[PATCH 4/7] tracing: Don't inc err_log entry count if entry allocation fails
From: Tom Zanussi
Date: Thu Jan 27 2022 - 16:44:41 EST
tr->n_err_log_entries should only be increased if entry allocation
succeeds.
Doing it when it fails won't cause any problems other than wasting an
entry, but should be fixed anyway.
Signed-off-by: Tom Zanussi <zanussi@xxxxxxxxxx>
---
kernel/trace/trace.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index a73d78dcda2c..addd8297874e 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
@@ -7740,7 +7740,8 @@ static struct tracing_log_err *get_tracing_log_err(struct trace_array *tr)
err = kzalloc(sizeof(*err), GFP_KERNEL);
if (!err)
err = ERR_PTR(-ENOMEM);
- tr->n_err_log_entries++;
+ else
+ tr->n_err_log_entries++;
return err;
}
--
2.17.1