Re: Committed: fix cris.md strict_low_part constraints (was: Re: A doubt about constraint modifiers)
> Date: Sun, 13 Apr 2008 16:18:04 +0530
> From: "Mohamed Shafi" <[EMAIL PROTECTED]>
> I am glad that the mistake has been rectified. But it would be of
> great help requirement of the '+' constraint for strict_low_part is
> mentioned somewhere in the gcc internals. Even though the mailing list
> helped me to solve the problem , i could have saved some time had it
> been documented.
It's not a requirement of strict_low_part per se, but for it
being an operator that keeps the rest of the register
unmodified, therefore an input to the insn. As Bernd mentioned,
it was when operand 0 was not *otherwise* marked as an input in
the constraints (by use of effectively "%0"), that it had to be
marked with "+". There's at least one other RTL operator that
is also keeping the rest of the register unmodified:
zero_extract. Once you understand that (which you have to),
IMHO the existing:
@table @samp
@cindex @samp{=} in constraint
@item =
Means that this operand is write-only for this instruction: the previous
value is discarded and replaced by output data.
@cindex @samp{+} in constraint
@item +
Means that this operand is both read and written by the instruction
should be sufficient. Maybe a note at the strict_low_part and
zero_extract documentation would help, but I'm not sure what.
Patches are welcome.
brgds, H-P
Re: Committed: fix cris.md strict_low_part constraints (was: Re: A doubt about constraint modifiers)
Hello all,
I am glad that the mistake has been rectified. But it would be of
great help requirement of the '+' constraint for strict_low_part is
mentioned somewhere in the gcc internals. Even though the mailing list
helped me to solve the problem , i could have saved some time had it
been documented.
Regards,
Shafi
On Sun, Apr 13, 2008 at 6:23 AM, Hans-Peter Nilsson
<[EMAIL PROTECTED]> wrote:
> > Date: Fri, 11 Apr 2008 15:32:02 +0200
> > From: Bernd Schmidt <[EMAIL PROTECTED]>
>
> > Mohamed Shafi wrote:
> > > In cris i saw this patten
> > >
> > > (define_insn "*andhi_lowpart"
> > > [(set (strict_low_part
> > > (match_operand:HI 0 "register_operand""=r,r, r,r,r,r"))
> > > (and:HI (match_operand:HI 1 "register_operand" "%0,0, 0,0,0,r")
> > > (match_operand:HI 2 "general_operand" "r,Q>,L,O,g,!To")))]
> > >
> > > Here = is used.
> >
> > This is incorrect since when the last alternative matches, nothing tells
> > the compiler that operand 0 is also an input.
>
> Oops, thanks for noticing. For consistency, I made all
> strict_low_part constraints use "+", including those where all
> alternatives had %0. Regtested cross to cris-elf and
> crisv32-elf.
>
> PS. I'm not a native speaker, but FWIW I agree with Joe Buck
> regarding usage of the word "doubt", but I see it's quite common
> with some non-native speakers, indicating the fault is at the
> teaching level.
>
> gcc:
> * config/cris/cris.md ("*andhi_lowpart_non_v32", "*andhi_lowpart_v32")
> ("*andqi_lowpart_non_v32", "*andqi_lowpart_v32"): Use "+" for the
> operand 0 constraint, not "=".
>
> Index: config/cris/cris.md
> ===
> --- config/cris/cris.md (revision 134224)
> +++ config/cris/cris.md (working copy)
> @@ -2958,7 +2958,7 @@ (define_insn "*expanded_andhi_v32"
>
> (define_insn "*andhi_lowpart_non_v32"
>[(set (strict_low_part
> -(match_operand:HI 0 "register_operand""=r,r, r,r,r,r"))
> +(match_operand:HI 0 "register_operand""+r,r, r,r,r,r"))
> (and:HI (match_operand:HI 1 "register_operand" "%0,0, 0,0,0,r")
> (match_operand:HI 2 "general_operand" "r,Q>,L,O,g,!To")))]
>"!TARGET_V32"
> @@ -2974,7 +2974,7 @@ (define_insn "*andhi_lowpart_non_v32"
>
> (define_insn "*andhi_lowpart_v32"
>[(set (strict_low_part
> -(match_operand:HI 0 "register_operand" "=r,r,r,r,r"))
> +(match_operand:HI 0 "register_operand" "+r,r,r,r,r"))
> (and:HI (match_operand:HI 1 "register_operand" "%0,0,0,0,0")
> (match_operand:HI 2 "general_operand" "r,Q>,L,O,g")))]
>"TARGET_V32"
> @@ -3025,7 +3025,7 @@ (define_insn "*andqi3_v32"
>
> (define_insn "*andqi_lowpart_non_v32"
>[(set (strict_low_part
> -(match_operand:QI 0 "register_operand""=r,r, r,r,r"))
> +(match_operand:QI 0 "register_operand""+r,r, r,r,r"))
> (and:QI (match_operand:QI 1 "register_operand" "%0,0, 0,0,r")
> (match_operand:QI 2 "general_operand" "r,Q>,O,g,!To")))]
>"!TARGET_V32"
> @@ -3040,7 +3040,7 @@ (define_insn "*andqi_lowpart_non_v32"
>
> (define_insn "*andqi_lowpart_v32"
>[(set (strict_low_part
> -(match_operand:QI 0 "register_operand" "=r,r,r,r"))
> +(match_operand:QI 0 "register_operand" "+r,r,r,r"))
> (and:QI (match_operand:QI 1 "register_operand" "%0,0,0,0")
> (match_operand:QI 2 "general_operand" "r,Q>,O,g")))]
>"TARGET_V32"
>
> brgds, H-P
>
Committed: fix cris.md strict_low_part constraints (was: Re: A doubt about constraint modifiers)
> Date: Fri, 11 Apr 2008 15:32:02 +0200
> From: Bernd Schmidt <[EMAIL PROTECTED]>
> Mohamed Shafi wrote:
> > In cris i saw this patten
> >
> > (define_insn "*andhi_lowpart"
> > [(set (strict_low_part
> > (match_operand:HI 0 "register_operand""=r,r, r,r,r,r"))
> > (and:HI (match_operand:HI 1 "register_operand" "%0,0, 0,0,0,r")
> > (match_operand:HI 2 "general_operand" "r,Q>,L,O,g,!To")))]
> >
> > Here = is used.
>
> This is incorrect since when the last alternative matches, nothing tells
> the compiler that operand 0 is also an input.
Oops, thanks for noticing. For consistency, I made all
strict_low_part constraints use "+", including those where all
alternatives had %0. Regtested cross to cris-elf and
crisv32-elf.
PS. I'm not a native speaker, but FWIW I agree with Joe Buck
regarding usage of the word "doubt", but I see it's quite common
with some non-native speakers, indicating the fault is at the
teaching level.
gcc:
* config/cris/cris.md ("*andhi_lowpart_non_v32", "*andhi_lowpart_v32")
("*andqi_lowpart_non_v32", "*andqi_lowpart_v32"): Use "+" for the
operand 0 constraint, not "=".
Index: config/cris/cris.md
===
--- config/cris/cris.md (revision 134224)
+++ config/cris/cris.md (working copy)
@@ -2958,7 +2958,7 @@ (define_insn "*expanded_andhi_v32"
(define_insn "*andhi_lowpart_non_v32"
[(set (strict_low_part
-(match_operand:HI 0 "register_operand""=r,r, r,r,r,r"))
+(match_operand:HI 0 "register_operand""+r,r, r,r,r,r"))
(and:HI (match_operand:HI 1 "register_operand" "%0,0, 0,0,0,r")
(match_operand:HI 2 "general_operand" "r,Q>,L,O,g,!To")))]
"!TARGET_V32"
@@ -2974,7 +2974,7 @@ (define_insn "*andhi_lowpart_non_v32"
(define_insn "*andhi_lowpart_v32"
[(set (strict_low_part
-(match_operand:HI 0 "register_operand" "=r,r,r,r,r"))
+(match_operand:HI 0 "register_operand" "+r,r,r,r,r"))
(and:HI (match_operand:HI 1 "register_operand" "%0,0,0,0,0")
(match_operand:HI 2 "general_operand" "r,Q>,L,O,g")))]
"TARGET_V32"
@@ -3025,7 +3025,7 @@ (define_insn "*andqi3_v32"
(define_insn "*andqi_lowpart_non_v32"
[(set (strict_low_part
-(match_operand:QI 0 "register_operand""=r,r, r,r,r"))
+(match_operand:QI 0 "register_operand""+r,r, r,r,r"))
(and:QI (match_operand:QI 1 "register_operand" "%0,0, 0,0,r")
(match_operand:QI 2 "general_operand" "r,Q>,O,g,!To")))]
"!TARGET_V32"
@@ -3040,7 +3040,7 @@ (define_insn "*andqi_lowpart_non_v32"
(define_insn "*andqi_lowpart_v32"
[(set (strict_low_part
-(match_operand:QI 0 "register_operand" "=r,r,r,r"))
+(match_operand:QI 0 "register_operand" "+r,r,r,r"))
(and:QI (match_operand:QI 1 "register_operand" "%0,0,0,0")
(match_operand:QI 2 "general_operand" "r,Q>,O,g")))]
"TARGET_V32"
brgds, H-P
Re: A doubt about constraint modifiers
On Fri, Apr 11, 2008 at 11:28:04AM +0530, Mohamed Shafi wrote: > [ another "doubt" ] You seem to be using the word "doubt" a lot whenever you don't completely understand something, but this is not what the word means. It means "to consider questionable or unlikely; to hesitate to believe; to distrust", as in "I doubt that she is telling the truth". So it makes it sound, not like you don't understand something, but that you think it is bad or wrong (not trustworthy). I know that this isn't your intention.
Re: A doubt about constraint modifiers
Mohamed Shafi wrote: In cris i saw this patten (define_insn "*andhi_lowpart" [(set (strict_low_part (match_operand:HI 0 "register_operand" "=r,r, r,r,r,r")) (and:HI (match_operand:HI 1 "register_operand" "%0,0, 0,0,0,r") (match_operand:HI 2 "general_operand" "r,Q>,L,O,g,!To")))] Here = is used. This is incorrect since when the last alternative matches, nothing tells the compiler that operand 0 is also an input. Bernd -- This footer brought to you by insane German lawmakers. Analog Devices GmbH Wilhelm-Wagenfeld-Str. 6 80807 Muenchen Sitz der Gesellschaft Muenchen, Registergericht Muenchen HRB 40368 Geschaeftsfuehrer Thomas Wessel, William A. Martin, Margaret Seif
Re: A doubt about constraint modifiers
On Fri, Apr 11, 2008 at 12:41 PM, Ian Lance Taylor <[EMAIL PROTECTED]> wrote: > "Mohamed Shafi" <[EMAIL PROTECTED]> writes: > > > I have noticed that when strict_low_part is used in a patten we need > > to use '+' as the constraint modifier if any constraints are used in > > the patterns. > > Why is this so? > > Using strict_low_part implies that the register or memory location is > neither a pure input nor a pure output. It is both an input and an > output. Therefore a '+' constraint is appropriate. Thanks for the reply. In cris i saw this patten (define_insn "*andhi_lowpart" [(set (strict_low_part (match_operand:HI 0 "register_operand""=r,r, r,r,r,r")) (and:HI (match_operand:HI 1 "register_operand" "%0,0, 0,0,0,r") (match_operand:HI 2 "general_operand" "r,Q>,L,O,g,!To")))] Here = is used. So when you say appropriate you mean to say that its not mandatory to use '+' but '=' also suffices? Regards, Shafi
Re: A doubt about constraint modifiers
"Mohamed Shafi" <[EMAIL PROTECTED]> writes: > I have noticed that when strict_low_part is used in a patten we need > to use '+' as the constraint modifier if any constraints are used in > the patterns. > Why is this so? Using strict_low_part implies that the register or memory location is neither a pure input nor a pure output. It is both an input and an output. Therefore a '+' constraint is appropriate. Ian
A doubt about constraint modifiers
Hello all, I have noticed that when strict_low_part is used in a patten we need to use '+' as the constraint modifier if any constraints are used in the patterns. Why is this so? Regards, Shafi
