The first is just to fix a failing test. For mv/i-3, the output file, "out" is empty on FreeBSD 8.1 Also, (and this is telling), it seems that no one is testing on non-linux kernels. On the first one I tried, just before I was planning to release, "make check" reported 30+ failing tests, all having to do with the new copying code.
>From ee74c7d36ddfc3758b2cdec540aab478eb724572 Mon Sep 17 00:00:00 2001 From: Jim Meyering <[email protected]> Date: Fri, 4 Feb 2011 14:59:03 +0100 Subject: [PATCH 1/2] tests: skip mv/i-3 on FreeBSD to avoid spurious failure * tests/mv/i-3: Skip when uname -s reports FreeBSD. --- tests/mv/i-3 | 2 ++ 1 files changed, 2 insertions(+), 0 deletions(-) diff --git a/tests/mv/i-3 b/tests/mv/i-3 index 80db5e8..7fba139 100755 --- a/tests/mv/i-3 +++ b/tests/mv/i-3 @@ -23,6 +23,8 @@ require_controlling_input_terminal_ skip_if_root_ trap '' TTIN # Ignore SIGTTIN +test "$(uname -s)" = FreeBSD && skip_ "known spurious failure on FreeBSD" + touch f g h i || framework_failure chmod 0 g i || framework_failure -- 1.7.4.2.g597a6 >From d5fa424a9c25db98619c47007f548aaef6155fce Mon Sep 17 00:00:00 2001 From: Jim Meyering <[email protected]> Date: Fri, 4 Feb 2011 15:01:39 +0100 Subject: [PATCH 2/2] cp: avoid spurious failure on any non-linux kernel * src/extent-scan.c (extent_scan_read) [!linux]: Always set scan->initial_scan_failed so caller knows not to report the failure. --- src/extent-scan.c | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --git a/src/extent-scan.c b/src/extent-scan.c index 9f38877..1ba59db 100644 --- a/src/extent-scan.c +++ b/src/extent-scan.c @@ -110,6 +110,7 @@ extent_scan_read (struct extent_scan *scan) extern bool extent_scan_read (struct extent_scan *scan ATTRIBUTE_UNUSED) { + scan->initial_scan_failed = true; errno = ENOTSUP; return false; } -- 1.7.4.2.g597a6
