| Message ID | 20240403234054.2020347-9-debug@rivosinc.com (mailing list archive) |
|---|---|
| State | RFC |
| Headers | show |
| Series | riscv control-flow integrity for usermode | expand |
| Context | Check | Description |
|---|---|---|
| conchuod/vmtest-fixes-PR | fail | merge-conflict |
On 04.04.24 01:34, Deepak Gupta wrote: > VM_SHADOW_STACK is defined by x86 as vm flag to mark a shadow stack vma. > > x86 uses VM_HIGH_ARCH_5 bit but that limits shadow stack vma to 64bit only. > arm64 follows same path (see links) > > To keep things simple, RISC-V follows the same. > This patch adds `ss` for shadow stack in process maps. > > Links: > https://lore.kernel.org/lkml/20231009-arm64-gcs-v6-12-78e55deaa4dd@kernel.org/#r > > Signed-off-by: Deepak Gupta <debug@rivosinc.com> > --- > fs/proc/task_mmu.c | 3 +++ > include/linux/mm.h | 11 ++++++++++- > 2 files changed, 13 insertions(+), 1 deletion(-) > > diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c > index 3f78ebbb795f..d9d63eb74f0d 100644 > --- a/fs/proc/task_mmu.c > +++ b/fs/proc/task_mmu.c > @@ -702,6 +702,9 @@ static void show_smap_vma_flags(struct seq_file *m, struct vm_area_struct *vma) > #endif /* CONFIG_HAVE_ARCH_USERFAULTFD_MINOR */ > #ifdef CONFIG_X86_USER_SHADOW_STACK > [ilog2(VM_SHADOW_STACK)] = "ss", > +#endif > +#ifdef CONFIG_RISCV_USER_CFI > + [ilog2(VM_SHADOW_STACK)] = "ss", > #endif > }; > size_t i; > diff --git a/include/linux/mm.h b/include/linux/mm.h > index f5a97dec5169..64109f6c70f5 100644 > --- a/include/linux/mm.h > +++ b/include/linux/mm.h > @@ -352,7 +352,16 @@ extern unsigned int kobjsize(const void *objp); > * for more details on the guard size. > */ > # define VM_SHADOW_STACK VM_HIGH_ARCH_5 > -#else > +#endif > + > +#ifdef CONFIG_RISCV_USER_CFI > +/* > + * RISC-V is going along with using VM_HIGH_ARCH_5 bit position for shadow stack > + */ Wow, really?! I could never have guesses that from the code :P Just drop that comment. Are them semantics the same as for the x86 variant documented? ("VM_SHADOW_STACK should not be set with VM_SHARED because of lack of") I assume so. So it might now make sense to merge both paths #if defined(CONFIG_X86_USER_SHADOW_STACK) || defined(CONFIG_RISCV_USER_CFI) or even introduce some ARCH_HAS_SHADOW_STACK so we can remove these arch-specific thingies here.
On Thu, Apr 04, 2024 at 08:58:06PM +0200, David Hildenbrand wrote: > or even introduce some ARCH_HAS_SHADOW_STACK so we can remove these > arch-specific thingies here. It would be convenient if you could pull the ARCH_HAS_USER_SHADOW_STACK patch out of my clone3 series to do that: https://lore.kernel.org/all/20240203-clone3-shadow-stack-v5-3-322c69598e4b@kernel.org/
On 04.04.24 21:04, Mark Brown wrote: > On Thu, Apr 04, 2024 at 08:58:06PM +0200, David Hildenbrand wrote: > >> or even introduce some ARCH_HAS_SHADOW_STACK so we can remove these >> arch-specific thingies here. > > It would be convenient if you could pull the ARCH_HAS_USER_SHADOW_STACK > patch out of my clone3 series to do that: > > https://lore.kernel.org/all/20240203-clone3-shadow-stack-v5-3-322c69598e4b@kernel.org/ Crazy, I completely forgot about that one. Yes!
On Thu, Apr 4, 2024 at 12:15 PM David Hildenbrand <david@redhat.com> wrote: > > On 04.04.24 21:04, Mark Brown wrote: > > On Thu, Apr 04, 2024 at 08:58:06PM +0200, David Hildenbrand wrote: > > > >> or even introduce some ARCH_HAS_SHADOW_STACK so we can remove these > >> arch-specific thingies here. > > > > It would be convenient if you could pull the ARCH_HAS_USER_SHADOW_STACK > > patch out of my clone3 series to do that: > > > > https://lore.kernel.org/all/20240203-clone3-shadow-stack-v5-3-322c69598e4b@kernel.org/ > > Crazy, I completely forgot about that one. Yes! I missed that. Roger. Will do that in the next series. Thanks.
diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index 3f78ebbb795f..d9d63eb74f0d 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -702,6 +702,9 @@ static void show_smap_vma_flags(struct seq_file *m, struct vm_area_struct *vma) #endif /* CONFIG_HAVE_ARCH_USERFAULTFD_MINOR */ #ifdef CONFIG_X86_USER_SHADOW_STACK [ilog2(VM_SHADOW_STACK)] = "ss", +#endif +#ifdef CONFIG_RISCV_USER_CFI + [ilog2(VM_SHADOW_STACK)] = "ss", #endif }; size_t i; diff --git a/include/linux/mm.h b/include/linux/mm.h index f5a97dec5169..64109f6c70f5 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -352,7 +352,16 @@ extern unsigned int kobjsize(const void *objp); * for more details on the guard size. */ # define VM_SHADOW_STACK VM_HIGH_ARCH_5 -#else +#endif + +#ifdef CONFIG_RISCV_USER_CFI +/* + * RISC-V is going along with using VM_HIGH_ARCH_5 bit position for shadow stack + */ +#define VM_SHADOW_STACK VM_HIGH_ARCH_5 +#endif + +#ifndef VM_SHADOW_STACK # define VM_SHADOW_STACK VM_NONE #endif
VM_SHADOW_STACK is defined by x86 as vm flag to mark a shadow stack vma. x86 uses VM_HIGH_ARCH_5 bit but that limits shadow stack vma to 64bit only. arm64 follows same path (see links) To keep things simple, RISC-V follows the same. This patch adds `ss` for shadow stack in process maps. Links: https://lore.kernel.org/lkml/20231009-arm64-gcs-v6-12-78e55deaa4dd@kernel.org/#r Signed-off-by: Deepak Gupta <debug@rivosinc.com> --- fs/proc/task_mmu.c | 3 +++ include/linux/mm.h | 11 ++++++++++- 2 files changed, 13 insertions(+), 1 deletion(-)