bpf: support to track BPF_JNE
Subsystems (4)
v3 cover bpf: support to track BPF_JNE
For now, the reg bounds is not handled for BPF_JNE case, which can cause the failure of following case:
/* The type of "a" is u16 */
if (a > 0 && a < 100) {
/* the range of the register for a is [0, 99], not [1, 99],
* and will cause the following error:
*
* invalid zero-sized read
*
* as a can be 0.
*/
bpf_skb_store_bytes(skb, xx, xx, a, 0);}
In the code above, "a > 0" will be compiled to "jmp xxx if a == 0". In the TRUE branch, the dst_reg will be marked as known to 0. However, in the fallthrough(FALSE) branch, the dst_reg will not be handled, which makes the [min, max] for a is [0, 99], not [1, 99].
In the 1st patch, we reduce the range of the dst reg if the src reg is a const and is exactly the edge of the dst reg For BPF_JNE.
In the 2nd patch, we just activate the test case for this logic in range_cond(), which is committed by Andrii in the
commit 8863238993e2 ("selftests/bpf: BPF register range bounds tester").
Changes since v2:
- fix a typo in the subject of the 1st patch
- add some comments to the 1st patch, as Eduard advised
- add some cases to the "crafted_cases"
Changes since v1:
- simplify the code in the 1st patch
- introduce the 2nd patch for the testing
Menglong Dong (2):
bpf: make the verifier tracks the "not equal" for regs selftests/bpf: activate the OP_NE login in range_cond()
kernel/bpf/verifier.c | 38 ++++++++++++++++++- .../selftests/bpf/prog_tests/reg_bounds.c | 25 +++++++++--- 2 files changed, 56 insertions(+), 7 deletions(-)
Signature · 1 line
2.39.2
PATCH v3 1/2 bpf: make the verifier tracks the "not equal" for regs
We can derive some new information for BPF_JNE in regs_refine_cond_op(). Take following code for example:
/* The type of "a" is u16 */
if (a > 0 && a < 100) {
/* the range of the register for a is [0, 99], not [1, 99],
* and will cause the following error:
*
* invalid zero-sized read
*
* as a can be 0.
*/
bpf_skb_store_bytes(skb, xx, xx, a, 0);}
In the code above, "a > 0" will be compiled to "jmp xxx if a == 0". In the TRUE branch, the dst_reg will be marked as known to 0. However, in the fallthrough(FALSE) branch, the dst_reg will not be handled, which makes the [min, max] for a is [0, 99], not [1, 99].
For BPF_JNE, we can reduce the range of the dst reg if the src reg is a const and is exactly the edge of the dst reg.
---
v2:
- fix a typo in the subject
- add some comments, as Eduard advised
--- kernel/bpf/verifier.c | 38 +++++++++++++++++++++++++++++++++++++- 1 file changed, 37 insertions(+), 1 deletion(-)
kernel/bpf/verifier.c +37 -1
@@ -14332,7 +14332,43 @@ static void regs_refine_cond_op(struct bpf_reg_state *reg1, struct bpf_reg_state }break ;case BPF_JNE : -/* we don't derive any new information for inequality yet */ +if (!is_reg_const (reg2, is_jmp32)) +swap (reg1, reg2); +if (!is_reg_const (reg2, is_jmp32)) +break ; + +/* try to recompute the bound of reg1 if reg2 is a const and +* is exactly the edge of reg1. +*/ + val =reg_const_value (reg2, is_jmp32); +if (is_jmp32) { +/* u32_min_value is not equal to 0xffffffff at this point, +* because otherwise u32_max_value is 0xffffffff as well, +* in such a case both reg1 and reg2 would be constants, +* jump would be predicted and reg_set_min_max() won't +* be called. +* +* Same reasoning works for all {u,s}{min,max}{32,64} cases +* below. +*/ +if (reg1->u32_min_value == (u32 )val) + reg1->u32_min_value++; +if (reg1->u32_max_value == (u32 )val) + reg1->u32_max_value--; +if (reg1->s32_min_value == (s32 )val) + reg1->s32_min_value++; +if (reg1->s32_max_value == (s32 )val) + reg1->s32_max_value--; + }else { +if (reg1->umin_value == (u64 )val) + reg1->umin_value++; +if (reg1->umax_value == (u64 )val) + reg1->umax_value--; +if (reg1->smin_value == (s64 )val) + reg1->smin_value++; +if (reg1->smax_value == (s64 )val) + reg1->smax_value--; + }break ;case BPF_JSET :if (!is_reg_const (reg2, is_jmp32))
Signature · 1 line
2.39.2
Re: [PATCH bpf-next v3 1/2] bpf: make the verifier tracks the "not equal" for regs
Quoted 20 lines · Menglong Dong
We can derive some new information for
BPF_JNEinregs_refine_cond_op(). Take following code for example:/* The type of "a" is u16 */
if (a > 0 && a < 100) { /* the range of the register for a is [0, 99], not [1, 99], * and will cause the following error: * * invalid zero-sized read * * as a can be 0. */ bpf_skb_store_bytes(skb, xx, xx, a, 0);}
In the code above, "a > 0" will be compiled to "jmp xxx if a == 0". In the TRUE branch, the
dst_regwill be marked as known to 0. However, in the fallthrough(FALSE) branch, thedst_regwill not be handled, which makes the [min, max] for a is [0, 99], not [1, 99].
Quoted 4 lines
For
BPF_JNE, we can reduce the range of the dst reg if the src reg is a const and is exactly the edge of the dst reg.Signed-off-by Menglong Dong <[email protected]>
Re: [PATCH bpf-next v3 1/2] bpf: make the verifier tracks the "not equal" for regs
In reply to Menglong DongQuoted 14 lines · Menglong Dong
We can derive some new information for
BPF_JNEinregs_refine_cond_op(). Take following code for example:/* The type of "a" is u16 */
if (a > 0 && a < 100) { /* the range of the register for a is [0, 99], not [1, 99], * and will cause the following error: * * invalid zero-sized read * * as a can be 0. */ bpf_skb_store_bytes(skb, xx, xx, a, 0);
Quoted 1 line
}
Please craft a selftest from above with inline asm (C might not work as compiler might optimize it)
Also we call:
/* fallthrough (FALSE) branch */
regs_refine_cond_op(false_reg1, false_reg2,
rev_opcode(opcode), is_jmp32);
/* jump (TRUE) branch */
regs_refine_cond_op(true_reg1, true_reg2, opcode, is_jmp32);so despite BPF_JNE is not handled explicitly it still should have caught above due to rev_opcode() ?
Re: [PATCH bpf-next v3 1/2] bpf: make the verifier tracks the "not equal" for regs
Hello,
Quoted 18 lines · Alexei Starovoitov
Quoted 14 lines · Menglong Dong
We can derive some new information for
BPF_JNEinregs_refine_cond_op(). Take following code for example:/* The type of "a" is u16 */
if (a > 0 && a < 100) { /* the range of the register for a is [0, 99], not [1, 99], * and will cause the following error: * * invalid zero-sized read * * as a can be 0. */ bpf_skb_store_bytes(skb, xx, xx, a, 0);Quoted 1 line
}
Quoted 2 lines
Please craft a selftest from above with inline asm (C might not work as compiler might optimize it)
Okay! Should I add this selftests to reg_bounds as a subtest, or add a "verifier_reg_edge.c" for verifier testing?
Quoted 9 lines
Also we call:
/* fallthrough (FALSE) branch */ regs_refine_cond_op(false_reg1, false_reg2, rev_opcode(opcode), is_jmp32); /* jump (TRUE) branch */ regs_refine_cond_op(true_reg1, true_reg2, opcode, is_jmp32);so despite
BPF_JNEis not handled explicitly it still should have caught above due torev_opcode()?
Ennn.....I'm a little confused. In this case, the TRUE path is handled properly, as the opcode is BPF_JEQ; and the FALSE is not handled properly, as the opcode is rev_opcode(BPF_JEQ), which is BPF_JNE. And the bpf_skb_store_bytes() will be called in the FALSE path. The origin state of false_reg* should be the same as true_reg*.
Re: [PATCH bpf-next v3 1/2] bpf: make the verifier tracks the "not equal" for regs
Quoted 23 lines · Menglong Dong
Hello,
Quoted 18 lines · Alexei Starovoitov
Quoted 15 lines · Menglong Dong
We can derive some new information for
BPF_JNEinregs_refine_cond_op(). Take following code for example:/* The type of "a" is u16 */
if (a > 0 && a < 100) { /* the range of the register for a is [0, 99], not [1, 99], * and will cause the following error: * * invalid zero-sized read * * as a can be 0. */ bpf_skb_store_bytes(skb, xx, xx, a, 0);}
Quoted 5 lines
Quoted 2 lines
Please craft a selftest from above with inline asm (C might not work as compiler might optimize it)
Okay! Should I add this selftests to
reg_boundsas a subtest, or add a "verifier_reg_edge.c" for verifier testing?
reg_bounds is for automated.
I think it will fit fine in the existing progs/verifier_bounds.c
Quoted 11 lines
Quoted 10 lines
Also we call:
/* fallthrough (FALSE) branch */ regs_refine_cond_op(false_reg1, false_reg2, rev_opcode(opcode), is_jmp32); /* jump (TRUE) branch */ regs_refine_cond_op(true_reg1, true_reg2, opcode, is_jmp32);so despite
BPF_JNEis not handled explicitly it still should have caught above due torev_opcode()?
Quoted 6 lines
Ennn.....I'm a little confused. In this case, the TRUE path is handled properly, as the opcode is
BPF_JEQ; and the FALSE is not handled properly, as the opcode isrev_opcode(BPF_JEQ), which isBPF_JNE. And thebpf_skb_store_bytes()will be called in the FALSE path. The origin state of false_reg* should be the same as true_reg*.
Ahh. I see.
It wasn't clear how 'a > 0 && a < 100' got to be JNE after optimizations.
Re: [PATCH bpf-next v3 1/2] bpf: make the verifier tracks the "not equal" for regs
In reply to Menglong DongQuoted 24 lines · Menglong Dong
We can derive some new information for
BPF_JNEinregs_refine_cond_op(). Take following code for example:/* The type of "a" is u16 */
if (a > 0 && a < 100) { /* the range of the register for a is [0, 99], not [1, 99], * and will cause the following error: * * invalid zero-sized read * * as a can be 0. */ bpf_skb_store_bytes(skb, xx, xx, a, 0);}
In the code above, "a > 0" will be compiled to "jmp xxx if a == 0". In the TRUE branch, the
dst_regwill be marked as known to 0. However, in the fallthrough(FALSE) branch, thedst_regwill not be handled, which makes the [min, max] for a is [0, 99], not [1, 99].For
BPF_JNE, we can reduce the range of the dst reg if the src reg is a const and is exactly the edge of the dst reg.
Quoted 9 lines
Signed-off-by Menglong Dong <[email protected]>---
v2:
- fix a typo in the subject
- add some comments, as Eduard advised--- kernel/bpf/verifier.c | 38 +++++++++++++++++++++++++++++++++++++- 1 file changed, 37 insertions(+), 1 deletion(-)
The logic looks good
Quoted 52 lines
kernel/bpf/verifier.c +37
-2@@ -14332,7 +14332,43 @@ static void regs_refine_cond_op(struct bpf_reg_state *reg1, struct bpf_reg_state }break ;case BPF_JNE : -/* we don't derive any new information for inequality yet */ +if (!is_reg_const (reg2, is_jmp32)) +swap (reg1, reg2); +if (!is_reg_const (reg2, is_jmp32)) +break ; + +/* try to recompute the bound of reg1 if reg2 is a const and +* is exactly the edge of reg1. +*/ + val =reg_const_value (reg2, is_jmp32); +if (is_jmp32) { +/* u32_min_value is not equal to 0xffffffff at this point, +* because otherwise u32_max_value is 0xffffffff as well, +* in such a case both reg1 and reg2 would be constants, +* jump would be predicted and reg_set_min_max() won't +* be called. +* +* Same reasoning works for all {u,s}{min,max}{32,64} cases +* below. +*/ +if (reg1->u32_min_value == (u32 )val) + reg1->u32_min_value++; +if (reg1->u32_max_value == (u32 )val) + reg1->u32_max_value--; +if (reg1->s32_min_value == (s32 )val) + reg1->s32_min_value++; +if (reg1->s32_max_value == (s32 )val) + reg1->s32_max_value--; + }else { +if (reg1->umin_value == (u64 )val) + reg1->umin_value++; +if (reg1->umax_value == (u64 )val) + reg1->umax_value--; +if (reg1->smin_value == (s64 )val) + reg1->smin_value++; +if (reg1->smax_value == (s64 )val) + reg1->smax_value--; + }break ;case BPF_JSET :if (!is_reg_const (reg2, is_jmp32)) --2.39.2
PATCH v3 2/2 selftests/bpf: activate the OP_NE login in range_cond()
In reply to Menglong DongThe edge range checking for the registers is supported by the verifier now, so we can activate the extended login in tools/testing/selftests/bpf/prog_tests/reg_bounds.c/range_cond() to test such logic.
Besides, I added some cases to the "crafted_cases" array for this logic. These cases are mainly used to test the edge of the src reg and dst reg.
---
v2:
- add some cases to the "crafted_cases"
--- .../selftests/bpf/prog_tests/reg_bounds.c | 25 ++++++++++++++----- 1 file changed, 19 insertions(+), 6 deletions(-)
tools/testing/selftests/bpf/prog_tests/reg_bounds.c +19 -6
@@ -590,12 +590,7 @@ static void range_cond(enum num_t t, struct range x, struct range y, *newy =range (t,max_t (t, x.a, y.a),min_t (t, x.b, y.b));break ;case OP_NE : -/* generic case, can't derive more information */ - *newx =range (t, x.a, x.b); - *newy =range (t, y.a, y.b); -break ; - -/* below extended logic is not supported by verifier just yet */ +/* below logic is supported by the verifier now */ if (x.a == x.b && x.a == y.a) {/* X is a constant matching left side of Y */ *newx =range (t, x.a, x.b);
@@ -2101,6 +2096,24 @@ static struct subtest_case crafted_cases[] = { {S32, S64, {(u32)(s32)S32_MIN, (u32)(s32)-255}, {(u32)(s32)-2, 0}}, {S32, S64, {0, 1}, {(u32)(s32)S32_MIN, (u32)(s32)S32_MIN}}, {S32, U32, {(u32)(s32)S32_MIN, (u32)(s32)S32_MIN}, {(u32)(s32)S32_MIN, (u32)(s32)S32_MIN}}, + +/* edge overlap testings for BPF_NE */ + {U64, U64, {1, 1}, {1, 0x80000000}}, + {U64, S64, {1, 1}, {1, 0x80000000}}, + {U64, U32, {1, 1}, {1, 0x80000000}}, + {U64, S32, {1, 1}, {1, 0x80000000}}, + {U64, U64, {0x80000000, 0x80000000}, {1, 0x80000000}}, + {U64, S64, {0x80000000, 0x80000000}, {1, 0x80000000}}, + {U64, U32, {0x80000000, 0x80000000}, {1, 0x80000000}}, + {U64, S32, {0x80000000, 0x80000000}, {1, 0x80000000}}, + {U64, U64, {1, 0x80000000}, {1, 1}}, + {U64, S64, {1, 0x80000000}, {1, 1}}, + {U64, U32, {1, 0x80000000}, {1, 1}}, + {U64, S32, {1, 0x80000000}, {1, 1}}, + {U64, U64, {1, 0x80000000}, {0x80000000, 0x80000000}}, + {U64, S64, {1, 0x80000000}, {0x80000000, 0x80000000}}, + {U64, U32, {1, 0x80000000}, {0x80000000, 0x80000000}}, + {U64, S32, {1, 0x80000000}, {0x80000000, 0x80000000}}, };/* Go over crafted hard-coded cases. This is fast, so we do it as part of
Signature · 1 line
2.39.2
Re: [PATCH bpf-next v3 2/2] selftests/bpf: activate the OP_NE login in range_cond()
Quoted 47 lines · Menglong Dong
The edge range checking for the registers is supported by the verifier now, so we can activate the extended login in tools/testing/selftests/bpf/prog_tests/reg_bounds.c/range_cond() to test such logic.
Besides, I added some cases to the "
crafted_cases" array for this logic. These cases are mainly used to test the edge of the src reg and dst reg.Signed-off-by Menglong Dong <[email protected]>---
v2:
- add some cases to the "crafted_cases"--- .../selftests/bpf/prog_tests/reg_bounds.c | 25 ++++++++++++++----- 1 file changed, 19 insertions(+), 6 deletions(-)tools/testing/selftests/bpf/prog_tests/reg_bounds.c +9
-6@@ -590,12 +590,7 @@ static void range_cond(enum num_t t, struct range x, struct range y, *newy =range (t,max_t (t, x.a, y.a),min_t (t, x.b, y.b));break ;case OP_NE : -/* generic case, can't derive more information */ - *newx =range (t, x.a, x.b); - *newy =range (t, y.a, y.b); -break ; - -/* below extended logic is not supported by verifier just yet */ +/* below logic is supported by the verifier now */ if (x.a == x.b && x.a == y.a) {/* X is a constant matching left side of Y */ *newx =range (t, x.a, x.b);@@ -2101,6 +2096,24 @@ static struct subtest_case crafted_cases[] = { {S32, S64, {(u32)(s32)S32_MIN, (u32)(s32)-255}, {(u32)(s32)-2, 0}}, {S32, S64, {0, 1}, {(u32)(s32)S32_MIN, (u32)(s32)S32_MIN}}, {S32, U32, {(u32)(s32)S32_MIN, (u32)(s32)S32_MIN}, {(u32)(s32)S32_MIN, (u32)(s32)S32_MIN}}, + +/* edge overlap testings for BPF_NE */ + {U64, U64, {1, 1}, {1, 0x80000000}}, + {U64, S64, {1, 1}, {1, 0x80000000}}, + {U64, U32, {1, 1}, {1, 0x80000000}}, + {U64, S32, {1, 1}, {1, 0x80000000}}, + {U64, U64, {0x80000000, 0x80000000}, {1, 0x80000000}}, + {U64, S64, {0x80000000, 0x80000000}, {1, 0x80000000}},
tools/testing/selftests/bpf/prog_tests/reg_bounds.c static struct subtest_case crafted_cases[] = { · line 2116
Quoted 10 lines
tools/testing/selftests/bpf/prog_tests/reg_bounds.c +10
-0+ {U64, U32, {0x80000000, 0x80000000}, {1, 0x80000000}}, + {U64, S32, {0x80000000, 0x80000000}, {1, 0x80000000}}, + {U64, U64, {1, 0x80000000}, {1, 1}}, + {U64, S64, {1, 0x80000000}, {1, 1}}, + {U64, U32, {1, 0x80000000}, {1, 1}}, + {U64, S32, {1, 0x80000000}, {1, 1}}, + {U64, U64, {1, 0x80000000}, {0x80000000, 0x80000000}}, + {U64, S64, {1, 0x80000000}, {0x80000000, 0x80000000}}, + {U64, U32, {1, 0x80000000}, {0x80000000, 0x80000000}}, + {U64, S32, {1, 0x80000000}, {0x80000000, 0x80000000}},
JNE and JEQ are sign-agnostic, so there is no need to use both U64 and S64 variants for comparison. As for the choice of values. Wouldn't it make sense to use really a boundary conditions:
0, 0xffffffffffffffff, and 0x80000000000000 for 64-bit and
0, 0xffffffff, and 0x80000000 for 32-bit? For this one use U32 as the init type?
BTW, all these cases should be tested with auto-generated tests, so please make sure to run
sudo SLOW_TESTS=1 ./test_progs -t reg_bounds_gen -j
locally. It will take a bit of time, but should help to get confidence in that everything is working and nothing regressed.
Quoted 6 lines
tools/testing/selftests/bpf/prog_tests/reg_bounds.c +0
-1};/* Go over crafted hard-coded cases. This is fast, so we do it as part of -- 2.39.2
Re: [PATCH bpf-next v3 2/2] selftests/bpf: activate the OP_NE login in range_cond()
Quoted 64 lines · Andrii Nakryiko
Quoted 47 lines · Menglong Dong
The edge range checking for the registers is supported by the verifier now, so we can activate the extended login in tools/testing/selftests/bpf/prog_tests/reg_bounds.c/range_cond() to test such logic.
Besides, I added some cases to the "
crafted_cases" array for this logic. These cases are mainly used to test the edge of the src reg and dst reg.Signed-off-by Menglong Dong <[email protected]>---
v2:
- add some cases to the "crafted_cases"--- .../selftests/bpf/prog_tests/reg_bounds.c | 25 ++++++++++++++----- 1 file changed, 19 insertions(+), 6 deletions(-)tools/testing/selftests/bpf/prog_tests/reg_bounds.c +9
-6@@ -590,12 +590,7 @@ static void range_cond(enum num_t t, struct range x, struct range y, *newy =range (t,max_t (t, x.a, y.a),min_t (t, x.b, y.b));break ;case OP_NE : -/* generic case, can't derive more information */ - *newx =range (t, x.a, x.b); - *newy =range (t, y.a, y.b); -break ; - -/* below extended logic is not supported by verifier just yet */ +/* below logic is supported by the verifier now */ if (x.a == x.b && x.a == y.a) {/* X is a constant matching left side of Y */ *newx =range (t, x.a, x.b);@@ -2101,6 +2096,24 @@ static struct subtest_case crafted_cases[] = { {S32, S64, {(u32)(s32)S32_MIN, (u32)(s32)-255}, {(u32)(s32)-2, 0}}, {S32, S64, {0, 1}, {(u32)(s32)S32_MIN, (u32)(s32)S32_MIN}}, {S32, U32, {(u32)(s32)S32_MIN, (u32)(s32)S32_MIN}, {(u32)(s32)S32_MIN, (u32)(s32)S32_MIN}}, + +/* edge overlap testings for BPF_NE */ + {U64, U64, {1, 1}, {1, 0x80000000}}, + {U64, S64, {1, 1}, {1, 0x80000000}}, + {U64, U32, {1, 1}, {1, 0x80000000}}, + {U64, S32, {1, 1}, {1, 0x80000000}}, + {U64, U64, {0x80000000, 0x80000000}, {1, 0x80000000}}, + {U64, S64, {0x80000000, 0x80000000}, {1, 0x80000000}},tools/testing/selftests/bpf/prog_tests/reg_bounds.c static struct subtest_case crafted_cases[] = { · line 2116
Quoted 10 lines
tools/testing/selftests/bpf/prog_tests/reg_bounds.c +10
-0+ {U64, U32, {0x80000000, 0x80000000}, {1, 0x80000000}}, + {U64, S32, {0x80000000, 0x80000000}, {1, 0x80000000}}, + {U64, U64, {1, 0x80000000}, {1, 1}}, + {U64, S64, {1, 0x80000000}, {1, 1}}, + {U64, U32, {1, 0x80000000}, {1, 1}}, + {U64, S32, {1, 0x80000000}, {1, 1}}, + {U64, U64, {1, 0x80000000}, {0x80000000, 0x80000000}}, + {U64, S64, {1, 0x80000000}, {0x80000000, 0x80000000}}, + {U64, U32, {1, 0x80000000}, {0x80000000, 0x80000000}}, + {U64, S32, {1, 0x80000000}, {0x80000000, 0x80000000}},JNE and JEQ are sign-agnostic, so there is no need to use both U64 and S64 variants for comparison. As for the choice of values. Wouldn't it make sense to use really a boundary conditions:
Quoted 3 lines
0,
0xffffffffffffffff, and0x80000000000000for 64-bit and
0,0xffffffff, and0x80000000for 32-bit? For this one use U32 as the init type?
Yeah, this makes sense.
Quoted 8 lines
BTW, all these cases should be tested with auto-generated tests, so please make sure to run
sudo SLOW_TESTS=1 ./test_progs -t
reg_bounds_gen-jlocally. It will take a bit of time, but should help to get confidence in that everything is working and nothing regressed.
I have already run the slow testing (it indeed takes some time) and everything works well. I'll add the test results to the commit log in the next version too.
Thanks!
Menglong Dong
Quoted 6 lines
Quoted 6 lines
tools/testing/selftests/bpf/prog_tests/reg_bounds.c +0
-1};/* Go over crafted hard-coded cases. This is fast, so we do it as part of -- 2.39.2