arch/x86/kernel/fpu/xstate.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
When XCR0[9] is set, PKRU can be read and written from userspace with
XSAVE and XRSTOR, even when CR4.PKE is clear.
Clear XCR0[9] when protection keys are disabled.
Reported-by: Tavis Ormandy <taviso@google.com>
Signed-off-by: Jim Mattson <jmattson@google.com>
---
arch/x86/kernel/fpu/xstate.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/arch/x86/kernel/fpu/xstate.c b/arch/x86/kernel/fpu/xstate.c
index 1afbc4866b10..a27b4f7b9365 100644
--- a/arch/x86/kernel/fpu/xstate.c
+++ b/arch/x86/kernel/fpu/xstate.c
@@ -71,7 +71,7 @@ static unsigned short xsave_cpuid_features[] __initdata = {
[XFEATURE_ZMM_Hi256] = X86_FEATURE_AVX512F,
[XFEATURE_Hi16_ZMM] = X86_FEATURE_AVX512F,
[XFEATURE_PT_UNIMPLEMENTED_SO_FAR] = X86_FEATURE_INTEL_PT,
- [XFEATURE_PKRU] = X86_FEATURE_PKU,
+ [XFEATURE_PKRU] = X86_FEATURE_OSPKE,
[XFEATURE_PASID] = X86_FEATURE_ENQCMD,
[XFEATURE_XTILE_CFG] = X86_FEATURE_AMX_TILE,
[XFEATURE_XTILE_DATA] = X86_FEATURE_AMX_TILE,
--
2.42.0.283.g2d96d420d3-goog
On 8/30/23 21:32, Jim Mattson wrote: > When XCR0[9] is set, PKRU can be read and written from userspace with > XSAVE and XRSTOR, even when CR4.PKE is clear. > > Clear XCR0[9] when protection keys are disabled. > > Reported-by: Tavis Ormandy <taviso@google.com> > Signed-off-by: Jim Mattson <jmattson@google.com> Is there any way to trigger this other than "nopku" on the command-line? I'm not sure how scary this particular covert channel is, but it does make sense to do this even if it's only to avoid wasting XSAVE space on a feature that nobody can use (for things other than covert channels). Acked-by: Dave Hansen <dave.hansen@linux.intel.com>
On Thu, Aug 31, 2023 at 12:12 PM Dave Hansen <dave.hansen@intel.com> wrote: > > On 8/30/23 21:32, Jim Mattson wrote: > > When XCR0[9] is set, PKRU can be read and written from userspace with > > XSAVE and XRSTOR, even when CR4.PKE is clear. > > > > Clear XCR0[9] when protection keys are disabled. > > > > Reported-by: Tavis Ormandy <taviso@google.com> > > Signed-off-by: Jim Mattson <jmattson@google.com> > > Is there any way to trigger this other than "nopku" on the command-line? Or by configuration option: CONFIG_X86_INTEL_MEMORY_PROTECTION_KEYS=n > I'm not sure how scary this particular covert channel is, but it does > make sense to do this even if it's only to avoid wasting XSAVE space on > a feature that nobody can use (for things other than covert channels). > > Acked-by: Dave Hansen <dave.hansen@linux.intel.com>
The following commit has been merged into the x86/urgent branch of tip:
Commit-ID: 18032b47adf1db7b7f5fb2d1344e65aafe6417df
Gitweb: https://git.kernel.org/tip/18032b47adf1db7b7f5fb2d1344e65aafe6417df
Author: Jim Mattson <jmattson@google.com>
AuthorDate: Wed, 30 Aug 2023 21:32:21 -07:00
Committer: Ingo Molnar <mingo@kernel.org>
CommitterDate: Thu, 31 Aug 2023 23:29:49 +02:00
x86/fpu/xstate: Fix PKRU covert channel
When XCR0[9] is set, PKRU can be read and written from userspace with
XSAVE and XRSTOR, even when CR4.PKE is clear.
Clear XCR0[9] when protection keys are disabled.
Reported-by: Tavis Ormandy <taviso@google.com>
Signed-off-by: Jim Mattson <jmattson@google.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Dave Hansen <dave.hansen@linux.intel.com>
Link: https://lore.kernel.org/r/20230831043228.1194256-1-jmattson@google.com
---
arch/x86/kernel/fpu/xstate.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/arch/x86/kernel/fpu/xstate.c b/arch/x86/kernel/fpu/xstate.c
index 1afbc48..a27b4f7 100644
--- a/arch/x86/kernel/fpu/xstate.c
+++ b/arch/x86/kernel/fpu/xstate.c
@@ -71,7 +71,7 @@ static unsigned short xsave_cpuid_features[] __initdata = {
[XFEATURE_ZMM_Hi256] = X86_FEATURE_AVX512F,
[XFEATURE_Hi16_ZMM] = X86_FEATURE_AVX512F,
[XFEATURE_PT_UNIMPLEMENTED_SO_FAR] = X86_FEATURE_INTEL_PT,
- [XFEATURE_PKRU] = X86_FEATURE_PKU,
+ [XFEATURE_PKRU] = X86_FEATURE_OSPKE,
[XFEATURE_PASID] = X86_FEATURE_ENQCMD,
[XFEATURE_XTILE_CFG] = X86_FEATURE_AMX_TILE,
[XFEATURE_XTILE_DATA] = X86_FEATURE_AMX_TILE,
© 2016 - 2025 Red Hat, Inc.