The 32-bit to 64-bit conversions need to have the 32-bit
data source elements aligned to 64-bit but only with doubles as
destination type.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=99660
Signed-off-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Tested-by: Mark Janes <mark.a.janes@intel.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
case nir_op_f2d:
case nir_op_i2d:
case nir_op_u2d:
- case nir_op_f2i64:
- case nir_op_f2u64:
- case nir_op_i2i64:
- case nir_op_i2u64:
- case nir_op_u2i64:
- case nir_op_u2u64:
- case nir_op_b2i64:
/* CHV PRM, vol07, 3D Media GPGPU Engine, Register Region Restrictions:
*
* "When source or destination is 64b (...), regioning in Align1
break;
}
/* fallthrough */
+ case nir_op_f2i64:
+ case nir_op_f2u64:
+ case nir_op_i2i64:
+ case nir_op_i2u64:
+ case nir_op_u2i64:
+ case nir_op_u2u64:
+ case nir_op_b2i64:
case nir_op_d2f:
case nir_op_d2i:
case nir_op_d2u: