[PATCH elfutils 0/2] fix two /proc/pid/maps inode parsing issues

[Yonghong Song]

The inode number in /proc/pid/maps has type "unsigned long".
The current parsing in libdwfl/linux-proc-maps.c and
tests/backtrace-data.c is not correct and it triggered
the following four test failures in one x64 box.
   FAIL: dwfl-bug-fd-leak 
   FAIL: run-backtrace-data.sh
   FAIL: run-backtrace-dwarf.sh
   FAIL: vdsosyms

The offending map entry
  7f269223d000-7f269226b000 r-xp  00:50 10224326387095067468   
/home/...
has an inode number beyond the valid range of type "long".

This patch set fixed the problem in libdwfl and tests respectively.

Yonghong Song (2):
  [libdwfl] parse inode in /proc/pid/maps correctly
  [tests] parse inode in /proc/pid/maps correctly in
run-backtrace-data.sh

 libdwfl/linux-proc-maps.c | 2 +-
 tests/backtrace-data.c| 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

-- 
2.17.1

[PATCH elfutils 2/2] [tests] parse inode in /proc/pid/maps correctly in run-backtrace-data.sh

[Yonghong Song]

The backtrace-data.c parsed the inode in /proc/pid/maps with
format "%*x".
This caused failure if inode is big. For example,
  7f269223d000-7f269226b000 r-xp  00:50 10224326387095067468   
/home/...

The correct format should be "%*lu" to reflect inode "unsigned long" type.
But that caused the following compilation error.
  acktrace-data.c: In function ‘maps_lookup’:
  backtrace-data.c:109:22: error: use of assignment suppression and length 
modifier
  together in gnu_scanf format [-Werror=format=]
   i = fscanf (f, "%lx-%lx %*s %lx %*x:%*x %*lu", &start, &end, &offset);

Fix the test with inode format string "%*s" then.

Signed-off-by: Yonghong Song 
---
 tests/backtrace-data.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tests/backtrace-data.c b/tests/backtrace-data.c
index 3a91c664..85ae9729 100644
--- a/tests/backtrace-data.c
+++ b/tests/backtrace-data.c
@@ -106,7 +106,7 @@ maps_lookup (pid_t pid, Dwarf_Addr addr, GElf_Addr *basep)
 {
   // 37e3c22000-37e3c23000 rw-p 00022000 00:11 49532 /lib64/ld-2.14.90.so 
*/
   unsigned long start, end, offset;
-  i = fscanf (f, "%lx-%lx %*s %lx %*x:%*x %*x", &start, &end, &offset);
+  i = fscanf (f, "%lx-%lx %*s %lx %*x:%*x %*s", &start, &end, &offset);
   assert (errno == 0);
   if (i != 3)
   break;
-- 
2.17.1

[PATCH elfutils 1/2] [libdwfl] parse inode in /proc/pid/maps correctly

[Yonghong Song]

The inode number in /proc/pid/maps is displayed as "unsigned long"
type.

In one of our x64 system, we have inode number exceeding valid "long"
type range, which caused the following test failure:
   FAIL: dwfl-bug-fd-leak
   FAIL: run-backtrace-dwarf.sh
   FAIL: vdsosyms

The offending map entry:
  7f269246b000-7f269246c000 rw-p 0002e000 00:50 10224326387095067468 /home/...

This patch changed sscanf inode number type from PRIi64 to PRIu64
and fixed the problem.

Signed-off-by: Yonghong Song 
---
 libdwfl/linux-proc-maps.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/libdwfl/linux-proc-maps.c b/libdwfl/linux-proc-maps.c
index c4438c0f..719cba68 100644
--- a/libdwfl/linux-proc-maps.c
+++ b/libdwfl/linux-proc-maps.c
@@ -217,7 +217,7 @@ proc_maps_report (Dwfl *dwfl, FILE *f, GElf_Addr 
sysinfo_ehdr, pid_t pid)
   uint64_t ino;
   int nread = -1;
   if (sscanf (line, "%" PRIx64 "-%" PRIx64 " %*s %" PRIx64
- " %x:%x %" PRIi64 " %n",
+ " %x:%x %" PRIu64 " %n",
  &start, &end, &offset, &dmajor, &dminor, &ino, &nread) < 6
  || nread <= 0)
{
-- 
2.17.1