Avoid false-positive error in gnulib tests 'test-perror2' and 'test-strerror_r'.

On openSUSE OBS, the above gnulib tests fail on armv7l.

Corresponding report on the gnulib mailing list:

    https://lists.gnu.org/r/bug-gnulib/2020-08/msg00220.html

    From:	Florian Weimer
    Date:	Thu, 27 Aug 2020 09:41:34 +0200
    Subject:	Use-after-free in test-perror2, test-strerror_r

    The problem is visible with glibc 2.32 under valgrind:

    ==20== Invalid read of size 1
    ==20==    at 0x483DAB4: strcmp (vg_replace_strmem.c:847)
    ==20==    by 0x109414: main (test-perror2.c:84)
    ==20==  Address 0x4a1a3d0 is 0 bytes inside a block of size 17 free'd
    ==20==    at 0x483A9F5: free (vg_replace_malloc.c:538)
    ==20==    by 0x48E2134: strerror_l (in /usr/lib64/libc-2.32.so)
    ==20==    by 0x109328: main (test-perror2.c:72)
    ==20==  Block was alloc'd at
    ==20==    at 0x4839809: malloc (vg_replace_malloc.c:307)
    ==20==    by 0x48CA03F: __vasprintf_internal (in /usr/lib64/libc-2.32.so)
    ==20==    by 0x48A46F9: asprintf (in /usr/lib64/libc-2.32.so)
    ==20==    by 0x48E2184: strerror_l (in /usr/lib64/libc-2.32.so)
    ==20==    by 0x1092E2: main (test-perror2.c:67)
    ==20==
    ==20== Invalid read of size 1
    ==20==    at 0x483DAC8: strcmp (vg_replace_strmem.c:847)
    ==20==    by 0x109414: main (test-perror2.c:84)
    ==20==  Address 0x4a1a3d1 is 1 bytes inside a block of size 17 free'd
    ==20==    at 0x483A9F5: free (vg_replace_malloc.c:538)
    ==20==    by 0x48E2134: strerror_l (in /usr/lib64/libc-2.32.so)
    ==20==    by 0x109328: main (test-perror2.c:72)
    ==20==  Block was alloc'd at
    ==20==    at 0x4839809: malloc (vg_replace_malloc.c:307)
    ==20==    by 0x48CA03F: __vasprintf_internal (in /usr/lib64/libc-2.32.so)
    ==20==    by 0x48A46F9: asprintf (in /usr/lib64/libc-2.32.so)
    ==20==    by 0x48E2184: strerror_l (in /usr/lib64/libc-2.32.so)
    ==20==    by 0x1092E2: main (test-perror2.c:67)

    I think it's the test that's invalid.

    This was reported as an actual grep test failure (without valgrind) on
    32-bit Arm, where glibc malloc happens to return a different buffer
    address for the internal allocation (so that msg3 != msg4).

    test-strerror_r has the same issue.

    Thanks,
    Florian

Upstream patch:

    https://git.sv.gnu.org/cgit/gnulib.git/commit/?id=175e0bc72808

    From 175e0bc72808d564074c4adcc72aeadb74adfcc6 Mon Sep 17 00:00:00 2001
    From: Paul Eggert <eggert@cs.ucla.edu>
    Date: Thu, 27 Aug 2020 17:52:58 -0700
    Subject: [PATCH] perror, strerror_r: remove unportable tests

    Problem reported by Florian Weimer in:
    https://lists.gnu.org/r/bug-gnulib/2020-08/msg00220.html
    * tests/test-perror2.c (main):
    * tests/test-strerror_r.c (main): Omit unportable tests.

This downstream patch is identical to upstream one modulo the ChangeLog entry.
---
 gnulib-tests/test-perror2.c    |    3 ---
 gnulib-tests/test-strerror_r.c |    3 ---
 2 files changed, 6 deletions(-)

Index: gnulib-tests/test-perror2.c
===================================================================
--- gnulib-tests/test-perror2.c.orig
+++ gnulib-tests/test-perror2.c
@@ -79,9 +79,6 @@ main (void)
     errno = -5;
     perror ("");
     ASSERT (!ferror (stderr));
-    ASSERT (msg1 == msg2 || msg1 == msg4 || STREQ (msg1, str1));
-    ASSERT (msg2 == msg4 || STREQ (msg2, str2));
-    ASSERT (msg3 == msg4 || STREQ (msg3, str3));
     ASSERT (STREQ (msg4, str4));
 
     free (str1);
Index: gnulib-tests/test-strerror_r.c
===================================================================
--- gnulib-tests/test-strerror_r.c.orig
+++ gnulib-tests/test-strerror_r.c
@@ -165,9 +165,6 @@ main (void)
 
     strerror_r (EACCES, buf, sizeof buf);
     strerror_r (-5, buf, sizeof buf);
-    ASSERT (msg1 == msg2 || msg1 == msg4 || STREQ (msg1, str1));
-    ASSERT (msg2 == msg4 || STREQ (msg2, str2));
-    ASSERT (msg3 == msg4 || STREQ (msg3, str3));
     ASSERT (STREQ (msg4, str4));
 
     free (str1);