Mesa (master): 22 new commits

Tue Jul 19 19:19:44 UTC 2016

URL:    http://cgit.freedesktop.org/mesa/mesa/commit/?id=0b626d75241213f2c5535b38aee233d6cdbed781
Author: Ian Romanick <ian.d.romanick at intel.com>
Date:   Mon Jun 27 11:38:15 2016 -0700

    nir/algebraic: Optimize fabs(u2f(x))
    
    I noticed this when I tried to do frexp(float(some_unsigned)) in the
    ir_unop_find_lsb lowering pass.  The code generated for frexp() uses
    fabs, and this resulted in an extra instruction.  Ultimately I ended up
    not using frexp.
    
    Signed-off-by: Ian Romanick <ian.d.romanick at intel.com>
    Reviewed-by: Matt Turner <mattst88 at gmail.com>

URL:    http://cgit.freedesktop.org/mesa/mesa/commit/?id=94296be276bf9c0a574df262b5d45f1bf38918ff
Author: Ian Romanick <ian.d.romanick at intel.com>
Date:   Fri Jun 24 01:58:18 2016 -0700

    st/mesa: Enable MESA_shader_integer_functions on all GLSL 1.30 platforms
    
    Signed-off-by: Ian Romanick <ian.d.romanick at intel.com>
    Reviewed-by: Matt Turner <mattst88 at gmail.com>

URL:    http://cgit.freedesktop.org/mesa/mesa/commit/?id=7cb49b1bd7f15d53e08d3716e9a835c42dc11c09
Author: Ian Romanick <ian.d.romanick at intel.com>
Date:   Thu Jun 23 16:19:45 2016 -0700

    i965: Enable MESA_shader_integer_functions on all GLSL 1.30 platforms
    
    Signed-off-by: Ian Romanick <ian.d.romanick at intel.com>
    Reviewed-by: Matt Turner <mattst88 at gmail.com>

URL:    http://cgit.freedesktop.org/mesa/mesa/commit/?id=5726e57f13e46ac1555861704f4318a58202b5ad
Author: Ian Romanick <ian.d.romanick at intel.com>
Date:   Thu Jun 9 16:35:30 2016 -0700

    i965: Don't lower uaddCarry and usubBorrow in both GLSL IR and NIR
    
    Signed-off-by: Ian Romanick <ian.d.romanick at intel.com>
    Reviewed-by: Matt Turner <mattst88 at gmail.com>

URL:    http://cgit.freedesktop.org/mesa/mesa/commit/?id=d7a47a76e0c6ccd1765f4c10c390e7d4f5f86414
Author: Ian Romanick <ian.d.romanick at intel.com>
Date:   Tue Jun 28 14:48:22 2016 -0700

    i965: Update assertion to account for Gen < 7
    
    Previously SHADER_OPCODE_MULH could only exist on Gen7+, so the
    assertion assumed the Gen7+ accumulator rules.  A future patch will
    allow this instruction on at least Gen6, so update the assertion.
    
    v2: Use get_lowered_simd_width instead of open coding it.  Suggested by
    Curro.
    
    Signed-off-by: Ian Romanick <ian.d.romanick at intel.com>
    Reviewed-by: Matt Turner <mattst88 at gmail.com> [v1]

URL:    http://cgit.freedesktop.org/mesa/mesa/commit/?id=3e7cebc8da5c9f16fa1b9a25ea72b8d31c86a440
Author: Ian Romanick <ian.d.romanick at intel.com>
Date:   Wed Jun 22 13:12:26 2016 -0700

    i965: Use LZD to implement nir_op_find_lsb on Gen < 7
    
    v2: Rebase on changes to previous two patches.
    
    Signed-off-by: Ian Romanick <ian.d.romanick at intel.com>
    Reviewed-by: Matt Turner <mattst88 at gmail.com>

URL:    http://cgit.freedesktop.org/mesa/mesa/commit/?id=c2019c6c261d5c46a4e5d3edc88836bcedf75f30
Author: Ian Romanick <ian.d.romanick at intel.com>
Date:   Tue Jun 21 17:18:04 2016 -0700

    i965: Use LZD to implement nir_op_ifind_msb on Gen < 7
    
    v2: Retype LZD source as UD to avoid potential problems with 0x80000000.
    Suggested by Matt.  Also update comment about problem values with
    LZD(abs(x)).  Suggested by Curro.
    
    Signed-off-by: Ian Romanick <ian.d.romanick at intel.com>
    Reviewed-by: Matt Turner <mattst88 at gmail.com>

URL:    http://cgit.freedesktop.org/mesa/mesa/commit/?id=de20086eed47e6bfe7c25835d72383114f99c7a9
Author: Ian Romanick <ian.d.romanick at intel.com>
Date:   Tue Jun 21 15:14:03 2016 -0700

    i965: Use LZD to implement nir_op_ufind_msb
    
    This uses one less instruction.
    
    v2: Move emit_find_msb_using_lzd out of the visitor classes.  Suggested
    by Curro.
    
    Signed-off-by: Ian Romanick <ian.d.romanick at intel.com>
    Reviewed-by: Matt Turner <mattst88 at gmail.com>

URL:    http://cgit.freedesktop.org/mesa/mesa/commit/?id=26c7f04d4a55b694623a389a564c9516d91c0026
Author: Ian Romanick <ian.d.romanick at intel.com>
Date:   Tue Jun 21 11:14:58 2016 -0700

    i965: Always enable GL_ARB_shading_language_packing
    
    With the existing lowering passes, the functions from this extension
    become a bunch of bit twiddling operations that have always been
    supported.
    
    Signed-off-by: Ian Romanick <ian.d.romanick at intel.com>
    Reviewed-by: Matt Turner <mattst88 at gmail.com>

URL:    http://cgit.freedesktop.org/mesa/mesa/commit/?id=4b2b6d4d4d62985679f164eef7d02a552e221941
Author: Ian Romanick <ian.d.romanick at intel.com>
Date:   Tue Jun 21 10:56:23 2016 -0700

    i965: Move enable of EXT_shader_integer_mix
    
    This extension does not depend on the Gen.  It only depends on the
    availability of GLSL 1.30.
    
    Signed-off-by: Ian Romanick <ian.d.romanick at intel.com>
    Reviewed-by: Matt Turner <mattst88 at gmail.com>

URL:    http://cgit.freedesktop.org/mesa/mesa/commit/?id=a2379e44aa7752c2a1fae42c8a8a6e8e2fb84e87
Author: Ian Romanick <ian.d.romanick at intel.com>
Date:   Mon Jun 27 15:21:21 2016 -0700

    glsl: Add lowering pass for ir_bin_imul_high
    
    This isn't the lowering pass you want.  Most GPUs that can support GLSL
    1.30 have a multiply unit that can do something more interesting than
    32x32->32.  Many have 32x16->48.  Any GPU that does, should do the
    lowering in the backend.  This is just the thing that will always work.
    
    Signed-off-by: Ian Romanick <ian.d.romanick at intel.com>
    Reviewed-by: Matt Turner <mattst88 at gmail.com>

URL:    http://cgit.freedesktop.org/mesa/mesa/commit/?id=1b5477668a7b1c448f64a69fbb488bcf47c3e029
Author: Ian Romanick <ian.d.romanick at intel.com>
Date:   Fri Jun 24 01:53:33 2016 -0700

    glsl: Add lowering pass for ir_unop_find_msb
    
    Signed-off-by: Ian Romanick <ian.d.romanick at intel.com>
    Reviewed-by: Matt Turner <mattst88 at gmail.com>

URL:    http://cgit.freedesktop.org/mesa/mesa/commit/?id=2a381a3c73be2f2df06f3feee708bf928645cf63
Author: Ian Romanick <ian.d.romanick at intel.com>
Date:   Fri Jun 24 01:17:23 2016 -0700

    glsl: Add lowering pass for ir_unop_find_lsb
    
    Signed-off-by: Ian Romanick <ian.d.romanick at intel.com>
    Reviewed-by: Matt Turner <mattst88 at gmail.com>

URL:    http://cgit.freedesktop.org/mesa/mesa/commit/?id=ad9acb19c39292de220b4d7dfdd2b5673129517a
Author: Ian Romanick <ian.d.romanick at intel.com>
Date:   Fri Jun 24 00:11:26 2016 -0700

    glsl: Add lowering pass for ir_unop_bitfield_reverse
    
    Signed-off-by: Ian Romanick <ian.d.romanick at intel.com>
    Reviewed-by: Matt Turner <mattst88 at gmail.com>

URL:    http://cgit.freedesktop.org/mesa/mesa/commit/?id=3079dcb00c19aa4773d1a1133bd8c4d1a48e4375
Author: Ian Romanick <ian.d.romanick at intel.com>
Date:   Thu Jun 23 23:16:10 2016 -0700

    glsl: Add lowering pass for ir_quadop_bitfield_insert
    
    Signed-off-by: Ian Romanick <ian.d.romanick at intel.com>
    Reviewed-by: Matt Turner <mattst88 at gmail.com>

URL:    http://cgit.freedesktop.org/mesa/mesa/commit/?id=4d6d219b58d43f6e14a6f9d7e49a00d437d65343
Author: Ian Romanick <ian.d.romanick at intel.com>
Date:   Thu Jun 23 16:57:23 2016 -0700

    glsl: Add lowering pass for ir_triop_bitfield_extract
    
    Signed-off-by: Ian Romanick <ian.d.romanick at intel.com>
    Reviewed-by: Matt Turner <mattst88 at gmail.com>

URL:    http://cgit.freedesktop.org/mesa/mesa/commit/?id=7340be8a01ca2df2d8578fa93589dbe5dea37e18
Author: Ian Romanick <ian.d.romanick at intel.com>
Date:   Thu Jun 23 16:16:21 2016 -0700

    glsl: Add lowering pass for ir_unop_bit_count
    
    Signed-off-by: Ian Romanick <ian.d.romanick at intel.com>
    Reviewed-by: Matt Turner <mattst88 at gmail.com>

URL:    http://cgit.freedesktop.org/mesa/mesa/commit/?id=806add360ff2ffbf70c6a52c64115fcb7ef718bd
Author: Ian Romanick <ian.d.romanick at intel.com>
Date:   Tue Jun 21 10:29:49 2016 -0700

    MESA_shader_integer_functions: Allow new function overload matching rules
    
    Signed-off-by: Ian Romanick <ian.d.romanick at intel.com>
    Reviewed-by: Matt Turner <mattst88 at gmail.com>

URL:    http://cgit.freedesktop.org/mesa/mesa/commit/?id=90537e1a0e97737c123742fb88fce7efd2284c04
Author: Ian Romanick <ian.d.romanick at intel.com>
Date:   Tue Jun 21 10:26:34 2016 -0700

    MESA_shader_integer_functions: Allow implicit int->uint conversions
    
    Signed-off-by: Ian Romanick <ian.d.romanick at intel.com>
    Reviewed-by: Matt Turner <mattst88 at gmail.com>

URL:    http://cgit.freedesktop.org/mesa/mesa/commit/?id=65b0346fdbaf22eb2a6fc33e426b672a40495800
Author: Ian Romanick <ian.d.romanick at intel.com>
Date:   Tue Jun 21 10:21:47 2016 -0700

    MESA_shader_integer_functions: Expose new built-in functions
    
    Signed-off-by: Ian Romanick <ian.d.romanick at intel.com>
    Reviewed-by: Matt Turner <mattst88 at gmail.com>

URL:    http://cgit.freedesktop.org/mesa/mesa/commit/?id=15c4ae461dba101e12924fc8e3e9099b3e700bab
Author: Ian Romanick <ian.d.romanick at intel.com>
Date:   Mon Jun 20 16:35:32 2016 -0700

    MESA_shader_integer_functions: Boiler plate extension tracking
    
    Signed-off-by: Ian Romanick <ian.d.romanick at intel.com>
    Reviewed-by: Matt Turner <mattst88 at gmail.com>

URL:    http://cgit.freedesktop.org/mesa/mesa/commit/?id=91482ef226de7686350202cfbdfda4358d9cea86
Author: Ian Romanick <ian.d.romanick at intel.com>
Date:   Mon Jun 20 16:28:34 2016 -0700

    MESA_shader_integer_functions: Add extension specification
    
    v2: Fix typo in #extension line noticed by Ken.
    
    v3: Update spec status.
    
    Signed-off-by: Ian Romanick <ian.d.romanick at intel.com>
    Reviewed-by: Matt Turner <mattst88 at gmail.com>