Message ID | 20210316151320.6123-10-yu-cheng.yu@intel.com (mailing list archive) |
---|---|
State | New, archived |
Headers | show |
Series | Control-flow Enforcement: Indirect Branch Tracking | expand |
On Tue, Mar 16, 2021 at 08:13:19AM -0700, Yu-cheng Yu wrote: > ENDBR is a special new instruction for the Indirect Branch Tracking (IBT) > component of CET. IBT prevents attacks by ensuring that (most) indirect > branches and function calls may only land at ENDBR instructions. Branches > that don't follow the rules will result in control flow (#CF) exceptions. > > ENDBR is a noop when IBT is unsupported or disabled. Most ENDBR > instructions are inserted automatically by the compiler, but branch > targets written in assembly must have ENDBR added manually. > > Add ENDBR to __vdso_sgx_enter_enclave() branch targets. > > Signed-off-by: Yu-cheng Yu <yu-cheng.yu@intel.com> > Cc: Andy Lutomirski <luto@kernel.org> > Cc: Borislav Petkov <bp@alien8.de> > Cc: Dave Hansen <dave.hansen@linux.intel.com> > Cc: Jarkko Sakkinen <jarkko@kernel.org> > Cc: Peter Zijlstra <peterz@infradead.org> > --- > arch/x86/entry/vdso/vsgx.S | 4 ++++ > 1 file changed, 4 insertions(+) > > diff --git a/arch/x86/entry/vdso/vsgx.S b/arch/x86/entry/vdso/vsgx.S > index 86a0e94f68df..1baa9b49053e 100644 > --- a/arch/x86/entry/vdso/vsgx.S > +++ b/arch/x86/entry/vdso/vsgx.S > @@ -6,6 +6,7 @@ > #include <asm/enclu.h> > > #include "extable.h" > +#include "../calling.h" > > /* Relative to %rbp. */ > #define SGX_ENCLAVE_OFFSET_OF_RUN 16 > @@ -27,6 +28,7 @@ > SYM_FUNC_START(__vdso_sgx_enter_enclave) > /* Prolog */ > .cfi_startproc > + ENDBR > push %rbp > .cfi_adjust_cfa_offset 8 > .cfi_rel_offset %rbp, 0 > @@ -62,6 +64,7 @@ SYM_FUNC_START(__vdso_sgx_enter_enclave) > .Lasync_exit_pointer: > .Lenclu_eenter_eresume: > enclu > + ENDBR > > /* EEXIT jumps here unless the enclave is doing something fancy. */ > mov SGX_ENCLAVE_OFFSET_OF_RUN(%rbp), %rbx > @@ -91,6 +94,7 @@ SYM_FUNC_START(__vdso_sgx_enter_enclave) > jmp .Lout > > .Lhandle_exception: > + ENDBR > mov SGX_ENCLAVE_OFFSET_OF_RUN(%rbp), %rbx > > /* Set the exception info. */ > -- > 2.21.0 > > Looks good to me. Reviewed-by: Jarkko Sakkinen <jarkko@kernel.org> /Jarkko
On 3/16/2021 12:22 PM, Jarkko Sakkinen wrote: > On Tue, Mar 16, 2021 at 08:13:19AM -0700, Yu-cheng Yu wrote: >> ENDBR is a special new instruction for the Indirect Branch Tracking (IBT) >> component of CET. IBT prevents attacks by ensuring that (most) indirect >> branches and function calls may only land at ENDBR instructions. Branches >> that don't follow the rules will result in control flow (#CF) exceptions. >> >> ENDBR is a noop when IBT is unsupported or disabled. Most ENDBR >> instructions are inserted automatically by the compiler, but branch >> targets written in assembly must have ENDBR added manually. >> >> Add ENDBR to __vdso_sgx_enter_enclave() branch targets. >> >> Signed-off-by: Yu-cheng Yu <yu-cheng.yu@intel.com> >> Cc: Andy Lutomirski <luto@kernel.org> >> Cc: Borislav Petkov <bp@alien8.de> >> Cc: Dave Hansen <dave.hansen@linux.intel.com> >> Cc: Jarkko Sakkinen <jarkko@kernel.org> >> Cc: Peter Zijlstra <peterz@infradead.org> >> --- >> arch/x86/entry/vdso/vsgx.S | 4 ++++ >> 1 file changed, 4 insertions(+) >> >> diff --git a/arch/x86/entry/vdso/vsgx.S b/arch/x86/entry/vdso/vsgx.S >> index 86a0e94f68df..1baa9b49053e 100644 >> --- a/arch/x86/entry/vdso/vsgx.S >> +++ b/arch/x86/entry/vdso/vsgx.S >> @@ -6,6 +6,7 @@ >> #include <asm/enclu.h> >> >> #include "extable.h" >> +#include "../calling.h" >> >> /* Relative to %rbp. */ >> #define SGX_ENCLAVE_OFFSET_OF_RUN 16 >> @@ -27,6 +28,7 @@ >> SYM_FUNC_START(__vdso_sgx_enter_enclave) >> /* Prolog */ >> .cfi_startproc >> + ENDBR >> push %rbp >> .cfi_adjust_cfa_offset 8 >> .cfi_rel_offset %rbp, 0 >> @@ -62,6 +64,7 @@ SYM_FUNC_START(__vdso_sgx_enter_enclave) >> .Lasync_exit_pointer: >> .Lenclu_eenter_eresume: >> enclu >> + ENDBR >> >> /* EEXIT jumps here unless the enclave is doing something fancy. */ >> mov SGX_ENCLAVE_OFFSET_OF_RUN(%rbp), %rbx >> @@ -91,6 +94,7 @@ SYM_FUNC_START(__vdso_sgx_enter_enclave) >> jmp .Lout >> >> .Lhandle_exception: >> + ENDBR >> mov SGX_ENCLAVE_OFFSET_OF_RUN(%rbp), %rbx >> >> /* Set the exception info. */ >> -- >> 2.21.0 >> >> > > Looks good to me. > > Reviewed-by: Jarkko Sakkinen <jarkko@kernel.org> Thanks for reviewing. In response to Dave's and Boris' comments, I will replace ENDBR macro with _CET_ENDBR that comes from the compiler. Can I still keep the Reviewed-by? > > /Jarkko >
On Tue, Mar 16, 2021 at 12:27:19PM -0700, Yu, Yu-cheng wrote: > On 3/16/2021 12:22 PM, Jarkko Sakkinen wrote: > > On Tue, Mar 16, 2021 at 08:13:19AM -0700, Yu-cheng Yu wrote: > > > ENDBR is a special new instruction for the Indirect Branch Tracking (IBT) > > > component of CET. IBT prevents attacks by ensuring that (most) indirect > > > branches and function calls may only land at ENDBR instructions. Branches > > > that don't follow the rules will result in control flow (#CF) exceptions. > > > > > > ENDBR is a noop when IBT is unsupported or disabled. Most ENDBR > > > instructions are inserted automatically by the compiler, but branch > > > targets written in assembly must have ENDBR added manually. > > > > > > Add ENDBR to __vdso_sgx_enter_enclave() branch targets. > > > > > > Signed-off-by: Yu-cheng Yu <yu-cheng.yu@intel.com> > > > Cc: Andy Lutomirski <luto@kernel.org> > > > Cc: Borislav Petkov <bp@alien8.de> > > > Cc: Dave Hansen <dave.hansen@linux.intel.com> > > > Cc: Jarkko Sakkinen <jarkko@kernel.org> > > > Cc: Peter Zijlstra <peterz@infradead.org> > > > --- > > > arch/x86/entry/vdso/vsgx.S | 4 ++++ > > > 1 file changed, 4 insertions(+) > > > > > > diff --git a/arch/x86/entry/vdso/vsgx.S b/arch/x86/entry/vdso/vsgx.S > > > index 86a0e94f68df..1baa9b49053e 100644 > > > --- a/arch/x86/entry/vdso/vsgx.S > > > +++ b/arch/x86/entry/vdso/vsgx.S > > > @@ -6,6 +6,7 @@ > > > #include <asm/enclu.h> > > > #include "extable.h" > > > +#include "../calling.h" > > > /* Relative to %rbp. */ > > > #define SGX_ENCLAVE_OFFSET_OF_RUN 16 > > > @@ -27,6 +28,7 @@ > > > SYM_FUNC_START(__vdso_sgx_enter_enclave) > > > /* Prolog */ > > > .cfi_startproc > > > + ENDBR > > > push %rbp > > > .cfi_adjust_cfa_offset 8 > > > .cfi_rel_offset %rbp, 0 > > > @@ -62,6 +64,7 @@ SYM_FUNC_START(__vdso_sgx_enter_enclave) > > > .Lasync_exit_pointer: > > > .Lenclu_eenter_eresume: > > > enclu > > > + ENDBR > > > /* EEXIT jumps here unless the enclave is doing something fancy. */ > > > mov SGX_ENCLAVE_OFFSET_OF_RUN(%rbp), %rbx > > > @@ -91,6 +94,7 @@ SYM_FUNC_START(__vdso_sgx_enter_enclave) > > > jmp .Lout > > > .Lhandle_exception: > > > + ENDBR > > > mov SGX_ENCLAVE_OFFSET_OF_RUN(%rbp), %rbx > > > /* Set the exception info. */ > > > -- > > > 2.21.0 > > > > > > > > > > Looks good to me. > > > > Reviewed-by: Jarkko Sakkinen <jarkko@kernel.org> > > Thanks for reviewing. In response to Dave's and Boris' comments, I will > replace ENDBR macro with _CET_ENDBR that comes from the compiler. Can I > still keep the Reviewed-by? I'll rather re-ack, thanks. Most likely give reviewed-by but I always prefer to see the code change before doing that. /Jarkko
diff --git a/arch/x86/entry/vdso/vsgx.S b/arch/x86/entry/vdso/vsgx.S index 86a0e94f68df..1baa9b49053e 100644 --- a/arch/x86/entry/vdso/vsgx.S +++ b/arch/x86/entry/vdso/vsgx.S @@ -6,6 +6,7 @@ #include <asm/enclu.h> #include "extable.h" +#include "../calling.h" /* Relative to %rbp. */ #define SGX_ENCLAVE_OFFSET_OF_RUN 16 @@ -27,6 +28,7 @@ SYM_FUNC_START(__vdso_sgx_enter_enclave) /* Prolog */ .cfi_startproc + ENDBR push %rbp .cfi_adjust_cfa_offset 8 .cfi_rel_offset %rbp, 0 @@ -62,6 +64,7 @@ SYM_FUNC_START(__vdso_sgx_enter_enclave) .Lasync_exit_pointer: .Lenclu_eenter_eresume: enclu + ENDBR /* EEXIT jumps here unless the enclave is doing something fancy. */ mov SGX_ENCLAVE_OFFSET_OF_RUN(%rbp), %rbx @@ -91,6 +94,7 @@ SYM_FUNC_START(__vdso_sgx_enter_enclave) jmp .Lout .Lhandle_exception: + ENDBR mov SGX_ENCLAVE_OFFSET_OF_RUN(%rbp), %rbx /* Set the exception info. */
ENDBR is a special new instruction for the Indirect Branch Tracking (IBT) component of CET. IBT prevents attacks by ensuring that (most) indirect branches and function calls may only land at ENDBR instructions. Branches that don't follow the rules will result in control flow (#CF) exceptions. ENDBR is a noop when IBT is unsupported or disabled. Most ENDBR instructions are inserted automatically by the compiler, but branch targets written in assembly must have ENDBR added manually. Add ENDBR to __vdso_sgx_enter_enclave() branch targets. Signed-off-by: Yu-cheng Yu <yu-cheng.yu@intel.com> Cc: Andy Lutomirski <luto@kernel.org> Cc: Borislav Petkov <bp@alien8.de> Cc: Dave Hansen <dave.hansen@linux.intel.com> Cc: Jarkko Sakkinen <jarkko@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> --- arch/x86/entry/vdso/vsgx.S | 4 ++++ 1 file changed, 4 insertions(+)