From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:46177) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1cfxoj-0002Gj-4x for qemu-devel@nongnu.org; Mon, 20 Feb 2017 18:56:30 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1cfxog-0005VX-2Z for qemu-devel@nongnu.org; Mon, 20 Feb 2017 18:56:29 -0500 Received: from mail-yw0-x242.google.com ([2607:f8b0:4002:c05::242]:35893) by eggs.gnu.org with esmtps (TLS1.0:RSA_AES_128_CBC_SHA1:16) (Exim 4.71) (envelope-from ) id 1cfxof-0005VT-Tg for qemu-devel@nongnu.org; Mon, 20 Feb 2017 18:56:25 -0500 Received: by mail-yw0-x242.google.com with SMTP id 2so3089515ywn.3 for ; Mon, 20 Feb 2017 15:56:25 -0800 (PST) References: <20170220145335.591-1-bobby.prani@gmail.com> <87fuj8x2gi.fsf@linaro.org> From: Pranith Kumar In-reply-to: <87fuj8x2gi.fsf@linaro.org> Date: Mon, 20 Feb 2017 18:56:23 -0500 Message-ID: <87k28ko1qw.fsf@gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 8bit Subject: Re: [Qemu-devel] [PATCH] mttcg: Set jmp_env to handle exit from tb_gen_code List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Alex =?utf-8?Q?Benn=C3=A9e?= Cc: qemu-devel@nongnu.org, rth@twiddle.net Alex Bennée writes: > Pranith Kumar writes: > >> tb_gen_code() can exit execution using cpu_exit_loop() when it cannot >> allocate new tb's. To handle this, we need to properly set the jmp_env >> pointer ahead of calling tb_gen_code(). >> >> CC:Alex Bennée >> CC: Richard Henderson >> Signed-off-by: Pranith Kumar >> --- >> cpu-exec.c | 23 +++++++++++------------ >> 1 file changed, 11 insertions(+), 12 deletions(-) >> >> diff --git a/cpu-exec.c b/cpu-exec.c >> index 97d79612d9..4b70988b24 100644 >> --- a/cpu-exec.c >> +++ b/cpu-exec.c >> @@ -236,23 +236,22 @@ static void cpu_exec_step(CPUState *cpu) >> >> cpu_get_tb_cpu_state(env, &pc, &cs_base, &flags); >> tb_lock(); >> - tb = tb_gen_code(cpu, pc, cs_base, flags, >> - 1 | CF_NOCACHE | CF_IGNORE_ICOUNT); >> - tb->orig_tb = NULL; >> - tb_unlock(); >> - >> - cc->cpu_exec_enter(cpu); >> - > > It occurs to me we are also diverging in our locking pattern from > tb_find which takes mmap_lock first. This is a NOP for system emulation > but needed for user-emulation (for which we can do cpu_exec_step but not > cpu_exec_nocache). Right. So we have to take the mmap_lock() before calling tb_gen_code(). However, this lock is released in the error path before calling cpu_loop_exit() if allocation of a new tb fails. The following is what I have after merging with the previous EXCP_ATOMIC handling patch. diff --git a/cpu-exec.c b/cpu-exec.c index a8e04bffbf..2bb3ba3672 100644 --- a/cpu-exec.c +++ b/cpu-exec.c @@ -228,6 +228,7 @@ static void cpu_exec_nocache(CPUState *cpu, int max_cycles, static void cpu_exec_step(CPUState *cpu) { + CPUClass *cc = CPU_GET_CLASS(cpu); CPUArchState *env = (CPUArchState *)cpu->env_ptr; TranslationBlock *tb; target_ulong cs_base, pc; @@ -235,16 +236,24 @@ static void cpu_exec_step(CPUState *cpu) cpu_get_tb_cpu_state(env, &pc, &cs_base, &flags); tb_lock(); - tb = tb_gen_code(cpu, pc, cs_base, flags, - 1 | CF_NOCACHE | CF_IGNORE_ICOUNT); - tb->orig_tb = NULL; - tb_unlock(); - /* execute the generated code */ - trace_exec_tb_nocache(tb, pc); - cpu_tb_exec(cpu, tb); - tb_lock(); - tb_phys_invalidate(tb, -1); - tb_free(tb); + if (sigsetjmp(cpu->jmp_env, 0) == 0) { + mmap_lock(); + tb = tb_gen_code(cpu, pc, cs_base, flags, + 1 | CF_NOCACHE | CF_IGNORE_ICOUNT); + tb->orig_tb = NULL; + mmap_unlock(); + tb_unlock(); + + cc->cpu_exec_enter(cpu); + /* execute the generated code */ + trace_exec_tb_nocache(tb, pc); + cpu_tb_exec(cpu, tb); + cc->cpu_exec_exit(cpu); + + tb_lock(); + tb_phys_invalidate(tb, -1); + tb_free(tb); + } tb_unlock(); } diff --git a/cpus.c b/cpus.c index 77bba08f9a..b39408b4b1 100644 --- a/cpus.c +++ b/cpus.c @@ -1347,6 +1347,11 @@ static void *qemu_tcg_rr_cpu_thread_fn(void *arg) if (r == EXCP_DEBUG) { cpu_handle_guest_debug(cpu); break; + } else if (r == EXCP_ATOMIC) { + qemu_mutex_unlock_iothread(); + cpu_exec_step_atomic(cpu); + qemu_mutex_lock_iothread(); + break; } } else if (cpu->stop) { if (cpu->unplug) { @@ -1457,6 +1462,10 @@ static void *qemu_tcg_cpu_thread_fn(void *arg) */ g_assert(cpu->halted); break; + case EXCP_ATOMIC: + qemu_mutex_unlock_iothread(); + cpu_exec_step_atomic(cpu); + qemu_mutex_lock_iothread(); default: /* Ignore everything else? */ break; -- Pranith