[PATCH bpf v3 0/2] Fix kprobe_multi interface issues for 5.18

[PATCH bpf v3 0/2] Fix kprobe_multi interface issues for 5.18

[Eugene Syromiatnikov]

Hello.

While [1] seems to require additional work[2] due to changes
in the interface (and it has already been re-targeted for bpf-next),
I would like to ask to consider the following two patches, that fix
possible out-of-bounds write and properly disable the interface
for 32-bit compat user space, for the 5.18 release.  Thank you.

[1] https://lore.kernel.org/lkml/cover.1652772731.git.esyr@redhat.com/
[2] https://lore.kernel.org/lkml/YoTXiAk1EpZ0rLKE@krava/

Eugene Syromiatnikov (2):
  bpf_trace: check size for overflow in bpf_kprobe_multi_link_attach
  bpf_trace: bail out from bpf_kprobe_multi_link_attach when in compat

 kernel/trace/bpf_trace.c | 15 +++++++++------
 1 file changed, 9 insertions(+), 6 deletions(-)

-- 
2.1.4

[PATCH bpf v3 1/2] bpf_trace: check size for overflow in bpf_kprobe_multi_link_attach

[Eugene Syromiatnikov]

Check that size would not overflow before calculation (and return
-EOVERFLOW if it will), to prevent potential out-of-bounds write
with the following copy_from_user.  Add the same check
to kprobe_multi_resolve_syms in case it will be called from elsewhere
in the future.

Fixes: 0dcac272540613d4 ("bpf: Add multi kprobe link")
Signed-off-by: Eugene Syromiatnikov <esyr@redhat.com>
---
 kernel/trace/bpf_trace.c | 13 ++++++++-----
 1 file changed, 8 insertions(+), 5 deletions(-)

diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c
index d8553f4..212faa4 100644
--- a/kernel/trace/bpf_trace.c
+++ b/kernel/trace/bpf_trace.c
@@ -2352,13 +2352,15 @@ static int
 kprobe_multi_resolve_syms(const void __user *usyms, u32 cnt,
 			  unsigned long *addrs)
 {
-	unsigned long addr, size;
+	unsigned long addr, sym_size;
+	u32 size;
 	const char __user **syms;
 	int err = -ENOMEM;
 	unsigned int i;
 	char *func;
 
-	size = cnt * sizeof(*syms);
+	if (check_mul_overflow(cnt, (u32)sizeof(*syms), &size))
+		return -EOVERFLOW;
 	syms = kvzalloc(size, GFP_KERNEL);
 	if (!syms)
 		return -ENOMEM;
@@ -2382,9 +2384,9 @@ kprobe_multi_resolve_syms(const void __user *usyms, u32 cnt,
 		addr = kallsyms_lookup_name(func);
 		if (!addr)
 			goto error;
-		if (!kallsyms_lookup_size_offset(addr, &size, NULL))
+		if (!kallsyms_lookup_size_offset(addr, &sym_size, NULL))
 			goto error;
-		addr = ftrace_location_range(addr, addr + size - 1);
+		addr = ftrace_location_range(addr, addr + sym_size - 1);
 		if (!addr)
 			goto error;
 		addrs[i] = addr;
@@ -2429,7 +2431,8 @@ int bpf_kprobe_multi_link_attach(const union bpf_attr *attr, struct bpf_prog *pr
 	if (!cnt)
 		return -EINVAL;
 
-	size = cnt * sizeof(*addrs);
+	if (check_mul_overflow(cnt, (u32)sizeof(*addrs), &size))
+		return -EOVERFLOW;
 	addrs = kvmalloc(size, GFP_KERNEL);
 	if (!addrs)
 		return -ENOMEM;
-- 
2.1.4

[PATCH bpf v3 2/2] bpf_trace: bail out from bpf_kprobe_multi_link_attach when in compat

[Eugene Syromiatnikov]

Since bpf_kprobe_multi_link_attach doesn't support 32-bit kernels
for whatever reason, having it enabled for compat processes on 64-bit
kernels makes even less sense due to discrepances in the type sizes
that it does not handle.

Fixes: 0dcac272540613d4 ("bpf: Add multi kprobe link")
Signed-off-by: Eugene Syromiatnikov <esyr@redhat.com>
---
 kernel/trace/bpf_trace.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c
index 212faa4..2f83489 100644
--- a/kernel/trace/bpf_trace.c
+++ b/kernel/trace/bpf_trace.c
@@ -2412,7 +2412,7 @@ int bpf_kprobe_multi_link_attach(const union bpf_attr *attr, struct bpf_prog *pr
 	int err;
 
 	/* no support for 32bit archs yet */
-	if (sizeof(u64) != sizeof(void *))
+	if (sizeof(u64) != sizeof(void *) || in_compat_syscall())
 		return -EOPNOTSUPP;
 
 	if (prog->expected_attach_type != BPF_TRACE_KPROBE_MULTI)
-- 
2.1.4

Re: [PATCH bpf v3 1/2] bpf_trace: check size for overflow in bpf_kprobe_multi_link_attach

[Yonghong Song]

On 5/18/22 5:22 AM, Eugene Syromiatnikov wrote:
> Check that size would not overflow before calculation (and return
> -EOVERFLOW if it will), to prevent potential out-of-bounds write
> with the following copy_from_user.  Add the same check
> to kprobe_multi_resolve_syms in case it will be called from elsewhere
> in the future.
> 
> Fixes: 0dcac272540613d4 ("bpf: Add multi kprobe link")
> Signed-off-by: Eugene Syromiatnikov <esyr@redhat.com>
> ---
>   kernel/trace/bpf_trace.c | 13 ++++++++-----
>   1 file changed, 8 insertions(+), 5 deletions(-)
> 
> diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c
> index d8553f4..212faa4 100644
> --- a/kernel/trace/bpf_trace.c
> +++ b/kernel/trace/bpf_trace.c
> @@ -2352,13 +2352,15 @@ static int
>   kprobe_multi_resolve_syms(const void __user *usyms, u32 cnt,
>   			  unsigned long *addrs)
>   {
> -	unsigned long addr, size;
> +	unsigned long addr, sym_size;
> +	u32 size;
>   	const char __user **syms;
>   	int err = -ENOMEM;
>   	unsigned int i;
>   	char *func;
>   
> -	size = cnt * sizeof(*syms);
> +	if (check_mul_overflow(cnt, (u32)sizeof(*syms), &size))
> +		return -EOVERFLOW;

In mm/util.c kvmalloc_node(), we have

         /* Don't even allow crazy sizes */
         if (unlikely(size > INT_MAX)) {
                 WARN_ON_ONCE(!(flags & __GFP_NOWARN));
                 return NULL;
         }

Basically the maximum size to be allocated in INT_MAX.

Here, we have 'size' as u32, which means if the size is 0xffff0000,
the check_mul_overflow will return false (no overflow) but
kvzalloc will still have a warning.

I think we should change the type of 'size' to be 'int' which
should catch the above case and be consistent with
what kvmalloc_node() intends to warn.

>   	syms = kvzalloc(size, GFP_KERNEL);
>   	if (!syms)
>   		return -ENOMEM;
> @@ -2382,9 +2384,9 @@ kprobe_multi_resolve_syms(const void __user *usyms, u32 cnt,
>   		addr = kallsyms_lookup_name(func);
>   		if (!addr)
>   			goto error;
> -		if (!kallsyms_lookup_size_offset(addr, &size, NULL))
> +		if (!kallsyms_lookup_size_offset(addr, &sym_size, NULL))
>   			goto error;
> -		addr = ftrace_location_range(addr, addr + size - 1);
> +		addr = ftrace_location_range(addr, addr + sym_size - 1);
>   		if (!addr)
>   			goto error;
>   		addrs[i] = addr;
> @@ -2429,7 +2431,8 @@ int bpf_kprobe_multi_link_attach(const union bpf_attr *attr, struct bpf_prog *pr
>   	if (!cnt)
>   		return -EINVAL;
>   
> -	size = cnt * sizeof(*addrs);
> +	if (check_mul_overflow(cnt, (u32)sizeof(*addrs), &size))
> +		return -EOVERFLOW;
>   	addrs = kvmalloc(size, GFP_KERNEL);
>   	if (!addrs)
>   		return -ENOMEM;

Re: [PATCH bpf v3 2/2] bpf_trace: bail out from bpf_kprobe_multi_link_attach when in compat

[Yonghong Song]

On 5/18/22 5:22 AM, Eugene Syromiatnikov wrote:
> Since bpf_kprobe_multi_link_attach doesn't support 32-bit kernels
> for whatever reason, having it enabled for compat processes on 64-bit
> kernels makes even less sense due to discrepances in the type sizes
> that it does not handle.

If I understand correctly, the reason is due to
in libbpf we have
struct bpf_link_create_opts {
         size_t sz; /* size of this struct for forward/backward 
compatibility */
         __u32 flags;
         union bpf_iter_link_info *iter_info;
         __u32 iter_info_len;
         __u32 target_btf_id;
         union {
                 struct {
                         __u64 bpf_cookie;
                 } perf_event;
                 struct {
                         __u32 flags;
                         __u32 cnt;
                         const char **syms;
                         const unsigned long *addrs;
                         const __u64 *cookies;
                 } kprobe_multi;
         };
         size_t :0;
};

Note that we have `const unsigned long *addrs;`

If we have 32-bit user space application and 64bit kernel,
and we will have userspace 32-bit pointers and kernel as
64bit pointers and current kernel doesn't handle 32-bit
user pointer properly.

Consider this may involve libbpf uapi change, maybe
we should change "const unsigned long *addrs;" to
"const __u64 *addrs;" considering we haven't freeze
libbpf UAPI yet.

Otherwise, we stick to current code with this patch,
it will make it difficult to support 32-bit app with
64-bit kernel for kprobe_multi in the future due to
uapi issues.

WDYT?

> 
> Fixes: 0dcac272540613d4 ("bpf: Add multi kprobe link")
> Signed-off-by: Eugene Syromiatnikov <esyr@redhat.com>
> ---
>   kernel/trace/bpf_trace.c | 2 +-
>   1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c
> index 212faa4..2f83489 100644
> --- a/kernel/trace/bpf_trace.c
> +++ b/kernel/trace/bpf_trace.c
> @@ -2412,7 +2412,7 @@ int bpf_kprobe_multi_link_attach(const union bpf_attr *attr, struct bpf_prog *pr
>   	int err;
>   
>   	/* no support for 32bit archs yet */
> -	if (sizeof(u64) != sizeof(void *))
> +	if (sizeof(u64) != sizeof(void *) || in_compat_syscall())
>   		return -EOPNOTSUPP;
>   
>   	if (prog->expected_attach_type != BPF_TRACE_KPROBE_MULTI)

Re: [PATCH bpf v3 1/2] bpf_trace: check size for overflow in bpf_kprobe_multi_link_attach

[Eugene Syromiatnikov]

On Wed, May 18, 2022 at 09:34:22AM -0700, Yonghong Song wrote:
> On 5/18/22 5:22 AM, Eugene Syromiatnikov wrote:
> >-	size = cnt * sizeof(*syms);
> >+	if (check_mul_overflow(cnt, (u32)sizeof(*syms), &size))
> >+		return -EOVERFLOW;
> 
> In mm/util.c kvmalloc_node(), we have
> 
>         /* Don't even allow crazy sizes */
>         if (unlikely(size > INT_MAX)) {
>                 WARN_ON_ONCE(!(flags & __GFP_NOWARN));
>                 return NULL;
>         }
> 
> Basically the maximum size to be allocated in INT_MAX.
> 
> Here, we have 'size' as u32, which means if the size is 0xffff0000,
> the check_mul_overflow will return false (no overflow) but
> kvzalloc will still have a warning.
> 
> I think we should change the type of 'size' to be 'int' which
> should catch the above case and be consistent with
> what kvmalloc_node() intends to warn.

Huh, it's a bitmore complicated as check_mul_overflow requires types to
match; what do you think about

+	if (check_mul_overflow(cnt, (u32)sizeof(*syms), &size) || size > INT_MAX)

?

Re: [PATCH bpf v3 2/2] bpf_trace: bail out from bpf_kprobe_multi_link_attach when in compat

[Eugene Syromiatnikov]

On Wed, May 18, 2022 at 09:55:05AM -0700, Yonghong Song wrote:
> 
> 
> On 5/18/22 5:22 AM, Eugene Syromiatnikov wrote:
> >Since bpf_kprobe_multi_link_attach doesn't support 32-bit kernels
> >for whatever reason, having it enabled for compat processes on 64-bit
> >kernels makes even less sense due to discrepances in the type sizes
> >that it does not handle.
> 
> If I understand correctly, the reason is due to
> in libbpf we have
> struct bpf_link_create_opts {
>         size_t sz; /* size of this struct for forward/backward compatibility
> */
>         __u32 flags;
>         union bpf_iter_link_info *iter_info;
>         __u32 iter_info_len;
>         __u32 target_btf_id;
>         union {
>                 struct {
>                         __u64 bpf_cookie;
>                 } perf_event;
>                 struct {
>                         __u32 flags;
>                         __u32 cnt;
>                         const char **syms;
>                         const unsigned long *addrs;
>                         const __u64 *cookies;
>                 } kprobe_multi;
>         };
>         size_t :0;
> };
> 
> Note that we have `const unsigned long *addrs;`
> 
> If we have 32-bit user space application and 64bit kernel,
> and we will have userspace 32-bit pointers and kernel as
> 64bit pointers and current kernel doesn't handle 32-bit
> user pointer properly.
> 
> Consider this may involve libbpf uapi change, maybe
> we should change "const unsigned long *addrs;" to
> "const __u64 *addrs;" considering we haven't freeze
> libbpf UAPI yet.
> 
> Otherwise, we stick to current code with this patch,
> it will make it difficult to support 32-bit app with
> 64-bit kernel for kprobe_multi in the future due to
> uapi issues.
> 
> WDYT?

As 32 bit arches are "unsupported" currently, the change would be more
a semantic one rather then practical;  I don't mind having it here (basically,
the tools/* part of [1]), though (assuming it is still possible to get it
in 5.18).

[1] https://lore.kernel.org/lkml/6ef675aeeea442fa8fc168cd1cb4e4e474f65a3f.1652772731.git.esyr@redhat.com/

> >
> >Fixes: 0dcac272540613d4 ("bpf: Add multi kprobe link")
> >Signed-off-by: Eugene Syromiatnikov <esyr@redhat.com>
> >---
> >  kernel/trace/bpf_trace.c | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> >
> >diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c
> >index 212faa4..2f83489 100644
> >--- a/kernel/trace/bpf_trace.c
> >+++ b/kernel/trace/bpf_trace.c
> >@@ -2412,7 +2412,7 @@ int bpf_kprobe_multi_link_attach(const union bpf_attr *attr, struct bpf_prog *pr
> >  	int err;
> >  	/* no support for 32bit archs yet */
> >-	if (sizeof(u64) != sizeof(void *))
> >+	if (sizeof(u64) != sizeof(void *) || in_compat_syscall())
> >  		return -EOPNOTSUPP;
> >  	if (prog->expected_attach_type != BPF_TRACE_KPROBE_MULTI)
> 

Re: [PATCH bpf v3 1/2] bpf_trace: check size for overflow in bpf_kprobe_multi_link_attach

[Yonghong Song]

On 5/18/22 1:00 PM, Eugene Syromiatnikov wrote:
> On Wed, May 18, 2022 at 09:34:22AM -0700, Yonghong Song wrote:
>> On 5/18/22 5:22 AM, Eugene Syromiatnikov wrote:
>>> -	size = cnt * sizeof(*syms);
>>> +	if (check_mul_overflow(cnt, (u32)sizeof(*syms), &size))
>>> +		return -EOVERFLOW;
>>
>> In mm/util.c kvmalloc_node(), we have
>>
>>          /* Don't even allow crazy sizes */
>>          if (unlikely(size > INT_MAX)) {
>>                  WARN_ON_ONCE(!(flags & __GFP_NOWARN));
>>                  return NULL;
>>          }
>>
>> Basically the maximum size to be allocated in INT_MAX.
>>
>> Here, we have 'size' as u32, which means if the size is 0xffff0000,
>> the check_mul_overflow will return false (no overflow) but
>> kvzalloc will still have a warning.
>>
>> I think we should change the type of 'size' to be 'int' which
>> should catch the above case and be consistent with
>> what kvmalloc_node() intends to warn.
> 
> Huh, it's a bitmore complicated as check_mul_overflow requires types to
> match; what do you think about
> 
> +	if (check_mul_overflow(cnt, (u32)sizeof(*syms), &size) || size > INT_MAX)
> 
> ?

This works for me.

Re: [PATCH bpf v3 2/2] bpf_trace: bail out from bpf_kprobe_multi_link_attach when in compat

[Yonghong Song]

On 5/18/22 1:03 PM, Eugene Syromiatnikov wrote:
> On Wed, May 18, 2022 at 09:55:05AM -0700, Yonghong Song wrote:
>>
>>
>> On 5/18/22 5:22 AM, Eugene Syromiatnikov wrote:
>>> Since bpf_kprobe_multi_link_attach doesn't support 32-bit kernels
>>> for whatever reason, having it enabled for compat processes on 64-bit
>>> kernels makes even less sense due to discrepances in the type sizes
>>> that it does not handle.
>>
>> If I understand correctly, the reason is due to
>> in libbpf we have
>> struct bpf_link_create_opts {
>>          size_t sz; /* size of this struct for forward/backward compatibility
>> */
>>          __u32 flags;
>>          union bpf_iter_link_info *iter_info;
>>          __u32 iter_info_len;
>>          __u32 target_btf_id;
>>          union {
>>                  struct {
>>                          __u64 bpf_cookie;
>>                  } perf_event;
>>                  struct {
>>                          __u32 flags;
>>                          __u32 cnt;
>>                          const char **syms;
>>                          const unsigned long *addrs;
>>                          const __u64 *cookies;
>>                  } kprobe_multi;
>>          };
>>          size_t :0;
>> };
>>
>> Note that we have `const unsigned long *addrs;`
>>
>> If we have 32-bit user space application and 64bit kernel,
>> and we will have userspace 32-bit pointers and kernel as
>> 64bit pointers and current kernel doesn't handle 32-bit
>> user pointer properly.
>>
>> Consider this may involve libbpf uapi change, maybe
>> we should change "const unsigned long *addrs;" to
>> "const __u64 *addrs;" considering we haven't freeze
>> libbpf UAPI yet.
>>
>> Otherwise, we stick to current code with this patch,
>> it will make it difficult to support 32-bit app with
>> 64-bit kernel for kprobe_multi in the future due to
>> uapi issues.
>>
>> WDYT?
> 
> As 32 bit arches are "unsupported" currently, the change would be more
> a semantic one rather then practical;  I don't mind having it here (basically,
> the tools/* part of [1]), though (assuming it is still possible to get it
> in 5.18).
> 
> [1] https://lore.kernel.org/lkml/6ef675aeeea442fa8fc168cd1cb4e4e474f65a3f.1652772731.git.esyr@redhat.com/

I think for patch [1], we only need libbpf and selftest change, no
kernel change is needed since we
explicitly does not support 32bit kernel in the
beginning of function bpf_kprobe_multi_link_attach():

         /* no support for 32bit archs yet */
         if (sizeof(u64) != sizeof(void *))
                 return -EOPNOTSUPP;

and in kernel, address (pointer) size will be considered
long (64bit) which is exactly the libbpf change did that.

> 
>>>
>>> Fixes: 0dcac272540613d4 ("bpf: Add multi kprobe link")
>>> Signed-off-by: Eugene Syromiatnikov <esyr@redhat.com>
>>> ---
>>>   kernel/trace/bpf_trace.c | 2 +-
>>>   1 file changed, 1 insertion(+), 1 deletion(-)
>>>
>>> diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c
>>> index 212faa4..2f83489 100644
>>> --- a/kernel/trace/bpf_trace.c
>>> +++ b/kernel/trace/bpf_trace.c
>>> @@ -2412,7 +2412,7 @@ int bpf_kprobe_multi_link_attach(const union bpf_attr *attr, struct bpf_prog *pr
>>>   	int err;
>>>   	/* no support for 32bit archs yet */
>>> -	if (sizeof(u64) != sizeof(void *))
>>> +	if (sizeof(u64) != sizeof(void *) || in_compat_syscall())
>>>   		return -EOPNOTSUPP;
>>>   	if (prog->expected_attach_type != BPF_TRACE_KPROBE_MULTI)
>>
>