Re: [PATCH v3] x86/uaccess: Use pointer masking to limit uaccess speculation

From: Borislav Petkov
Date: Mon Sep 14 2020 - 15:22:11 EST


On Mon, Sep 14, 2020 at 11:48:55AM -0700, Dan Williams wrote:
> > Err, stupid question: can this macro then be folded into access_ok() so
> > that you don't have to touch so many places and the check can happen
> > automatically?
>
> I think that ends up with more changes because it changes the flow of
> access_ok() from returning a boolean to returning a modified user
> address that can be used in the speculative path.

I mean something like the totally untested, only to show intent hunk
below? (It is late here so I could very well be missing an aspect):

diff --git a/arch/x86/include/asm/uaccess.h b/arch/x86/include/asm/uaccess.h
index 2bffba2a1b23..c94e1589682c 100644
--- a/arch/x86/include/asm/uaccess.h
+++ b/arch/x86/include/asm/uaccess.h
@@ -7,6 +7,7 @@
#include <linux/compiler.h>
#include <linux/kasan-checks.h>
#include <linux/string.h>
+#include <linux/nospec.h>
#include <asm/asm.h>
#include <asm/page.h>
#include <asm/smap.h>
@@ -92,8 +93,15 @@ static inline bool pagefault_disabled(void);
*/
#define access_ok(addr, size) \
({ \
+ bool range; \
+ typeof(addr) a = addr, b; \
+ \
WARN_ON_IN_IRQ(); \
- likely(!__range_not_ok(addr, size, user_addr_max())); \
+ \
+ range = __range_not_ok(addr, size, user_addr_max()); \
+ b = (typeof(addr)) array_index_nospec((__force unsigned long)addr, TASK_SIZE_MAX); \
+ \
+ likely(!range && a == b); \
})

/*

--
Regards/Gruss,
Boris.

https://people.kernel.org/tglx/notes-about-netiquette