[LTP] [PATCH v2] mmap04.c: Avoid vma merging

Avinesh Kumar akumar@suse.de
Thu Jan 25 09:14:42 CET 2024


Hi Petr,

On Wednesday, January 24, 2024 6:05:47 PM CET Petr Vorel wrote:
> Hi Avinesh, Martin,
> 
> > Hi,
> > Reviewed-by: Martin Doucha <mdoucha@suse.cz>
> > 
> > On 24. 01. 24 14:26, Avinesh Kumar wrote:
> > > We hit a scenario where new mapping was merged with existing mapping of
> > > same permission and the return address from the mmap was hidden in the
> > > merged mapping in /proc/self/maps, causing the test to fail.
> > > To avoid this, we first  create a 2-page mapping with the different
> > > permissions, and then remap the 2nd page with the perms being tested.
> > > 
> > > Signed-off-by: Avinesh Kumar <akumar@suse.de>
> > > Reported-by: Martin Doucha <mdoucha@suse.cz>
> > > Signed-off-by: Avinesh Kumar <akumar@suse.de>
> > > ---
> > > 
> > >   testcases/kernel/syscalls/mmap/mmap04.c | 23 ++++++++++++++++-------
> > >   1 file changed, 16 insertions(+), 7 deletions(-)
> > > 
> > > diff --git a/testcases/kernel/syscalls/mmap/mmap04.c
> > > b/testcases/kernel/syscalls/mmap/mmap04.c index f6f4f7c98..fa85deed1
> > > 100644
> > > --- a/testcases/kernel/syscalls/mmap/mmap04.c
> > > +++ b/testcases/kernel/syscalls/mmap/mmap04.c
> > > @@ -17,8 +17,8 @@
> > > 
> > >   #include "tst_test.h"
> > >   #include <stdio.h>
> > > 
> > > -#define MMAPSIZE 1024
> > > -static char *addr;
> > > +static char *addr1;
> > > +static char *addr2;
> > > 
> > >   static struct tcase {
> > >   
> > >   	int prot;
> > > 
> > > @@ -44,14 +44,23 @@ static struct tcase {
> > > 
> > >   static void run(unsigned int i)
> > >   {
> > >   
> > >   	struct tcase *tc = &tcases[i];
> > > 
> > > -	char addr_str[20];
> > > 
> > >   	char perms[8];
> > >   	char fmt[1024];
> > > 
> > > +	unsigned int pagesize;
> > > +	int flag;
> > > -	addr = SAFE_MMAP(NULL, MMAPSIZE, tc->prot, tc->flags, -1, 0);
> > > +	pagesize = SAFE_SYSCONF(_SC_PAGESIZE);
> > > -	sprintf(addr_str, "%" PRIxPTR, (uintptr_t)addr);
> > > -	sprintf(fmt, "%s-%%*x %%s", addr_str);
> > > +	/* To avoid new mapping getting merged with existing mappings, we
> > > first
> > > +	 * create a 2-page mapping with the different permissions, and then
> > > remap
> > > +	 * the 2nd page with the perms being tested.
> > > +	 */
> > > +	flag = (tc->flags & MAP_PRIVATE) ? MAP_SHARED : MAP_PRIVATE;
> > > +	addr1 = SAFE_MMAP(NULL, pagesize * 2, PROT_NONE, MAP_ANONYMOUS | 
flag,
> > > -1, 0); +
> > > +	addr2 = SAFE_MMAP(addr1 + pagesize, pagesize, tc->prot, tc->flags |
> > > MAP_FIXED, -1, 0); +
> > > +	sprintf(fmt, "%" PRIxPTR "-%%*x %%s", (uintptr_t)addr2);
> > > 
> > >   	SAFE_FILE_LINES_SCANF("/proc/self/maps", fmt, perms);
> > >   	if (!strcmp(perms, tc->exp_perms)) {
> > > 
> > > @@ -61,7 +70,7 @@ static void run(unsigned int i)
> > > 
> > >   						tc-
>exp_perms, perms);
> > >   	
> > >   	}
> > > 
> > > -	SAFE_MUNMAP(addr, MMAPSIZE);
> > > +	SAFE_MUNMAP(addr1, pagesize * 2);
> 
> Shouldn't there be also second munmap()?
> SAFE_MUNMAP(addr2, pagesize);
No, we are unmapping both the mappings ( 2 pages ) together.

Regards,
Avinesh

> 
> Kind regards,
> Petr
> 
> > >   }
> > >   static struct tst_test test = {






More information about the ltp mailing list