This is the mail archive of the libc-alpha@sourceware.org mailing list for the glibc project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Fix p_secstodate overflow handling (bug 22463)


The resolv/res_debug.c function p_secstodate does:

        struct tm timebuf;
        time = __gmtime_r(&clock, &timebuf);
        time->tm_year += 1900;
        time->tm_mon += 1;
        sprintf(output, "%04d%02d%02d%02d%02d%02d",
                time->tm_year, time->tm_mon, time->tm_mday,
                time->tm_hour, time->tm_min, time->tm_sec);

If __gmtime_r returns NULL (because the year overflows the range of
int), this will dereference a null pointer.  Otherwise, if the
computed year does not fit in four characters, this will cause a
buffer overrun of the fixed-size 15-byte buffer.  With current GCC
mainline, there is a compilation failure because of the possible
buffer overrun.

I couldn't find a specification for how this function is meant to
behave.  This patch makes the overflow cases use <overflow> as the
output string and set errno to EOVERFLOW (__gmtime_r will already have
done so in the case of NULL return).  I couldn't persuade GCC to
recognize the ranges of the struct tm fields by adding explicit range
checks with a call to __builtin_unreachable if outside the range, so
having added a check for the year, this patch then disables the
-Wformat-overflow= warning for the sprintf call.

I do not know why this build failure (which I observed when testing a
fix for the regcomp.c build failure) should be new with mainline GCC
(that is, I don't know what GCC change might have introduced it, when
the basic functionality for such warnings was already in GCC 7).

I do not know if this is a security issue (that is, if there are
plausible ways in which a date before -999 or after 9999 from an
untrusted source might end up in this function).  The system clock is
arguably an untrusted source (in that e.g. NTP is insecure), but
probably not to that extent (NTP can't communicate such wild
timestamps).

Tested with build-many-glibcs.py that, in conjunction with a fix for
regcomp.c compilation, this restores the build for arm with mainline
GCC.  Also tested for x86_64.

2017-11-21  Joseph Myers  <joseph@codesourcery.com>

	[BZ #22463]
	* resolv/res_debug.c: Include <libc-diag.h>.
	(p_secstodate): Use "<overflow>" as output and set errno to
	EOVERFLOW if year would not fit in four characters.
	(p_secstodate) [__GNUC_PREREQ (7, 0)]: Disable -Wformat-overflow=
	for sprintf call.

diff --git a/resolv/res_debug.c b/resolv/res_debug.c
index 4114c2d..8a9ac77 100644
--- a/resolv/res_debug.c
+++ b/resolv/res_debug.c
@@ -107,6 +107,7 @@
 #include <string.h>
 #include <time.h>
 #include <shlib-compat.h>
+#include <libc-diag.h>
 
 #ifdef SPRINTF_CHAR
 # define SPRINTF(x) strlen(sprintf/**/x)
@@ -1064,11 +1065,28 @@ p_secstodate (u_long secs) {
 
 	struct tm timebuf;
 	time = __gmtime_r(&clock, &timebuf);
+	if (time == NULL
+	    || time->tm_year > 9999 - 1900
+	    || time->tm_year < -999 - 1900) {
+		strcpy (output, "<overflow>");
+		__set_errno (EOVERFLOW);
+		return output;
+	}
 	time->tm_year += 1900;
 	time->tm_mon += 1;
+	/* The struct tm fields, given the above check of tm_year,
+	   must have values that mean this sprintf exactly fills the
+	   buffer.  But as of GCC 8 of 2017-11-21, GCC cannot tell
+	   that, even given range checks on all fields with
+	   __builtin_unreachable called for out-of-range values.  */
+	DIAG_PUSH_NEEDS_COMMENT;
+#if __GNUC_PREREQ (7, 0)
+	DIAG_IGNORE_NEEDS_COMMENT (8, "-Wformat-overflow=");
+#endif
 	sprintf(output, "%04d%02d%02d%02d%02d%02d",
 		time->tm_year, time->tm_mon, time->tm_mday,
 		time->tm_hour, time->tm_min, time->tm_sec);
+	DIAG_POP_NEEDS_COMMENT;
 	return (output);
 }
 libresolv_hidden_def (__p_secstodate)

-- 
Joseph S. Myers
joseph@codesourcery.com


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]