rs6000 and ffs

[Andrew Pinski]

Okay now they look the same as the other patterns in rs600.md.
OK for b-i-b?


Thanks,
Andrew Pinski

-------------- next part --------------
A non-text attachment was scrubbed...
Name: ffs.diff
Type: application/octet-stream
Size: 2764 bytes
Desc: not available
URL: <http://gcc.gnu.org/pipermail/gcc-patches/attachments/20021202/a5503b85/attachment.obj>
-------------- next part --------------

On Monday, Dec 2, 2002, at 11:18 US/Pacific, David Edelsohn wrote:

> 	More style issues: final parenthesis and brackets should be on the
> same line, not a new line:
>
> (define_expand "ffssi2"
>   [(set (match_dup 2)
>         (neg:SI (match_operand:SI 1 "gpc_reg_operand" "r")))
>    (parallel [(set (match_dup 3)
> 		   (and:SI (match_dup 1) (match_dup 2)))
> 	      (clobber (scratch:CC))])
>    (set (match_dup 4) (unspec:SI [(match_dup 3)] 21))
>    (set (match_operand:SI 0 "gpc_reg_operand" "=r")
>         (minus:SI (const_int 32) (match_dup 4)))]
> ...
>
> (define_insn "cntlzw2"
>   [(set (match_operand:SI 0 "gpc_reg_operand" "=r")
>         (unspec:SI [(match_operand:SI 1 "gpc_reg_operand" "r")] 21))]
>   ""
>   "{cntlz|cntlzw} %0,%1")
>
>
> Just look at any of the other patterns in rs6000.md.  One should not be
> able to distinguish code you added from the style in the rest of the 
> file.
>
> David
>