[igt-dev] [PATCH i-g-t 1/1] runner: Flush igt_runner outputs before forking

* [igt-dev] [PATCH i-g-t 1/1] runner: Flush igt_runner outputs before forking
@ 2018-09-19 11:49 Petri Latvala
  2018-09-19 13:23 ` [igt-dev] ✓ Fi.CI.BAT: success for series starting with [i-g-t,1/1] " Patchwork
                   ` (2 more replies)
  0 siblings, 3 replies; 4+ messages in thread
From: Petri Latvala @ 2018-09-19 11:49 UTC (permalink / raw)
  To: igt-dev; +Cc: Tomi Sarvela

If the output of igt_runner is piped or redirected, buffered prints
could be left lingering and read as test executable output if execv()
fails. This can happen easily if CI for example generates a testlist
with an incorrect binary name, or an optional test binary (say,
kms_chamelium) is not built for the deployment in question.

Signed-off-by: Petri Latvala <petri.latvala@intel.com>
Cc: Tomi Sarvela <tomi.p.sarvela@intel.com>
Cc: Arkadiusz Hiler <arkadiusz.hiler@intel.com>
---
 runner/executor.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/runner/executor.c b/runner/executor.c
index 701ca80d..fd262eb4 100644
--- a/runner/executor.c
+++ b/runner/executor.c
@@ -780,6 +780,14 @@ static int execute_entry(size_t idx,
 		printf("\n");
 	}
 
+	/*
+	 * Flush outputs before forking so our (buffered) output won't
+	 * end up in the test outputs.
+	 */
+
+	fflush(stdout);
+	fflush(stderr);
+
 	if ((child = fork())) {
 		int outfd = outpipe[0];
 		int errfd = errpipe[0];
-- 
2.18.0

_______________________________________________
igt-dev mailing list
igt-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/igt-dev

^ permalink raw reply related	[flat|nested] 4+ messages in thread