All of lore.kernel.org
 help / color / mirror / Atom feed
From: Sergey Fedorov <serge.fdrv@gmail.com>
To: Peter Maydell <peter.maydell@linaro.org>,
	QEMU Developers <qemu-devel@nongnu.org>
Cc: Laurent Desnogues <laurent.desnogues@gmail.com>,
	qemu-arm@nongnu.org, Patch Tracking <patches@linaro.org>
Subject: Re: [Qemu-arm] [PATCH] target-arm/translate-a64.c: Correct unallocated checks for ldst_excl
Date: Mon, 23 Nov 2015 21:54:53 +0300	[thread overview]
Message-ID: <565360FD.6010301@gmail.com> (raw)
In-Reply-To: <56535E25.90305@gmail.com>

On 23.11.2015 21:42, Sergey Fedorov wrote:
> On 23.11.2015 19:49, Peter Maydell wrote:
>> Ping? I forgot to mark this for-2.5, and given how long the bug's
>> been hanging around there's not much urgency to fixing it, but
>> we might as well put the fix into 2.5 if it gets reviewed.
>>
> Hi, Peter. I'm going to review this carefully in a few days :)
> For now, I see that the comment for this function should be updated to
> match new code.
>

However, I've just checked the logic of the patch was correct. Just the
comment needs to be adjusted.

Best regards,
Sergey

>> On 16 November 2015 at 18:28, Peter Maydell <peter.maydell@linaro.org> wrote:
>>> The checks for the unallocated encodings in the ldst_excl group
>>> (exclusives and load-acquire/store-release) were not correct. This
>>> error meant that in turn we ended up with code attempting to handle
>>> the non-existent case of "non-exclusive load-acquire/store-release
>>> pair". Delete that broken and now unreachable code.
>>>
>>> Reported-by: Laurent Desnogues <laurent.desnogues@gmail.com>
>>> Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
>>> ---
>>> The easiest way to validate that we have the unallocated
>>> conditions correct now is to look at C4.4.6 "load/store exclusive"
>>> in the v8 ARM ARM rev A.h: our three conditions correspond
>>> to the three "unallocated" rows in the decode table.
>>>
>>> PS: Laurent originally reported this way back in 2014:
>>> http://lists.gnu.org/archive/html/qemu-devel/2014-10/msg01255.html
>>>
>>>  target-arm/translate-a64.c | 12 ++----------
>>>  1 file changed, 2 insertions(+), 10 deletions(-)
>>>
>>> diff --git a/target-arm/translate-a64.c b/target-arm/translate-a64.c
>>> index fe485a4..890ace4 100644
>>> --- a/target-arm/translate-a64.c
>>> +++ b/target-arm/translate-a64.c
>>> @@ -1833,7 +1833,8 @@ static void disas_ldst_excl(DisasContext *s, uint32_t insn)
>>>      int size = extract32(insn, 30, 2);
>>>      TCGv_i64 tcg_addr;
>>>
>>> -    if ((!is_excl && !is_lasr) ||
>>> +    if ((!is_excl && !is_pair && !is_lasr) ||
>>> +        (!is_excl && is_pair) ||
>>>          (is_pair && size < 2)) {
>>>          unallocated_encoding(s);
>>>          return;
>>> @@ -1862,15 +1863,6 @@ static void disas_ldst_excl(DisasContext *s, uint32_t insn)
>>>          } else {
>>>              do_gpr_ld(s, tcg_rt, tcg_addr, size, false, false);
>>>          }
>>> -        if (is_pair) {
>>> -            TCGv_i64 tcg_rt2 = cpu_reg(s, rt);
>>> -            tcg_gen_addi_i64(tcg_addr, tcg_addr, 1 << size);
>>> -            if (is_store) {
>>> -                do_gpr_st(s, tcg_rt2, tcg_addr, size);
>>> -            } else {
>>> -                do_gpr_ld(s, tcg_rt2, tcg_addr, size, false, false);
>>> -            }
>>> -        }
>>>      }
>>>  }
>>>
>>> --
>>> 1.9.1
>>>


WARNING: multiple messages have this Message-ID (diff)
From: Sergey Fedorov <serge.fdrv@gmail.com>
To: Peter Maydell <peter.maydell@linaro.org>,
	QEMU Developers <qemu-devel@nongnu.org>
Cc: Laurent Desnogues <laurent.desnogues@gmail.com>,
	qemu-arm@nongnu.org, Patch Tracking <patches@linaro.org>
Subject: Re: [Qemu-devel] [Qemu-arm] [PATCH] target-arm/translate-a64.c: Correct unallocated checks for ldst_excl
Date: Mon, 23 Nov 2015 21:54:53 +0300	[thread overview]
Message-ID: <565360FD.6010301@gmail.com> (raw)
In-Reply-To: <56535E25.90305@gmail.com>

On 23.11.2015 21:42, Sergey Fedorov wrote:
> On 23.11.2015 19:49, Peter Maydell wrote:
>> Ping? I forgot to mark this for-2.5, and given how long the bug's
>> been hanging around there's not much urgency to fixing it, but
>> we might as well put the fix into 2.5 if it gets reviewed.
>>
> Hi, Peter. I'm going to review this carefully in a few days :)
> For now, I see that the comment for this function should be updated to
> match new code.
>

However, I've just checked the logic of the patch was correct. Just the
comment needs to be adjusted.

Best regards,
Sergey

>> On 16 November 2015 at 18:28, Peter Maydell <peter.maydell@linaro.org> wrote:
>>> The checks for the unallocated encodings in the ldst_excl group
>>> (exclusives and load-acquire/store-release) were not correct. This
>>> error meant that in turn we ended up with code attempting to handle
>>> the non-existent case of "non-exclusive load-acquire/store-release
>>> pair". Delete that broken and now unreachable code.
>>>
>>> Reported-by: Laurent Desnogues <laurent.desnogues@gmail.com>
>>> Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
>>> ---
>>> The easiest way to validate that we have the unallocated
>>> conditions correct now is to look at C4.4.6 "load/store exclusive"
>>> in the v8 ARM ARM rev A.h: our three conditions correspond
>>> to the three "unallocated" rows in the decode table.
>>>
>>> PS: Laurent originally reported this way back in 2014:
>>> http://lists.gnu.org/archive/html/qemu-devel/2014-10/msg01255.html
>>>
>>>  target-arm/translate-a64.c | 12 ++----------
>>>  1 file changed, 2 insertions(+), 10 deletions(-)
>>>
>>> diff --git a/target-arm/translate-a64.c b/target-arm/translate-a64.c
>>> index fe485a4..890ace4 100644
>>> --- a/target-arm/translate-a64.c
>>> +++ b/target-arm/translate-a64.c
>>> @@ -1833,7 +1833,8 @@ static void disas_ldst_excl(DisasContext *s, uint32_t insn)
>>>      int size = extract32(insn, 30, 2);
>>>      TCGv_i64 tcg_addr;
>>>
>>> -    if ((!is_excl && !is_lasr) ||
>>> +    if ((!is_excl && !is_pair && !is_lasr) ||
>>> +        (!is_excl && is_pair) ||
>>>          (is_pair && size < 2)) {
>>>          unallocated_encoding(s);
>>>          return;
>>> @@ -1862,15 +1863,6 @@ static void disas_ldst_excl(DisasContext *s, uint32_t insn)
>>>          } else {
>>>              do_gpr_ld(s, tcg_rt, tcg_addr, size, false, false);
>>>          }
>>> -        if (is_pair) {
>>> -            TCGv_i64 tcg_rt2 = cpu_reg(s, rt);
>>> -            tcg_gen_addi_i64(tcg_addr, tcg_addr, 1 << size);
>>> -            if (is_store) {
>>> -                do_gpr_st(s, tcg_rt2, tcg_addr, size);
>>> -            } else {
>>> -                do_gpr_ld(s, tcg_rt2, tcg_addr, size, false, false);
>>> -            }
>>> -        }
>>>      }
>>>  }
>>>
>>> --
>>> 1.9.1
>>>

  reply	other threads:[~2015-11-23 18:55 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-11-16 18:28 [PATCH] target-arm/translate-a64.c: Correct unallocated checks for ldst_excl Peter Maydell
2015-11-16 18:28 ` [Qemu-devel] " Peter Maydell
2015-11-23 16:49 ` [Qemu-arm] " Peter Maydell
2015-11-23 16:49   ` [Qemu-devel] " Peter Maydell
2015-11-23 18:42   ` Sergey Fedorov
2015-11-23 18:42     ` [Qemu-devel] " Sergey Fedorov
2015-11-23 18:54     ` Sergey Fedorov [this message]
2015-11-23 18:54       ` Sergey Fedorov
2015-11-24 11:03       ` Peter Maydell
2015-11-24 11:03         ` [Qemu-devel] " Peter Maydell
2015-11-24 12:14         ` Sergey Fedorov
2015-11-24 12:14           ` [Qemu-devel] " Sergey Fedorov

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=565360FD.6010301@gmail.com \
    --to=serge.fdrv@gmail.com \
    --cc=laurent.desnogues@gmail.com \
    --cc=patches@linaro.org \
    --cc=peter.maydell@linaro.org \
    --cc=qemu-arm@nongnu.org \
    --cc=qemu-devel@nongnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.