Skip to contentSkip to conversation

safe_hlt() & NMIs

PATCH Chuck Ebbert 1 participant 1 message Dec 17, 2004

lore
Share & tools
lore
#0 lore raw
Quoted 5 lines · On Tue, 14 Dec 2004 at 15:00:56 -0800 (PST) Linus Torvalds wrote:

Checking for kernel CS also requires checking that it's not vm86 mode, btw. So that's not just a "regs->xcs & 0xffff == __KERNEL_CS" either.

But something like

Quoted 8 lines
      static inline int kernel_mode(struct pt_regs *regs)
      {
              return !((regs->eflags & VM_MASK) | (regs->xcs & 3));
      }

should DTRT.

Can you pls double-check my thinking, and test?

There is already a user_mode() macro in asm-i386/ptrace.h but it's slow.

x86_64's macro is ugly but at least there's no logical-or in it.

Your i386 code is better, so how about applying this patch (boots/runs/is faster
on my tests):

include/asm-i386/ptrace.h 2004-12-17 16:59:39.956099664 -0500 renamed from include/asm-i386/ptrace.h 2004-10-19 15:28:18.000000000 -0400 +5 -1
@@ -55,7 +55,11 @@ struct pt_regs {
 #define PTRACE_SET_THREAD_AREA    26
 
 #ifdef __KERNEL__
-#define user_mode(regs) ((VM_MASK & (regs)->eflags) || (3 & (regs)->xcs))
+static inline int kernel_mode(struct pt_regs *regs)
+{
+       return !((3 & (regs)->xcs) | (VM_MASK & (regs)->eflags));
+}
+#define user_mode(regs) (!kernel_mode(regs))
 #define instruction_pointer(regs) ((regs)->eip)
 #if defined(CONFIG_SMP) && defined(CONFIG_FRAME_POINTER)
 extern unsigned long profile_pc(struct pt_regs *regs);

--
Please take it as a sign of my infinite respect for you, that I insist on you doing all the work.
-- Rusty Russell