On Sun, Dec 3, 2023 at 7:58 AM Bruno Haible wrote:
>
> In a testdir of module 'argp', built with clang 17 and -fsanitize=address,
> I see this test failure:
>
> $ cat test-argp-2.sh.log
> --- expected2023-12-03 12:52:58.892318895 +0100
> +++ out 2023-12-03 12:52:58.904318928 +0100
> @@ -27,9 +
On Sun, Dec 03, 2023 at 11:03:30AM +0100, Bruno Haible wrote:
> There is no guarantee, but the malloc() result has an alignment of at least 4
> on all of today's platforms. [1]
It may be legitimate to have such assumptions. But it is likewise
legitimate that compilers and the associated libraries
In a testdir of module 'argp', built with clang 17 and -fsanitize=address,
I see this test failure:
$ cat test-argp-2.sh.log
--- expected2023-12-03 12:52:58.892318895 +0100
+++ out 2023-12-03 12:52:58.904318928 +0100
@@ -27,9 +27,9 @@
-l, --limerick create a limerick
-p, --p
Hi Paul,
In a testdir of the 'mcel' module, compiled with "gcc -fsanitize=address"
(and CFLAGS="-O0 -fno-omit-frame-pointer -ggdb" and
ASAN_OPTIONS="detect_leaks=0 abort_on_error=1"), I see this test failure:
ERROR: AddressSanitizer: stack-buffer-overflow on address 0x7ffc6de32ad3 at pc
0x558658
Andreas F. Borchert wrote:
> > foo *p = (foo *) m; /* CAST */
> ...
> This line is not guaranteed to be ok due to § 6.3.2.3, (7):
>
>A pointer to an object type may be converted to a pointer to a different
>object type. If the resulting pointer is not correctly aligned for the
>r
> foo.c
>
> #include
> #include
> #ifdef __GNUC__
> # define unreachable() __builtin_unreachable ()
> #endif
> typedef struct { int a,b,c,d,e,f,g,h,i,j,k,l,m,n,o,p,q,r,s,t; } foo;
> int main ()
> {
> char *m = malloc (1)
Am So., 3. Dez. 2023 um 10:05 Uhr schrieb Paul Eggert :
> On 2023-12-02 01:04, Bruno Haible wrote:
> > On the contrary, I will try to find and eliminate such alarms.
>
> Please don't complicate and/or slow down shared Gnulib code just to
> pacify this particular false alarm from Clang. The obstack
On 2023-12-02 01:04, Bruno Haible wrote:
On the contrary, I will try to find and eliminate such alarms.
Please don't complicate and/or slow down shared Gnulib code just to
pacify this particular false alarm from Clang. The obstack fix was fine,
because it made obstack clearer and no slower. B
