calibrate: use div_u64_rem for old compiler compat

Under CERN CO current standard compiler gcc (GCC) 4.1.2 20080704 (Red Hat 4.1.2-52) a long, 64 bit division is generated even if the divisor is a power of 2 (sigh). So we need yet another div_u64_rem.

calibrate: use div_u64_rem for old compiler compat
Under CERN CO current standard compiler gcc (GCC) 4.1.2 20080704 (Red Hat 4.1.2-52) a long, 64 bit division is generated even if the divisor is a power of 2 (sigh). So we need yet another div_u64_rem.
be36dd09 · Juan David González Cobas · 24d7d843 · be36dd09
Commit be36dd09 authored Jul 05, 2012 by Juan David González Cobas
Hide whitespace changes
Inline Side-by-side

Showing with 2 additions and 1 deletion

calibrate.c calibrate.c +2 -1

No files found.
--- a/calibrate.c
+++ b/calibrate.c
@@ -64,6 +64,7 @@ static uint64_t output_delay_ps(struct spec_fd *fd, int ch, int fine, int n,
 	int i;
 	uint64_t *results;
 	uint64_t res, acc = 0;
+	int rem;

 	results = kmalloc(n * sizeof(*results), GFP_KERNEL);
 	if (!results)
@@ -119,7 +120,7 @@ static uint64_t output_delay_ps(struct spec_fd *fd, int ch, int fine, int n,
 	fd_ch_writel(fd, ch, 0, FD_REG_DCR);

 	/* Calculate avg, min max */
-	acc = (acc + n / 2) / n;
+	acc = div_u64_rem((acc + n / 2), n, &rem);
 	if (stats) {
 		stats->avg = acc;
 		stats->min = ~0LL;