From dd1473f5825dd65355ff8261c06473c88dc16ae1 Mon Sep 17 00:00:00 2001 From: Richard Henderson Date: Sat, 3 Mar 2018 17:16:21 -0500 Subject: [PATCH] tcg: Increase hit rate of lookup_tb_ptr We can call tb_htable_lookup even when the tb_jmp_cache is completely empty. Therefore, un-nest most of the code dependent on tb != NULL from the read from the cache. This improves the hit rate of lookup_tb_ptr; for instance, when booting and immediately shutting down debian-arm, the hit rate improves from 93.2% to 99.4%. Backports commit b97a879de980e99452063851597edb98e7e8039c from qemu --- qemu/tcg-runtime.c | 27 ++++++++++++++------------- 1 file changed, 14 insertions(+), 13 deletions(-) diff --git a/qemu/tcg-runtime.c b/qemu/tcg-runtime.c index 5fb953b4..8cba0715 100644 --- a/qemu/tcg-runtime.c +++ b/qemu/tcg-runtime.c @@ -149,23 +149,24 @@ void *HELPER(lookup_tb_ptr)(CPUArchState *env, target_ulong addr) CPUState *cpu = ENV_GET_CPU(env); TranslationBlock *tb; target_ulong cs_base, pc; - uint32_t flags; + uint32_t flags, addr_hash; - tb = atomic_rcu_read(&cpu->tb_jmp_cache[tb_jmp_cache_hash_func(addr)]); - if (likely(tb)) { - cpu_get_tb_cpu_state(env, &pc, &cs_base, &flags); - if (likely(tb->pc == addr && tb->cs_base == cs_base && - tb->flags == flags)) { - goto found; - } + addr_hash = tb_jmp_cache_hash_func(addr); + // Unicorn: atomic_read used instead of atomic_rcu_read + tb = atomic_read(&cpu->tb_jmp_cache[addr_hash]); + cpu_get_tb_cpu_state(env, &pc, &cs_base, &flags); + + if (unlikely(!(tb + && tb->pc == addr + && tb->cs_base == cs_base + && tb->flags == flags))) { tb = tb_htable_lookup(cpu, addr, cs_base, flags); - if (likely(tb)) { - atomic_set(&cpu->tb_jmp_cache[tb_jmp_cache_hash_func(addr)], tb); - goto found; + if (!tb) { + return tcg_ctx->code_gen_epilogue; } + atomic_set(&cpu->tb_jmp_cache[addr_hash], tb); } - return tcg_ctx->code_gen_epilogue; - found: + // Unicorn: commented out //qemu_log_mask_and_addr(CPU_LOG_EXEC, addr, // "Chain %p [%d: " TARGET_FMT_lx "] %s\n",