[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: fprintf-posix on macOS
From: |
Bruno Haible |
Subject: |
Re: fprintf-posix on macOS |
Date: |
Thu, 03 Dec 2020 21:59:22 +0100 |
User-agent: |
KMail/5.1.3 (Linux/4.4.0-193-generic; KDE/5.18.0; x86_64; ; ) |
Martin Storsjö wrote:
> FAIL: test-fprintf-posix3.sh
This patch fixes it. It now produces a diagnostic
"Skipping test: cannot trust address space size on this platform"
2020-12-03 Bruno Haible <bruno@clisp.org>
fprintf-posix-tests: Avoid a test failure on macOS 10.13.
Reported by Martin Storsjö <martin@martin.st> in
<https://lists.gnu.org/archive/html/bug-gnulib/2020-12/msg00003.html>.
* tests/test-fprintf-posix3.c: Skip the test on macOS.
(main): Return a different exit code at each point. Allow 100 KB extra
memory consumption.
* tests/test-fprintf-posix3.sh: Update. Remove the "get_rusage_as()
doesn't work" diagnostic.
diff --git a/tests/test-fprintf-posix3.c b/tests/test-fprintf-posix3.c
index 67bcb92..5521839 100644
--- a/tests/test-fprintf-posix3.c
+++ b/tests/test-fprintf-posix3.c
@@ -20,6 +20,24 @@
#include <stdio.h>
+#if defined __APPLE__ && defined __MACH__ /* macOS
*/
+
+/* On macOS 10.13, this test fails, because the address space size increases
+ by 10 MB to 42 MB during the test's execution. But it's not a malloc
+ leak, as can be seen by running the 'leaks' program. And it does not fail
+ if the test's output is redirected to /dev/null. Probably piping a lot
+ of output to stdout, when not redirected to /dev/null, allocates
intermediate
+ buffers in the virtual address space. */
+
+int
+main ()
+{
+ fprintf (stderr, "Skipping test: cannot trust address space size on this
platform\n");
+ return 78;
+}
+
+#else
+
#include <stdlib.h>
#include <string.h>
#include <errno.h>
@@ -84,7 +102,7 @@ main (int argc, char *argv[])
if (memory == NULL)
return 1;
memset (memory, 17, MAX_ALLOC_TOTAL);
- result = 78;
+ result = 80;
}
else
{
@@ -98,14 +116,16 @@ main (int argc, char *argv[])
but should not result in a permanent memory allocation. */
if (fprintf (stdout, "%011000d\n", 17) == -1
&& errno == ENOMEM)
- return 1;
+ return 2;
}
result = 0;
}
- if (get_rusage_as () > initial_rusage_as + MAX_ALLOC_TOTAL)
- return 1;
+ if (get_rusage_as () > initial_rusage_as + MAX_ALLOC_TOTAL + 100000)
+ return 3;
return result;
}
+
+#endif /* !macOS */
diff --git a/tests/test-fprintf-posix3.sh b/tests/test-fprintf-posix3.sh
index 351d375..96609c7 100755
--- a/tests/test-fprintf-posix3.sh
+++ b/tests/test-fprintf-posix3.sh
@@ -4,7 +4,7 @@
(${CHECKER} ./test-fprintf-posix3${EXEEXT} 0
result=$?
- if test $result != 77 && test $result != 78; then result=1; fi
+ if test $result != 77 && test $result != 78 && test $result != 80; then
result=1; fi
exit $result
) 2>/dev/null
malloc_result=$?
@@ -12,6 +12,10 @@ if test $malloc_result = 77; then
echo "Skipping test: no way to determine address space size"
exit 77
fi
+if test $malloc_result = 78; then
+ echo "Skipping test: cannot trust address space size on this platform"
+ exit 77
+fi
${CHECKER} ./test-fprintf-posix3${EXEEXT} 1 > /dev/null
result=$?
@@ -23,9 +27,4 @@ if test $result != 0; then
exit 1
fi
-if test $malloc_result = 78; then
- echo "Skipping test: get_rusage_as() doesn't work"
- exit 77
-fi
-
exit 0
- Issues with posix functions on modern macOS/Xcode, Martin Storsjö, 2020/12/01
- Re: Issues with posix functions on modern macOS/Xcode, Martin Storsjö, 2020/12/01
- Re: Issues with posix functions on modern macOS/Xcode, Bruno Haible, 2020/12/01
- Re: Issues with posix functions on modern macOS/Xcode, Bruno Haible, 2020/12/02
- Re: strsignal on macOS, Bruno Haible, 2020/12/02
- Re: fprintf-posix on macOS,
Bruno Haible <=
- Re: utime on macOS, Bruno Haible, 2020/12/04