XX +XXX,XX @@ static const TranslatorOps hexagon_tr_ops = { .tb_stop = hexagon_tr_tb_stop, }; -void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int *max_insns, - vaddr pc, void *host_pc) +void hexagon_translate_code(CPUState *cs, TranslationBlock *tb, + int *max_insns, vaddr pc, void *host_pc) { DisasContext ctx; diff --git a/target/hppa/cpu.c b/target/hppa/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/hppa/cpu.c +++ b/target/hppa/cpu.c @@ -XXX,XX +XXX,XX @@ static const struct SysemuCPUOps hppa_sysemu_ops = { static const TCGCPUOps hppa_tcg_ops = { .initialize = hppa_translate_init, + .translate_code = hppa_translate_code, .synchronize_from_tb = hppa_cpu_synchronize_from_tb, .restore_state_to_opc = hppa_restore_state_to_opc, diff --git a/target/hppa/translate.c b/target/hppa/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/hppa/translate.c +++ b/target/hppa/translate.c @@ -XXX,XX +XXX,XX @@ static const TranslatorOps hppa_tr_ops = { #endif }; -void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int *max_insns, - vaddr pc, void *host_pc) +void hppa_translate_code(CPUState *cs, TranslationBlock *tb, + int *max_insns, vaddr pc, void *host_pc) { DisasContext ctx = { }; translator_loop(cs, tb, max_insns, pc, host_pc, &hppa_tr_ops, &ctx.base); diff --git a/target/i386/tcg/tcg-cpu.c b/target/i386/tcg/tcg-cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/i386/tcg/tcg-cpu.c +++ b/target/i386/tcg/tcg-cpu.c @@ -XXX,XX +XXX,XX @@ static bool x86_debug_check_breakpoint(CPUState *cs) static const TCGCPUOps x86_tcg_ops = { .initialize = tcg_x86_init, + .translate_code = x86_translate_code, .synchronize_from_tb = x86_cpu_synchronize_from_tb, .restore_state_to_opc = x86_restore_state_to_opc, .cpu_exec_enter = x86_cpu_exec_enter, diff --git a/target/i386/tcg/translate.c b/target/i386/tcg/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/i386/tcg/translate.c +++ b/target/i386/tcg/translate.c @@ -XXX,XX +XXX,XX @@ static const TranslatorOps i386_tr_ops = { .tb_stop = i386_tr_tb_stop, }; -/* generate intermediate code for basic block 'tb'. */ -void gen_intermediate_code(CPUState *cpu, TranslationBlock *tb, int *max_insns, - vaddr pc, void *host_pc) +void x86_translate_code(CPUState *cpu, TranslationBlock *tb, + int *max_insns, vaddr pc, void *host_pc) { DisasContext dc; diff --git a/target/loongarch/cpu.c b/target/loongarch/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/loongarch/cpu.c +++ b/target/loongarch/cpu.c @@ -XXX,XX +XXX,XX @@ static void loongarch_cpu_dump_state(CPUState *cs, FILE *f, int flags) static const TCGCPUOps loongarch_tcg_ops = { .initialize = loongarch_translate_init, + .translate_code = loongarch_translate_code, .synchronize_from_tb = loongarch_cpu_synchronize_from_tb, .restore_state_to_opc = loongarch_restore_state_to_opc, diff --git a/target/loongarch/tcg/translate.c b/target/loongarch/tcg/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/loongarch/tcg/translate.c +++ b/target/loongarch/tcg/translate.c @@ -XXX,XX +XXX,XX @@ static const TranslatorOps loongarch_tr_ops = { .tb_stop = loongarch_tr_tb_stop, }; -void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int *max_insns, - vaddr pc, void *host_pc) +void loongarch_translate_code(CPUState *cs, TranslationBlock *tb, + int *max_insns, vaddr pc, void *host_pc) { DisasContext ctx; diff --git a/target/m68k/cpu.c b/target/m68k/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/m68k/cpu.c +++ b/target/m68k/cpu.c @@ -XXX,XX +XXX,XX @@ static const struct SysemuCPUOps m68k_sysemu_ops = { static const TCGCPUOps m68k_tcg_ops = { .initialize = m68k_tcg_init, + .translate_code = m68k_translate_code, .restore_state_to_opc = m68k_restore_state_to_opc, #ifndef CONFIG_USER_ONLY diff --git a/target/m68k/translate.c b/target/m68k/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/m68k/translate.c +++ b/target/m68k/translate.c @@ -XXX,XX +XXX,XX @@ static const TranslatorOps m68k_tr_ops = { .tb_stop = m68k_tr_tb_stop, }; -void gen_intermediate_code(CPUState *cpu, TranslationBlock *tb, int *max_insns, - vaddr pc, void *host_pc) +void m68k_translate_code(CPUState *cpu, TranslationBlock *tb, + int *max_insns, vaddr pc, void *host_pc) { DisasContext dc; translator_loop(cpu, tb, max_insns, pc, host_pc, &m68k_tr_ops, &dc.base); diff --git a/target/microblaze/cpu.c b/target/microblaze/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/microblaze/cpu.c +++ b/target/microblaze/cpu.c @@ -XXX,XX +XXX,XX @@ static const struct SysemuCPUOps mb_sysemu_ops = { static const TCGCPUOps mb_tcg_ops = { .initialize = mb_tcg_init, + .translate_code = mb_translate_code, .synchronize_from_tb = mb_cpu_synchronize_from_tb, .restore_state_to_opc = mb_restore_state_to_opc, diff --git a/target/microblaze/translate.c b/target/microblaze/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/microblaze/translate.c +++ b/target/microblaze/translate.c @@ -XXX,XX +XXX,XX @@ static const TranslatorOps mb_tr_ops = { .tb_stop = mb_tr_tb_stop, }; -void gen_intermediate_code(CPUState *cpu, TranslationBlock *tb, int *max_insns, - vaddr pc, void *host_pc) +void mb_translate_code(CPUState *cpu, TranslationBlock *tb, + int *max_insns, vaddr pc, void *host_pc) { DisasContext dc; translator_loop(cpu, tb, max_insns, pc, host_pc, &mb_tr_ops, &dc.base); diff --git a/target/mips/cpu.c b/target/mips/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/mips/cpu.c +++ b/target/mips/cpu.c @@ -XXX,XX +XXX,XX @@ static const Property mips_cpu_properties[] = { #include "hw/core/tcg-cpu-ops.h" static const TCGCPUOps mips_tcg_ops = { .initialize = mips_tcg_init, + .translate_code = mips_translate_code, .synchronize_from_tb = mips_cpu_synchronize_from_tb, .restore_state_to_opc = mips_restore_state_to_opc, diff --git a/target/mips/tcg/translate.c b/target/mips/tcg/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/mips/tcg/translate.c +++ b/target/mips/tcg/translate.c @@ -XXX,XX +XXX,XX @@ static const TranslatorOps mips_tr_ops = { .tb_stop = mips_tr_tb_stop, }; -void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int *max_insns, - vaddr pc, void *host_pc) +void mips_translate_code(CPUState *cs, TranslationBlock *tb, + int *max_insns, vaddr pc, void *host_pc) { DisasContext ctx; diff --git a/target/openrisc/cpu.c b/target/openrisc/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/openrisc/cpu.c +++ b/target/openrisc/cpu.c @@ -XXX,XX +XXX,XX @@ static const struct SysemuCPUOps openrisc_sysemu_ops = { static const TCGCPUOps openrisc_tcg_ops = { .initialize = openrisc_translate_init, + .translate_code = openrisc_translate_code, .synchronize_from_tb = openrisc_cpu_synchronize_from_tb, .restore_state_to_opc = openrisc_restore_state_to_opc, diff --git a/target/openrisc/translate.c b/target/openrisc/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/openrisc/translate.c +++ b/target/openrisc/translate.c @@ -XXX,XX +XXX,XX @@ static const TranslatorOps openrisc_tr_ops = { .tb_stop = openrisc_tr_tb_stop, }; -void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int *max_insns, - vaddr pc, void *host_pc) +void openrisc_translate_code(CPUState *cs, TranslationBlock *tb, + int *max_insns, vaddr pc, void *host_pc) { DisasContext ctx; diff --git a/target/ppc/cpu_init.c b/target/ppc/cpu_init.c index XXXXXXX..XXXXXXX 100644 --- a/target/ppc/cpu_init.c +++ b/target/ppc/cpu_init.c @@ -XXX,XX +XXX,XX @@ static const struct SysemuCPUOps ppc_sysemu_ops = { static const TCGCPUOps ppc_tcg_ops = { .initialize = ppc_translate_init, + .translate_code = ppc_translate_code, .restore_state_to_opc = ppc_restore_state_to_opc, #ifdef CONFIG_USER_ONLY diff --git a/target/ppc/translate.c b/target/ppc/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/ppc/translate.c +++ b/target/ppc/translate.c @@ -XXX,XX +XXX,XX @@ static const TranslatorOps ppc_tr_ops = { .tb_stop = ppc_tr_tb_stop, }; -void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int *max_insns, - vaddr pc, void *host_pc) +void ppc_translate_code(CPUState *cs, TranslationBlock *tb, + int *max_insns, vaddr pc, void *host_pc) { DisasContext ctx; diff --git a/target/riscv/tcg/tcg-cpu.c b/target/riscv/tcg/tcg-cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/tcg/tcg-cpu.c +++ b/target/riscv/tcg/tcg-cpu.c @@ -XXX,XX +XXX,XX @@ static void riscv_restore_state_to_opc(CPUState *cs, static const TCGCPUOps riscv_tcg_ops = { .initialize = riscv_translate_init, + .translate_code = riscv_translate_code, .synchronize_from_tb = riscv_cpu_synchronize_from_tb, .restore_state_to_opc = riscv_restore_state_to_opc, diff --git a/target/riscv/translate.c b/target/riscv/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/riscv/translate.c +++ b/target/riscv/translate.c @@ -XXX,XX +XXX,XX @@ static const TranslatorOps riscv_tr_ops = { .tb_stop = riscv_tr_tb_stop, }; -void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int *max_insns, - vaddr pc, void *host_pc) +void riscv_translate_code(CPUState *cs, TranslationBlock *tb, + int *max_insns, vaddr pc, void *host_pc) { DisasContext ctx; diff --git a/target/rx/cpu.c b/target/rx/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/rx/cpu.c +++ b/target/rx/cpu.c @@ -XXX,XX +XXX,XX @@ static const struct SysemuCPUOps rx_sysemu_ops = { static const TCGCPUOps rx_tcg_ops = { .initialize = rx_translate_init, + .translate_code = rx_translate_code, .synchronize_from_tb = rx_cpu_synchronize_from_tb, .restore_state_to_opc = rx_restore_state_to_opc, .tlb_fill = rx_cpu_tlb_fill, diff --git a/target/rx/translate.c b/target/rx/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/rx/translate.c +++ b/target/rx/translate.c @@ -XXX,XX +XXX,XX @@ static const TranslatorOps rx_tr_ops = { .tb_stop = rx_tr_tb_stop, }; -void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int *max_insns, - vaddr pc, void *host_pc) +void rx_translate_code(CPUState *cs, TranslationBlock *tb, + int *max_insns, vaddr pc, void *host_pc) { DisasContext dc; diff --git a/target/s390x/cpu.c b/target/s390x/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/s390x/cpu.c +++ b/target/s390x/cpu.c @@ -XXX,XX +XXX,XX @@ void cpu_get_tb_cpu_state(CPUS390XState *env, vaddr *pc, static const TCGCPUOps s390_tcg_ops = { .initialize = s390x_translate_init, + .translate_code = s390x_translate_code, .restore_state_to_opc = s390x_restore_state_to_opc, #ifdef CONFIG_USER_ONLY diff --git a/target/s390x/tcg/translate.c b/target/s390x/tcg/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/s390x/tcg/translate.c +++ b/target/s390x/tcg/translate.c @@ -XXX,XX +XXX,XX @@ static const TranslatorOps s390x_tr_ops = { .disas_log = s390x_tr_disas_log, }; -void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int *max_insns, - vaddr pc, void *host_pc) +void s390x_translate_code(CPUState *cs, TranslationBlock *tb, + int *max_insns, vaddr pc, void *host_pc) { DisasContext dc; diff --git a/target/sh4/cpu.c b/target/sh4/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/sh4/cpu.c +++ b/target/sh4/cpu.c @@ -XXX,XX +XXX,XX @@ static const struct SysemuCPUOps sh4_sysemu_ops = { static const TCGCPUOps superh_tcg_ops = { .initialize = sh4_translate_init, + .translate_code = sh4_translate_code, .synchronize_from_tb = superh_cpu_synchronize_from_tb, .restore_state_to_opc = superh_restore_state_to_opc, diff --git a/target/sh4/translate.c b/target/sh4/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/sh4/translate.c +++ b/target/sh4/translate.c @@ -XXX,XX +XXX,XX @@ static const TranslatorOps sh4_tr_ops = { .tb_stop = sh4_tr_tb_stop, }; -void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int *max_insns, - vaddr pc, void *host_pc) +void sh4_translate_code(CPUState *cs, TranslationBlock *tb, + int *max_insns, vaddr pc, void *host_pc) { DisasContext ctx; diff --git a/target/sparc/cpu.c b/target/sparc/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/sparc/cpu.c +++ b/target/sparc/cpu.c @@ -XXX,XX +XXX,XX @@ static const struct SysemuCPUOps sparc_sysemu_ops = { static const TCGCPUOps sparc_tcg_ops = { .initialize = sparc_tcg_init, + .translate_code = sparc_translate_code, .synchronize_from_tb = sparc_cpu_synchronize_from_tb, .restore_state_to_opc = sparc_restore_state_to_opc, diff --git a/target/sparc/translate.c b/target/sparc/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/sparc/translate.c +++ b/target/sparc/translate.c @@ -XXX,XX +XXX,XX @@ static const TranslatorOps sparc_tr_ops = { .tb_stop = sparc_tr_tb_stop, }; -void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int *max_insns, - vaddr pc, void *host_pc) +void sparc_translate_code(CPUState *cs, TranslationBlock *tb, + int *max_insns, vaddr pc, void *host_pc) { DisasContext dc = {}; diff --git a/target/tricore/cpu.c b/target/tricore/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/tricore/cpu.c +++ b/target/tricore/cpu.c @@ -XXX,XX +XXX,XX @@ static const struct SysemuCPUOps tricore_sysemu_ops = { static const TCGCPUOps tricore_tcg_ops = { .initialize = tricore_tcg_init, + .translate_code = tricore_translate_code, .synchronize_from_tb = tricore_cpu_synchronize_from_tb, .restore_state_to_opc = tricore_restore_state_to_opc, .tlb_fill = tricore_cpu_tlb_fill, diff --git a/target/tricore/translate.c b/target/tricore/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/tricore/translate.c +++ b/target/tricore/translate.c @@ -XXX,XX +XXX,XX @@ static const TranslatorOps tricore_tr_ops = { .tb_stop = tricore_tr_tb_stop, }; - -void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int *max_insns, - vaddr pc, void *host_pc) +void tricore_translate_code(CPUState *cs, TranslationBlock *tb, + int *max_insns, vaddr pc, void *host_pc) { DisasContext ctx; translator_loop(cs, tb, max_insns, pc, host_pc, diff --git a/target/xtensa/cpu.c b/target/xtensa/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/xtensa/cpu.c +++ b/target/xtensa/cpu.c @@ -XXX,XX +XXX,XX @@ static const struct SysemuCPUOps xtensa_sysemu_ops = { static const TCGCPUOps xtensa_tcg_ops = { .initialize = xtensa_translate_init, + .translate_code = xtensa_translate_code, .debug_excp_handler = xtensa_breakpoint_handler, .restore_state_to_opc = xtensa_restore_state_to_opc, diff --git a/target/xtensa/translate.c b/target/xtensa/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/xtensa/translate.c +++ b/target/xtensa/translate.c @@ -XXX,XX +XXX,XX @@ static const TranslatorOps xtensa_translator_ops = { .tb_stop = xtensa_tr_tb_stop, }; -void gen_intermediate_code(CPUState *cpu, TranslationBlock *tb, int *max_insns, - vaddr pc, void *host_pc) +void xtensa_translate_code(CPUState *cpu, TranslationBlock *tb, + int *max_insns, vaddr pc, void *host_pc) { DisasContext dc = {}; translator_loop(cpu, tb, max_insns, pc, host_pc, -- 2.43.0