copy_siginfo_to_user correctly strips off the high bits of si->si_code
before copying it to userspace.
Unfortunately, ptrace32_siginfo calls copy_siginfo_to_user to put the
siginfo into a temporary userspace buffer and counts on the high bits
still being present so that it may initialize the union in the final
32-bit siginfo.
This patch works around this problem by having copy_siginfo_to_user
copy si->si_code intact to userspace. This introduces a bug in native
64-bit PTRACE_GETSIGINFO, but I haven't noticed anything being
bothered by it.
diff --git a/kernel/signal.c b/kernel/signal.c
index 1e067a1..28813b1 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -2088,7 +2088,7 @@ int copy_siginfo_to_user(siginfo_t __user *to, siginfo_t
*from)
*/
err = __put_user(from->si_signo, &to->si_signo);
err |= __put_user(from->si_errno, &to->si_errno);
- err |= __put_user((short)from->si_code, &to->si_code);
+ err |= __put_user(from->si_code, &to->si_code);
switch (from->si_code & __SI_MASK) {
case __SI_KILL:
err |= __put_user(from->si_pid, &to->si_pid);
-------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace
_______________________________________________
User-mode-linux-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel
