From: Matthew Fernandez <matthew.fernandez@gmail.com>
Date: Fri, 21 May 2021 02:34:36 +0000 (-0700)
Subject: optimize two calls to gvprintf with single characters
X-Git-Tag: 2.47.3~31^2~2
X-Git-Url: https://granicus.if.org/sourcecode?a=commitdiff_plain;h=c54470e710964d566bce944befcbd048c0581567;p=graphviz

optimize two calls to gvprintf with single characters

The expression `gvprintf(j, "%c", x)` is equivalent to `gvwrite(j, &x, 1)`.
However, it seems modern compilers, even with link-time optimization enabled,
are not clever enough to see this equivalence. By unraveling the gvprintf call
to what it eventually bottoms out to, we can accelerate SVG generation.

On tests/regression_tests/large/long_chain, this drops the number of gvprintf
calls from 297008 to 165008, reducing the amount of the trace for which gvprintf
is responsible from 3.27% to 2.24%. Total executed instructions are reduced from
8160974807 to 8098098396, a speed up of ~1%.
---

diff --git a/plugin/core/gvrender_core_svg.c b/plugin/core/gvrender_core_svg.c
index 8a6c4989b..a852b4db1 100644
--- a/plugin/core/gvrender_core_svg.c
+++ b/plugin/core/gvrender_core_svg.c
@@ -59,7 +59,7 @@ static void svg_bzptarray(GVJ_t * job, pointf * A, int n)
     if (A[0].x <= A[n-1].x) {
 #endif
 	for (i = 0; i < n; i++) {
-	    gvprintf(job, "%c", c);
+	    gvwrite(job, &c, 1);
             gvprintdouble(job, A[i].x);
             gvputs(job, ",");
             gvprintdouble(job, -A[i].y);
@@ -71,7 +71,7 @@ static void svg_bzptarray(GVJ_t * job, pointf * A, int n)
 #if EDGEALIGN
     } else {
 	for (i = n-1; i >= 0; i--) {
-	    gvprintf(job, "%c", c);
+	    gvwrite(job, &c, 1);
             gvprintdouble(job, A[i].x);
             gvputs(job, ",");
             gvprintdouble(job, -A[i].y);