Christopher S. Aker wrote:
I just rolled a 2.6.11-rc4+cfq-ts-21 host kernel and I'll be testing this today.


2.6.11 doesn't seem to be ready for host duty yet:

Yes. There is a bug in the v8-rc2 patch for 2.6.11
Try the attached fix, please.

      Bodo


To BlaisorBlade: I tested the v8-rc2 for 2.6.9, too. It works fine for me.
From: Bodo Stroesser <[EMAIL PROTECTED]>

This patch fixes a bug in SKAS3 host patch v8-rc2 for 2.6.11,
that causes syscalls not to be intercepted, when PTRACE_SYSEMU
is used.

Signed-off-by: Bodo Stroesser <[EMAIL PROTECTED]>
---

--- linux-2.6.11-rc4/arch/i386/kernel/ptrace.c.orig	2005-02-15 15:33:26.000000000 +0100
+++ linux-2.6.11-rc4/arch/i386/kernel/ptrace.c	2005-02-15 15:58:18.000000000 +0100
@@ -802,7 +802,7 @@ int do_syscall_trace(struct pt_regs *reg
 	if (is_singlestep)
 		send_sigtrap(current, regs, 0);
 
-	if (!test_thread_flag(TIF_SYSCALL_TRACE))
+	if (!test_thread_flag(TIF_SYSCALL_TRACE) && !is_sysemu)
 		return 0;
 
 	/* the 0x80 provides a way for the tracing parent to distinguish

Reply via email to