Toolchain/binutils-gdb
1
0
Fork 0
forked from Imagelibrary/binutils-gdb
Code Pull Requests Activity

gdb, gdbserver: make target_waitstatus safe

Browse Source 
I stumbled on a bug caused by the fact that a code path read
target_waitstatus::value::sig (expecting it to contain a gdb_signal
value) while target_waitstatus::kind was TARGET_WAITKIND_FORKED.  This
meant that the active union field was in fact
target_waitstatus::value::related_pid, and contained a ptid.  The read
signal value was therefore garbage, and that caused GDB to crash soon
after.  Or, since that GDB was built with ubsan, this nice error
message:

    /home/simark/src/binutils-gdb/gdb/linux-nat.c:1271:12: runtime error: load of value 2686365, which is not a valid value for type 'gdb_signal'

Despite being a large-ish change, I think it would be nice to make
target_waitstatus safe against that kind of bug.  As already done
elsewhere (e.g. dynamic_prop), validate that the type of value read from
the union matches what is supposed to be the active field.

 - Make the kind and value of target_waitstatus private.
 - Make the kind initialized to TARGET_WAITKIND_IGNORE on
   target_waitstatus construction.  This is what most users appear to do
   explicitly.
 - Add setters, one for each kind.  Each setter takes as a parameter the
   data associated to that kind, if any.  This makes it impossible to
   forget to attach the associated data.
 - Add getters, one for each associated data type.  Each getter
   validates that the data type fetched by the user matches the wait
   status kind.
 - Change "integer" to "exit_status", "related_pid" to "child_ptid",
   just because that's more precise terminology.
 - Fix all users.

That last point is semi-mechanical.  There are a lot of obvious changes,
but some less obvious ones.  For example, it's not possible to set the
kind at some point and the associated data later, as some users did.
But in any case, the intent of the code should not change in this patch.

This was tested on x86-64 Linux (unix, native-gdbserver and
native-extended-gdbserver boards).  It was built-tested on x86-64
FreeBSD, NetBSD, MinGW and macOS.  The rest of the changes to native
files was done as a best effort.  If I forgot any place to update in
these files, it should be easy to fix (unless the change happens to
reveal an actual bug).

Change-Id: I0ae967df1ff6e28de78abbe3ac9b4b2ff4ad03b7
This commit is contained in:
Simon Marchi
2021-10-21 16:12:04 -04:00
parent c360a4732b
commit 183be22290
44 changed files with 826 additions and 705 deletions
Download Patch File Download Diff File Expand all files Collapse all files

33
gdb/fbsd-nat.c
View File

@@ -1181,7 +1181,7 @@ fbsd_nat_target::wait (ptid_t ptid, struct target_waitstatus *ourstatus,
}
#endif
wptid = inf_ptrace_target::wait (ptid, ourstatus, target_options);
if (ourstatus->kind == TARGET_WAITKIND_STOPPED)
if (ourstatus->kind () == TARGET_WAITKIND_STOPPED)
{
struct ptrace_lwpinfo pl;
pid_t pid;
@@ -1252,7 +1252,7 @@ fbsd_nat_target::wait (ptid_t ptid, struct target_waitstatus *ourstatus,
pl.pl_lwpid);
add_thread (this, wptid);
}
ourstatus->kind = TARGET_WAITKIND_SPURIOUS;
ourstatus->set_spurious ();
return wptid;
}
#endif
@@ -1263,14 +1263,14 @@ fbsd_nat_target::wait (ptid_t ptid, struct target_waitstatus *ourstatus,
#ifndef PTRACE_VFORK
struct kinfo_proc kp;
#endif
bool is_vfork = false;
ptid_t child_ptid;
pid_t child;
child = pl.pl_child_pid;
ourstatus->kind = TARGET_WAITKIND_FORKED;
#ifdef PTRACE_VFORK
if (pl.pl_flags & PL_FLAG_VFORKED)
ourstatus->kind = TARGET_WAITKIND_VFORKED;
is_vfork = true;
#endif
/* Make sure the other end of the fork is stopped too. */
@@ -1299,12 +1299,16 @@ fbsd_nat_target::wait (ptid_t ptid, struct target_waitstatus *ourstatus,
if (fbsd_fetch_kinfo_proc (child, &kp))
{
if (kp.ki_flag & P_PPWAIT)
ourstatus->kind = TARGET_WAITKIND_VFORKED;
is_vfork = true;
}
else
warning (_("Failed to fetch process information"));
#endif
ourstatus->value.related_pid = child_ptid;
if (is_vfork)
ourstatus->set_vforked (child_ptid);
else
ourstatus->set_forked (child_ptid);
return wptid;
}
@@ -1321,7 +1325,7 @@ fbsd_nat_target::wait (ptid_t ptid, struct target_waitstatus *ourstatus,
#ifdef PTRACE_VFORK
if (pl.pl_flags & PL_FLAG_VFORK_DONE)
{
ourstatus->kind = TARGET_WAITKIND_VFORK_DONE;
ourstatus->set_vfork_done ();
return wptid;
}
#endif
@@ -1329,9 +1333,8 @@ fbsd_nat_target::wait (ptid_t ptid, struct target_waitstatus *ourstatus,
if (pl.pl_flags & PL_FLAG_EXEC)
{
ourstatus->kind = TARGET_WAITKIND_EXECD;
ourstatus->value.execd_pathname
= xstrdup (pid_to_exec_file (pid));
ourstatus->set_execd
(make_unique_xstrdup (pid_to_exec_file (pid)));
return wptid;
}
@@ -1348,7 +1351,7 @@ fbsd_nat_target::wait (ptid_t ptid, struct target_waitstatus *ourstatus,
SIGTRAP, so only treat SIGTRAP events as system call
entry/exit events. */
if (pl.pl_flags & (PL_FLAG_SCE | PL_FLAG_SCX)
&& ourstatus->value.sig == SIGTRAP)
&& ourstatus->sig () == SIGTRAP)
{
#ifdef HAVE_STRUCT_PTRACE_LWPINFO_PL_SYSCALL_CODE
if (catch_syscall_enabled ())
@@ -1356,10 +1359,10 @@ fbsd_nat_target::wait (ptid_t ptid, struct target_waitstatus *ourstatus,
if (catching_syscall_number (pl.pl_syscall_code))
{
if (pl.pl_flags & PL_FLAG_SCE)
ourstatus->kind = TARGET_WAITKIND_SYSCALL_ENTRY;
ourstatus->set_syscall_entry (pl.pl_syscall_code);
else
ourstatus->kind = TARGET_WAITKIND_SYSCALL_RETURN;
ourstatus->value.syscall_number = pl.pl_syscall_code;
ourstatus->set_syscall_return (pl.pl_syscall_code);
return wptid;
}
}
@@ -1490,7 +1493,7 @@ fbsd_nat_target::follow_fork (inferior *child_inf, ptid_t child_ptid,
perror_with_name (("ptrace"));
#ifndef PTRACE_VFORK
if (fork_kind == TARGET_WAITKIND_VFORKED)
if (fork_kind () == TARGET_WAITKIND_VFORKED)
{
/* We can't insert breakpoints until the child process has
finished with the shared memory region. The parent