states.c:range_within() must be updated to properly check if
cnum-based range in an old state is a superset of a range in the cur
state. Currently it makes the decision using min/max accessors:
reg_umin(old) <= reg_umin(cur) <= reg_umax(old)
This is wrong for cnums that cross both UT_MAX/0 and ST_MAX/ST_MIN
boundaries. Consider cnum32{base=0x7FFFFFF0, size=0x80000020},
which represents values [0x7FFFFFF0, ..., U32_MAX, 0, ..., 0x10].
Its projections are u32_min/max=0/U32_MAX, s32_min/max=S32_MIN/MAX.
A register with range [0x100, 0x200] (which lies entirely in the gap
of the wrapping range) would pass the min/max check despite having no
overlap with the actual cnum arc.
This commit replaces min/max comparison with cnum{32,64}_is_subset()
operation. The operation implementation is verified using cbmc model
checker in [1].
[1] https://github.com/eddyz87/cnum-verif/
Fixes: bbc6310855 ("bpf: replace min/max fields with struct cnum{32,64}")
Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20260425-cnum-range-within-v1-1-2fdca70cb09d@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
This commit adds basic definitions for cnum32/cnum64.
This is a unified numeric range representation for signed and unsigned
domains. Inspired by an old post from Shung-Hsi Yu [1] and paper [2].
Operations correctness is verified using cbmc model checker,
tests source code can be found in a separate repo [3].
The cnum64_cnum32_intersect() function is notable, because it handled
several cases verifier.c:deduce_bounds_64_from_32() does not.
Given:
- a is a 64-bit range
- b is a 32-bit range
- t is a refined 64-bit range, such that ∀ v ∈ a, (u32)v ∈ b: v ∈ t.
cnum64_cnum32_intersect() makes the following deductions:
(A): 'b' is a sub-range of the first or the last 32-bit
sub-range of 'a':
64-bit number axis --->
N*2^32 (N+1)*2^32 (N+2)*2^32 (N+3)*2^32
||------|---|=====|-------||----------|=====|-------||----------|=====|----|--||
| |< b >| |< b >| |< b >| |
| | | |
|<--+--------------------------- a ---------------------------+--->|
| |
|<-------------------------- t -------------------------->|
(B) 'b' does not intersect with the first of the last 32-bit
sub-range of 'a':
N*2^32 (N+1)*2^32 (N+2)*2^32 (N+3)*2^32
||--|=====|----|----------||--|=====|---------------||--|=====|------------|--||
|< b >| | |< b >| |< b >| |
| | | |
|<-------------+--------- a -------------------|----------->|
| |
|<-------- t ------------------>|
(C) 'b' crosses 0/U32_MAX boundary:
N*2^32 (N+1)*2^32 (N+2)*2^32 (N+3)*2^32
||===|---------|------|===||===|----------------|===||===|---------|------|===||
|b >| | |< b||b >| |< b||b >| | |< b|
| | | |
|<-----+----------------- a --------------+-------->|
| |
|<---------------- t ------------->|
Current implementation of deduce_bounds_64_from_32() only handles
case (A).
[1] https://lore.kernel.org/all/ZTZxoDJJbX9mrQ9w@u94a/
[2] https://jorgenavas.github.io/papers/ACM-TOPLAS-wrapped.pdf
[3] https://github.com/eddyz87/cnum-verif/tree/master
Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20260424-cnums-everywhere-rfc-v1-v3-1-ca434b39a486@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
