From: Mark Stapp <mjs@voltanet.io>
Date: Fri, 12 Jun 2020 20:31:22 +0000 (-0400)
Subject: zebra: avoid an atomic builtin that clang doesn't like
X-Git-Tag: base_7.5~273^2
X-Git-Url: https://git.puffer.fish/?a=commitdiff_plain;h=refs%2Fpull%2F6575%2Fhead;p=mirror%2Ffrr.git

zebra: avoid an atomic builtin that clang doesn't like

We had special-cased use of atomic_add_fetch, because clang just
does not like that builtin. Just use atomic_fetch_add instead.

Signed-off-by: Mark Stapp <mjs@voltanet.io>
---

diff --git a/zebra/zebra_dplane.c b/zebra/zebra_dplane.c
index eb3d48d784..b9163cfaab 100644
--- a/zebra/zebra_dplane.c
+++ b/zebra/zebra_dplane.c
@@ -2158,17 +2158,12 @@ static int dplane_update_enqueue(struct zebra_dplane_ctx *ctx)
 	}
 	DPLANE_UNLOCK();
 
-	curr = atomic_add_fetch_explicit(
-#ifdef __clang__
-		/* TODO -- issue with the clang atomic/intrinsics currently;
-		 * casting away the 'Atomic'-ness of the variable works.
-		 */
-		(uint32_t *)&(zdplane_info.dg_routes_queued),
-#else
+	curr = atomic_fetch_add_explicit(
 		&(zdplane_info.dg_routes_queued),
-#endif
 		1, memory_order_seq_cst);
 
+	curr++;	/* We got the pre-incremented value */
+
 	/* Maybe update high-water counter also */
 	high = atomic_load_explicit(&zdplane_info.dg_routes_queued_max,
 				    memory_order_seq_cst);