Re: [Mesa-dev] [PATCH (gles3) 15/20] glsl: Add lowering pass for GLSL ES 3.00 pack/unpack operations (v2)

2013-01-23 Thread Ian Romanick
On 01/22/2013 10:36 PM, Matt Turner wrote: On Mon, Jan 21, 2013 at 12:49 AM, Chad Versace chad.vers...@linux.intel.com wrote: + /** +* \brief Lower a packSnorm2x16 expression. +* +* \param vec2_rval is packSnorm2x16's input +* \return packSnorm2x16's output as a uint rvalue +

Re: [Mesa-dev] [PATCH (gles3) 15/20] glsl: Add lowering pass for GLSL ES 3.00 pack/unpack operations (v2)

2013-01-23 Thread Matt Turner
On Wed, Jan 23, 2013 at 6:06 AM, Ian Romanick i...@freedesktop.org wrote: On 01/22/2013 10:36 PM, Matt Turner wrote: On Mon, Jan 21, 2013 at 12:49 AM, Chad Versace chad.vers...@linux.intel.com wrote: + /** +* \brief Lower a packSnorm2x16 expression. +* +* \param vec2_rval is

Re: [Mesa-dev] [PATCH (gles3) 15/20] glsl: Add lowering pass for GLSL ES 3.00 pack/unpack operations (v2)

2013-01-23 Thread Chad Versace
On 01/22/2013 07:36 PM, Matt Turner wrote: On Mon, Jan 21, 2013 at 12:49 AM, Chad Versace chad.vers...@linux.intel.com wrote: Lower them to arithmetic and bit manipulation expressions. v2: - Rewrite using ir_builder. [for idr] - In lowering packHalf2x16, don't truncate subnormal float16

Re: [Mesa-dev] [PATCH (gles3) 15/20] glsl: Add lowering pass for GLSL ES 3.00 pack/unpack operations (v2)

2013-01-23 Thread Chad Versace
On 01/22/2013 09:19 PM, Paul Berry wrote: On 21 January 2013 00:49, Chad Versace chad.vers...@linux.intel.com wrote: Lower them to arithmetic and bit manipulation expressions. v2: - Rewrite using ir_builder. [for idr] - In lowering packHalf2x16, don't truncate subnormal float16 values

Re: [Mesa-dev] [PATCH (gles3) 15/20] glsl: Add lowering pass for GLSL ES 3.00 pack/unpack operations (v2)

2013-01-23 Thread Chad Versace
On 01/23/2013 06:06 AM, Ian Romanick wrote: On 01/22/2013 10:36 PM, Matt Turner wrote: On Mon, Jan 21, 2013 at 12:49 AM, Chad Versace chad.vers...@linux.intel.com wrote: + /** +* \brief Lower a packSnorm2x16 expression. +* +* \param vec2_rval is packSnorm2x16's input +*

Re: [Mesa-dev] [PATCH (gles3) 15/20] glsl: Add lowering pass for GLSL ES 3.00 pack/unpack operations (v2)

2013-01-23 Thread Paul Berry
On 23 January 2013 12:44, Chad Versace chad.vers...@linux.intel.com wrote: On 01/22/2013 09:19 PM, Paul Berry wrote: On 21 January 2013 00:49, Chad Versace chad.vers...@linux.intel.com wrote: Lower them to arithmetic and bit manipulation expressions. v2: - Rewrite using

Re: [Mesa-dev] [PATCH (gles3) 15/20] glsl: Add lowering pass for GLSL ES 3.00 pack/unpack operations (v2)

2013-01-23 Thread Ian Romanick
On 01/23/2013 03:48 PM, Chad Versace wrote: On 01/23/2013 06:06 AM, Ian Romanick wrote: On 01/22/2013 10:36 PM, Matt Turner wrote: On Mon, Jan 21, 2013 at 12:49 AM, Chad Versace chad.vers...@linux.intel.com wrote: + /** +* \brief Lower a packSnorm2x16 expression. +* +* \param

Re: [Mesa-dev] [PATCH (gles3) 15/20] glsl: Add lowering pass for GLSL ES 3.00 pack/unpack operations (v2)

2013-01-22 Thread Matt Turner
On Mon, Jan 21, 2013 at 12:49 AM, Chad Versace chad.vers...@linux.intel.com wrote: Lower them to arithmetic and bit manipulation expressions. v2: - Rewrite using ir_builder. [for idr] - In lowering packHalf2x16, don't truncate subnormal float16 values to zero. And round to even

Re: [Mesa-dev] [PATCH (gles3) 15/20] glsl: Add lowering pass for GLSL ES 3.00 pack/unpack operations (v2)

2013-01-22 Thread Paul Berry
On 21 January 2013 00:49, Chad Versace chad.vers...@linux.intel.com wrote: Lower them to arithmetic and bit manipulation expressions. v2: - Rewrite using ir_builder. [for idr] - In lowering packHalf2x16, don't truncate subnormal float16 values to zero. And round to even rather than

[Mesa-dev] [PATCH (gles3) 15/20] glsl: Add lowering pass for GLSL ES 3.00 pack/unpack operations (v2)

2013-01-21 Thread Chad Versace
Lower them to arithmetic and bit manipulation expressions. v2: - Rewrite using ir_builder. [for idr] - In lowering packHalf2x16, don't truncate subnormal float16 values to zero. And round to even rather than to zero. [for stereotype441] CC: Ian Romanick i...@freedesktop.org CC: Paul