Skip to contentSkip to conversation

bpftool: Enable aarch64 ISA extensions for JIT disassembly

PATCH Superseded Puranjay Mohan 3 participants 3 messages Mar 6 – Mar 9

Subsystems (2)

v1 of 4 Superseded Latest is v4 →

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 bpftool: Enable aarch64 ISA extensions for JIT disassembly

The LLVM disassembler needs ISA extension features enabled to correctly decode instructions from those extensions. On aarch64, without these features, instructions like LSE atomics (e.g. ldaddal) are silently decoded as incorrect instructions and disassembly is truncated.

Use LLVMCreateDisasmCPUFeatures() with "+all" features for aarch64 targets so that the disassembler can handle any instruction the kernel JIT might emit.

Before:

int bench_trigger_uprobe(void * ctx):
bpf_prog_538c6a43d1c6b84c_bench_trigger_uprobe:
; int cpu = bpf_get_smp_processor_id();

   0:   mov     x9, x30
   4:   nop
   8:   stp     x29, x30, [sp, #-16]!
   c:   mov     x29, sp
  10:   stp     xzr, x26, [sp, #-16]!
  14:   mov     x26, sp
  18:   mrs     x10, SP_EL0
  1c:   ldr     w7, [x10, #16]
; __sync_add_and_fetch(&hits[cpu & CPU_MASK].value, 1);
  20:   and     w7, w7, #0xff
; __sync_add_and_fetch(&hits[cpu & CPU_MASK].value, 1);
  24:   lsl     x7, x7, #7
  28:   mov     x0, #-281474976710656
  2c:   movk    x0, #32768, lsl #32
  30:   movk    x0, #35407, lsl #16
  34:   add     x0, x0, x7
  38:   mov     x1, #1

; __sync_add_and_fetch(&hits[cpu & CPU_MASK].value, 1);
3c: mov x1, #1

After:

int bench_trigger_uprobe(void * ctx):
bpf_prog_538c6a43d1c6b84c_bench_trigger_uprobe:
; int cpu = bpf_get_smp_processor_id();

   0:   mov     x9, x30
   4:   nop
   8:   stp     x29, x30, [sp, #-16]!
   c:   mov     x29, sp
  10:   stp     xzr, x26, [sp, #-16]!
  14:   mov     x26, sp
  18:   mrs     x10, SP_EL0
  1c:   ldr     w7, [x10, #16]
; __sync_add_and_fetch(&hits[cpu & CPU_MASK].value, 1);
  20:   and     w7, w7, #0xff
; __sync_add_and_fetch(&hits[cpu & CPU_MASK].value, 1);
  24:   lsl     x7, x7, #7
  28:   mov     x0, #-281474976710656
  2c:   movk    x0, #32768, lsl #32
  30:   movk    x0, #35407, lsl #16
  34:   add     x0, x0, x7
  38:   mov     x1, #1
; __sync_add_and_fetch(&hits[cpu & CPU_MASK].value, 1);
  3c:   ldaddal x1, x1, [x0]
; return 0;
  40:   mov     w7, #0
  44:   ldp     xzr, x26, [sp], #16
  48:   ldp     x29, x30, [sp], #16
  4c:   mov     x0, x7
  50:   ret
  54:   nop
  58:   ldr     x10, #8
  5c:   br      x10
---
 tools/bpf/bpftool/jit_disasm.c | 13 ++++++++++++-
 1 file changed, 12 insertions(+), 1 deletion(-)
tools/bpf/bpftool/jit_disasm.c +12 -1
index 8895b4e1f690..317d810a30ba 100644
@@ -93,7 +93,18 @@ init_context(disasm_ctx_t *ctx, const char *arch,
 		p_err("Failed to retrieve triple");
 		return -1;
 	}
-	*ctx = LLVMCreateDisasm(triple, NULL, 0, NULL, symbol_lookup_callback);
+
+	/*
+	 * Enable all aarch64 ISA extensions so the disassembler can handle any
+	 * instruction the kernel JIT might emit (e.g. ARM64 LSE atomics).
+	 */
+	if (strstr(triple, "aarch64"))
+		*ctx = LLVMCreateDisasmCPUFeatures(triple, "", "+all",
+						   NULL, 0, NULL,
+						   symbol_lookup_callback);
+	else
+		*ctx = LLVMCreateDisasm(triple, NULL, 0, NULL,
+					symbol_lookup_callback);
 	LLVMDisposeMessage(triple);
 
 	if (!*ctx) {

base-commit: 6dd780f973816133f189efec04118c1e6b1b443d

Signature · 1 line
2.47.3
maintainer
#1 lore raw
Quoted 68 lines · Puranjay Mohan

The LLVM disassembler needs ISA extension features enabled to correctly decode instructions from those extensions. On aarch64, without these features, instructions like LSE atomics (e.g. ldaddal) are silently decoded as incorrect instructions and disassembly is truncated.

Use LLVMCreateDisasmCPUFeatures() with "+all" features for aarch64 targets so that the disassembler can handle any instruction the kernel JIT might emit.

Before:

int bench_trigger_uprobe(void * ctx):
bpf_prog_538c6a43d1c6b84c_bench_trigger_uprobe:

; int cpu = bpf_get_smp_processor_id();
   0:   mov     x9, x30
   4:   nop
   8:   stp     x29, x30, [sp, #-16]!
   c:   mov     x29, sp
  10:   stp     xzr, x26, [sp, #-16]!
  14:   mov     x26, sp
  18:   mrs     x10, SP_EL0
  1c:   ldr     w7, [x10, #16]
; __sync_add_and_fetch(&hits[cpu & CPU_MASK].value, 1);
  20:   and     w7, w7, #0xff
; __sync_add_and_fetch(&hits[cpu & CPU_MASK].value, 1);
  24:   lsl     x7, x7, #7
  28:   mov     x0, #-281474976710656
  2c:   movk    x0, #32768, lsl #32
  30:   movk    x0, #35407, lsl #16
  34:   add     x0, x0, x7
  38:   mov     x1, #1

; __sync_add_and_fetch(&hits[cpu & CPU_MASK].value, 1);
3c: mov x1, #1

After:

int bench_trigger_uprobe(void * ctx):
bpf_prog_538c6a43d1c6b84c_bench_trigger_uprobe:

; int cpu = bpf_get_smp_processor_id();
   0:   mov     x9, x30
   4:   nop
   8:   stp     x29, x30, [sp, #-16]!
   c:   mov     x29, sp
  10:   stp     xzr, x26, [sp, #-16]!
  14:   mov     x26, sp
  18:   mrs     x10, SP_EL0
  1c:   ldr     w7, [x10, #16]
; __sync_add_and_fetch(&hits[cpu & CPU_MASK].value, 1);
  20:   and     w7, w7, #0xff
; __sync_add_and_fetch(&hits[cpu & CPU_MASK].value, 1);
  24:   lsl     x7, x7, #7
  28:   mov     x0, #-281474976710656
  2c:   movk    x0, #32768, lsl #32
  30:   movk    x0, #35407, lsl #16
  34:   add     x0, x0, x7
  38:   mov     x1, #1
; __sync_add_and_fetch(&hits[cpu & CPU_MASK].value, 1);
  3c:   ldaddal x1, x1, [x0]
; return 0;
  40:   mov     w7, #0
  44:   ldp     xzr, x26, [sp], #16
  48:   ldp     x29, x30, [sp], #16
  4c:   mov     x0, x7
  50:   ret
  54:   nop
  58:   ldr     x10, #8
  5c:   br      x10
Quoted 1 line

Thanks! The change looks OK to me, I just have a few nits below.

Quoted 9 lines
---
 tools/bpf/bpftool/jit_disasm.c | 13 ++++++++++++-
 1 file changed, 12 insertions(+), 1 deletion(-)
tools/bpf/bpftool/jit_disasm.c +0 -0
index 8895b4e1f690..317d810a30ba 100644

tools/bpf/bpftool/jit_disasm.c init_context(disasm_ctx_t *ctx, const char *arch, · line 101

Quoted 10 lines
tools/bpf/bpftool/jit_disasm.c +6 -1
@@ -93,7 +93,18 @@ init_context(disasm_ctx_t *ctx, const char *arch,
 		p_err("Failed to retrieve triple");
 		return -1;
 	}
-	*ctx = LLVMCreateDisasm(triple, NULL, 0, NULL, symbol_lookup_callback);
+
+	/*
+	 * Enable all aarch64 ISA extensions so the disassembler can handle any
+	 * instruction the kernel JIT might emit (e.g. ARM64 LSE atomics).
+	 */
+	if (strstr(triple, "aarch64"))

I don't think the arch name can appear in the middle of the triple, can it? So maybe just strncmp(triple, "aarch64", 7)?

tools/bpf/bpftool/jit_disasm.c init_context(disasm_ctx_t *ctx, const char *arch, · line 107

Quoted 6 lines
tools/bpf/bpftool/jit_disasm.c +6 -0
+		*ctx = LLVMCreateDisasmCPUFeatures(triple, "", "+all",
+						   NULL, 0, NULL,
+						   symbol_lookup_callback);
+	else
+		*ctx = LLVMCreateDisasm(triple, NULL, 0, NULL,
+					symbol_lookup_callback);

Formatting nit: The line length limit is 100 char so you could have these LLVMCreateDisasm...() calls on 2 and 1 lines (rather than 3 and 2 lines), respectively.

Thanks,
Quentin

reviewer Acked-by on v1 1/1
#2 lore raw
In reply to Puranjay Mohan
Quoted 68 lines · Puranjay Mohan

The LLVM disassembler needs ISA extension features enabled to correctly decode instructions from those extensions. On aarch64, without these features, instructions like LSE atomics (e.g. ldaddal) are silently decoded as incorrect instructions and disassembly is truncated.

Use LLVMCreateDisasmCPUFeatures() with "+all" features for aarch64 targets so that the disassembler can handle any instruction the kernel JIT might emit.

Before:

int bench_trigger_uprobe(void * ctx):
bpf_prog_538c6a43d1c6b84c_bench_trigger_uprobe:

; int cpu = bpf_get_smp_processor_id();
    0:   mov     x9, x30
    4:   nop
    8:   stp     x29, x30, [sp, #-16]!
    c:   mov     x29, sp
   10:   stp     xzr, x26, [sp, #-16]!
   14:   mov     x26, sp
   18:   mrs     x10, SP_EL0
   1c:   ldr     w7, [x10, #16]
; __sync_add_and_fetch(&hits[cpu & CPU_MASK].value, 1);
   20:   and     w7, w7, #0xff
; __sync_add_and_fetch(&hits[cpu & CPU_MASK].value, 1);
   24:   lsl     x7, x7, #7
   28:   mov     x0, #-281474976710656
   2c:   movk    x0, #32768, lsl #32
   30:   movk    x0, #35407, lsl #16
   34:   add     x0, x0, x7
   38:   mov     x1, #1
; __sync_add_and_fetch(&hits[cpu & CPU_MASK].value, 1);
   3c:   mov     x1, #1

After:

int bench_trigger_uprobe(void * ctx):
bpf_prog_538c6a43d1c6b84c_bench_trigger_uprobe:

; int cpu = bpf_get_smp_processor_id();
    0:   mov     x9, x30
    4:   nop
    8:   stp     x29, x30, [sp, #-16]!
    c:   mov     x29, sp
   10:   stp     xzr, x26, [sp, #-16]!
   14:   mov     x26, sp
   18:   mrs     x10, SP_EL0
   1c:   ldr     w7, [x10, #16]
; __sync_add_and_fetch(&hits[cpu & CPU_MASK].value, 1);
   20:   and     w7, w7, #0xff
; __sync_add_and_fetch(&hits[cpu & CPU_MASK].value, 1);
   24:   lsl     x7, x7, #7
   28:   mov     x0, #-281474976710656
   2c:   movk    x0, #32768, lsl #32
   30:   movk    x0, #35407, lsl #16
   34:   add     x0, x0, x7
   38:   mov     x1, #1
; __sync_add_and_fetch(&hits[cpu & CPU_MASK].value, 1);
   3c:   ldaddal x1, x1, [x0]
; return 0;
   40:   mov     w7, #0
   44:   ldp     xzr, x26, [sp], #16
   48:   ldp     x29, x30, [sp], #16
   4c:   mov     x0, x7
   50:   ret
   54:   nop
   58:   ldr     x10, #8
   5c:   br      x10
Quoted 1 line
Quoted 9 lines
---
  tools/bpf/bpftool/jit_disasm.c | 13 ++++++++++++-
  1 file changed, 12 insertions(+), 1 deletion(-)
tools/bpf/bpftool/jit_disasm.c +0 -0
index 8895b4e1f690..317d810a30ba 100644

tools/bpf/bpftool/jit_disasm.c init_context(disasm_ctx_t *ctx, const char *arch, · line 101

Quoted 10 lines
tools/bpf/bpftool/jit_disasm.c +6 -1
@@ -93,7 +93,18 @@ init_context(disasm_ctx_t *ctx, const char *arch,
  		p_err("Failed to retrieve triple");
  		return -1;
  	}
-	*ctx = LLVMCreateDisasm(triple, NULL, 0, NULL, symbol_lookup_callback);
+
+	/*
+	 * Enable all aarch64 ISA extensions so the disassembler can handle any
+	 * instruction the kernel JIT might emit (e.g. ARM64 LSE atomics).
+	 */
+	if (strstr(triple, "aarch64"))

As mentioned by Quentin, "aarch64" should be in the leading triple, so it would be great to check with that instead of strstr.

There are another variant "arm64" which is used for Apple platformas, so we can ignore them.

Quoted 11 lines
tools/bpf/bpftool/jit_disasm.c +6 -0
+		*ctx = LLVMCreateDisasmCPUFeatures(triple, "", "+all",
+						   NULL, 0, NULL,
+						   symbol_lookup_callback);
+	else
+		*ctx = LLVMCreateDisasm(triple, NULL, 0, NULL,
+					symbol_lookup_callback);
  	LLVMDisposeMessage(triple);
 
  	if (!*ctx) {

base-commit: 6dd780f973816133f189efec04118c1e6b1b443d