From: Chen Gang <chengang@emindsoft.com.cn>
To: Paolo Bonzini <pbonzini@redhat.com>
Cc: qemu-devel@nongnu.org, Chen Gang <gang.chen.5i5j@gmail.com>,
ehabkost@redhat.com, rth@twiddle.net
Subject: Re: [PATCH] target: i386: Check float overflow about register stack
Date: Fri, 21 Feb 2020 22:09:11 +0800 [thread overview]
Message-ID: <900fd511-72f0-675d-4a7e-d228b2ade9c7@emindsoft.com.cn> (raw)
In-Reply-To: <a5533719-7ef1-938b-e52c-20711e65417f@redhat.com>
On 2020/2/21 下午4:58, Paolo Bonzini wrote:
> On 21/02/20 04:45, chengang@emindsoft.com.cn wrote:
>> static inline void fpush(CPUX86State *env)
>> {
>> - env->fpstt = (env->fpstt - 1) & 7;
>> - env->fptags[env->fpstt] = 0; /* validate stack entry */
>> + set_fpstt(env, env->fpstt - 1, false, true);
>
> On overflow fpstt is ~0, so this does:
>
> env->foverflow = true;
> env->fpstt = 7;
> env->fptags[7] = 0; /* validate stack entry */
>
> Is this correct? You are going to set ST0 so the register should not be
> marked empty.
>
Originally, I wanted to add foverflow to mark the stack overflow only,
and kept another things no touch.
But I think what you said above is correct, for me, if fpush/f[i]ld*_STO
are overflow, the env->fpstt, env->fpregs and env->fptags should be kept
no touch, and foverflow is set true, so there is no negative effect.
Welcome your idea.
>> void helper_fdecstp(CPUX86State *env)
>> {
>> - env->fpstt = (env->fpstt - 1) & 7;
>> + set_fpstt(env, env->fpstt - 1, false, false);
>
> This is clearing env->foverflow. But after 8 consecutive fdecstp or
> fincstp the result of FXAM should not change.
>
>> env->fpus &= ~0x4700;
>> }
>>
>> void helper_fincstp(CPUX86State *env)
>> {
>> - env->fpstt = (env->fpstt + 1) & 7;
>> + set_fpstt(env, env->fpstt + 1, true, false);
>
> Same here.
>
OK. thanks.
Now if foverflow is only for fpush/f[i]ld*_ST0, I guess fincstp/fdecstp
can clear foverflow. The env->fptags are only for fpop, which keep no
touch in fincstp/fdecstp.
> The actual bug is hinted in helper_fxam_ST0:
>
> /* XXX: test fptags too */
>
> I think the correct fix should be something like
>
> diff --git a/target/i386/fpu_helper.c b/target/i386/fpu_helper.c
> index 99f28f267f..792a128a6d 100644
> --- a/target/i386/fpu_helper.c
> +++ b/target/i386/fpu_helper.c
> @@ -991,7 +991,11 @@ void helper_fxam_ST0(CPUX86State *env)
> env->fpus |= 0x200; /* C1 <-- 1 */
> }
>
> - /* XXX: test fptags too */
> + if (env->fptags[env->fpstt]) {
> + env->fpus |= 0x4100; /* Empty */
> + return;
> + }
> +
For fpop overflow, this fix is enough, but for me, we still need
foverflow to check fpush/fld*_ST0 overflow.
Don't you think we need check fpush/f[i]ld*_ST0 overflow?
Thanks
next prev parent reply other threads:[~2020-02-21 14:10 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-02-21 3:45 [PATCH] target: i386: Check float overflow about register stack chengang
2020-02-21 8:58 ` Paolo Bonzini
2020-02-21 14:09 ` Chen Gang [this message]
2020-02-21 16:18 ` Paolo Bonzini
2020-02-22 2:10 ` Chen Gang
2020-02-22 2:27 ` Chen Gang
2020-02-22 7:37 ` Paolo Bonzini
2020-02-22 12:25 ` Chen Gang
2020-02-24 12:43 ` Paolo Bonzini
2020-02-25 1:22 ` Chen Gang
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=900fd511-72f0-675d-4a7e-d228b2ade9c7@emindsoft.com.cn \
--to=chengang@emindsoft.com.cn \
--cc=ehabkost@redhat.com \
--cc=gang.chen.5i5j@gmail.com \
--cc=pbonzini@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=rth@twiddle.net \
/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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).