x86/tlb: Fix tlb flushing when lguest clears PGE

Fengguang reported random corruptions from various locations on x86-32
after commits d2852a2 ("arch: add ARCH_HAS_SET_MEMORY config") and
9d876e7 ("bpf: fix unlocking of jited image when module ronx not set")
that uses the former. While x86-32 doesn't have a JIT like x86_64, the
bpf_prog_lock_ro() and bpf_prog_unlock_ro() got enabled due to
ARCH_HAS_SET_MEMORY, whereas Fengguang's test kernel doesn't have module
support built in and therefore never had the DEBUG_SET_MODULE_RONX setting
enabled.

After investigating the crashes further, it turned out that using
set_memory_ro() and set_memory_rw() didn't have the desired effect, for
example, setting the pages as read-only on x86-32 would still let
probe_kernel_write() succeed without error. This behavior would manifest
itself in situations where the vmalloc'ed buffer was accessed prior to
set_memory_*() such as in case of bpf_prog_alloc(). In cases where it
wasn't, the page attribute changes seemed to have taken effect, leading to
the conclusion that a TLB invalidate didn't happen. Moreover, it turned out
that this issue reproduced with qemu in "-cpu kvm64" mode, but not for
"-cpu host". When the issue occurs, change_page_attr_set_clr() did trigger
a TLB flush as expected via __flush_tlb_all() through cpa_flush_range(),
though.

There are 3 variants for issuing a TLB flush: invpcid_flush_all() (depends
on CPU feature bits X86_FEATURE_INVPCID, X86_FEATURE_PGE), cr4 based flush
(depends on X86_FEATURE_PGE), and cr3 based flush.  For "-cpu host" case in
my setup, the flush used invpcid_flush_all() variant, whereas for "-cpu
kvm64", the flush was cr4 based. Switching the kvm64 case to cr3 manually
worked fine, and further investigating the cr4 one turned out that
X86_CR4_PGE bit was not set in cr4 register, meaning the
__native_flush_tlb_global_irq_disabled() wrote cr4 twice with the same
value instead of clearing X86_CR4_PGE in the first write to trigger the
flush.

It turned out that X86_CR4_PGE was cleared from cr4 during init from
lguest_arch_host_init() via adjust_pge(). The X86_FEATURE_PGE bit is also
cleared from there due to concerns of using PGE in guest kernel that can
lead to hard to trace bugs (see bff672e ("lguest: documentation V:
Host") in init()). The CPU feature bits are cleared in dynamic
boot_cpu_data, but they never propagated to __flush_tlb_all() as it uses
static_cpu_has() instead of boot_cpu_has() for testing which variant of TLB
flushing to use, meaning they still used the old setting of the host
kernel.

Clearing via setup_clear_cpu_cap(X86_FEATURE_PGE) so this would propagate
to static_cpu_has() checks is too late at this point as sections have been
patched already, so for now, it seems reasonable to switch back to
boot_cpu_has(X86_FEATURE_PGE) as it was prior to commit c109bf9
("x86/cpufeature: Remove cpu_has_pge"). This lets the TLB flush trigger via
cr3 as originally intended, properly makes the new page attributes visible
and thus fixes the crashes seen by Fengguang.

Fixes: c109bf9 ("x86/cpufeature: Remove cpu_has_pge")
Reported-by: Fengguang Wu <[email protected]>
Signed-off-by: Daniel Borkmann <[email protected]>
Cc: [email protected]
Cc: Kees Cook <[email protected]>
Cc: "David S. Miller" <[email protected]>
Cc: [email protected]
Cc: Rusty Russell <[email protected]>
Cc: Alexei Starovoitov <[email protected]>
Cc: Linus Torvalds <[email protected]>
Cc: [email protected]
Cc: Laura Abbott <[email protected]>
Cc: [email protected]
Link: http://lkml.kernrl.org/r/[email protected]
Link: http://lkml.kernel.org/r/25c41ad9eca164be4db9ad84f768965b7eb19d9e.1489191673.git.daniel@iogearbox.net
Signed-off-by: Thomas Gleixner <[email protected]>

Author: borkmann

arch/x86/include/asm/tlbflush.h

@@ -188,7 +188,7 @@ static inline void __native_flush_tlb_single(unsigned long addr)
 
 static inline void __flush_tlb_all(void)
 {
-	if (static_cpu_has(X86_FEATURE_PGE))
+	if (boot_cpu_has(X86_FEATURE_PGE))
 		__flush_tlb_global();
 	else
 		__flush_tlb();