https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=206128
--- Comment #11 from Phil Shafer <p...@freebsd.org> ---
Was finally able to reproduce this and have a fix. I was using the wrong value
when checking for available buffer space:
diff --git a/libxo/libxo.c b/libxo/libxo.c
index 9df67de..7a714b1 100644
--- a/libxo/libxo.c
+++ b/libxo/libxo.c
@@ -2816,7 +2816,7 @@ xo_format_string (xo_handle_t *xop, xo_buffer_t *xbp,
xo_xff_flags_t flags,
* but if we did the work ourselves, then we need to do it.
*/
int delta = xfp->xf_width[XF_WIDTH_MIN] - cols;
- if (!xo_buf_has_room(xbp, delta))
+ if (!xo_buf_has_room(xbp, xfp->xf_width[XF_WIDTH_MIN]))
goto bail;
/*
In addition, I'm adding a high-water mark that will trigger a flush:
@@ -6128,6 +6128,12 @@ xo_do_emit_fields (xo_handle_t *xop, xo_field_info_t
*fields,
XOIF_CLEAR(xop, XOIF_REORDER);
+ /*
+ * If we've got enough data, flush it.
+ */
+ if (xo_buf_offset(&xop->xo_data) > XO_BUF_HIGH_WATER)
+ flush = 1;
+
/* If we don't have an anchor, write the text out */
if (flush && !XOIF_ISSET(xop, XOIF_ANCHOR)) {
if (xo_write(xop) < 0)
I'll roll a new version of libxo tomorrow.
Apologies for taking so long with this, but it was mostly just an issue of
reproducing it. I wasn't seeing enough churn in my "vmstat -z" value to
trigger the issue, so I munged one of the regression tests to help.
Thanks,
Phil
--
You are receiving this mail because:
You are the assignee for the bug.
_______________________________________________
freebsd-bugs@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-bugs
To unsubscribe, send any mail to "freebsd-bugs-unsubscr...@freebsd.org"
