arm64 has a feature called Top Byte Ignore, which allows to embed pointer
tags into the top byte of each pointer. Userspace programs (such as
HWASan, a memory debugging tool [1]) might use this feature and pass
tagged user pointers to the kernel through syscalls or other interfaces.
Right now the kernel is already able to handle user faults with tagged
pointers, due to these patches:
1. 81cddd65 ("arm64: traps: fix userspace cache maintenance emulation on a
tagged pointer")
2. 7dcd9dd8 ("arm64: hw_breakpoint: fix watchpoint matching for tagged
pointers")
3. 276e9327 ("arm64: entry: improve data abort handling of tagged
pointers")
When passing tagged pointers to syscalls, there's a special case of such a
pointer being passed to one of the memory syscalls (mmap, mprotect, etc.).
These syscalls don't do memory accesses but rather deal with memory
ranges, hence an untagged pointer is better suited.
This patchset extends tagged pointer support to non-memory syscalls. This
is done by reusing the untagged_addr macro to untag user pointers when the
kernel performs pointer checking to find out whether the pointer comes
from userspace (most notably in access_ok).
The following testing approaches has been taken to find potential issues
with user pointer untagging:
1. Static testing (with sparse [2] and separately with a custom static
analyzer based on Clang) to track casts of __user pointers to integer
types to find places where untagging needs to be done.
2. Dynamic testing: adding BUG_ON(has_tag(addr)) to find_vma() and running
a modified syzkaller version that passes tagged pointers to the kernel.
Based on the results of the testing the requried patches have been added
to the patchset.
This patchset is a prerequisite for ARM's memory tagging hardware feature
support [3].
Thanks!
[1] http://clang.llvm.org/docs/HardwareAssistedAddressSanitizerDesign.html
[2] https://github.com/lucvoo/sparse-dev/commit/5f960cb10f56ec2017c128ef9d16060e0145f292
[3] https://community.arm.com/processors/b/blog/posts/arm-a-profile-architecture-2018-developments-armv85a
Changes in v7:
- Rebased onto 17b57b18 (4.19-rc6).
- Dropped the "arm64: untag user address in __do_user_fault" patch, since
the existing patches already handle user faults properly.
- Dropped the "usb, arm64: untag user addresses in devio" patch, since the
passed pointer must come from a vma and therefore be untagged.
- Dropped the "arm64: annotate user pointers casts detected by sparse"
patch (see the discussion to the replies of the v6 of this patchset).
- Added more context to the cover letter.
- Updated Documentation/arm64/tagged-pointers.txt.
Changes in v6:
- Added annotations for user pointer casts found by sparse.
- Rebased onto 050cdc6c (4.19-rc1+).
Changes in v5:
- Added 3 new patches that add untagging to places found with static
analysis.
- Rebased onto 44c929e1 (4.18-rc8).
Changes in v4:
- Added a selftest for checking that passing tagged pointers to the
kernel succeeds.
- Rebased onto 81e97f013 (4.18-rc1+).
Changes in v3:
- Rebased onto e5c51f30 (4.17-rc6+).
- Added linux-arch@ to the list of recipients.
Changes in v2:
- Rebased onto 2d618bdf (4.17-rc3+).
- Removed excessive untagging in gup.c.
- Removed untagging pointers returned from __uaccess_mask_ptr.
Changes in v1:
- Rebased onto 4.17-rc1.
Changes in RFC v2:
- Added "#ifndef untagged_addr..." fallback in linux/uaccess.h instead of
defining it for each arch individually.
- Updated Documentation/arm64/tagged-pointers.txt.
- Dropped "mm, arm64: untag user addresses in memory syscalls".
- Rebased onto 3eb2ce82 (4.16-rc7).
Andrey Konovalov (8):
arm64: add type casts to untagged_addr macro
uaccess: add untagged_addr definition for other arches
arm64: untag user addresses in access_ok and __uaccess_mask_ptr
mm, arm64: untag user addresses in mm/gup.c
lib, arm64: untag addrs passed to strncpy_from_user and strnlen_user
fs, arm64: untag user address in copy_mount_options
arm64: update Documentation/arm64/tagged-pointers.txt
selftests, arm64: add a selftest for passing tagged pointers to kernel
Documentation/arm64/tagged-pointers.txt | 24 +++++++++++--------
arch/arm64/include/asm/uaccess.h | 14 +++++++----
fs/namespace.c | 2 +-
include/linux/uaccess.h | 4 ++++
lib/strncpy_from_user.c | 2 ++
lib/strnlen_user.c | 2 ++
mm/gup.c | 4 ++++
tools/testing/selftests/arm64/.gitignore | 1 +
tools/testing/selftests/arm64/Makefile | 11 +++++++++
.../testing/selftests/arm64/run_tags_test.sh | 12 ++++++++++
tools/testing/selftests/arm64/tags_test.c | 19 +++++++++++++++
11 files changed, 79 insertions(+), 16 deletions(-)
create mode 100644 tools/testing/selftests/arm64/.gitignore
create mode 100644 tools/testing/selftests/arm64/Makefile
create mode 100755 tools/testing/selftests/arm64/run_tags_test.sh
create mode 100644 tools/testing/selftests/arm64/tags_test.c
--
2.19.0.605.g01d371f741-goog
This patch adds a simple test, that calls the uname syscall with a
tagged user pointer as an argument. Without the kernel accepting tagged
user pointers the test fails with EFAULT.
Signed-off-by: Andrey Konovalov <[email protected]>
---
tools/testing/selftests/arm64/.gitignore | 1 +
tools/testing/selftests/arm64/Makefile | 11 +++++++++++
.../testing/selftests/arm64/run_tags_test.sh | 12 ++++++++++++
tools/testing/selftests/arm64/tags_test.c | 19 +++++++++++++++++++
4 files changed, 43 insertions(+)
create mode 100644 tools/testing/selftests/arm64/.gitignore
create mode 100644 tools/testing/selftests/arm64/Makefile
create mode 100755 tools/testing/selftests/arm64/run_tags_test.sh
create mode 100644 tools/testing/selftests/arm64/tags_test.c
diff --git a/tools/testing/selftests/arm64/.gitignore b/tools/testing/selftests/arm64/.gitignore
new file mode 100644
index 000000000000..e8fae8d61ed6
--- /dev/null
+++ b/tools/testing/selftests/arm64/.gitignore
@@ -0,0 +1 @@
+tags_test
diff --git a/tools/testing/selftests/arm64/Makefile b/tools/testing/selftests/arm64/Makefile
new file mode 100644
index 000000000000..a61b2e743e99
--- /dev/null
+++ b/tools/testing/selftests/arm64/Makefile
@@ -0,0 +1,11 @@
+# SPDX-License-Identifier: GPL-2.0
+
+# ARCH can be overridden by the user for cross compiling
+ARCH ?= $(shell uname -m 2>/dev/null || echo not)
+
+ifneq (,$(filter $(ARCH),aarch64 arm64))
+TEST_GEN_PROGS := tags_test
+TEST_PROGS := run_tags_test.sh
+endif
+
+include ../lib.mk
diff --git a/tools/testing/selftests/arm64/run_tags_test.sh b/tools/testing/selftests/arm64/run_tags_test.sh
new file mode 100755
index 000000000000..745f11379930
--- /dev/null
+++ b/tools/testing/selftests/arm64/run_tags_test.sh
@@ -0,0 +1,12 @@
+#!/bin/sh
+# SPDX-License-Identifier: GPL-2.0
+
+echo "--------------------"
+echo "running tags test"
+echo "--------------------"
+./tags_test
+if [ $? -ne 0 ]; then
+ echo "[FAIL]"
+else
+ echo "[PASS]"
+fi
diff --git a/tools/testing/selftests/arm64/tags_test.c b/tools/testing/selftests/arm64/tags_test.c
new file mode 100644
index 000000000000..1452ed7d33f9
--- /dev/null
+++ b/tools/testing/selftests/arm64/tags_test.c
@@ -0,0 +1,19 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <stdio.h>
+#include <unistd.h>
+#include <stdint.h>
+#include <sys/utsname.h>
+
+#define SHIFT_TAG(tag) ((uint64_t)(tag) << 56)
+#define SET_TAG(ptr, tag) (((uint64_t)(ptr) & ~SHIFT_TAG(0xff)) | \
+ SHIFT_TAG(tag))
+
+int main(void)
+{
+ struct utsname utsname;
+ void *ptr = &utsname;
+ void *tagged_ptr = (void *)SET_TAG(ptr, 0x42);
+ int err = uname(tagged_ptr);
+ return err;
+}
--
2.19.0.605.g01d371f741-goog
Document the changes in Documentation/arm64/tagged-pointers.txt.
Signed-off-by: Andrey Konovalov <[email protected]>
---
Documentation/arm64/tagged-pointers.txt | 24 ++++++++++++++----------
1 file changed, 14 insertions(+), 10 deletions(-)
diff --git a/Documentation/arm64/tagged-pointers.txt b/Documentation/arm64/tagged-pointers.txt
index a25a99e82bb1..ae877d185fdb 100644
--- a/Documentation/arm64/tagged-pointers.txt
+++ b/Documentation/arm64/tagged-pointers.txt
@@ -17,13 +17,21 @@ this byte for application use.
Passing tagged addresses to the kernel
--------------------------------------
-All interpretation of userspace memory addresses by the kernel assumes
-an address tag of 0x00.
+Some initial work for supporting non-zero address tags passed to the
+kernel has been done. As of now, the kernel supports tags in:
-This includes, but is not limited to, addresses found in:
+ - user fault addresses
- - pointer arguments to system calls, including pointers in structures
- passed to system calls,
+ - pointer arguments (including pointers in structures), which don't
+ describe virtual memory ranges, passed to system calls
+
+All other interpretations of userspace memory addresses by the kernel
+assume an address tag of 0x00. This includes, but is not limited to,
+addresses found in:
+
+ - pointer arguments (including pointers in structures), which describe
+ virtual memory ranges, passed to memory system calls (mmap, mprotect,
+ etc.)
- the stack pointer (sp), e.g. when interpreting it to deliver a
signal,
@@ -33,11 +41,7 @@ This includes, but is not limited to, addresses found in:
Using non-zero address tags in any of these locations may result in an
error code being returned, a (fatal) signal being raised, or other modes
-of failure.
-
-For these reasons, passing non-zero address tags to the kernel via
-system calls is forbidden, and using a non-zero address tag for sp is
-strongly discouraged.
+of failure. Using a non-zero address tag for sp is strongly discouraged.
Programs maintaining a frame pointer and frame records that use non-zero
address tags may suffer impaired or inaccurate debug and profiling
--
2.19.0.605.g01d371f741-goog
mm/gup.c provides a kernel interface that accepts user addresses and
manipulates user pages directly (for example get_user_pages, that is used
by the futex syscall). Since a user can provided tagged addresses, we need
to handle such case.
Add untagging to gup.c functions that use user addresses for vma lookup.
Signed-off-by: Andrey Konovalov <[email protected]>
---
mm/gup.c | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/mm/gup.c b/mm/gup.c
index 1abc8b4afff6..6f09132c654e 100644
--- a/mm/gup.c
+++ b/mm/gup.c
@@ -666,6 +666,8 @@ static long __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
if (!nr_pages)
return 0;
+ start = untagged_addr(start);
+
VM_BUG_ON(!!pages != !!(gup_flags & FOLL_GET));
/*
@@ -820,6 +822,8 @@ int fixup_user_fault(struct task_struct *tsk, struct mm_struct *mm,
struct vm_area_struct *vma;
vm_fault_t ret, major = 0;
+ address = untagged_addr(address);
+
if (unlocked)
fault_flags |= FAULT_FLAG_ALLOW_RETRY;
--
2.19.0.605.g01d371f741-goog
strncpy_from_user and strnlen_user accept user addresses as arguments, and
do not go through the same path as copy_from_user and others, so here we
need to handle the case of tagged user addresses separately.
Untag user pointers passed to these functions.
Signed-off-by: Andrey Konovalov <[email protected]>
---
lib/strncpy_from_user.c | 2 ++
lib/strnlen_user.c | 2 ++
2 files changed, 4 insertions(+)
diff --git a/lib/strncpy_from_user.c b/lib/strncpy_from_user.c
index b53e1b5d80f4..97467cd2bc59 100644
--- a/lib/strncpy_from_user.c
+++ b/lib/strncpy_from_user.c
@@ -106,6 +106,8 @@ long strncpy_from_user(char *dst, const char __user *src, long count)
if (unlikely(count <= 0))
return 0;
+ src = untagged_addr(src);
+
max_addr = user_addr_max();
src_addr = (unsigned long)src;
if (likely(src_addr < max_addr)) {
diff --git a/lib/strnlen_user.c b/lib/strnlen_user.c
index 60d0bbda8f5e..8b5f56466e00 100644
--- a/lib/strnlen_user.c
+++ b/lib/strnlen_user.c
@@ -108,6 +108,8 @@ long strnlen_user(const char __user *str, long count)
if (unlikely(count <= 0))
return 0;
+ str = untagged_addr(str);
+
max_addr = user_addr_max();
src_addr = (unsigned long)str;
if (likely(src_addr < max_addr)) {
--
2.19.0.605.g01d371f741-goog
copy_from_user (and a few other similar functions) are used to copy data
from user memory into the kernel memory or vice versa. Since a user can
provided a tagged pointer to one of the syscalls that use copy_from_user,
we need to correctly handle such pointers.
Do this by untagging user pointers in access_ok and in __uaccess_mask_ptr,
before performing access validity checks.
Signed-off-by: Andrey Konovalov <[email protected]>
---
arch/arm64/include/asm/uaccess.h | 11 +++++++----
1 file changed, 7 insertions(+), 4 deletions(-)
diff --git a/arch/arm64/include/asm/uaccess.h b/arch/arm64/include/asm/uaccess.h
index 2d6451cbaa86..fa7318d3d7d5 100644
--- a/arch/arm64/include/asm/uaccess.h
+++ b/arch/arm64/include/asm/uaccess.h
@@ -105,7 +105,8 @@ static inline unsigned long __range_ok(const void __user *addr, unsigned long si
#define untagged_addr(addr) \
((__typeof__(addr))sign_extend64((__u64)(addr), 55))
-#define access_ok(type, addr, size) __range_ok(addr, size)
+#define access_ok(type, addr, size) \
+ __range_ok(untagged_addr(addr), size)
#define user_addr_max get_fs
#define _ASM_EXTABLE(from, to) \
@@ -237,7 +238,8 @@ static inline void uaccess_enable_not_uao(void)
/*
* Sanitise a uaccess pointer such that it becomes NULL if above the
- * current addr_limit.
+ * current addr_limit. In case the pointer is tagged (has the top byte set),
+ * untag the pointer before checking.
*/
#define uaccess_mask_ptr(ptr) (__typeof__(ptr))__uaccess_mask_ptr(ptr)
static inline void __user *__uaccess_mask_ptr(const void __user *ptr)
@@ -245,10 +247,11 @@ static inline void __user *__uaccess_mask_ptr(const void __user *ptr)
void __user *safe_ptr;
asm volatile(
- " bics xzr, %1, %2\n"
+ " bics xzr, %3, %2\n"
" csel %0, %1, xzr, eq\n"
: "=&r" (safe_ptr)
- : "r" (ptr), "r" (current_thread_info()->addr_limit)
+ : "r" (ptr), "r" (current_thread_info()->addr_limit),
+ "r" (untagged_addr(ptr))
: "cc");
csdb();
--
2.19.0.605.g01d371f741-goog
To allow arm64 syscalls accept tagged pointers from userspace, we must
untag them when they are passed to the kernel. Since untagging is done in
generic parts of the kernel, the untagged_addr macro needs to be defined
for all architectures.
Define it as a noop for other architectures besides arm64.
Signed-off-by: Andrey Konovalov <[email protected]>
---
include/linux/uaccess.h | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/include/linux/uaccess.h b/include/linux/uaccess.h
index efe79c1cdd47..c045b4eff95e 100644
--- a/include/linux/uaccess.h
+++ b/include/linux/uaccess.h
@@ -13,6 +13,10 @@
#include <asm/uaccess.h>
+#ifndef untagged_addr
+#define untagged_addr(addr) addr
+#endif
+
/*
* Architectures should provide two primitives (raw_copy_{to,from}_user())
* and get rid of their private instances of copy_{to,from}_user() and
--
2.19.0.605.g01d371f741-goog
In copy_mount_options a user address is being subtracted from TASK_SIZE.
If the address is lower than TASK_SIZE, the size is calculated to not
allow the exact_copy_from_user() call to cross TASK_SIZE boundary.
However if the address is tagged, then the size will be calculated
incorrectly.
Untag the address before subtracting.
Signed-off-by: Andrey Konovalov <[email protected]>
---
fs/namespace.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/fs/namespace.c b/fs/namespace.c
index 99186556f8d3..51f763fb9430 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -2672,7 +2672,7 @@ void *copy_mount_options(const void __user * data)
* the remainder of the page.
*/
/* copy_from_user cannot cross TASK_SIZE ! */
- size = TASK_SIZE - (unsigned long)data;
+ size = TASK_SIZE - (unsigned long)untagged_addr(data);
if (size > PAGE_SIZE)
size = PAGE_SIZE;
--
2.19.0.605.g01d371f741-goog
This patch makes the untagged_addr macro accept all kinds of address types
(void *, unsigned long, etc.) and allows not to specify type casts in each
place where it is used. This is done by using __typeof__.
Signed-off-by: Andrey Konovalov <[email protected]>
---
arch/arm64/include/asm/uaccess.h | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/arch/arm64/include/asm/uaccess.h b/arch/arm64/include/asm/uaccess.h
index e66b0fca99c2..2d6451cbaa86 100644
--- a/arch/arm64/include/asm/uaccess.h
+++ b/arch/arm64/include/asm/uaccess.h
@@ -102,7 +102,8 @@ static inline unsigned long __range_ok(const void __user *addr, unsigned long si
* up with a tagged userland pointer. Clear the tag to get a sane pointer to
* pass on to access_ok(), for instance.
*/
-#define untagged_addr(addr) sign_extend64(addr, 55)
+#define untagged_addr(addr) \
+ ((__typeof__(addr))sign_extend64((__u64)(addr), 55))
#define access_ok(type, addr, size) __range_ok(addr, size)
#define user_addr_max get_fs
--
2.19.0.605.g01d371f741-goog
On Tue, Oct 02, 2018 at 03:12:35PM +0200, Andrey Konovalov wrote:
...
> Changes in v7:
> - Rebased onto 17b57b18 (4.19-rc6).
> - Dropped the "arm64: untag user address in __do_user_fault" patch, since
> the existing patches already handle user faults properly.
> - Dropped the "usb, arm64: untag user addresses in devio" patch, since the
> passed pointer must come from a vma and therefore be untagged.
> - Dropped the "arm64: annotate user pointers casts detected by sparse"
> patch (see the discussion to the replies of the v6 of this patchset).
> - Added more context to the cover letter.
> - Updated Documentation/arm64/tagged-pointers.txt.
Hi,
I'm quite hapy now with what concerns me (sparse).
Please, feel free to add my:
Reviewed-by: Luc Van Oostenryck <[email protected]>
Cheers,
-- Luc Van Oostenryck
On Tue, Oct 02, 2018 at 03:12:42PM +0200, Andrey Konovalov wrote:
> diff --git a/Documentation/arm64/tagged-pointers.txt b/Documentation/arm64/tagged-pointers.txt
> index a25a99e82bb1..ae877d185fdb 100644
> --- a/Documentation/arm64/tagged-pointers.txt
> +++ b/Documentation/arm64/tagged-pointers.txt
> @@ -17,13 +17,21 @@ this byte for application use.
> Passing tagged addresses to the kernel
> --------------------------------------
>
> -All interpretation of userspace memory addresses by the kernel assumes
> -an address tag of 0x00.
> +Some initial work for supporting non-zero address tags passed to the
> +kernel has been done. As of now, the kernel supports tags in:
With my maintainer hat on, the above statement leads me to think this
new ABI is work in progress, so not yet suitable for upstream.
Also, how is user space supposed to know that it can now pass tagged
pointers into the kernel? An ABI change (or relaxation), needs to be
advertised by the kernel, usually via a new HWCAP bit (e.g. HWCAP_TBI).
Once we have a HWCAP bit in place, we need to be pretty clear about
which syscalls can and cannot cope with tagged pointers. The "as of now"
implies potential further relaxation which, again, would need to be
advertised to user in some (additional) way.
> -This includes, but is not limited to, addresses found in:
> + - user fault addresses
While the kernel currently supports this in some way (by clearing the
tag exception entry, el0_da), the above implies (at least to me) that
sigcontext.fault_address would contain the tagged address. That's not
the case (unless I missed it in your patches).
> - - pointer arguments to system calls, including pointers in structures
> - passed to system calls,
> + - pointer arguments (including pointers in structures), which don't
> + describe virtual memory ranges, passed to system calls
I think we need to be more precise here...
> +All other interpretations of userspace memory addresses by the kernel
> +assume an address tag of 0x00. This includes, but is not limited to,
> +addresses found in:
> +
> + - pointer arguments (including pointers in structures), which describe
> + virtual memory ranges, passed to memory system calls (mmap, mprotect,
> + etc.)
...and probably a full list here.
--
Catalin
On Wed, Oct 3, 2018 at 7:32 PM, Catalin Marinas <[email protected]> wrote:
> On Tue, Oct 02, 2018 at 03:12:42PM +0200, Andrey Konovalov wrote:
>> diff --git a/Documentation/arm64/tagged-pointers.txt b/Documentation/arm64/tagged-pointers.txt
>> index a25a99e82bb1..ae877d185fdb 100644
>> --- a/Documentation/arm64/tagged-pointers.txt
>> +++ b/Documentation/arm64/tagged-pointers.txt
>> @@ -17,13 +17,21 @@ this byte for application use.
>> Passing tagged addresses to the kernel
>> --------------------------------------
>>
>> -All interpretation of userspace memory addresses by the kernel assumes
>> -an address tag of 0x00.
>> +Some initial work for supporting non-zero address tags passed to the
>> +kernel has been done. As of now, the kernel supports tags in:
>
> With my maintainer hat on, the above statement leads me to think this
> new ABI is work in progress, so not yet suitable for upstream.
OK, I think we can just say "The kernel supports tags in:" here. Will do in v8.
>
> Also, how is user space supposed to know that it can now pass tagged
> pointers into the kernel? An ABI change (or relaxation), needs to be
> advertised by the kernel, usually via a new HWCAP bit (e.g. HWCAP_TBI).
> Once we have a HWCAP bit in place, we need to be pretty clear about
> which syscalls can and cannot cope with tagged pointers. The "as of now"
> implies potential further relaxation which, again, would need to be
> advertised to user in some (additional) way.
How exactly should I do that? Something like this [1]? Or is it only
for hardware specific things and for this patchset I need to do
something else?
[1] https://github.com/torvalds/linux/commit/7206dc93a58fb76421c4411eefa3c003337bcb2d
>
>> -This includes, but is not limited to, addresses found in:
>> + - user fault addresses
>
> While the kernel currently supports this in some way (by clearing the
> tag exception entry, el0_da), the above implies (at least to me) that
> sigcontext.fault_address would contain the tagged address. That's not
> the case (unless I missed it in your patches).
I'll update the doc to reflect this in v8.
>
>> - - pointer arguments to system calls, including pointers in structures
>> - passed to system calls,
>> + - pointer arguments (including pointers in structures), which don't
>> + describe virtual memory ranges, passed to system calls
>
> I think we need to be more precise here...
In what way?
>
>> +All other interpretations of userspace memory addresses by the kernel
>> +assume an address tag of 0x00. This includes, but is not limited to,
>> +addresses found in:
>> +
>> + - pointer arguments (including pointers in structures), which describe
>> + virtual memory ranges, passed to memory system calls (mmap, mprotect,
>> + etc.)
>
> ...and probably a full list here.
Will add a full list in v8.
Hi Andrey,
On 02/10/2018 14:12, Andrey Konovalov wrote:
> arm64 has a feature called Top Byte Ignore, which allows to embed pointer
> tags into the top byte of each pointer. Userspace programs (such as
> HWASan, a memory debugging tool [1]) might use this feature and pass
> tagged user pointers to the kernel through syscalls or other interfaces.
>
> Right now the kernel is already able to handle user faults with tagged
> pointers, due to these patches:
>
> 1. 81cddd65 ("arm64: traps: fix userspace cache maintenance emulation on a
> tagged pointer")
> 2. 7dcd9dd8 ("arm64: hw_breakpoint: fix watchpoint matching for tagged
> pointers")
> 3. 276e9327 ("arm64: entry: improve data abort handling of tagged
> pointers")
>
> When passing tagged pointers to syscalls, there's a special case of such a
> pointer being passed to one of the memory syscalls (mmap, mprotect, etc.).
> These syscalls don't do memory accesses but rather deal with memory
> ranges, hence an untagged pointer is better suited.
>
> This patchset extends tagged pointer support to non-memory syscalls. This
> is done by reusing the untagged_addr macro to untag user pointers when the
> kernel performs pointer checking to find out whether the pointer comes
> from userspace (most notably in access_ok).
>
> The following testing approaches has been taken to find potential issues
> with user pointer untagging:
>
> 1. Static testing (with sparse [2] and separately with a custom static
> analyzer based on Clang) to track casts of __user pointers to integer
> types to find places where untagging needs to be done.
>
> 2. Dynamic testing: adding BUG_ON(has_tag(addr)) to find_vma() and running
> a modified syzkaller version that passes tagged pointers to the kernel.
>
...
I have been thinking a bit lately on how to address the problem of user tagged pointers passed to the kernel through syscalls, and IMHO probably the best way we have to catch them all and make sure that the approach is maintainable in the long term is to introduce shims that tag/untag the pointers passed to the kernel.
In details, what I am proposing can live either in userspace (preferred solution so that we do not have to relax the ABI) or in kernel space and can be summarized as follows:
- A shim is specific to a syscall and is called by the libc when it needs to invoke the respective syscall.
- It is required only if the syscall accepts pointers.
- It saves the tags of a pointers passed to the syscall in memory (same approach if the we are passing a struct that contains pointers to the kernel, with the difference that all the tags of the pointers in the struct need to be saved singularly)
- Untags the pointers
- Invokes the syscall
- Retags the pointers with the tags stored in memory
- Returns
What do you think?
--
Regards,
Vincenzo
On Wed, Oct 17, 2018 at 4:06 PM, Vincenzo Frascino
<[email protected]> wrote:
> Hi Andrey,
> I have been thinking a bit lately on how to address the problem of user tagged pointers passed to the kernel through syscalls, and IMHO probably the best way we have to catch them all and make sure that the approach is maintainable in the long term is to introduce shims that tag/untag the pointers passed to the kernel.
>
> In details, what I am proposing can live either in userspace (preferred solution so that we do not have to relax the ABI) or in kernel space and can be summarized as follows:
> - A shim is specific to a syscall and is called by the libc when it needs to invoke the respective syscall.
> - It is required only if the syscall accepts pointers.
> - It saves the tags of a pointers passed to the syscall in memory (same approach if the we are passing a struct that contains pointers to the kernel, with the difference that all the tags of the pointers in the struct need to be saved singularly)
> - Untags the pointers
> - Invokes the syscall
> - Retags the pointers with the tags stored in memory
> - Returns
>
> What do you think?
Hi Vincenzo,
If I correctly understand what you are proposing, I'm not sure if that
would work with the countless number of different ioctl calls. For
example when an ioctl accepts a struct with a bunch of pointer fields.
In this case a shim like the one you propose can't live in userspace,
since libc doesn't know about the interface of all ioctls, so it can't
know which fields to untag. The kernel knows about those interfaces
(since the kernel implements them), but then we would need a custom
shim for each ioctl variation, which doesn't seem practical.
Thanks!
On Wed, Oct 17, 2018 at 7:20 AM, Andrey Konovalov <[email protected]> wrote:
> On Wed, Oct 17, 2018 at 4:06 PM, Vincenzo Frascino
> <[email protected]> wrote:
>> Hi Andrey,
>> I have been thinking a bit lately on how to address the problem of user tagged pointers passed to the kernel through syscalls, and IMHO probably the best way we have to catch them all and make sure that the approach is maintainable in the long term is to introduce shims that tag/untag the pointers passed to the kernel.
>>
>> In details, what I am proposing can live either in userspace (preferred solution so that we do not have to relax the ABI) or in kernel space and can be summarized as follows:
>> - A shim is specific to a syscall and is called by the libc when it needs to invoke the respective syscall.
>> - It is required only if the syscall accepts pointers.
>> - It saves the tags of a pointers passed to the syscall in memory (same approach if the we are passing a struct that contains pointers to the kernel, with the difference that all the tags of the pointers in the struct need to be saved singularly)
>> - Untags the pointers
>> - Invokes the syscall
>> - Retags the pointers with the tags stored in memory
>> - Returns
>>
>> What do you think?
>
> Hi Vincenzo,
>
> If I correctly understand what you are proposing, I'm not sure if that
> would work with the countless number of different ioctl calls. For
> example when an ioctl accepts a struct with a bunch of pointer fields.
> In this case a shim like the one you propose can't live in userspace,
> since libc doesn't know about the interface of all ioctls, so it can't
> know which fields to untag. The kernel knows about those interfaces
> (since the kernel implements them), but then we would need a custom
> shim for each ioctl variation, which doesn't seem practical.
The current patchset handles majority of pointers in a just a few
common places, like copy_from_user. Userspace shims will need to untag
& retag all pointer arguments - we are looking at hundreds if not
thousands of shims. They will also be located in a different code base
from the syscall / ioctl implementations, which would make them
impossible to keep up to date.
On Wed, Oct 10, 2018 at 04:09:25PM +0200, Andrey Konovalov wrote:
> On Wed, Oct 3, 2018 at 7:32 PM, Catalin Marinas <[email protected]> wrote:
> > On Tue, Oct 02, 2018 at 03:12:42PM +0200, Andrey Konovalov wrote:
[...]
> > Also, how is user space supposed to know that it can now pass tagged
> > pointers into the kernel? An ABI change (or relaxation), needs to be
> > advertised by the kernel, usually via a new HWCAP bit (e.g. HWCAP_TBI).
> > Once we have a HWCAP bit in place, we need to be pretty clear about
> > which syscalls can and cannot cope with tagged pointers. The "as of now"
> > implies potential further relaxation which, again, would need to be
> > advertised to user in some (additional) way.
>
> How exactly should I do that? Something like this [1]? Or is it only
> for hardware specific things and for this patchset I need to do
> something else?
>
> [1] https://github.com/torvalds/linux/commit/7206dc93a58fb76421c4411eefa3c003337bcb2d
Thinking some more on this, we should probably keep the HWCAP_* bits for
actual hardware features. Maybe someone else has a better idea (the
linux-abi list?). An option would be to make use of AT_FLAGS auxv
(currently 0) in Linux. I've seen some MIPS patches in the past but
nothing upstream.
Yet another option would be for the user to probe on some innocuous
syscall currently returning -EFAULT on tagged pointer arguments but I
don't particularly like this.
> >> - - pointer arguments to system calls, including pointers in structures
> >> - passed to system calls,
> >> + - pointer arguments (including pointers in structures), which don't
> >> + describe virtual memory ranges, passed to system calls
> >
> > I think we need to be more precise here...
>
> In what way?
In the way of being explicit about which syscalls support tagged
pointers, unless we find a good reason to support tagged pointers on all
syscalls and avoid any lists.
--
Catalin
On Wed, Oct 17, 2018 at 01:25:42PM -0700, Evgenii Stepanov wrote:
> On Wed, Oct 17, 2018 at 7:20 AM, Andrey Konovalov <[email protected]> wrote:
> > On Wed, Oct 17, 2018 at 4:06 PM, Vincenzo Frascino
> > <[email protected]> wrote:
> >> I have been thinking a bit lately on how to address the problem of
> >> user tagged pointers passed to the kernel through syscalls, and
> >> IMHO probably the best way we have to catch them all and make sure
> >> that the approach is maintainable in the long term is to introduce
> >> shims that tag/untag the pointers passed to the kernel.
> >>
> >> In details, what I am proposing can live either in userspace
> >> (preferred solution so that we do not have to relax the ABI) or in
> >> kernel space and can be summarized as follows:
> >> - A shim is specific to a syscall and is called by the libc when
> >> it needs to invoke the respective syscall.
> >> - It is required only if the syscall accepts pointers.
> >> - It saves the tags of a pointers passed to the syscall in memory
> >> (same approach if the we are passing a struct that contains
> >> pointers to the kernel, with the difference that all the tags of
> >> the pointers in the struct need to be saved singularly)
> >> - Untags the pointers
> >> - Invokes the syscall
> >> - Retags the pointers with the tags stored in memory
> >> - Returns
> >>
> >> What do you think?
> >
> > If I correctly understand what you are proposing, I'm not sure if that
> > would work with the countless number of different ioctl calls. For
> > example when an ioctl accepts a struct with a bunch of pointer fields.
> > In this case a shim like the one you propose can't live in userspace,
> > since libc doesn't know about the interface of all ioctls, so it can't
> > know which fields to untag. The kernel knows about those interfaces
> > (since the kernel implements them), but then we would need a custom
> > shim for each ioctl variation, which doesn't seem practical.
>
> The current patchset handles majority of pointers in a just a few
> common places, like copy_from_user. Userspace shims will need to untag
> & retag all pointer arguments - we are looking at hundreds if not
> thousands of shims. They will also be located in a different code base
> from the syscall / ioctl implementations, which would make them
> impossible to keep up to date.
I think ioctls are a good reason not to attempt such user-space shim
layer (though it would have been much easier for the kernel ;)).
--
Catalin
On 10/17/18 9:25 PM, Evgenii Stepanov wrote:
> On Wed, Oct 17, 2018 at 7:20 AM, Andrey Konovalov <[email protected]> wrote:
>> On Wed, Oct 17, 2018 at 4:06 PM, Vincenzo Frascino
>> <[email protected]> wrote:
>>> Hi Andrey,
>>> I have been thinking a bit lately on how to address the problem of user tagged pointers passed to the kernel through syscalls, and IMHO probably the best way we have to catch them all and make sure that the approach is maintainable in the long term is to introduce shims that tag/untag the pointers passed to the kernel.
>>>
>>> In details, what I am proposing can live either in userspace (preferred solution so that we do not have to relax the ABI) or in kernel space and can be summarized as follows:
>>> - A shim is specific to a syscall and is called by the libc when it needs to invoke the respective syscall.
>>> - It is required only if the syscall accepts pointers.
>>> - It saves the tags of a pointers passed to the syscall in memory (same approach if the we are passing a struct that contains pointers to the kernel, with the difference that all the tags of the pointers in the struct need to be saved singularly)
>>> - Untags the pointers
>>> - Invokes the syscall
>>> - Retags the pointers with the tags stored in memory
>>> - Returns
>>>
>>> What do you think?
>>
>> Hi Vincenzo,
>>
>> If I correctly understand what you are proposing, I'm not sure if that
>> would work with the countless number of different ioctl calls. For
>> example when an ioctl accepts a struct with a bunch of pointer fields.
>> In this case a shim like the one you propose can't live in userspace,
>> since libc doesn't know about the interface of all ioctls, so it can't
>> know which fields to untag. The kernel knows about those interfaces
>> (since the kernel implements them), but then we would need a custom
>> shim for each ioctl variation, which doesn't seem practical.
>
> The current patchset handles majority of pointers in a just a few
> common places, like copy_from_user. Userspace shims will need to untag
> & retag all pointer arguments - we are looking at hundreds if not
> thousands of shims. They will also be located in a different code base
> from the syscall / ioctl implementations, which would make them
> impossible to keep up to date.
>
I agree with both of you, ioctl is the real show stopper for this approach. Thanks for pointing this out.
--
Regards,
Vincenzo