Skip to contentSkip to conversation

bpf: report arena faults to BPF streams

PATCH Superseded Puranjay Mohan 3 participants 11 messages Sep 8, 2025 – Sep 9, 2025

Subsystems (8)
x86 Kselftest BPF arm64 +4 more subsystems

v6 of 7 Superseded Latest is v7 →

lore
Share & tools
lore

Puranjay Mohan

author maintainer
#0 lore raw

v6 cover bpf: report arena faults to BPF streams

Changes in v5->v6:
v5: https://lore.kernel.org/all/[email protected]/
- Introduces __stderr and __stdout for easy testing of bpf streams (Eduard)
- Add more test cases for arena fault reporting (subprog and callback)
- Fix main_prog_aux usage and return main_prog from find_from_stack_cb (Kumar)
- Properly fix the build issue reported by kernel test robot

Changes in v4->v5:
v4: https://lore.kernel.org/all/[email protected]/
- Added patch 2 to introducing main_prog_aux for easier access to streams.
- Fixed bug in fault handlers when arena_reg == dst_reg
- Updated selftest to check test above edge case.
- Added comments about the usage of barrier_var() in code and commit message.

Changes in v3->v4:
v3: https://lore.kernel.org/all/[email protected]/
- Fixed a build issue when CONFIG_BPF_JIT=y and # CONFIG_BPF_SYSCALL is not set

Changes in v2->v3:
v2: https://lore.kernel.org/all/[email protected]/
- Improved the selftest to check the exact fault address
- Dropped BPF_NO_KFUNC_PROTOTYPES and bpf_arena_alloc/free_pages() usage
- Rebased on bpf-next/master

Changes in v1->v2:
v1: https://lore.kernel.org/all/[email protected]/
- Changed variable and mask names for consistency (Yonghong)
- Added Acked-by: Yonghong Song <[email protected]> on two patches

This set adds the support of reporting page faults inside arena to BPF stderr stream. The reported address is the one that a user would expect to see if they pass it to bpf_printk();

Here is an example output from the stderr stream and bpf_printk()

ERROR: Arena WRITE access at unmapped address 0xdeaddead0000

CPU: 9 UID: 0 PID: 502 Comm: test_progs
Call trace:
bpf_stream_stage_dump_stack+0xc0/0x150
bpf_prog_report_arena_violation+0x98/0xf0
ex_handler_bpf+0x5c/0x78
fixup_exception+0xf8/0x160
__do_kernel_fault+0x40/0x188
do_bad_area+0x70/0x88
do_translation_fault+0x54/0x98
do_mem_abort+0x4c/0xa8
el1_abort+0x44/0x70
el1h_64_sync_handler+0x50/0x108
el1h_64_sync+0x6c/0x70
bpf_prog_a64a9778d31b8e88_stream_arena_write_fault+0x84/0xc8
  *(page) = 1; @ stream.c:100
bpf_prog_test_run_syscall+0x100/0x328
__sys_bpf+0x508/0xb98
__arm64_sys_bpf+0x2c/0x48
invoke_syscall+0x50/0x120
el0_svc_common.constprop.0+0x48/0xf8
do_el0_svc+0x28/0x40
el0_svc+0x48/0xf8
el0t_64_sync_handler+0xa0/0xe8
el0t_64_sync+0x198/0x1a0

Same address is printed by bpf_printk():

1389.078831: bpf_trace_printk: Read Address: 0xdeaddead0000

To make this possible, some extra metadata has to be passed to the bpf exception handler, so the bpf exception handling mechanism for both x86-64 and arm64 have been improved in this set.

The streams selftest has been updated to test this new feature.

Puranjay Mohan (5):
bpf: arm64: simplify exception table handling
bpf: core: introduce main_prog_aux for stream access
bpf: Report arena faults to BPF stderr
selftests: bpf: introduce __stderr and __stdout selftests/bpf: Add tests for arena fault reporting

 arch/arm64/net/bpf_jit_comp.c                 |  77 ++++++---
 arch/x86/net/bpf_jit_comp.c                   |  76 ++++++++-
 include/linux/bpf.h                           |   7 +
 kernel/bpf/arena.c                            |  30 ++++
 kernel/bpf/core.c                             |   6 +-
 kernel/bpf/verifier.c                         |   1 +
 .../testing/selftests/bpf/prog_tests/stream.c |  53 ++----
 tools/testing/selftests/bpf/progs/bpf_misc.h  |  10 ++
 tools/testing/selftests/bpf/progs/stream.c    | 158 ++++++++++++++++++
 tools/testing/selftests/bpf/test_loader.c     |  90 ++++++++++
 10 files changed, 443 insertions(+), 65 deletions(-)
Signature · 1 line
2.47.3
author maintainer
#1 lore raw

PATCH v6 1/5 bpf: arm64: simplify exception table handling

BPF loads with BPF_PROBE_MEM(SX) can load from unsafe pointers and the JIT adds an exception table entry for the JITed instruction which allows the exeption handler to set the destination register of the load to zero and continue execution from the next instruction.

As all arm64 instructions are AARCH64_INSN_SIZE size, the exception handler can just increment the pc by AARCH64_INSN_SIZE without needing the exact address of the instruction following the the faulting instruction.

Simplify the exception table usage in arm64 JIT by only saving the destination register in ex->fixup and drop everything related to the fixup_offset. The fault handler is modified to add AARCH64_INSN_SIZE to the pc.

---
 arch/arm64/net/bpf_jit_comp.c | 25 +++----------------------
 1 file changed, 3 insertions(+), 22 deletions(-)
arch/arm64/net/bpf_jit_comp.c +3 -22
index f0b1cb2c3bc48..e6d1fdc1e6f52 100644
@@ -1066,19 +1066,18 @@ static void build_epilogue(struct jit_ctx *ctx, bool was_classic)
 	emit(A64_RET(A64_LR), ctx);
 }
 
-#define BPF_FIXUP_OFFSET_MASK	GENMASK(26, 0)
 #define BPF_FIXUP_REG_MASK	GENMASK(31, 27)
 #define DONT_CLEAR 5 /* Unused ARM64 register from BPF's POV */
 
 bool ex_handler_bpf(const struct exception_table_entry *ex,
 		    struct pt_regs *regs)
 {
-	off_t offset = FIELD_GET(BPF_FIXUP_OFFSET_MASK, ex->fixup);
 	int dst_reg = FIELD_GET(BPF_FIXUP_REG_MASK, ex->fixup);
 
 	if (dst_reg != DONT_CLEAR)
 		regs->regs[dst_reg] = 0;
-	regs->pc = (unsigned long)&ex->fixup - offset;
+	/* Skip the faulting instruction */
+	regs->pc += AARCH64_INSN_SIZE;
 	return true;
 }
 
@@ -1088,7 +1087,6 @@ static int add_exception_handler(const struct bpf_insn *insn,
 				 int dst_reg)
 {
 	off_t ins_offset;
-	off_t fixup_offset;
 	unsigned long pc;
 	struct exception_table_entry *ex;
 
@@ -1119,22 +1117,6 @@ static int add_exception_handler(const struct bpf_insn *insn,
 	if (WARN_ON_ONCE(ins_offset >= 0 || ins_offset < INT_MIN))
 		return -ERANGE;
 
-	/*
-	 * Since the extable follows the program, the fixup offset is always
-	 * negative and limited to BPF_JIT_REGION_SIZE. Store a positive value
-	 * to keep things simple, and put the destination register in the upper
-	 * bits. We don't need to worry about buildtime or runtime sort
-	 * modifying the upper bits because the table is already sorted, and
-	 * isn't part of the main exception table.
-	 *
-	 * The fixup_offset is set to the next instruction from the instruction
-	 * that may fault. The execution will jump to this after handling the
-	 * fault.
-	 */
-	fixup_offset = (long)&ex->fixup - (pc + AARCH64_INSN_SIZE);
-	if (!FIELD_FIT(BPF_FIXUP_OFFSET_MASK, fixup_offset))
-		return -ERANGE;
-
 	/*
 	 * The offsets above have been calculated using the RO buffer but we
 	 * need to use the R/W buffer for writes.
@@ -1147,8 +1129,7 @@ static int add_exception_handler(const struct bpf_insn *insn,
 	if (BPF_CLASS(insn->code) != BPF_LDX)
 		dst_reg = DONT_CLEAR;
 
-	ex->fixup = FIELD_PREP(BPF_FIXUP_OFFSET_MASK, fixup_offset) |
-		    FIELD_PREP(BPF_FIXUP_REG_MASK, dst_reg);
+	ex->fixup = FIELD_PREP(BPF_FIXUP_REG_MASK, dst_reg);
 
 	ex->type = EX_TYPE_BPF;
 
Signature · 1 line
2.47.3
author maintainer
#2 lore raw

PATCH v6 2/5 bpf: core: introduce main_prog_aux for stream access

In reply to Puranjay Mohan

BPF streams are only valid for the main programs, to make it easier to access streams from subprogs, introduce main_prog_aux in struct bpf_prog_aux.

prog->aux->main_prog_aux = prog->aux, for main programs and prog->aux->main_prog_aux = main_prog->aux, for subprograms.

Make bpf_prog_find_from_stack() use the added main_prog_aux to return the mainprog when a subprog is found on the stack.

---
 include/linux/bpf.h   | 1 +
 kernel/bpf/core.c     | 6 +++---
 kernel/bpf/verifier.c | 1 +
 3 files changed, 5 insertions(+), 3 deletions(-)
include/linux/bpf.h +1 -0
index 8f6e87f0f3a89..d133171c4d2a9 100644
@@ -1633,6 +1633,7 @@ struct bpf_prog_aux {
 	/* function name for valid attach_btf_id */
 	const char *attach_func_name;
 	struct bpf_prog **func;
+	struct bpf_prog_aux *main_prog_aux;
 	void *jit_data; /* JIT specific data. arch dependent */
 	struct bpf_jit_poke_descriptor *poke_tab;
 	struct bpf_kfunc_desc_tab *kfunc_tab;
kernel/bpf/core.c +3 -3
index ef01cc644a965..c732737e6b735 100644
@@ -120,6 +120,7 @@ struct bpf_prog *bpf_prog_alloc_no_stats(unsigned int size, gfp_t gfp_extra_flag
 
 	fp->pages = size / PAGE_SIZE;
 	fp->aux = aux;
+	fp->aux->main_prog_aux = aux;
 	fp->aux->prog = fp;
 	fp->jit_requested = ebpf_jit_enabled();
 	fp->blinding_requested = bpf_jit_blinding_enabled(fp);
@@ -3292,9 +3293,8 @@ static bool find_from_stack_cb(void *cookie, u64 ip, u64 sp, u64 bp)
 	rcu_read_unlock();
 	if (!prog)
 		return true;
-	if (bpf_is_subprog(prog))
-		return true;
-	ctxp->prog = prog;
+	/* Make sure we return the main prog if we found a subprog */
+	ctxp->prog = prog->aux->main_prog_aux->prog;
 	return false;
 }
 
kernel/bpf/verifier.c +1 -0
index b9394f8fac0ed..0162c5ad682ed 100644
@@ -21597,6 +21597,7 @@ static int jit_subprogs(struct bpf_verifier_env *env)
 		func[i]->aux->func_info_cnt = prog->aux->func_info_cnt;
 		func[i]->aux->poke_tab = prog->aux->poke_tab;
 		func[i]->aux->size_poke_tab = prog->aux->size_poke_tab;
+		func[i]->aux->main_prog_aux = prog->aux;
 
 		for (j = 0; j < prog->aux->size_poke_tab; j++) {
 			struct bpf_jit_poke_descriptor *poke;
Signature · 1 line
2.47.3
author maintainer
#3 lore raw

PATCH v6 3/5 bpf: Report arena faults to BPF stderr

In reply to Puranjay Mohan

Begin reporting arena page faults and the faulting address to BPF program's stderr, this patch adds support in the arm64 and x86-64 JITs, support for other archs can be added later.

The fault handlers receive the 32 bit address in the arena region so the upper 32 bits of user_vm_start is added to it before printing the address. This is what the user would expect to see as this is what is printed by bpf_printk() is you pass it an address returned by bpf_arena_alloc_pages();

---
 arch/arm64/net/bpf_jit_comp.c | 52 ++++++++++++++++++++++++
 arch/x86/net/bpf_jit_comp.c   | 76 ++++++++++++++++++++++++++++++++---
 include/linux/bpf.h           |  6 +++
 kernel/bpf/arena.c            | 30 ++++++++++++++
 4 files changed, 159 insertions(+), 5 deletions(-)
arch/arm64/net/bpf_jit_comp.c +52 -0
index e6d1fdc1e6f52..556ab2fd222d8 100644
@@ -1066,6 +1066,30 @@ static void build_epilogue(struct jit_ctx *ctx, bool was_classic)
 	emit(A64_RET(A64_LR), ctx);
 }
 
+/*
+ * Metadata encoding for exception handling in JITed code.
+ *
+ * Format of `fixup` field in `struct exception_table_entry`:
+ *
+ * Bit layout of `fixup` (32-bit):
+ *
+ * +-----------+--------+-----------+-----------+----------+
+ * |   31-27   | 26-22  |     21    |   20-16   |   15-0   |
+ * |           |        |           |           |          |
+ * | FIXUP_REG | Unused | ARENA_ACC | ARENA_REG |  OFFSET  |
+ * +-----------+--------+-----------+-----------+----------+
+ *
+ * - OFFSET (16 bits): Offset used to compute address for Load/Store instruction.
+ * - ARENA_REG (5 bits): Register that is used to calculate the address for load/store when
+ *                       accessing the arena region.
+ * - ARENA_ACCESS (1 bit): This bit is set when the faulting instruction accessed the arena region.
+ * - FIXUP_REG (5 bits): Destination register for the load instruction (cleared on fault) or set to
+ *                       DONT_CLEAR if it is a store instruction.
+ */
+
+#define BPF_FIXUP_OFFSET_MASK      GENMASK(15, 0)
+#define BPF_FIXUP_ARENA_REG_MASK   GENMASK(20, 16)
+#define BPF_ARENA_ACCESS           BIT(21)
 #define BPF_FIXUP_REG_MASK	GENMASK(31, 27)
 #define DONT_CLEAR 5 /* Unused ARM64 register from BPF's POV */
 
@@ -1073,11 +1097,22 @@ bool ex_handler_bpf(const struct exception_table_entry *ex,
 		    struct pt_regs *regs)
 {
 	int dst_reg = FIELD_GET(BPF_FIXUP_REG_MASK, ex->fixup);
+	s16 off = FIELD_GET(BPF_FIXUP_OFFSET_MASK, ex->fixup);
+	int arena_reg = FIELD_GET(BPF_FIXUP_ARENA_REG_MASK, ex->fixup);
+	bool is_arena = !!(ex->fixup & BPF_ARENA_ACCESS);
+	bool is_write = (dst_reg == DONT_CLEAR);
+	unsigned long addr;
+
+	if (is_arena) {
+		addr = regs->regs[arena_reg] + off;
+		bpf_prog_report_arena_violation(is_write, addr, regs->pc);
+	}
 
 	if (dst_reg != DONT_CLEAR)
 		regs->regs[dst_reg] = 0;
 	/* Skip the faulting instruction */
 	regs->pc += AARCH64_INSN_SIZE;
+
 	return true;
 }
 
@@ -1087,6 +1122,9 @@ static int add_exception_handler(const struct bpf_insn *insn,
 				 int dst_reg)
 {
 	off_t ins_offset;
+	s16 off = insn->off;
+	bool is_arena;
+	int arena_reg;
 	unsigned long pc;
 	struct exception_table_entry *ex;
 
@@ -1100,6 +1138,9 @@ static int add_exception_handler(const struct bpf_insn *insn,
 				BPF_MODE(insn->code) != BPF_PROBE_ATOMIC)
 		return 0;
 
+	is_arena = (BPF_MODE(insn->code) == BPF_PROBE_MEM32) ||
+		   (BPF_MODE(insn->code) == BPF_PROBE_ATOMIC);
+
 	if (!ctx->prog->aux->extable ||
 	    WARN_ON_ONCE(ctx->exentry_idx >= ctx->prog->aux->num_exentries))
 		return -EINVAL;
@@ -1131,6 +1172,17 @@ static int add_exception_handler(const struct bpf_insn *insn,
 
 	ex->fixup = FIELD_PREP(BPF_FIXUP_REG_MASK, dst_reg);
 
+	if (is_arena) {
+		ex->fixup |= BPF_ARENA_ACCESS;
+		if (BPF_CLASS(insn->code) == BPF_LDX)
+			arena_reg = bpf2a64[insn->src_reg];
+		else
+			arena_reg = bpf2a64[insn->dst_reg];
+
+		ex->fixup |=  FIELD_PREP(BPF_FIXUP_OFFSET_MASK, off) |
+			      FIELD_PREP(BPF_FIXUP_ARENA_REG_MASK, arena_reg);
+	}
+
 	ex->type = EX_TYPE_BPF;
 
 	ctx->exentry_idx++;
arch/x86/net/bpf_jit_comp.c +71 -5

114 lines not on this page · raw message

include/linux/bpf.h +6 -0
index d133171c4d2a9..41f776071ff51 100644
@@ -2881,6 +2881,7 @@ void bpf_dynptr_init(struct bpf_dynptr_kern *ptr, void *data,
 		     enum bpf_dynptr_type type, u32 offset, u32 size);
 void bpf_dynptr_set_null(struct bpf_dynptr_kern *ptr);
 void bpf_dynptr_set_rdonly(struct bpf_dynptr_kern *ptr);
+void bpf_prog_report_arena_violation(bool write, unsigned long addr, unsigned long fault_ip);
 
 #else /* !CONFIG_BPF_SYSCALL */
 static inline struct bpf_prog *bpf_prog_get(u32 ufd)
@@ -3168,6 +3169,11 @@ static inline void bpf_dynptr_set_null(struct bpf_dynptr_kern *ptr)
 static inline void bpf_dynptr_set_rdonly(struct bpf_dynptr_kern *ptr)
 {
 }
+
+static inline void bpf_prog_report_arena_violation(bool write, unsigned long addr,
+						   unsigned long fault_ip)
+{
+}
 #endif /* CONFIG_BPF_SYSCALL */
 
 static __always_inline int
kernel/bpf/arena.c +30 -0
index 5b37753799d20..d0b31b40d3826 100644
@@ -633,3 +633,33 @@ static int __init kfunc_init(void)
 	return register_btf_kfunc_id_set(BPF_PROG_TYPE_UNSPEC, &common_kfunc_set);
 }
 late_initcall(kfunc_init);
+
+void bpf_prog_report_arena_violation(bool write, unsigned long addr, unsigned long fault_ip)
+{
+	struct bpf_stream_stage ss;
+	struct bpf_prog *prog;
+	u64 user_vm_start;
+
+	/*
+	 * The RCU read lock is held to safely traverse the latch tree, but we
+	 * don't need its protection when accessing the prog, since it will not
+	 * disappear while we are handling the fault.
+	 */
+	rcu_read_lock();
+	prog = bpf_prog_ksym_find(fault_ip);
+	rcu_read_unlock();
+	if (!prog)
+		return;
+
+	/* Use main prog for stream access */
+	prog = prog->aux->main_prog_aux->prog;
+
+	user_vm_start = bpf_arena_get_user_vm_start(prog->aux->arena);
+	addr += (user_vm_start >> 32) << 32;
+
+	bpf_stream_stage(ss, prog, BPF_STDERR, ({
+		bpf_stream_printk(ss, "ERROR: Arena %s access at unmapped address 0x%lx\n",
+				  write ? "WRITE" : "READ", addr);
+		bpf_stream_dump_stack(ss);
+	}));
+}
Signature · 1 line
2.47.3
#4 lore raw

Re: [PATCH bpf-next v6 3/5] bpf: Report arena faults to BPF stderr

Quoted 99 lines · Puranjay Mohan

Begin reporting arena page faults and the faulting address to BPF program's stderr, this patch adds support in the arm64 and x86-64 JITs, support for other archs can be added later.

The fault handlers receive the 32 bit address in the arena region so the upper 32 bits of user_vm_start is added to it before printing the address. This is what the user would expect to see as this is what is printed by bpf_printk() is you pass it an address returned by bpf_arena_alloc_pages();

---
 arch/arm64/net/bpf_jit_comp.c | 52 ++++++++++++++++++++++++
 arch/x86/net/bpf_jit_comp.c   | 76 ++++++++++++++++++++++++++++++++---
 include/linux/bpf.h           |  6 +++
 kernel/bpf/arena.c            | 30 ++++++++++++++
 4 files changed, 159 insertions(+), 5 deletions(-)
arch/arm64/net/bpf_jit_comp.c +41 -0
index e6d1fdc1e6f52..556ab2fd222d8 100644
@@ -1066,6 +1066,30 @@ static void build_epilogue(struct jit_ctx *ctx, bool was_classic)
        emit(A64_RET(A64_LR), ctx);
 }
 
+/*
+ * Metadata encoding for exception handling in JITed code.
+ *
+ * Format of `fixup` field in `struct exception_table_entry`:
+ *
+ * Bit layout of `fixup` (32-bit):
+ *
+ * +-----------+--------+-----------+-----------+----------+
+ * |   31-27   | 26-22  |     21    |   20-16   |   15-0   |
+ * |           |        |           |           |          |
+ * | FIXUP_REG | Unused | ARENA_ACC | ARENA_REG |  OFFSET  |
+ * +-----------+--------+-----------+-----------+----------+
+ *
+ * - OFFSET (16 bits): Offset used to compute address for Load/Store instruction.
+ * - ARENA_REG (5 bits): Register that is used to calculate the address for load/store when
+ *                       accessing the arena region.
+ * - ARENA_ACCESS (1 bit): This bit is set when the faulting instruction accessed the arena region.
+ * - FIXUP_REG (5 bits): Destination register for the load instruction (cleared on fault) or set to
+ *                       DONT_CLEAR if it is a store instruction.
+ */
+
+#define BPF_FIXUP_OFFSET_MASK      GENMASK(15, 0)
+#define BPF_FIXUP_ARENA_REG_MASK   GENMASK(20, 16)
+#define BPF_ARENA_ACCESS           BIT(21)
 #define BPF_FIXUP_REG_MASK     GENMASK(31, 27)
 #define DONT_CLEAR 5 /* Unused ARM64 register from BPF's POV */
 
@@ -1073,11 +1097,22 @@ bool ex_handler_bpf(const struct exception_table_entry *ex,
                    struct pt_regs *regs)
 {
        int dst_reg = FIELD_GET(BPF_FIXUP_REG_MASK, ex->fixup);
+       s16 off = FIELD_GET(BPF_FIXUP_OFFSET_MASK, ex->fixup);
+       int arena_reg = FIELD_GET(BPF_FIXUP_ARENA_REG_MASK, ex->fixup);
+       bool is_arena = !!(ex->fixup & BPF_ARENA_ACCESS);
+       bool is_write = (dst_reg == DONT_CLEAR);
+       unsigned long addr;
+
+       if (is_arena) {
+               addr = regs->regs[arena_reg] + off;
+               bpf_prog_report_arena_violation(is_write, addr, regs->pc);
+       }
 
        if (dst_reg != DONT_CLEAR)
                regs->regs[dst_reg] = 0;
        /* Skip the faulting instruction */
        regs->pc += AARCH64_INSN_SIZE;
+
        return true;
 }
 
@@ -1087,6 +1122,9 @@ static int add_exception_handler(const struct bpf_insn *insn,
                                 int dst_reg)
 {
        off_t ins_offset;
+       s16 off = insn->off;
+       bool is_arena;
+       int arena_reg;
        unsigned long pc;
        struct exception_table_entry *ex;
 
@@ -1100,6 +1138,9 @@ static int add_exception_handler(const struct bpf_insn *insn,
                                BPF_MODE(insn->code) != BPF_PROBE_ATOMIC)
                return 0;
 
+       is_arena = (BPF_MODE(insn->code) == BPF_PROBE_MEM32) ||
+                  (BPF_MODE(insn->code) == BPF_PROBE_ATOMIC);
+
        if (!ctx->prog->aux->extable ||
            WARN_ON_ONCE(ctx->exentry_idx >= ctx->prog->aux->num_exentries))
                return -EINVAL;

arch/arm64/net/bpf_jit_comp.c static int add_exception_handler(const struct bpf_insn *insn, · line 1180

Quoted 9 lines
arch/arm64/net/bpf_jit_comp.c +6 -0
@@ -1131,6 +1172,17 @@ static int add_exception_handler(const struct bpf_insn *insn,
 
        ex->fixup = FIELD_PREP(BPF_FIXUP_REG_MASK, dst_reg);
 
+       if (is_arena) {
+               ex->fixup |= BPF_ARENA_ACCESS;
+               if (BPF_CLASS(insn->code) == BPF_LDX)
+                       arena_reg = bpf2a64[insn->src_reg];
+               else
+                       arena_reg = bpf2a64[insn->dst_reg];

I think this is correct, since insn->src/dst_reg is the register before the adjustment of:
emit(A64_ADD(1, tmp2, src, arena_vm_base)

so ex handler doing regs->regs[arena_reg]
should read 64-bit with upper 32-bit being zero. right?
If so, please add a comment describing this subtle logic.

Quoted 101 lines
arch/arm64/net/bpf_jit_comp.c +5 -0
+
+               ex->fixup |=  FIELD_PREP(BPF_FIXUP_OFFSET_MASK, off) |
+                             FIELD_PREP(BPF_FIXUP_ARENA_REG_MASK, arena_reg);
+       }
+
        ex->type = EX_TYPE_BPF;
 
        ctx->exentry_idx++;
arch/x86/net/bpf_jit_comp.c +55 -2
index 7e3fca1646203..007c273f3deea 100644
@@ -8,6 +8,7 @@
 #include <linux/netdevice.h>
 #include <linux/filter.h>
 #include <linux/if_vlan.h>
+#include <linux/bitfield.h>
 #include <linux/bpf.h>
 #include <linux/memory.h>
 #include <linux/sort.h>
@@ -1388,16 +1389,67 @@ static int emit_atomic_ld_st_index(u8 **pprog, u32 atomic_op, u32 size,
        return 0;
 }
 
+/*
+ * Metadata encoding for exception handling in JITed code.
+ *
+ * Format of `fixup` and `data` fields in `struct exception_table_entry`:
+ *
+ * Bit layout of `fixup` (32-bit):
+ *
+ * +-----------+--------+-----------+---------+----------+
+ * | 31        | 30-24  |   23-16   |   15-8  |    7-0   |
+ * |           |        |           |         |          |
+ * | ARENA_ACC | Unused | ARENA_REG | DST_REG | INSN_LEN |
+ * +-----------+--------+-----------+---------+----------+
+ *
+ * - INSN_LEN (8 bits): Length of faulting insn (max x86 insn = 15 bytes (fits in 8 bits)).
+ * - DST_REG  (8 bits): Offset of dst_reg from reg2pt_regs[] (max offset = 112 (fits in 8 bits)).
+ *                      This is set to DONT_CLEAR if the insn is a store.
+ * - ARENA_REG (8 bits): Offset of the register that is used to calculate the
+ *                       address for load/store when accessing the arena region.
+ * - ARENA_ACCESS (1 bit): This bit is set when the faulting instruction accessed the arena region.
+ *
+ * Bit layout of `data` (32-bit):
+ *
+ * +--------------+--------+--------------+
+ * |   31-16     |  15-8  |     7-0      |
+ * |              |       |              |
+ * | ARENA_OFFSET | Unused |  EX_TYPE_BPF |
+ * +--------------+--------+--------------+
+ *
+ * - ARENA_OFFSET (16 bits): Offset used to calculate the address for load/store when
+ *                           accessing the arena region.
+ */
+
 #define DONT_CLEAR 1
+#define FIXUP_INSN_LEN_MASK    GENMASK(7, 0)
+#define FIXUP_REG_MASK         GENMASK(15, 8)
+#define FIXUP_ARENA_REG_MASK   GENMASK(23, 16)
+#define FIXUP_ARENA_ACCESS     BIT(31)
+#define DATA_ARENA_OFFSET_MASK GENMASK(31, 16)
 
 bool ex_handler_bpf(const struct exception_table_entry *x, struct pt_regs *regs)
 {
-       u32 reg = x->fixup >> 8;
+       u32 reg = FIELD_GET(FIXUP_REG_MASK, x->fixup);
+       u32 insn_len = FIELD_GET(FIXUP_INSN_LEN_MASK, x->fixup);
+       bool is_arena = !!(x->fixup & FIXUP_ARENA_ACCESS);
+       bool is_write = (reg == DONT_CLEAR);
+       unsigned long addr;
+       s16 off;
+       u32 arena_reg;
+
+       if (is_arena) {
+               arena_reg = FIELD_GET(FIXUP_ARENA_REG_MASK, x->fixup);
+               off = FIELD_GET(DATA_ARENA_OFFSET_MASK, x->data);
+               addr = *(unsigned long *)((void *)regs + arena_reg) + off;
+               bpf_prog_report_arena_violation(is_write, addr, regs->ip);
+       }
 
        /* jump over faulting load and clear dest register */
        if (reg != DONT_CLEAR)
                *(unsigned long *)((void *)regs + reg) = 0;
-       regs->ip += x->fixup & 0xff;
+       regs->ip += insn_len;
+
        return true;
 }
 
@@ -2070,6 +2122,7 @@ st:                       if (is_imm8(insn->off))
                        {
                                struct exception_table_entry *ex;
                                u8 *_insn = image + proglen + (start_of_ldx - temp);
+                               u32 arena_reg, fixup_reg;
                                s64 delta;
 
                                if (!bpf_prog->aux->extable)
@@ -2089,8 +2142,20 @@ st:                      if (is_imm8(insn->off))
 
                                ex->data = EX_TYPE_BPF;

arch/x86/net/bpf_jit_comp.c st: if (is_imm8(insn->off)) · line 2151

Quoted 10 lines
arch/x86/net/bpf_jit_comp.c +7 -2
 
-                               ex->fixup = (prog - start_of_ldx) |
-                                       ((BPF_CLASS(insn->code) == BPF_LDX ? reg2pt_regs[dst_reg] : DONT_CLEAR) << 8);
+                               if (BPF_CLASS(insn->code) == BPF_LDX) {
+                                       arena_reg = reg2pt_regs[src_reg];
+                                       fixup_reg = reg2pt_regs[dst_reg];
+                               } else {
+                                       arena_reg = reg2pt_regs[dst_reg];
+                                       fixup_reg = DONT_CLEAR;
+                               }

here it's probably also correct, since x86 jit is using r12 to add kern_vm_start.
A comment is necessary.

Quoted 65 lines
arch/x86/net/bpf_jit_comp.c +9 -1
+
+                               ex->fixup = FIELD_PREP(FIXUP_INSN_LEN_MASK, prog - start_of_ldx) |
+                                           FIELD_PREP(FIXUP_ARENA_REG_MASK, arena_reg) |
+                                           FIELD_PREP(FIXUP_REG_MASK, fixup_reg);
+                               ex->fixup |= FIXUP_ARENA_ACCESS;
+
+                               ex->data |= FIELD_PREP(DATA_ARENA_OFFSET_MASK, insn->off);
                        }
                        break;
 
@@ -2208,7 +2273,8 @@ st:                       if (is_imm8(insn->off))
                                 * End result: x86 insn "mov rbx, qword ptr [rax+0x14]"
                                 * of 4 bytes will be ignored and rbx will be zero inited.
                                 */
-                               ex->fixup = (prog - start_of_ldx) | (reg2pt_regs[dst_reg] << 8);
+                               ex->fixup = FIELD_PREP(FIXUP_INSN_LEN_MASK, prog - start_of_ldx) |
+                                           FIELD_PREP(FIXUP_REG_MASK, reg2pt_regs[dst_reg]);
                        }
                        break;
 
include/linux/bpf.h +6 -0
index d133171c4d2a9..41f776071ff51 100644
@@ -2881,6 +2881,7 @@ void bpf_dynptr_init(struct bpf_dynptr_kern *ptr, void *data,
                     enum bpf_dynptr_type type, u32 offset, u32 size);
 void bpf_dynptr_set_null(struct bpf_dynptr_kern *ptr);
 void bpf_dynptr_set_rdonly(struct bpf_dynptr_kern *ptr);
+void bpf_prog_report_arena_violation(bool write, unsigned long addr, unsigned long fault_ip);
 
 #else /* !CONFIG_BPF_SYSCALL */
 static inline struct bpf_prog *bpf_prog_get(u32 ufd)
@@ -3168,6 +3169,11 @@ static inline void bpf_dynptr_set_null(struct bpf_dynptr_kern *ptr)
 static inline void bpf_dynptr_set_rdonly(struct bpf_dynptr_kern *ptr)
 {
 }
+
+static inline void bpf_prog_report_arena_violation(bool write, unsigned long addr,
+                                                  unsigned long fault_ip)
+{
+}
 #endif /* CONFIG_BPF_SYSCALL */
 
 static __always_inline int
kernel/bpf/arena.c +13 -0
index 5b37753799d20..d0b31b40d3826 100644
@@ -633,3 +633,33 @@ static int __init kfunc_init(void)
        return register_btf_kfunc_id_set(BPF_PROG_TYPE_UNSPEC, &common_kfunc_set);
 }
 late_initcall(kfunc_init);
+
+void bpf_prog_report_arena_violation(bool write, unsigned long addr, unsigned long fault_ip)
+{
+       struct bpf_stream_stage ss;
+       struct bpf_prog *prog;
+       u64 user_vm_start;
+
+       /*
+        * The RCU read lock is held to safely traverse the latch tree, but we
+        * don't need its protection when accessing the prog, since it will not
+        * disappear while we are handling the fault.
+        */
+       rcu_read_lock();

kernel/bpf/arena.c static int __init kfunc_init(void) · line 658

Quoted 10 lines
kernel/bpf/arena.c +10 -0
+       prog = bpf_prog_ksym_find(fault_ip);
+       rcu_read_unlock();
+       if (!prog)
+               return;
+
+       /* Use main prog for stream access */
+       prog = prog->aux->main_prog_aux->prog;
+
+       user_vm_start = bpf_arena_get_user_vm_start(prog->aux->arena);
+       addr += (user_vm_start >> 32) << 32;

in arena.c we already have:
static u64 clear_lo32(u64 val)

{
        return val & ~(u64)~0U;
}

pls use it.

pw-bot: cr

author maintainer
#5 lore raw

Re: [PATCH bpf-next v6 3/5] bpf: Report arena faults to BPF stderr

Quoted 114 lines · Alexei Starovoitov
Quoted 99 lines · Puranjay Mohan

Begin reporting arena page faults and the faulting address to BPF program's stderr, this patch adds support in the arm64 and x86-64 JITs, support for other archs can be added later.

The fault handlers receive the 32 bit address in the arena region so the upper 32 bits of user_vm_start is added to it before printing the address. This is what the user would expect to see as this is what is printed by bpf_printk() is you pass it an address returned by bpf_arena_alloc_pages();

---
 arch/arm64/net/bpf_jit_comp.c | 52 ++++++++++++++++++++++++
 arch/x86/net/bpf_jit_comp.c   | 76 ++++++++++++++++++++++++++++++++---
 include/linux/bpf.h           |  6 +++
 kernel/bpf/arena.c            | 30 ++++++++++++++
 4 files changed, 159 insertions(+), 5 deletions(-)
arch/arm64/net/bpf_jit_comp.c +41 -0
index e6d1fdc1e6f52..556ab2fd222d8 100644
@@ -1066,6 +1066,30 @@ static void build_epilogue(struct jit_ctx *ctx, bool was_classic)
        emit(A64_RET(A64_LR), ctx);
 }
 
+/*
+ * Metadata encoding for exception handling in JITed code.
+ *
+ * Format of `fixup` field in `struct exception_table_entry`:
+ *
+ * Bit layout of `fixup` (32-bit):
+ *
+ * +-----------+--------+-----------+-----------+----------+
+ * |   31-27   | 26-22  |     21    |   20-16   |   15-0   |
+ * |           |        |           |           |          |
+ * | FIXUP_REG | Unused | ARENA_ACC | ARENA_REG |  OFFSET  |
+ * +-----------+--------+-----------+-----------+----------+
+ *
+ * - OFFSET (16 bits): Offset used to compute address for Load/Store instruction.
+ * - ARENA_REG (5 bits): Register that is used to calculate the address for load/store when
+ *                       accessing the arena region.
+ * - ARENA_ACCESS (1 bit): This bit is set when the faulting instruction accessed the arena region.
+ * - FIXUP_REG (5 bits): Destination register for the load instruction (cleared on fault) or set to
+ *                       DONT_CLEAR if it is a store instruction.
+ */
+
+#define BPF_FIXUP_OFFSET_MASK      GENMASK(15, 0)
+#define BPF_FIXUP_ARENA_REG_MASK   GENMASK(20, 16)
+#define BPF_ARENA_ACCESS           BIT(21)
 #define BPF_FIXUP_REG_MASK     GENMASK(31, 27)
 #define DONT_CLEAR 5 /* Unused ARM64 register from BPF's POV */
 
@@ -1073,11 +1097,22 @@ bool ex_handler_bpf(const struct exception_table_entry *ex,
                    struct pt_regs *regs)
 {
        int dst_reg = FIELD_GET(BPF_FIXUP_REG_MASK, ex->fixup);
+       s16 off = FIELD_GET(BPF_FIXUP_OFFSET_MASK, ex->fixup);
+       int arena_reg = FIELD_GET(BPF_FIXUP_ARENA_REG_MASK, ex->fixup);
+       bool is_arena = !!(ex->fixup & BPF_ARENA_ACCESS);
+       bool is_write = (dst_reg == DONT_CLEAR);
+       unsigned long addr;
+
+       if (is_arena) {
+               addr = regs->regs[arena_reg] + off;
+               bpf_prog_report_arena_violation(is_write, addr, regs->pc);
+       }
 
        if (dst_reg != DONT_CLEAR)
                regs->regs[dst_reg] = 0;
        /* Skip the faulting instruction */
        regs->pc += AARCH64_INSN_SIZE;
+
        return true;
 }
 
@@ -1087,6 +1122,9 @@ static int add_exception_handler(const struct bpf_insn *insn,
                                 int dst_reg)
 {
        off_t ins_offset;
+       s16 off = insn->off;
+       bool is_arena;
+       int arena_reg;
        unsigned long pc;
        struct exception_table_entry *ex;
 
@@ -1100,6 +1138,9 @@ static int add_exception_handler(const struct bpf_insn *insn,
                                BPF_MODE(insn->code) != BPF_PROBE_ATOMIC)
                return 0;
 
+       is_arena = (BPF_MODE(insn->code) == BPF_PROBE_MEM32) ||
+                  (BPF_MODE(insn->code) == BPF_PROBE_ATOMIC);
+
        if (!ctx->prog->aux->extable ||
            WARN_ON_ONCE(ctx->exentry_idx >= ctx->prog->aux->num_exentries))
                return -EINVAL;

arch/arm64/net/bpf_jit_comp.c static int add_exception_handler(const struct bpf_insn *insn, · line 1180

Quoted 9 lines
arch/arm64/net/bpf_jit_comp.c +6 -0
@@ -1131,6 +1172,17 @@ static int add_exception_handler(const struct bpf_insn *insn,
 
        ex->fixup = FIELD_PREP(BPF_FIXUP_REG_MASK, dst_reg);
 
+       if (is_arena) {
+               ex->fixup |= BPF_ARENA_ACCESS;
+               if (BPF_CLASS(insn->code) == BPF_LDX)
+                       arena_reg = bpf2a64[insn->src_reg];
+               else
+                       arena_reg = bpf2a64[insn->dst_reg];

I think this is correct, since insn->src/dst_reg is the register before the adjustment of:
emit(A64_ADD(1, tmp2, src, arena_vm_base)

Quoted 3 lines

so ex handler doing regs->regs[arena_reg]
should read 64-bit with upper 32-bit being zero.
right?

Yes, this is the un-adjusted address with upper 32-bits being zero because of a preceeding addr_space_cast(r<n>, 0x0, 0x1) instruction. ex_handler_bpf will get this 32-bit address and pass it to bpf_prog_report_arena_violation() which will add upper 32-bits of user_vm_start to it before printing it in stderr stream.

Quoted 1 line

If so, please add a comment describing this subtle logic.

I will add a comment in both x86 and arm64 JITs

Quoted 195 lines
Quoted 102 lines
arch/arm64/net/bpf_jit_comp.c +5 -0
+
+               ex->fixup |=  FIELD_PREP(BPF_FIXUP_OFFSET_MASK, off) |
+                             FIELD_PREP(BPF_FIXUP_ARENA_REG_MASK, arena_reg);
+       }
+
        ex->type = EX_TYPE_BPF;
 
        ctx->exentry_idx++;
arch/x86/net/bpf_jit_comp.c +55 -2
index 7e3fca1646203..007c273f3deea 100644
@@ -8,6 +8,7 @@
 #include <linux/netdevice.h>
 #include <linux/filter.h>
 #include <linux/if_vlan.h>
+#include <linux/bitfield.h>
 #include <linux/bpf.h>
 #include <linux/memory.h>
 #include <linux/sort.h>
@@ -1388,16 +1389,67 @@ static int emit_atomic_ld_st_index(u8 **pprog, u32 atomic_op, u32 size,
        return 0;
 }
 
+/*
+ * Metadata encoding for exception handling in JITed code.
+ *
+ * Format of `fixup` and `data` fields in `struct exception_table_entry`:
+ *
+ * Bit layout of `fixup` (32-bit):
+ *
+ * +-----------+--------+-----------+---------+----------+
+ * | 31        | 30-24  |   23-16   |   15-8  |    7-0   |
+ * |           |        |           |         |          |
+ * | ARENA_ACC | Unused | ARENA_REG | DST_REG | INSN_LEN |
+ * +-----------+--------+-----------+---------+----------+
+ *
+ * - INSN_LEN (8 bits): Length of faulting insn (max x86 insn = 15 bytes (fits in 8 bits)).
+ * - DST_REG  (8 bits): Offset of dst_reg from reg2pt_regs[] (max offset = 112 (fits in 8 bits)).
+ *                      This is set to DONT_CLEAR if the insn is a store.
+ * - ARENA_REG (8 bits): Offset of the register that is used to calculate the
+ *                       address for load/store when accessing the arena region.
+ * - ARENA_ACCESS (1 bit): This bit is set when the faulting instruction accessed the arena region.
+ *
+ * Bit layout of `data` (32-bit):
+ *
+ * +--------------+--------+--------------+
+ * |   31-16     |  15-8  |     7-0      |
+ * |              |       |              |
+ * | ARENA_OFFSET | Unused |  EX_TYPE_BPF |
+ * +--------------+--------+--------------+
+ *
+ * - ARENA_OFFSET (16 bits): Offset used to calculate the address for load/store when
+ *                           accessing the arena region.
+ */
+
 #define DONT_CLEAR 1
+#define FIXUP_INSN_LEN_MASK    GENMASK(7, 0)
+#define FIXUP_REG_MASK         GENMASK(15, 8)
+#define FIXUP_ARENA_REG_MASK   GENMASK(23, 16)
+#define FIXUP_ARENA_ACCESS     BIT(31)
+#define DATA_ARENA_OFFSET_MASK GENMASK(31, 16)
 
 bool ex_handler_bpf(const struct exception_table_entry *x, struct pt_regs *regs)
 {
-       u32 reg = x->fixup >> 8;
+       u32 reg = FIELD_GET(FIXUP_REG_MASK, x->fixup);
+       u32 insn_len = FIELD_GET(FIXUP_INSN_LEN_MASK, x->fixup);
+       bool is_arena = !!(x->fixup & FIXUP_ARENA_ACCESS);
+       bool is_write = (reg == DONT_CLEAR);
+       unsigned long addr;
+       s16 off;
+       u32 arena_reg;
+
+       if (is_arena) {
+               arena_reg = FIELD_GET(FIXUP_ARENA_REG_MASK, x->fixup);
+               off = FIELD_GET(DATA_ARENA_OFFSET_MASK, x->data);
+               addr = *(unsigned long *)((void *)regs + arena_reg) + off;
+               bpf_prog_report_arena_violation(is_write, addr, regs->ip);
+       }
 
        /* jump over faulting load and clear dest register */
        if (reg != DONT_CLEAR)
                *(unsigned long *)((void *)regs + reg) = 0;
-       regs->ip += x->fixup & 0xff;
+       regs->ip += insn_len;
+
        return true;
 }
 
@@ -2070,6 +2122,7 @@ st:                       if (is_imm8(insn->off))
                        {
                                struct exception_table_entry *ex;
                                u8 *_insn = image + proglen + (start_of_ldx - temp);
+                               u32 arena_reg, fixup_reg;
                                s64 delta;
 
                                if (!bpf_prog->aux->extable)
@@ -2089,8 +2142,20 @@ st:                      if (is_imm8(insn->off))
 
                                ex->data = EX_TYPE_BPF;

arch/x86/net/bpf_jit_comp.c st: if (is_imm8(insn->off)) · line 2151

Quoted 10 lines
arch/x86/net/bpf_jit_comp.c +7 -2
 
-                               ex->fixup = (prog - start_of_ldx) |
-                                       ((BPF_CLASS(insn->code) == BPF_LDX ? reg2pt_regs[dst_reg] : DONT_CLEAR) << 8);
+                               if (BPF_CLASS(insn->code) == BPF_LDX) {
+                                       arena_reg = reg2pt_regs[src_reg];
+                                       fixup_reg = reg2pt_regs[dst_reg];
+                               } else {
+                                       arena_reg = reg2pt_regs[dst_reg];
+                                       fixup_reg = DONT_CLEAR;
+                               }

here it's probably also correct, since x86 jit is using r12 to add kern_vm_start.
A comment is necessary.

Quoted 65 lines
arch/x86/net/bpf_jit_comp.c +9 -1
+
+                               ex->fixup = FIELD_PREP(FIXUP_INSN_LEN_MASK, prog - start_of_ldx) |
+                                           FIELD_PREP(FIXUP_ARENA_REG_MASK, arena_reg) |
+                                           FIELD_PREP(FIXUP_REG_MASK, fixup_reg);
+                               ex->fixup |= FIXUP_ARENA_ACCESS;
+
+                               ex->data |= FIELD_PREP(DATA_ARENA_OFFSET_MASK, insn->off);
                        }
                        break;
 
@@ -2208,7 +2273,8 @@ st:                       if (is_imm8(insn->off))
                                 * End result: x86 insn "mov rbx, qword ptr [rax+0x14]"
                                 * of 4 bytes will be ignored and rbx will be zero inited.
                                 */
-                               ex->fixup = (prog - start_of_ldx) | (reg2pt_regs[dst_reg] << 8);
+                               ex->fixup = FIELD_PREP(FIXUP_INSN_LEN_MASK, prog - start_of_ldx) |
+                                           FIELD_PREP(FIXUP_REG_MASK, reg2pt_regs[dst_reg]);
                        }
                        break;
 
include/linux/bpf.h +6 -0
index d133171c4d2a9..41f776071ff51 100644
@@ -2881,6 +2881,7 @@ void bpf_dynptr_init(struct bpf_dynptr_kern *ptr, void *data,
                     enum bpf_dynptr_type type, u32 offset, u32 size);
 void bpf_dynptr_set_null(struct bpf_dynptr_kern *ptr);
 void bpf_dynptr_set_rdonly(struct bpf_dynptr_kern *ptr);
+void bpf_prog_report_arena_violation(bool write, unsigned long addr, unsigned long fault_ip);
 
 #else /* !CONFIG_BPF_SYSCALL */
 static inline struct bpf_prog *bpf_prog_get(u32 ufd)
@@ -3168,6 +3169,11 @@ static inline void bpf_dynptr_set_null(struct bpf_dynptr_kern *ptr)
 static inline void bpf_dynptr_set_rdonly(struct bpf_dynptr_kern *ptr)
 {
 }
+
+static inline void bpf_prog_report_arena_violation(bool write, unsigned long addr,
+                                                  unsigned long fault_ip)
+{
+}
 #endif /* CONFIG_BPF_SYSCALL */
 
 static __always_inline int
kernel/bpf/arena.c +13 -0
index 5b37753799d20..d0b31b40d3826 100644
@@ -633,3 +633,33 @@ static int __init kfunc_init(void)
        return register_btf_kfunc_id_set(BPF_PROG_TYPE_UNSPEC, &common_kfunc_set);
 }
 late_initcall(kfunc_init);
+
+void bpf_prog_report_arena_violation(bool write, unsigned long addr, unsigned long fault_ip)
+{
+       struct bpf_stream_stage ss;
+       struct bpf_prog *prog;
+       u64 user_vm_start;
+
+       /*
+        * The RCU read lock is held to safely traverse the latch tree, but we
+        * don't need its protection when accessing the prog, since it will not
+        * disappear while we are handling the fault.
+        */
+       rcu_read_lock();

kernel/bpf/arena.c static int __init kfunc_init(void) · line 658

Quoted 10 lines
kernel/bpf/arena.c +10 -0
+       prog = bpf_prog_ksym_find(fault_ip);
+       rcu_read_unlock();
+       if (!prog)
+               return;
+
+       /* Use main prog for stream access */
+       prog = prog->aux->main_prog_aux->prog;
+
+       user_vm_start = bpf_arena_get_user_vm_start(prog->aux->arena);
+       addr += (user_vm_start >> 32) << 32;

in arena.c we already have:
static u64 clear_lo32(u64 val)

Quoted 5 lines
{
        return val & ~(u64)~0U;
}

pls use it.

Ack.

Thanks,
Puranjay

author maintainer
#6 lore raw

PATCH v6 4/5 selftests: bpf: introduce __stderr and __stdout

In reply to Puranjay Mohan

Add __stderr and __stdout to validate the output of BPF streams for bpf selftests. Similar to __xlated, __jited, etc., __stderr/out can be used in the BPF progs to compare a string (regex supported) to the output in the bpf streams.

---
 tools/testing/selftests/bpf/progs/bpf_misc.h | 10 +++
 tools/testing/selftests/bpf/test_loader.c    | 90 ++++++++++++++++++++
 2 files changed, 100 insertions(+)
tools/testing/selftests/bpf/progs/bpf_misc.h +10 -0
index 72c2d72a245e5..7905396c9cc4e 100644
@@ -35,6 +35,12 @@
  *                   inside the brackets.
  * __msg_unpriv      Same as __msg but for unprivileged mode.
  *
+ * __stderr          Message expected to be found in bpf stderr stream. The
+ *                   same regex rules apply like __msg.
+ * __stderr_unpriv   Same as __stderr but for unpriveleged mode.
+ * __stdout          Same as __stderr but for stdout stream.
+ * __stdout_unpriv   Same as __stdout but for unpriveleged mode.
+ *
  * __xlated          Expect a line in a disassembly log after verifier applies rewrites.
  *                   Multiple __xlated attributes could be specified.
  *                   Regular expressions could be specified same way as in __msg.
@@ -140,6 +146,10 @@
 #define __caps_unpriv(caps)	__attribute__((btf_decl_tag("comment:test_caps_unpriv=" EXPAND_QUOTE(caps))))
 #define __load_if_JITed()	__attribute__((btf_decl_tag("comment:load_mode=jited")))
 #define __load_if_no_JITed()	__attribute__((btf_decl_tag("comment:load_mode=no_jited")))
+#define __stderr(msg)		__attribute__((btf_decl_tag("comment:test_expect_stderr=" XSTR(__COUNTER__) "=" msg)))
+#define __stderr_unpriv(msg)	__attribute__((btf_decl_tag("comment:test_expect_stderr_unpriv=" XSTR(__COUNTER__) "=" msg)))
+#define __stdout(msg)		__attribute__((btf_decl_tag("comment:test_expect_stdout=" XSTR(__COUNTER__) "=" msg)))
+#define __stdout_unpriv(msg)	__attribute__((btf_decl_tag("comment:test_expect_stdout_unpriv=" XSTR(__COUNTER__) "=" msg)))
 
 /* Define common capabilities tested using __caps_unpriv */
 #define CAP_NET_ADMIN		12
tools/testing/selftests/bpf/test_loader.c +90 -0

144 lines not on this page · raw message

Signature · 1 line
2.47.3
maintainer Acked-by on v6 4/5
#7 lore raw

Re: [PATCH bpf-next v6 4/5] selftests: bpf: introduce __stderr and __stdout

Quoted 7 lines · Puranjay Mohan

Add __stderr and __stdout to validate the output of BPF streams for bpf selftests. Similar to __xlated, __jited, etc., __stderr/out can be used in the BPF progs to compare a string (regex supported) to the output in the bpf streams.

---

Thank you for adding this. I think it's a nice option for tests that had to invent a complex __retval() when checking for multiple conditions to be true in bpf program.

[...]

author maintainer
#8 lore raw

PATCH v6 5/5 selftests/bpf: Add tests for arena fault reporting

In reply to Puranjay Mohan

Add selftests for testing the reporting of arena page faults through BPF streams. Two new bpf programs are added that read and write to an unmapped arena address and the fault reporting is verified in the userspace through streams.

The added bpf programs need to access the user_vm_start in struct bpf_arena, this is done by casting &arena to struct bpf_arena *, but barrier_var() is used on this ptr before accessing ptr->user_vm_start; to stop GCC from issuing an out-of-bound access due to the cast from smaller map struct to larger "struct bpf_arena"

Other tests related to streams have been converted to use __stderr in progs/stream.c directly and test_stream_errors() in prog_tests/stream.c has been repurposed to validate the arena fault address printed in the stderr stream. This can't be directly validated using __stderr because the address is dynamic.

---
 .../testing/selftests/bpf/prog_tests/stream.c |  53 ++----
 tools/testing/selftests/bpf/progs/stream.c    | 158 ++++++++++++++++++
 2 files changed, 176 insertions(+), 35 deletions(-)
tools/testing/selftests/bpf/prog_tests/stream.c +18 -35
index 9d0e5d93edee7..61ab1da9b189b 100644
@@ -18,29 +18,9 @@ void test_stream_success(void)
 	return;
 }
 
-struct {
-	int prog_off;
-	const char *errstr;
-} stream_error_arr[] = {
-	{
-		offsetof(struct stream, progs.stream_cond_break),
-		"ERROR: Timeout detected for may_goto instruction\n"
-		"CPU: [0-9]+ UID: 0 PID: [0-9]+ Comm: .*\n"
-		"Call trace:\n"
-		"([a-zA-Z_][a-zA-Z0-9_]*\\+0x[0-9a-fA-F]+/0x[0-9a-fA-F]+\n"
-		"|[ \t]+[^\n]+\n)*",
-	},
-	{
-		offsetof(struct stream, progs.stream_deadlock),
-		"ERROR: AA or ABBA deadlock detected for bpf_res_spin_lock\n"
-		"Attempted lock   = (0x[0-9a-fA-F]+)\n"
-		"Total held locks = 1\n"
-		"Held lock\\[ 0\\] = \\1\n"  // Lock address must match
-		"CPU: [0-9]+ UID: 0 PID: [0-9]+ Comm: .*\n"
-		"Call trace:\n"
-		"([a-zA-Z_][a-zA-Z0-9_]*\\+0x[0-9a-fA-F]+/0x[0-9a-fA-F]+\n"
-		"|[ \t]+[^\n]+\n)*",
-	},
+int prog_off[] = {
+	offsetof(struct stream, progs.stream_arena_read_fault),
+	offsetof(struct stream, progs.stream_arena_write_fault),
 };
 
 static int match_regex(const char *pattern, const char *string)
@@ -56,34 +36,33 @@ static int match_regex(const char *pattern, const char *string)
 	return rc == 0 ? 1 : 0;
 }
 
-void test_stream_errors(void)
+void test_stream_arena_fault_address(void)
 {
 	LIBBPF_OPTS(bpf_test_run_opts, opts);
 	LIBBPF_OPTS(bpf_prog_stream_read_opts, ropts);
 	struct stream *skel;
 	int ret, prog_fd;
 	char buf[1024];
+	char fault_addr[64];
 
 	skel = stream__open_and_load();
 	if (!ASSERT_OK_PTR(skel, "stream__open_and_load"))
 		return;
 
-	for (int i = 0; i < ARRAY_SIZE(stream_error_arr); i++) {
+	for (int i = 0; i < ARRAY_SIZE(prog_off); i++) {
 		struct bpf_program **prog;
 
-		prog = (struct bpf_program **)(((char *)skel) + stream_error_arr[i].prog_off);
+		prog = (struct bpf_program **)(((char *)skel) + prog_off[i]);
 		prog_fd = bpf_program__fd(*prog);
 		ret = bpf_prog_test_run_opts(prog_fd, &opts);
 		ASSERT_OK(ret, "ret");
 		ASSERT_OK(opts.retval, "retval");
 
-#if !defined(__x86_64__) && !defined(__s390x__) && !defined(__aarch64__)
-		ASSERT_TRUE(1, "Timed may_goto unsupported, skip.");
-		if (i == 0) {
-			ret = bpf_prog_stream_read(prog_fd, 2, buf, sizeof(buf), &ropts);
-			ASSERT_EQ(ret, 0, "stream read");
-			continue;
-		}
+#if !defined(__x86_64__) && !defined(__aarch64__)
+		ASSERT_TRUE(1, "Arena fault reporting unsupported, skip.");
+		ret = bpf_prog_stream_read(prog_fd, 2, buf, sizeof(buf), &ropts);
+		ASSERT_EQ(ret, 0, "stream read");
+		continue;
 #endif
 
 		ret = bpf_prog_stream_read(prog_fd, BPF_STREAM_STDERR, buf, sizeof(buf), &ropts);
@@ -91,9 +70,13 @@ void test_stream_errors(void)
 		ASSERT_LE(ret, 1023, "len for buf");
 		buf[ret] = '\0';
 
-		ret = match_regex(stream_error_arr[i].errstr, buf);
-		if (!ASSERT_TRUE(ret == 1, "regex match"))
+		sprintf(fault_addr, "0x%lx", skel->bss->fault_addr);
+		ret = match_regex(fault_addr, buf);
+
+		if (!ASSERT_TRUE(ret == 1, "regex match")) {
 			fprintf(stderr, "Output from stream:\n%s\n", buf);
+			fprintf(stderr, "Fault Addr: 0x%lx\n", skel->bss->fault_addr);
+		}
 	}
 
 	stream__destroy(skel);
tools/testing/selftests/bpf/progs/stream.c +158 -0

191 lines not on this page · raw message

Signature · 1 line
2.47.3
#9 lore raw

Re: [PATCH bpf-next v6 5/5] selftests/bpf: Add tests for arena fault reporting

tools/testing/selftests/bpf/progs/stream.c int stream_syscall(void *ctx) · line 136

Quoted 13 lines · Puranjay Mohan
tools/testing/selftests/bpf/progs/stream.c +12 -0
+int stream_arena_write_fault(void *ctx)
+{
+       struct bpf_arena *ptr = (void *)&arena;
+       u64 user_vm_start;
+
+       /* Prevent GCC bounds warning: casting &arena to struct bpf_arena *
+        * triggers bounds checking since the map definition is smaller than struct
+        * bpf_arena. barrier_var() makes the pointer opaque to GCC, preventing the
+        * bounds analysis
+        */
+       barrier_var(ptr);
+       user_vm_start =  ptr->user_vm_start;

nit: Too many spaces.

maintainer
#10 lore raw

Re: [PATCH bpf-next v6 5/5] selftests/bpf: Add tests for arena fault reporting

In reply to Puranjay Mohan

On Mon, 2025-09-08 at 16:36 +0000, Puranjay Mohan wrote:

[...]

Quoted 21 lines
tools/testing/selftests/bpf/prog_tests/stream.c +0 -13
index 9d0e5d93edee7..61ab1da9b189b 100644
@@ -18,29 +18,9 @@ void test_stream_success(void)
 	return;
 }
 
-struct {
-	int prog_off;
-	const char *errstr;
-} stream_error_arr[] = {
-	{
-		offsetof(struct stream, progs.stream_cond_break),
-		"ERROR: Timeout detected for may_goto instruction\n"
-		"CPU: [0-9]+ UID: 0 PID: [0-9]+ Comm: .*\n"
-		"Call trace:\n"
-		"([a-zA-Z_][a-zA-Z0-9_]*\\+0x[0-9a-fA-F]+/0x[0-9a-fA-F]+\n"
-		"|[ \t]+[^\n]+\n)*",
-	},
-	{

tools/testing/selftests/bpf/prog_tests/stream.c void test_stream_success(void) · line 43

Quoted 10 lines
tools/testing/selftests/bpf/prog_tests/stream.c +0 -10
-		offsetof(struct stream, progs.stream_deadlock),
-		"ERROR: AA or ABBA deadlock detected for bpf_res_spin_lock\n"
-		"Attempted lock   = (0x[0-9a-fA-F]+)\n"
-		"Total held locks = 1\n"
-		"Held lock\\[ 0\\] = \\1\n"  // Lock address must match
-		"CPU: [0-9]+ UID: 0 PID: [0-9]+ Comm: .*\n"
-		"Call trace:\n"
-		"([a-zA-Z_][a-zA-Z0-9_]*\\+0x[0-9a-fA-F]+/0x[0-9a-fA-F]+\n"
-		"|[ \t]+[^\n]+\n)*",
-	},

Nit: maybe put an update of the old tests to a separate commit?

Quoted 16 lines
tools/testing/selftests/bpf/prog_tests/stream.c +4 -1
+int prog_off[] = {
+	offsetof(struct stream, progs.stream_arena_read_fault),
+	offsetof(struct stream, progs.stream_arena_write_fault),
 };
 
 static int match_regex(const char *pattern, const char *string)
@@ -56,34 +36,33 @@ static int match_regex(const char *pattern, const char *string)
 	return rc == 0 ? 1 : 0;
 }
 
-void test_stream_errors(void)
+void test_stream_arena_fault_address(void)
 {
 	LIBBPF_OPTS(bpf_test_run_opts, opts);
 	LIBBPF_OPTS(bpf_prog_stream_read_opts, ropts);
 	struct stream *skel;

tools/testing/selftests/bpf/prog_tests/stream.c static int match_regex(const char *pattern, const char *string) · line 52

Quoted 10 lines
tools/testing/selftests/bpf/prog_tests/stream.c +2 -1
 	int ret, prog_fd;
 	char buf[1024];
+	char fault_addr[64];
 
 	skel = stream__open_and_load();
 	if (!ASSERT_OK_PTR(skel, "stream__open_and_load"))
 		return;
 
-	for (int i = 0; i < ARRAY_SIZE(stream_error_arr); i++) {
+	for (int i = 0; i < ARRAY_SIZE(prog_off); i++) {

Nit: start a sub-test for each i?

Quoted 12 lines
tools/testing/selftests/bpf/prog_tests/stream.c +1 -4
 		struct bpf_program **prog;
 
-		prog = (struct bpf_program **)(((char *)skel) + stream_error_arr[i].prog_off);
+		prog = (struct bpf_program **)(((char *)skel) + prog_off[i]);
 		prog_fd = bpf_program__fd(*prog);
 		ret = bpf_prog_test_run_opts(prog_fd, &opts);
 		ASSERT_OK(ret, "ret");
 		ASSERT_OK(opts.retval, "retval");
 
-#if !defined(__x86_64__) && !defined(__s390x__) && !defined(__aarch64__)
-		ASSERT_TRUE(1, "Timed may_goto unsupported, skip.");
-		if (i == 0) {

tools/testing/selftests/bpf/prog_tests/stream.c static int match_regex(const char *pattern, const char *string) · line 66

Quoted 10 lines
tools/testing/selftests/bpf/prog_tests/stream.c +5 -4
-			ret = bpf_prog_stream_read(prog_fd, 2, buf, sizeof(buf), &ropts);
-			ASSERT_EQ(ret, 0, "stream read");
-			continue;
-		}
+#if !defined(__x86_64__) && !defined(__aarch64__)
+		ASSERT_TRUE(1, "Arena fault reporting unsupported, skip.");
+		ret = bpf_prog_stream_read(prog_fd, 2, buf, sizeof(buf), &ropts);
+		ASSERT_EQ(ret, 0, "stream read");
+		continue;
 #endif

Nit: move this #if !defined to the beginning of the function and add
test__skip() call there?

Quoted 18 lines
tools/testing/selftests/bpf/prog_tests/stream.c +6 -2
 		ret = bpf_prog_stream_read(prog_fd, BPF_STREAM_STDERR, buf, sizeof(buf), &ropts);
@@ -91,9 +70,13 @@ void test_stream_errors(void)
 		ASSERT_LE(ret, 1023, "len for buf");
 		buf[ret] = '\0';
 
-		ret = match_regex(stream_error_arr[i].errstr, buf);
-		if (!ASSERT_TRUE(ret == 1, "regex match"))
+		sprintf(fault_addr, "0x%lx", skel->bss->fault_addr);
+		ret = match_regex(fault_addr, buf);
+
+		if (!ASSERT_TRUE(ret == 1, "regex match")) {
 			fprintf(stderr, "Output from stream:\n%s\n", buf);
+			fprintf(stderr, "Fault Addr: 0x%lx\n", skel->bss->fault_addr);
+		}
 	}
 
 	stream__destroy(skel);