tracing: Add back FORTIFY_SOURCE logic to kernel_stack event structure
Steven Rostedt
maintainerPATCH 1/1 tracing: Add back FORTIFY_SOURCE logic to kernel_stack event structure
From: "Steven Rostedt (Google)" <[email protected]>
For backward compatibility, older tooling expects to see the kernel_stack event with a "caller" field that is a fixed size array of 8 addresses. The code now supports more than 8 with an added "size" field that states the real number of entries. But the "caller" field still just looks like a fixed size to user space.
Since the tracing macros that create the user space format files also creates the structures that those files represent, the kernel_stack event structure had its "caller" field a fixed size of 8, but in reality, when it is allocated on the ring buffer, it can hold more if the stack trace is bigger that 8 functions. The copying of these entries was simply done with a memcpy():
size = nr_entries * sizeof(unsigned long); memcpy(entry->caller, fstack->calls, size);
The FORTIFY_SOURCE logic noticed at runtime that when the nr_entries was larger than 8, that the memcpy() was writing more than what the structure stated it can hold and it complained about it. This is because the FORTIFY_SOURCE code is unaware that the amount allocated is actually enough to hold the size. It does not expect that a fixed size field will hold more than the fixed size.
This was originally solved by hiding the caller assignment with some pointer arithmetic.
ptr = ring_buffer_data(); entry = ptr; ptr += offsetof(typeof(*entry), caller); memcpy(ptr, fstack->calls, size);
But it is considered bad form to hide from kernel hardening. Instead, make it work nicely with FORTIFY_SOURCE by adding a new __stack_array() macro that is specific for this one special use case. The macro will take 4
arguments: type, item, len, field (whereas the __array() macro takes just the first three). This macro will act just like the __array() macro when creating the code to deal with the format file that is exposed to user space. But for the kernel, it will turn the caller field into:
type item[] __counted_by(field);
or for this instance:
unsigned long caller[] __counted_by(size);
Now the kernel code can expose the assignment of the caller to the FORTIFY_SOURCE and everyone is happy!
--- kernel/trace/trace.c | 25 ++++--------------------- kernel/trace/trace.h | 10 ++++++++++ kernel/trace/trace_entries.h | 2 +- kernel/trace/trace_export.c | 9 +++++++++ 4 files changed, 24 insertions(+), 22 deletions(-)
kernel/trace/trace.c +4 -21
@@ -3118,7 +3118,6 @@ static void __ftrace_trace_stack(struct trace_buffer *buffer,struct ftrace_stack *fstack;struct stack_entry *entry;int stackidx; -void *ptr;/* * Add one, for this function and the call to save_stack_trace()
@@ -3156,32 +3155,16 @@ static void __ftrace_trace_stack(struct trace_buffer *buffer, nr_entries =stack_trace_save (fstack->calls, size, skip); } - size = nr_entries *sizeof (unsigned long ); event =__trace_buffer_lock_reserve (buffer,TRACE_STACK , - (sizeof (*entry) -sizeof (entry->caller)) + size, +struct_size (entry, caller, nr_entries), trace_ctx);if (!event) goto out; - ptr =ring_buffer_event_data (event); - entry = ptr; - -/* -* For backward compatibility reasons, the entry->caller is an -* array of 8 slots to store the stack. This is also exported -* to user space. The amount allocated on the ring buffer actually -* holds enough for the stack specified by nr_entries. This will -* go into the location of entry->caller. Due to string fortifiers -* checking the size of the destination of memcpy() it triggers -* when it detects that size is greater than 8. To hide this from -* the fortifiers, we use "ptr" and pointer arithmetic to assign caller. -* -* The below is really just: -* memcpy(&entry->caller, fstack->calls, size); -*/ - ptr += offsetof(typeof(*entry), caller); -memcpy (ptr, fstack->calls, size); + entry =ring_buffer_event_data (event); entry->size = nr_entries; +memcpy (&entry->caller, fstack->calls, +flex_array_size (entry, caller, nr_entries));if (!call_filter_check_discard (call, entry, buffer, event))__buffer_unlock_commit (buffer, event);
kernel/trace/trace.h +10 -0
@@ -77,6 +77,16 @@ enum trace_type {#undef __array#define __array (type, item, size) type item[size]; +/* +* For backward compatibility, older user space expects to see the +* kernel_stack event with a fixed size caller field. But today the fix +* size is ignored by the kernel, and the real structure is dynamic. +* Expose to user space: "unsigned long caller[8];" but the real structure +* will be "unsigned long caller[] __counted_by(size)" +*/ +#undef __stack_array +#define __stack_array (type, item, size, field) type item[] __counted_by(field); +#undef __array_desc#define __array_desc (type, container, item, size)
kernel/trace/trace_entries.h +1 -1
@@ -190,7 +190,7 @@ FTRACE_ENTRY(kernel_stack, stack_entry,F_STRUCT (__field (int ,size ) - __array( unsigned long, caller, FTRACE_STACK_ENTRIES ) + __stack_array( unsigned long, caller, FTRACE_STACK_ENTRIES, size) ), F_printk("\t=> %ps\n\t=> %ps\n\t=> %ps\n"
kernel/trace/trace_export.c +9 -0
@@ -51,6 +51,9 @@ static int ftrace_event_register(struct trace_event_call *call,#undef __array#define __array (type, item, size) type item[size]; +#undef __stack_array +#define __stack_array (type, item, size, field) __array(type, item, size) +#undef __array_desc#define __array_desc (type, container, item, size) type item[size];
@@ -114,6 +117,9 @@ static void __always_unused ____ftrace_check_##name(void) \ is_signed_type(_type), .filter_type = FILTER_OTHER, \ .len = _len }, +#undef __stack_array +#define __stack_array (_type, _item, _len, _field) __array(_type, _item, _len) +#undef __array_desc#define __array_desc (_type, _container, _item, _len) __array(_type, _item, _len)
@@ -149,6 +155,9 @@ static struct trace_event_fields ftrace_event_fields_##name[] = { \#undef __array#define __array (type, item, len) +#undef __stack_array +#define __stack_array (type, item, len, field) +#undef __array_desc#define __array_desc (type, container, item, len)
Signature · 1 line
2.40.1
Quoted 48 lines · Steven Rostedt
From: "Steven Rostedt (Google)" <[email protected]>
For backward compatibility, older tooling expects to see the
kernel_stackevent with a "caller" field that is a fixed size array of 8 addresses. The code now supports more than 8 with an added "size" field that states the real number of entries. But the "caller" field still just looks like a fixed size to user space.Since the tracing macros that create the user space format files also creates the structures that those files represent, the
kernel_stackevent structure had its "caller" field a fixed size of 8, but in reality, when it is allocated on the ring buffer, it can hold more if the stack trace is bigger that 8 functions. The copying of these entries was simply done with amemcpy():size = nr_entries * sizeof(unsigned long); memcpy(entry->caller, fstack->calls, size);The
FORTIFY_SOURCElogic noticed at runtime that when thenr_entrieswas larger than 8, that thememcpy()was writing more than what the structure stated it can hold and it complained about it. This is because theFORTIFY_SOURCEcode is unaware that the amount allocated is actually enough to hold the size. It does not expect that a fixed size field will hold more than the fixed size.This was originally solved by hiding the caller assignment with some pointer arithmetic.
ptr = ring_buffer_data(); entry = ptr; ptr += offsetof(typeof(*entry), caller); memcpy(ptr, fstack->calls, size);But it is considered bad form to hide from kernel hardening. Instead, make it work nicely with
FORTIFY_SOURCEby adding a new__stack_array()macro that is specific for this one special use case. The macro will take 4
arguments: type, item, len, field (whereas the__array()macro takes just the first three). This macro will act just like the__array()macro when creating the code to deal with the format file that is exposed to user space. But for the kernel, it will turn the caller field into:type item[] __counted_by(field);or for this instance:
unsigned long caller[] __counted_by(size);
Quoted 7 lines
Now the kernel code can expose the assignment of the caller to the
FORTIFY_SOURCEand everyone is happy!Suggested-by Kees Cook <[email protected]>Signed-off-by Steven Rostedt (Google) <[email protected]>
Yay! This looks good. :)
Signature · 1 line
Kees Cook