Signed-off-by: Andrew Cooper <andrew.coop...@citrix.com>
---
CC: Jan Beulich <jbeul...@suse.com>
CC: Roger Pau Monné <roger....@citrix.com>
CC: Wei Liu <w...@xen.org>
RFC. I don't know if this is something we'd want to keep or not.
Getting extable handling working for test_nx_data is proving tricky, and while
I can't spot anything that should stop the extable from working with NX
faults, from a security hardening perspective, there really ought to
be.
(Spurious faults aside), there are no circumstances where an NX fault is
legitimate, and restricting extable's ability to interfere with the fatality
of an NX fault provides a better security posture.
---
xen/arch/x86/setup.c | 39 +++++++++++++++++++++++++++++++++++++++
1 file changed, 39 insertions(+)
diff --git a/xen/arch/x86/setup.c b/xen/arch/x86/setup.c
index 3bbc46f244b9..7cb530a7528f 100644
--- a/xen/arch/x86/setup.c
+++ b/xen/arch/x86/setup.c
@@ -668,6 +668,45 @@ static void noreturn init_done(void)
(unsigned long)&__2M_rodata_end,
PAGE_HYPERVISOR_RO);
+ if ( IS_ENABLED(CONFIG_DEBUG) )
+ {
+ static const char test_rodata = 1;
+ static char __ro_after_init test_ro_after_init = 1;
+
+#define PROBE(insn, c, p) \
+ ({ \
+ bool fault = 0; \
+ asm ( "1:" insn "[ptr]\n\t" \
+ "2:\n\t" \
+ ".section .fixup,\"ax\"\n\t" \
+ "3: movb $1, %[fault]\n\t" \
+ "jmp 2b\n\t" \
+ ".previous" \
+ _ASM_EXTABLE(1b, 3b) \
+ : [fault] "+r" (fault) \
+ : [ptr] c (p) \
+ ); \
+ fault; \
+ })
+
+ if ( !PROBE("notb %", "m", test_rodata) )
+ panic("No fault from test_rodata\n");
+
+ if ( !PROBE("notb %", "m", test_ro_after_init) )
+ panic("No fault from test_ro_after_init\n");
+
+ if ( !PROBE("notb %", "m", init_done) )
+ panic("No fault from modifying init_done\n");
+
+ if ( 0 /* RFC */ && cpu_has_nx )
+ {
+ static char test_nx_data[1] = { 0xc3 };
+
+ if ( !PROBE("call %c", "i", test_nx_data) )
+ panic("No fault from test_nx_data\n");
+ }
+ }
+
startup_cpu_idle_loop();
}
--
2.11.0
