gallivm/nir: fix integer divide SIGFPE

Blender was crashing with a SIGFPE even though the divide by 0
logic was kicking in. I'm not sure why TGSI doesn't get into this state.

The problem was is the numerator was INT_MIN we'd replace the div by
0 with a divide by -1, which is an exception for INT_MIN as INT_MIN/-1
== INT_MAX + 1 (too large for 32-bits). Instead for integer divides
just replace the mask values with 0x7fffffff. Also fix up the
result handling so it aligns with TGSI usage. (gives 0)

Fixes: c717ac1247c3 ("gallivm/nir: wrap idiv to avoid divide by 0 (v2)")
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3956>

diff --git a/src/gallium/auxiliary/gallivm/lp_bld_nir.c b/src/gallium/auxiliary/gallivm/lp_bld_nir.c
index f6e84855bf5d4fd9a1ae144e2eff251e98ad45ca..c9f67226dcc9444f7166fcd9ebd6a032a990e85b 100644 (file)
--- a/src/gallium/auxiliary/gallivm/lp_bld_nir.c
+++ b/src/gallium/auxiliary/gallivm/lp_bld_nir.c
@@ -389,15 +389,26 @@ do_int_divide(struct lp_build_nir_context *bld_base,
    struct gallivm_state *gallivm = bld_base->base.gallivm;
    LLVMBuilderRef builder = gallivm->builder;
    struct lp_build_context *int_bld = get_int_bld(bld_base, is_unsigned, src_bit_size);
-   LLVMValueRef div_mask = lp_build_cmp(int_bld, PIPE_FUNC_EQUAL, src2,
-                                        int_bld->zero);
+   struct lp_build_context *mask_bld = get_int_bld(bld_base, true, src_bit_size);
+   LLVMValueRef div_mask = lp_build_cmp(mask_bld, PIPE_FUNC_EQUAL, src2,
+                                        mask_bld->zero);
+
+   if (!is_unsigned) {
+      /* INT_MIN (0x80000000) / -1 (0xffffffff) causes sigfpe, seen with blender. */
+      div_mask = LLVMBuildAnd(builder, div_mask, lp_build_const_int_vec(gallivm, int_bld->type, 0x7fffffff), "");
+   }
    LLVMValueRef divisor = LLVMBuildOr(builder,
                                       div_mask,
                                       src2, "");
    LLVMValueRef result = lp_build_div(int_bld, src, divisor);
-   /* udiv by zero is guaranteed to return 0xffffffff at least with d3d10
-    * may as well do same for idiv */
-   return LLVMBuildOr(builder, div_mask, result, "");
+
+   if (!is_unsigned) {
+      LLVMValueRef not_div_mask = LLVMBuildNot(builder, div_mask, "");
+      return LLVMBuildAnd(builder, not_div_mask, result, "");
+   } else
+      /* udiv by zero is guaranteed to return 0xffffffff at least with d3d10
+       * may as well do same for idiv */
+      return LLVMBuildOr(builder, div_mask, result, "");
 }
 
 static LLVMValueRef do_alu_action(struct lp_build_nir_context *bld_base,