From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:52556) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1cfqdJ-0000q4-5x for qemu-devel@nongnu.org; Mon, 20 Feb 2017 11:16:14 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1cfqdF-00009s-8O for qemu-devel@nongnu.org; Mon, 20 Feb 2017 11:16:13 -0500 Received: from mail-wm0-x22c.google.com ([2a00:1450:400c:c09::22c]:36923) by eggs.gnu.org with esmtps (TLS1.0:RSA_AES_128_CBC_SHA1:16) (Exim 4.71) (envelope-from ) id 1cfqdF-00009M-0n for qemu-devel@nongnu.org; Mon, 20 Feb 2017 11:16:09 -0500 Received: by mail-wm0-x22c.google.com with SMTP id v77so56602795wmv.0 for ; Mon, 20 Feb 2017 08:16:08 -0800 (PST) References: <20170220145335.591-1-bobby.prani@gmail.com> From: Alex =?utf-8?Q?Benn=C3=A9e?= In-reply-to: <20170220145335.591-1-bobby.prani@gmail.com> Date: Mon, 20 Feb 2017 16:16:13 +0000 Message-ID: <87fuj8x2gi.fsf@linaro.org> 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: Pranith Kumar Cc: qemu-devel@nongnu.org, rth@twiddle.net 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). > if (sigsetjmp(cpu->jmp_env, 0) == 0) { > + 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); > /* execute the generated code */ > trace_exec_tb_nocache(tb, pc); > cpu_tb_exec(cpu, tb); > - } > + cc->cpu_exec_exit(cpu); > > - cc->cpu_exec_exit(cpu); > - tb_lock(); > - tb_phys_invalidate(tb, -1); > - tb_free(tb); > + tb_lock(); > + tb_phys_invalidate(tb, -1); > + tb_free(tb); > + } > tb_unlock(); > } -- Alex Bennée