BUG/MINOR: cli/fd: don't isolate the thread for each individual fd
It takes ages to proceed with "show fd" when there is sustained activity
because it uses the rendez-vous point for each and every file descriptor
in the loop. It's very common to see socat timeout there.
Instead of doing this, let's just isolate the function when entering the
loop. Its duration is limited by the number of FDs that may be emitted in
a single buffer anyway, so it's much lighter and responds much faster.
diff --git a/src/cli.c b/src/cli.c
index 23caac3..ae419b0 100644
--- a/src/cli.c
+++ b/src/cli.c
@@ -865,12 +865,18 @@
{
struct stream_interface *si = appctx->owner;
int fd = appctx->ctx.cli.i0;
+ int ret = 1;
if (unlikely(si_ic(si)->flags & (CF_WRITE_ERROR|CF_SHUTW)))
- return 1;
+ goto end;
chunk_reset(&trash);
+ /* isolate the threads once per round. We're limited to a buffer worth
+ * of output anyway, it cannot last very long.
+ */
+ thread_isolate();
+
/* we have two inner loops here, one for the proxy, the other one for
* the buffer.
*/
@@ -883,14 +889,10 @@
void *ctx = NULL;
uint32_t conn_flags = 0;
- thread_isolate();
-
fdt = fdtab[fd];
- if (!fdt.owner) {
- thread_release();
+ if (!fdt.owner)
goto skip; // closed
- }
if (fdt.iocb == conn_fd_handler) {
conn_flags = ((struct connection *)fdt.owner)->flags;
@@ -956,24 +958,26 @@
li->bind_conf->frontend->id);
}
- thread_release();
-
chunk_appendf(&trash, "\n");
if (ci_putchk(si_ic(si), &trash) == -1) {
si_rx_room_blk(si);
- return 0;
+ appctx->ctx.cli.i0 = fd;
+ ret = 0;
+ break;
}
skip:
if (appctx->st2 == STAT_ST_END)
break;
fd++;
- appctx->ctx.cli.i0 = fd;
}
+ end:
/* dump complete */
- return 1;
+
+ thread_release();
+ return ret;
}
/* This function dumps some activity counters used by developers and support to