Merge branch 'complete-bpf-verifier-precision-tracking-support-for-register-spills'

Andrii Nakryiko says:

====================
Complete BPF verifier precision tracking support for register spills

Add support to BPF verifier to track and support register spill/fill to/from
stack regardless if it was done through read-only R10 register (which is the
only form supported today), or through a general register after copying R10
into it, while also potentially modifying offset.

Once we add register this generic spill/fill support to precision
backtracking, we can take advantage of it to stop doing eager STACK_ZERO
conversion on register spill. Instead we can rely on (im)precision of spilled
const zero register to improve verifier state pruning efficiency. This
situation of using const zero register to initialize stack slots is very
common with __builtin_memset() usage or just zero-initializing variables on
the stack, and it causes unnecessary state duplication, as that STACK_ZERO
knowledge is often not necessary for correctness, as those zero values are
never used in precise context. Thus, relying on register imprecision helps
tremendously, especially in real-world BPF programs.

To make spilled const zero register behave completely equivalently to
STACK_ZERO, we need to improve few other small pieces, which is done in the
second part of the patch set. See individual patches for details. There are
also two small bug fixes spotted during STACK_ZERO debugging.

The patch set consists of logically three changes:
  - patch #1 (and corresponding tests in patch #2) is fixing/impoving precision
    propagation for stack spills/fills. This can be landed as a stand-alone
    improvement;
  - patches #3 through #9 is improving verification scalability by utilizing
    register (im)precision instead of eager STACK_ZERO. These changes depend
    on patch #1.
  - patch #10 is a memory efficiency improvement to how instruction/jump
    history is tracked and maintained. It depends on patch #1, but is not
    strictly speaking required, even though I believe it's a good long-term
    solution to have a path-dependent per-instruction information. Kind
    of like a path-dependent counterpart to path-agnostic insn_aux array.

v3->v3:
  - fixed up Fixes tag (Alexei);
  - fixed few more selftests to not use BPF_ST instruction in inline asm
    directly, checked with CI, it was happy (CI);
v2->v3:
  - BPF_ST instruction workaround (Eduard);
  - force dereference in added tests to catch problems (Eduard);
  - some commit message massaging (Alexei);
v1->v2:
  - clean ups, WARN_ONCE(), insn_flags helpers added (Eduard);
  - added more selftests for STACK_ZERO/STACK_MISC cases (Eduard);
  - a bit more detailed explanation of effect of avoiding STACK_ZERO in favor
    of register spill in patch #8 commit (Alexei);
  - global shared instruction history refactoring moved to be the last patch
    in the series to make it easier to revert it, if applied (Alexei).
====================

Link: https://lore.kernel.org/r/20231205184248.1502704-1-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

3 files changed, 226 insertions, 26 deletions

diff --git a/tools/testing/selftests/bpf/progs/verifier_spill_fill.c b/tools/testing/selftests/bpf/progs/verifier_spill_fill.c
index 6115520154e3..df4920da3472 100644
--- a/tools/testing/selftests/bpf/progs/verifier_spill_fill.c
+++ b/tools/testing/selftests/bpf/progs/verifier_spill_fill.c
@@ -4,6 +4,7 @@
 #include <linux/bpf.h>
 #include <bpf/bpf_helpers.h>
 #include "bpf_misc.h"
+#include <../../../tools/include/linux/filter.h>
 
 struct {
 	__uint(type, BPF_MAP_TYPE_RINGBUF);
@@ -450,4 +451,130 @@ l0_%=:	r1 >>= 16;					\
 	: __clobber_all);
 }
 
+SEC("raw_tp")
+__log_level(2)
+__success
+__msg("fp-8=0m??mmmm")
+__msg("fp-16=00mm??mm")
+__msg("fp-24=00mm???m")
+__naked void spill_subregs_preserve_stack_zero(void)
+{
+	asm volatile (
+		"call %[bpf_get_prandom_u32];"
+
+		/* 32-bit subreg spill with ZERO, MISC, and INVALID */
+		".8byte %[fp1_u8_st_zero];"   /* ZERO, LLVM-18+: *(u8 *)(r10 -1) = 0; */
+		"*(u8 *)(r10 -2) = r0;"       /* MISC */
+		/* fp-3 and fp-4 stay INVALID */
+		"*(u32 *)(r10 -8) = r0;"
+
+		/* 16-bit subreg spill with ZERO, MISC, and INVALID */
+		".8byte %[fp10_u16_st_zero];" /* ZERO, LLVM-18+: *(u16 *)(r10 -10) = 0; */
+		"*(u16 *)(r10 -12) = r0;"     /* MISC */
+		/* fp-13 and fp-14 stay INVALID */
+		"*(u16 *)(r10 -16) = r0;"
+
+		/* 8-bit subreg spill with ZERO, MISC, and INVALID */
+		".8byte %[fp18_u16_st_zero];" /* ZERO, LLVM-18+: *(u16 *)(r18 -10) = 0; */
+		"*(u16 *)(r10 -20) = r0;"     /* MISC */
+		/* fp-21, fp-22, and fp-23 stay INVALID */
+		"*(u8 *)(r10 -24) = r0;"
+
+		"r0 = 0;"
+		"exit;"
+	:
+	: __imm(bpf_get_prandom_u32),
+	  __imm_insn(fp1_u8_st_zero, BPF_ST_MEM(BPF_B, BPF_REG_FP, -1, 0)),
+	  __imm_insn(fp10_u16_st_zero, BPF_ST_MEM(BPF_H, BPF_REG_FP, -10, 0)),
+	  __imm_insn(fp18_u16_st_zero, BPF_ST_MEM(BPF_H, BPF_REG_FP, -18, 0))
+	: __clobber_all);
+}
+
+char single_byte_buf[1] SEC(".data.single_byte_buf");
+
+SEC("raw_tp")
+__log_level(2)
+__success
+/* make sure fp-8 is all STACK_ZERO */
+__msg("2: (7a) *(u64 *)(r10 -8) = 0          ; R10=fp0 fp-8_w=00000000")
+/* but fp-16 is spilled IMPRECISE zero const reg */
+__msg("4: (7b) *(u64 *)(r10 -16) = r0        ; R0_w=0 R10=fp0 fp-16_w=0")
+/* and now check that precision propagation works even for such tricky case */
+__msg("10: (71) r2 = *(u8 *)(r10 -9)         ; R2_w=P0 R10=fp0 fp-16_w=0")
+__msg("11: (0f) r1 += r2")
+__msg("mark_precise: frame0: last_idx 11 first_idx 0 subseq_idx -1")
+__msg("mark_precise: frame0: regs=r2 stack= before 10: (71) r2 = *(u8 *)(r10 -9)")
+__msg("mark_precise: frame0: regs= stack=-16 before 9: (bf) r1 = r6")
+__msg("mark_precise: frame0: regs= stack=-16 before 8: (73) *(u8 *)(r1 +0) = r2")
+__msg("mark_precise: frame0: regs= stack=-16 before 7: (0f) r1 += r2")
+__msg("mark_precise: frame0: regs= stack=-16 before 6: (71) r2 = *(u8 *)(r10 -1)")
+__msg("mark_precise: frame0: regs= stack=-16 before 5: (bf) r1 = r6")
+__msg("mark_precise: frame0: regs= stack=-16 before 4: (7b) *(u64 *)(r10 -16) = r0")
+__msg("mark_precise: frame0: regs=r0 stack= before 3: (b7) r0 = 0")
+__naked void partial_stack_load_preserves_zeros(void)
+{
+	asm volatile (
+		/* fp-8 is all STACK_ZERO */
+		".8byte %[fp8_st_zero];" /* LLVM-18+: *(u64 *)(r10 -8) = 0; */
+
+		/* fp-16 is const zero register */
+		"r0 = 0;"
+		"*(u64 *)(r10 -16) = r0;"
+
+		/* load single U8 from non-aligned STACK_ZERO slot */
+		"r1 = %[single_byte_buf];"
+		"r2 = *(u8 *)(r10 -1);"
+		"r1 += r2;"
+		"*(u8 *)(r1 + 0) = r2;" /* this should be fine */
+
+		/* load single U8 from non-aligned ZERO REG slot */
+		"r1 = %[single_byte_buf];"
+		"r2 = *(u8 *)(r10 -9);"
+		"r1 += r2;"
+		"*(u8 *)(r1 + 0) = r2;" /* this should be fine */
+
+		/* load single U16 from non-aligned STACK_ZERO slot */
+		"r1 = %[single_byte_buf];"
+		"r2 = *(u16 *)(r10 -2);"
+		"r1 += r2;"
+		"*(u8 *)(r1 + 0) = r2;" /* this should be fine */
+
+		/* load single U16 from non-aligned ZERO REG slot */
+		"r1 = %[single_byte_buf];"
+		"r2 = *(u16 *)(r10 -10);"
+		"r1 += r2;"
+		"*(u8 *)(r1 + 0) = r2;" /* this should be fine */
+
+		/* load single U32 from non-aligned STACK_ZERO slot */
+		"r1 = %[single_byte_buf];"
+		"r2 = *(u32 *)(r10 -4);"
+		"r1 += r2;"
+		"*(u8 *)(r1 + 0) = r2;" /* this should be fine */
+
+		/* load single U32 from non-aligned ZERO REG slot */
+		"r1 = %[single_byte_buf];"
+		"r2 = *(u32 *)(r10 -12);"
+		"r1 += r2;"
+		"*(u8 *)(r1 + 0) = r2;" /* this should be fine */
+
+		/* for completeness, load U64 from STACK_ZERO slot */
+		"r1 = %[single_byte_buf];"
+		"r2 = *(u64 *)(r10 -8);"
+		"r1 += r2;"
+		"*(u8 *)(r1 + 0) = r2;" /* this should be fine */
+
+		/* for completeness, load U64 from ZERO REG slot */
+		"r1 = %[single_byte_buf];"
+		"r2 = *(u64 *)(r10 -16);"
+		"r1 += r2;"
+		"*(u8 *)(r1 + 0) = r2;" /* this should be fine */
+
+		"r0 = 0;"
+		"exit;"
+	:
+	: __imm_ptr(single_byte_buf),
+	  __imm_insn(fp8_st_zero, BPF_ST_MEM(BPF_DW, BPF_REG_FP, -8, 0))
+	: __clobber_common);
+}
+
 char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/progs/verifier_subprog_precision.c b/tools/testing/selftests/bpf/progs/verifier_subprog_precision.c
index 0dfe3f8b69ac..6f5d19665cf6 100644
--- a/tools/testing/selftests/bpf/progs/verifier_subprog_precision.c
+++ b/tools/testing/selftests/bpf/progs/verifier_subprog_precision.c
@@ -589,11 +589,24 @@ static __u64 subprog_spill_reg_precise(void)
 
 SEC("?raw_tp")
 __success __log_level(2)
-/* precision backtracking can't currently handle stack access not through r10,
- * so we won't be able to mark stack slot fp-8 as precise, and so will
- * fallback to forcing all as precise
- */
-__msg("mark_precise: frame0: falling back to forcing all scalars precise")
+__msg("10: (0f) r1 += r7")
+__msg("mark_precise: frame0: last_idx 10 first_idx 7 subseq_idx -1")
+__msg("mark_precise: frame0: regs=r7 stack= before 9: (bf) r1 = r8")
+__msg("mark_precise: frame0: regs=r7 stack= before 8: (27) r7 *= 4")
+__msg("mark_precise: frame0: regs=r7 stack= before 7: (79) r7 = *(u64 *)(r10 -8)")
+__msg("mark_precise: frame0: parent state regs= stack=-8:  R0_w=2 R6_w=1 R8_rw=map_value(map=.data.vals,ks=4,vs=16) R10=fp0 fp-8_rw=P1")
+__msg("mark_precise: frame0: last_idx 18 first_idx 0 subseq_idx 7")
+__msg("mark_precise: frame0: regs= stack=-8 before 18: (95) exit")
+__msg("mark_precise: frame1: regs= stack= before 17: (0f) r0 += r2")
+__msg("mark_precise: frame1: regs= stack= before 16: (79) r2 = *(u64 *)(r1 +0)")
+__msg("mark_precise: frame1: regs= stack= before 15: (79) r0 = *(u64 *)(r10 -16)")
+__msg("mark_precise: frame1: regs= stack= before 14: (7b) *(u64 *)(r10 -16) = r2")
+__msg("mark_precise: frame1: regs= stack= before 13: (7b) *(u64 *)(r1 +0) = r2")
+__msg("mark_precise: frame1: regs=r2 stack= before 6: (85) call pc+6")
+__msg("mark_precise: frame0: regs=r2 stack= before 5: (bf) r2 = r6")
+__msg("mark_precise: frame0: regs=r6 stack= before 4: (07) r1 += -8")
+__msg("mark_precise: frame0: regs=r6 stack= before 3: (bf) r1 = r10")
+__msg("mark_precise: frame0: regs=r6 stack= before 2: (b7) r6 = 1")
 __naked int subprog_spill_into_parent_stack_slot_precise(void)
 {
 	asm volatile (
@@ -628,14 +641,68 @@ __naked int subprog_spill_into_parent_stack_slot_precise(void)
 	);
 }
 
-__naked __noinline __used
-static __u64 subprog_with_checkpoint(void)
+SEC("?raw_tp")
+__success __log_level(2)
+__msg("17: (0f) r1 += r0")
+__msg("mark_precise: frame0: last_idx 17 first_idx 0 subseq_idx -1")
+__msg("mark_precise: frame0: regs=r0 stack= before 16: (bf) r1 = r7")
+__msg("mark_precise: frame0: regs=r0 stack= before 15: (27) r0 *= 4")
+__msg("mark_precise: frame0: regs=r0 stack= before 14: (79) r0 = *(u64 *)(r10 -16)")
+__msg("mark_precise: frame0: regs= stack=-16 before 13: (7b) *(u64 *)(r7 -8) = r0")
+__msg("mark_precise: frame0: regs=r0 stack= before 12: (79) r0 = *(u64 *)(r8 +16)")
+__msg("mark_precise: frame0: regs= stack=-16 before 11: (7b) *(u64 *)(r8 +16) = r0")
+__msg("mark_precise: frame0: regs=r0 stack= before 10: (79) r0 = *(u64 *)(r7 -8)")
+__msg("mark_precise: frame0: regs= stack=-16 before 9: (7b) *(u64 *)(r10 -16) = r0")
+__msg("mark_precise: frame0: regs=r0 stack= before 8: (07) r8 += -32")
+__msg("mark_precise: frame0: regs=r0 stack= before 7: (bf) r8 = r10")
+__msg("mark_precise: frame0: regs=r0 stack= before 6: (07) r7 += -8")
+__msg("mark_precise: frame0: regs=r0 stack= before 5: (bf) r7 = r10")
+__msg("mark_precise: frame0: regs=r0 stack= before 21: (95) exit")
+__msg("mark_precise: frame1: regs=r0 stack= before 20: (bf) r0 = r1")
+__msg("mark_precise: frame1: regs=r1 stack= before 4: (85) call pc+15")
+__msg("mark_precise: frame0: regs=r1 stack= before 3: (bf) r1 = r6")
+__msg("mark_precise: frame0: regs=r6 stack= before 2: (b7) r6 = 1")
+__naked int stack_slot_aliases_precision(void)
 {
 	asm volatile (
-		"r0 = 0;"
-		/* guaranteed checkpoint if BPF_F_TEST_STATE_FREQ is used */
-		"goto +0;"
+		"r6 = 1;"
+		/* pass r6 through r1 into subprog to get it back as r0;
+		 * this whole chain will have to be marked as precise later
+		 */
+		"r1 = r6;"
+		"call identity_subprog;"
+		/* let's setup two registers that are aliased to r10 */
+		"r7 = r10;"
+		"r7 += -8;"			/* r7 = r10 - 8 */
+		"r8 = r10;"
+		"r8 += -32;"			/* r8 = r10 - 32 */
+		/* now spill subprog's return value (a r6 -> r1 -> r0 chain)
+		 * a few times through different stack pointer regs, making
+		 * sure to use r10, r7, and r8 both in LDX and STX insns, and
+		 * *importantly* also using a combination of const var_off and
+		 * insn->off to validate that we record final stack slot
+		 * correctly, instead of relying on just insn->off derivation,
+		 * which is only valid for r10-based stack offset
+		 */
+		"*(u64 *)(r10 - 16) = r0;"
+		"r0 = *(u64 *)(r7 - 8);"	/* r7 - 8 == r10 - 16 */
+		"*(u64 *)(r8 + 16) = r0;"	/* r8 + 16 = r10 - 16 */
+		"r0 = *(u64 *)(r8 + 16);"
+		"*(u64 *)(r7 - 8) = r0;"
+		"r0 = *(u64 *)(r10 - 16);"
+		/* get ready to use r0 as an index into array to force precision */
+		"r0 *= 4;"
+		"r1 = %[vals];"
+		/* here r0->r1->r6 chain is forced to be precise and has to be
+		 * propagated back to the beginning, including through the
+		 * subprog call and all the stack spills and loads
+		 */
+		"r1 += r0;"
+		"r0 = *(u32 *)(r1 + 0);"
 		"exit;"
+		:
+		: __imm_ptr(vals)
+		: __clobber_common, "r6"
 	);
 }
 
diff --git a/tools/testing/selftests/bpf/verifier/precise.c b/tools/testing/selftests/bpf/verifier/precise.c
index 0d84dd1f38b6..8a2ff81d8350 100644
--- a/tools/testing/selftests/bpf/verifier/precise.c
+++ b/tools/testing/selftests/bpf/verifier/precise.c
@@ -140,10 +140,11 @@
 	.result = REJECT,
 },
 {
-	"precise: ST insn causing spi > allocated_stack",
+	"precise: ST zero to stack insn is supported",
 	.insns = {
 	BPF_MOV64_REG(BPF_REG_3, BPF_REG_10),
 	BPF_JMP_IMM(BPF_JNE, BPF_REG_3, 123, 0),
+	/* not a register spill, so we stop precision propagation for R4 here */
 	BPF_ST_MEM(BPF_DW, BPF_REG_3, -8, 0),
 	BPF_LDX_MEM(BPF_DW, BPF_REG_4, BPF_REG_10, -8),
 	BPF_MOV64_IMM(BPF_REG_0, -1),
@@ -157,11 +158,11 @@
 	mark_precise: frame0: last_idx 4 first_idx 2\
 	mark_precise: frame0: regs=r4 stack= before 4\
 	mark_precise: frame0: regs=r4 stack= before 3\
-	mark_precise: frame0: regs= stack=-8 before 2\
-	mark_precise: frame0: falling back to forcing all scalars precise\
-	force_precise: frame0: forcing r0 to be precise\
 	mark_precise: frame0: last_idx 5 first_idx 5\
-	mark_precise: frame0: parent state regs= stack=:",
+	mark_precise: frame0: parent state regs=r0 stack=:\
+	mark_precise: frame0: last_idx 4 first_idx 2\
+	mark_precise: frame0: regs=r0 stack= before 4\
+	5: R0=-1 R4=0",
 	.result = VERBOSE_ACCEPT,
 	.retval = -1,
 },
@@ -169,6 +170,8 @@
 	"precise: STX insn causing spi > allocated_stack",
 	.insns = {
 	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_prandom_u32),
+	/* make later reg spill more interesting by having somewhat known scalar */
+	BPF_ALU64_IMM(BPF_AND, BPF_REG_0, 0xff),
 	BPF_MOV64_REG(BPF_REG_3, BPF_REG_10),
 	BPF_JMP_IMM(BPF_JNE, BPF_REG_3, 123, 0),
 	BPF_STX_MEM(BPF_DW, BPF_REG_3, BPF_REG_0, -8),
@@ -179,18 +182,21 @@
 	},
 	.prog_type = BPF_PROG_TYPE_XDP,
 	.flags = BPF_F_TEST_STATE_FREQ,
-	.errstr = "mark_precise: frame0: last_idx 6 first_idx 6\
+	.errstr = "mark_precise: frame0: last_idx 7 first_idx 7\
 	mark_precise: frame0: parent state regs=r4 stack=:\
-	mark_precise: frame0: last_idx 5 first_idx 3\
-	mark_precise: frame0: regs=r4 stack= before 5\
-	mark_precise: frame0: regs=r4 stack= before 4\
-	mark_precise: frame0: regs= stack=-8 before 3\
-	mark_precise: frame0: falling back to forcing all scalars precise\
-	force_precise: frame0: forcing r0 to be precise\
-	force_precise: frame0: forcing r0 to be precise\
-	force_precise: frame0: forcing r0 to be precise\
-	force_precise: frame0: forcing r0 to be precise\
-	mark_precise: frame0: last_idx 6 first_idx 6\
+	mark_precise: frame0: last_idx 6 first_idx 4\
+	mark_precise: frame0: regs=r4 stack= before 6: (b7) r0 = -1\
+	mark_precise: frame0: regs=r4 stack= before 5: (79) r4 = *(u64 *)(r10 -8)\
+	mark_precise: frame0: regs= stack=-8 before 4: (7b) *(u64 *)(r3 -8) = r0\
+	mark_precise: frame0: parent state regs=r0 stack=:\
+	mark_precise: frame0: last_idx 3 first_idx 3\
+	mark_precise: frame0: regs=r0 stack= before 3: (55) if r3 != 0x7b goto pc+0\
+	mark_precise: frame0: regs=r0 stack= before 2: (bf) r3 = r10\
+	mark_precise: frame0: regs=r0 stack= before 1: (57) r0 &= 255\
+	mark_precise: frame0: parent state regs=r0 stack=:\
+	mark_precise: frame0: last_idx 0 first_idx 0\
+	mark_precise: frame0: regs=r0 stack= before 0: (85) call bpf_get_prandom_u32#7\
+	mark_precise: frame0: last_idx 7 first_idx 7\
 	mark_precise: frame0: parent state regs= stack=:",
 	.result = VERBOSE_ACCEPT,
 	.retval = -1,