From: "Pierre Muller" <pierre.muller@ics-cnrs.unistra.fr>
To: "'Mark Kettenis'" <mark.kettenis@xs4all.nl>, <hjl.tools@gmail.com>
Cc: <gdb-patches@sourceware.org>
Subject: RE: PATCH: Support i386 without SSE
Date: Thu, 08 Apr 2010 20:33:00 -0000 [thread overview]
Message-ID: <000c01cad75a$b751a490$25f4edb0$@muller@ics-cnrs.unistra.fr> (raw)
In-Reply-To: <201004081908.o38J8D9H025274@glazunov.sibelius.xs4all.nl>
> -----Message d'origine-----
> De : gdb-patches-owner@sourceware.org [mailto:gdb-patches-
> owner@sourceware.org] De la part de Mark Kettenis
> Envoyé : Thursday, April 08, 2010 9:08 PM
> À : hjl.tools@gmail.com
> Cc : gdb-patches@sourceware.org
> Objet : Re: PATCH: Support i386 without SSE
>
> > Date: Wed, 7 Apr 2010 13:05:47 -0700
> > From: "H.J. Lu" <hongjiu.lu@intel.com>
> >
> > Hi,
> >
> > This patch supports i386 without SSE. Any comments/suggestions?
>
> Eli, Pierre, can you test wether this fixes the issues on Go32?
I tested the patch and it does
indeed fix the two problems that I
reported.
Nevertheless, there are still strange things:
if you use 'maint print register-groups'
you will find out that
the SSE xmmX registers are still present as registers 32 to 40
and the ymmX as 41 to 48, but with empty names and a strange int0_t type.
(see below).
As this is not user visible,
I think that the patch is OK for me...
Pierre
(top-gdb) maint print register-groups
Name Nr Rel Offset Size Type Groups
eax 0 0 0 4 int32_t
general,all,save,restore
ecx 1 1 4 4 int32_t
general,all,save,restore
edx 2 2 8 4 int32_t
general,all,save,restore
ebx 3 3 12 4 int32_t
general,all,save,restore
esp 4 4 16 4 *1
general,all,save,restore
ebp 5 5 20 4 *1
general,all,save,restore
esi 6 6 24 4 int32_t
general,all,save,restore
edi 7 7 28 4 int32_t
general,all,save,restore
eip 8 8 32 4 *1
general,all,save,restore
eflags 9 9 36 4 i386_eflags
general,all,save,restore
cs 10 10 40 4 int32_t
general,all,save,restore
ss 11 11 44 4 int32_t
general,all,save,restore
ds 12 12 48 4 int32_t
general,all,save,restore
es 13 13 52 4 int32_t
general,all,save,restore
fs 14 14 56 4 int32_t
general,all,save,restore
gs 15 15 60 4 int32_t
general,all,save,restore
st0 16 16 64 10 _i387_ext float,all,save,restore
st1 17 17 74 10 _i387_ext float,all,save,restore
st2 18 18 84 10 _i387_ext float,all,save,restore
st3 19 19 94 10 _i387_ext float,all,save,restore
st4 20 20 104 10 _i387_ext float,all,save,restore
st5 21 21 114 10 _i387_ext float,all,save,restore
st6 22 22 124 10 _i387_ext float,all,save,restore
st7 23 23 134 10 _i387_ext float,all,save,restore
fctrl 24 24 144 4 long float,all,save,restore
fstat 25 25 148 4 long float,all,save,restore
ftag 26 26 152 4 long float,all,save,restore
fiseg 27 27 156 4 long float,all,save,restore
fioff 28 28 160 4 long float,all,save,restore
foseg 29 29 164 4 long float,all,save,restore
fooff 30 30 168 4 long float,all,save,restore
fop 31 31 172 4 long float,all,save,restore
'' 32 32 176 0 int0_t general
'' 33 33 176 0 int0_t general
'' 34 34 176 0 int0_t general
'' 35 35 176 0 int0_t general
'' 36 36 176 0 int0_t general
'' 37 37 176 0 int0_t general
'' 38 38 176 0 int0_t general
'' 39 39 176 0 int0_t general
'' 40 40 176 0 int0_t general
'' 41 41 176 0 int0_t general
'' 42 42 176 0 int0_t general
'' 43 43 176 0 int0_t general
'' 44 44 176 0 int0_t general
'' 45 45 176 0 int0_t general
'' 46 46 176 0 int0_t general
'' 47 47 176 0 int0_t general
'' 48 48 176 0 int0_t general
al 49 0 176 1 int8_t
cl 50 1 177 1 int8_t
dl 51 2 178 1 int8_t
bl 52 3 179 1 int8_t
ah 53 4 180 1 int8_t
ch 54 5 181 1 int8_t
dh 55 6 182 1 int8_t
bh 56 7 183 1 int8_t
ax 57 8 184 2 int16_t
cx 58 9 186 2 int16_t
dx 59 10 188 2 int16_t
bx 60 11 190 2 int16_t
'' 61 12 192 2 int16_t
bp 62 13 194 2 int16_t
si 63 14 196 2 int16_t
di 64 15 198 2 int16_t
mm0 65 16 200 8 _vec64i mmx,all,vector
mm1 66 17 208 8 _vec64i mmx,all,vector
mm2 67 18 216 8 _vec64i mmx,all,vector
mm3 68 19 224 8 _vec64i mmx,all,vector
mm4 69 20 232 8 _vec64i mmx,all,vector
mm5 70 21 240 8 _vec64i mmx,all,vector
mm6 71 22 248 8 _vec64i mmx,all,vector
mm7 72 23 256 8 _vec64i mmx,all,vector
*1: Register type's name NULL.
There are two oddities here:
the void * type are not nicely reported,
and the missing registers are in the general group.
next prev parent reply other threads:[~2010-04-08 20:33 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-04-07 20:05 H.J. Lu
2010-04-07 20:12 ` H.J. Lu
2010-04-07 20:31 ` Eli Zaretskii
2010-04-08 19:08 ` Mark Kettenis
2010-04-08 20:33 ` Pierre Muller [this message]
[not found] ` <000c01cad75a$b751a490$25f4edb0$%muller@ics-cnrs.unistra.fr>
2010-04-08 21:37 ` Eli Zaretskii
2010-04-08 22:29 ` H.J. Lu
2010-04-09 7:41 ` Eli Zaretskii
2010-04-09 13:16 ` H.J. Lu
2010-04-09 15:08 ` Eli Zaretskii
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='000c01cad75a$b751a490$25f4edb0$@muller@ics-cnrs.unistra.fr' \
--to=pierre.muller@ics-cnrs.unistra.fr \
--cc=gdb-patches@sourceware.org \
--cc=hjl.tools@gmail.com \
--cc=mark.kettenis@xs4all.nl \
/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