From: Colin Walters <walters@verbum.org>
Date: Fri, 4 Aug 2023 02:49:48 +0000 (-0400)
Subject: deploy: Fix mutex locking for global sync timeout
X-Git-Tag: archive/raspbian/2023.7-3+rpi1~1^2~9^2^2~28^2
X-Git-Url: https://dgit.raspbian.org/?a=commitdiff_plain;h=402e04280b54c058ad47be99fe6a9326caf2ae00;p=ostree.git

deploy: Fix mutex locking for global sync timeout

The locking here was always too long - by holding the mutex
during the `sync()` call, it means `g_cond_wait_until()` can
never wake up (because its API requires the mutex to be locked).

Confusingly though of course we do still print the "timed out"
message, and I think that tricked us when we were doing testing
here.

We only need to lock the mutex when we're manipulating shared
state, which basically boils down to the `gboolean success`.
---

diff --git a/src/libostree/ostree-sysroot-deploy.c b/src/libostree/ostree-sysroot-deploy.c
index 974d2336..91bccdc0 100644
--- a/src/libostree/ostree-sysroot-deploy.c
+++ b/src/libostree/ostree-sysroot-deploy.c
@@ -1606,11 +1606,10 @@ static void *
 sync_in_thread (void *ptr)
 {
   SyncData *syncdata = ptr;
-  // Ensure that the caller is blocked waiting
-  g_mutex_lock (&syncdata->mutex);
   ot_journal_print (LOG_INFO, "Starting global sync()");
   sync ();
   ot_journal_print (LOG_INFO, "Completed global sync()");
+  g_mutex_lock (&syncdata->mutex);
   // Signal success
   syncdata->success = true;
   g_cond_broadcast (&syncdata->cond);