From: Marek Olšák <marek.olsak@amd.com>
Date: Thu, 25 Aug 2016 18:22:59 +0000 (+0200)
Subject: radeonsi: don't use allocas for arrays with LLVM 3.8
X-Git-Url: https://git.libre-soc.org/?a=commitdiff_plain;h=a491b9e945423a81f785220b85a9b395060952aa;p=mesa.git

radeonsi: don't use allocas for arrays with LLVM 3.8

It crashes.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=97413
---

diff --git a/src/gallium/drivers/radeon/radeon_setup_tgsi_llvm.c b/src/gallium/drivers/radeon/radeon_setup_tgsi_llvm.c
index 71f52fddbd3..4643e6d0ce7 100644
--- a/src/gallium/drivers/radeon/radeon_setup_tgsi_llvm.c
+++ b/src/gallium/drivers/radeon/radeon_setup_tgsi_llvm.c
@@ -559,8 +559,10 @@ static void emit_declaration(struct lp_build_tgsi_context *bld_base,
 			 * FIXME: We shouldn't need to have the non-alloca
 			 * code path for arrays. LLVM should be smart enough to
 			 * promote allocas into registers when profitable.
+			 *
+			 * LLVM 3.8 crashes with this.
 			 */
-			if (array_size > 16) {
+			if (HAVE_LLVM >= 0x0309 && array_size > 16) {
 				array_alloca = LLVMBuildAlloca(builder,
 					LLVMArrayType(bld_base->base.vec_type,
 						      array_size), "array");