[Mesa-stable] [PATCH 1/2] i965/fs: Fix single-precision to double-precision conversions for CHV/BSW

Samuel Iglesias Gonsálvez siglesias at igalia.com
Wed Jun 15 07:25:44 UTC 2016


From: Iago Toral Quiroga <itoral at igalia.com>

>From the Cherryview PRM, Volume 7, 3D Media GPGPU Engine,
Register Region Restrictions:

   "When source or destination is 64b (...), regioning in Align1
    must follow these rules:

    1. Source and destination horizontal stride must be aligned to
       the same qword.
    (...)"

Cc: "12.0" <mesa-stable at lists.freedesktop.org>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=95462
---
 src/mesa/drivers/dri/i965/brw_fs_nir.cpp | 27 +++++++++++++++++++++++++++
 1 file changed, 27 insertions(+)

diff --git a/src/mesa/drivers/dri/i965/brw_fs_nir.cpp b/src/mesa/drivers/dri/i965/brw_fs_nir.cpp
index b811953..c271e64 100644
--- a/src/mesa/drivers/dri/i965/brw_fs_nir.cpp
+++ b/src/mesa/drivers/dri/i965/brw_fs_nir.cpp
@@ -715,10 +715,37 @@ fs_visitor::nir_emit_alu(const fs_builder &bld, nir_alu_instr *instr)
    case nir_op_u2f:
       if (optimize_extract_to_float(instr, result))
          return;
+      inst = bld.MOV(result, op[0]);
+      inst->saturate = instr->dest.saturate;
+      break;
 
    case nir_op_f2d:
    case nir_op_i2d:
    case nir_op_u2d:
+      /* CHV PRM, vol07, 3D Media GPGPU Engine, Register Region Restrictions:
+       *
+       *    "When source or destination is 64b (...), regioning in Align1
+       *     must follow these rules:
+       *
+       *     1. Source and destination horizontal stride must be aligned to
+       *        the same qword.
+       *     (...)"
+       *
+       * This means that 32-bit to 64-bit conversions need to have the 32-bit
+       * data elements aligned to 64-bit. This restriction does not apply to
+       * BDW and later.
+       */
+      if (devinfo->is_cherryview) {
+         fs_reg tmp = bld.vgrf(result.type, 1);
+         tmp = subscript(tmp, op[0].type, 0);
+         inst = bld.MOV(tmp, op[0]);
+         inst->regs_written =
+            inst->dst.component_size(bld.dispatch_width()) / REG_SIZE;
+         inst = bld.MOV(result, tmp);
+         inst->saturate = instr->dest.saturate;
+         break;
+      }
+      /* fallthrough */
    case nir_op_d2f:
    case nir_op_d2i:
    case nir_op_d2u:
-- 
2.8.1



More information about the mesa-stable mailing list