[Beignet] [PATCH] improve the build performance of vector type built-in function.

[Zhigang Gong]

In practice for the vector which has <= 4 components, it's better
to keep the original method.

For a 4 or less components vector, Use the complex union
and the for loop may bring too much overhead.

What's your opinion?

On Thu, Jul 24, 2014 at 04:23:24AM +0800, xionghu.luo at intel.com wrote:
> From: LuoXionghu <xionghu.luo at intel.com>
> 
> expand the gentypen with loop to reduce the redundant inline.
> 
> Signed-off-by: LuoXionghu <xionghu.luo at intel.com>
> ---
>  backend/src/gen_builtin_vector.py | 42 +++++++++++++++++++++++++++++++++------
>  1 file changed, 36 insertions(+), 6 deletions(-)
> 
> diff --git a/backend/src/gen_builtin_vector.py b/backend/src/gen_builtin_vector.py
> index b100bbf..83e2bcb 100755
> --- a/backend/src/gen_builtin_vector.py
> +++ b/backend/src/gen_builtin_vector.py
> @@ -283,9 +283,39 @@ class builtinProto():
>                  formatStr += '{0} {1}param{2}'.format(ptype[0], pointerStr, n)
>  
>          formatStr += ')'
> -        formatStr = self.append(formatStr, '{{return ({0}{1})('.format(vtype[0], vtype[1]))
> -        self.indent = len(formatStr)
> -        for j in range(0, vtype[1]):
> +        if self.functionName != 'select' and ptypeSeqs[0] == ptypeSeqs[self.paramCount-1]:
> +            formatStr += '\n{ \n  union{'
> +            formatStr = self.append(formatStr, '    {0} va[{1}];'.format(vtype[0], vtype[1]))
> +            formatStr = self.append(formatStr, '    {0}{1} vv{2};'.format(vtype[0], vtype[1], vtype[1]))
> +            formatStr += '\n  }uret;'
> +            formatStr += '\n  union{'
> +            formatStr = self.append(formatStr, '    {0} pa[{1}];'.format(ptype[0], ptype[1]))
> +            formatStr = self.append(formatStr, '    {0}{1} pv{2};'.format(ptype[0], ptype[1], ptype[1]))
> +            formatStr += '\n  }'
> +            for n in range(0, self.paramCount):
> +              formatStr += 'usrc{0}'.format(n)
> +              if n+1 != self.paramCount:
> +                formatStr +=', '
> +            formatStr += ';'
> +
> +            for n in range(0, self.paramCount):
> +              formatStr = self.append(formatStr, '  usrc{0}.pv{1} = param{2};'.format(n, ptype[1], n))
> +            formatStr = self.append(formatStr, '  for(int i =0; i < {0}; i++)'.format(ptype[1]))
> +            formatStr = self.append(formatStr, '    uret.va[i] = {0}('.format(self.functionName))
> +
> +            for n in range(0, self.paramCount):
> +              formatStr += 'usrc{0}.pa[i]'.format(n)
> +              if n+1 != self.paramCount:
> +                formatStr +=', '
> +            formatStr += ');'
> +            formatStr = self.append(formatStr, ' return uret.vv{0};'.format(vtype[1]))
> +            formatStr += '\n}'
> +            formatStr = self.append(formatStr)
> +            return formatStr
> +        else:
> +          formatStr = self.append(formatStr, '{{return ({0}{1})('.format(vtype[0], vtype[1]))
> +          self.indent = len(formatStr)
> +          for j in range(0, vtype[1]):
>              if (j != 0):
>                  formatStr += ','
>                  if (j + 1) % 2 == 0:
> @@ -320,10 +350,10 @@ class builtinProto():
>  
>              formatStr += ')'
>  
> -        formatStr += '); }\n'
> -        self.append(formatStr)
> +          formatStr += '); }\n'
> +          self.append(formatStr)
>  
> -        return formatStr
> +          return formatStr
>  
>      def output(self):
>          for line in self.outputStr:
> -- 
> 1.8.1.2
> 
> _______________________________________________
> Beignet mailing list
> Beignet at lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/beignet