From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from eggs.gnu.org ([2001:4830:134:3::10]:44464)
	by lists.gnu.org with esmtp (Exim 4.71)
	(envelope-from <rth7680@gmail.com>) id 1Upi9X-0004Li-1h
	for qemu-devel@nongnu.org; Thu, 20 Jun 2013 12:56:10 -0400
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)
	(envelope-from <rth7680@gmail.com>) id 1Upi9U-0001uZ-AV
	for qemu-devel@nongnu.org; Thu, 20 Jun 2013 12:56:06 -0400
Received: from mail-qc0-x22f.google.com ([2607:f8b0:400d:c01::22f]:47738)
	by eggs.gnu.org with esmtp (Exim 4.71)
	(envelope-from <rth7680@gmail.com>) id 1Upi9U-0001uO-6t
	for qemu-devel@nongnu.org; Thu, 20 Jun 2013 12:56:04 -0400
Received: by mail-qc0-f175.google.com with SMTP id k14so3811952qcv.6
	for <qemu-devel@nongnu.org>; Thu, 20 Jun 2013 09:56:03 -0700 (PDT)
Sender: Richard Henderson <rth7680@gmail.com>
Message-ID: <51C3341F.3020801@twiddle.net>
Date: Thu, 20 Jun 2013 09:55:59 -0700
From: Richard Henderson <rth@twiddle.net>
MIME-Version: 1.0
References: <51C2DDFD.4030702@huawei.com> <51C2DF23.3020202@huawei.com>
	<51C3189A.5010206@huawei.com>
In-Reply-To: <51C3189A.5010206@huawei.com>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
Subject: Re: [Qemu-devel] [PATCH 1/1] tcg/aarch64: Implement tlb lookup fast
	path
List-Id: <qemu-devel.nongnu.org>
List-Unsubscribe: <https://lists.nongnu.org/mailman/options/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=unsubscribe>
List-Archive: <http://lists.nongnu.org/archive/html/qemu-devel>
List-Post: <mailto:qemu-devel@nongnu.org>
List-Help: <mailto:qemu-devel-request@nongnu.org?subject=help>
List-Subscribe: <https://lists.nongnu.org/mailman/listinfo/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=subscribe>
To: Claudio Fontana <claudio.fontana@huawei.com>
Cc: Peter Maydell <peter.maydell@linaro.org>, Jani Kokkonen <Jani.Kokkonen@huawei.com>, "qemu-devel@nongnu.org" <qemu-devel@nongnu.org>

On 06/20/2013 07:58 AM, Claudio Fontana wrote:
>> > +    tcg_out_ldst(s, TARGET_LONG_BITS == 64 ? LDST_64 : LDST_32,
>> > +                  LDST_LD, TCG_REG_X0, TCG_REG_X2, tlb_offset & 0xfff);
>> > +    tcg_out_ldst(s, LDST_64, LDST_LD, TCG_REG_X1, TCG_REG_X2,
>> > +        (tlb_offset & 0xfff) + (offsetof(CPUTLBEntry, addend) -
>> > +             (is_read ? offsetof(CPUTLBEntry, addr_read) :
>> > +                   offsetof(CPUTLBEntry, addr_write))));
>> > +
>> > +    tcg_out_cmp(s, 1, TCG_REG_X0, TCG_REG_X3, 0);
>> > +    *label_ptr = s->code_ptr;
>> > +    tcg_out_goto_cond_noaddr(s, TCG_COND_NE);
>> > +}
> hmm should not the compare and branch actually be before the loading of the addend?
> If we jump to the slow path we don't need to load the addend do we?
> 

No, but it's the slow path, and we don't care if we do extra work.
What's more important is minimizing the memory load delay for the
fast path.


r~