Skip to contentSkip to conversation

net: bpf: arm64: address randomize and write protect JIT code

PATCH In review Daniel Borkmann 3 participants 4 messages Sep 16, 2014

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

PATCH v4 1/1 net: bpf: arm64: address randomize and write protect JIT code

This is the ARM64 variant for 314beb9bcab ("x86: bpf_jit_comp: secure bpf jit against spraying attacks").

Thanks to commit 11d91a770f1f ("arm64: Add CONFIG_DEBUG_SET_MODULE_RONX support") which added necessary infrastructure, we can now implement RO marking of eBPF generated JIT image pages and randomize start offset for the JIT code, so that it does not reside directly on a page boundary anymore. Likewise, the holes are filled with illegal instructions: here we use BRK #0x100 (opcode 0xd4202000) to trigger a fault in the kernel (unallocated BRKs would trigger a fault through do_debug_exception). This seems more reliable as we don't have a guaranteed undefined instruction space on ARM64.

This is basically the ARM64 variant of what we already have in ARM via commit 55309dd3d4cd ("net: bpf: arm: address randomize and write protect JIT code"). Moreover, this commit also presents a merge resolution due to conflicts with commit 60a3b2253c41 ("net: bpf: make eBPF interpreter images read-only") as we don't use kfree() in bpf_jit_free() anymore to release the locked bpf_prog structure, but instead bpf_prog_unlock_free() through a different allocator.

JIT tested on aarch64 with BPF test suite.

Reference: http://mainisusuallyafunction.blogspot.com/2012/11/attacking-hardened-linux-systems-with.html

Cc Zi Shen Lim <[email protected]>
Cc Will Deacon <[email protected]>
Cc Catalin Marinas <[email protected]>
Cc David S. Miller <[email protected]>
Cc Alexei Starovoitov <[email protected]>

---
v3->v4:
- Inclusion into debug-monitors.h as suggested by Will
v2->v3:
- Use cpu_to_le32() as suggested by Zi/Will
v1->v2:
- Use brk insn as suggested by Catalin
Note:
- This patch depends on net-next being merged to mainline due to the mentioned merge conflict.

 arch/arm64/include/asm/debug-monitors.h |  8 +++++++
 arch/arm64/net/bpf_jit_comp.c           | 39 +++++++++++++++++++++++++--------
 2 files changed, 38 insertions(+), 9 deletions(-)
arch/arm64/include/asm/debug-monitors.h +8 -0
index 7fb3437..230132f 100644
@@ -48,9 +48,11 @@
 /*
  * #imm16 values used for BRK instruction generation
  * Allowed values for kgbd are 0x400 - 0x7ff
+ * 0x100: for triggering a fault on purpose (reserved)
  * 0x400: for dynamic BRK instruction
  * 0x401: for compile time BRK instruction
  */
+#define FAULT_BRK_IMM			0x100
 #define KGDB_DYN_DGB_BRK_IMM		0x400
 #define KDBG_COMPILED_DBG_BRK_IMM	0x401
 
@@ -61,6 +63,12 @@
 #define AARCH64_BREAK_MON	0xd4200000
 
 /*
+ * BRK instruction for provoking a fault on purpose
+ * Unlike kgdb, #imm16 value with unallocated handler is used for faulting.
+ */
+#define AARCH64_BREAK_FAULT	(AARCH64_BREAK_MON | (FAULT_BRK_IMM << 5))
+
+/*
  * Extract byte from BRK instruction
  */
 #define KGDB_DYN_DGB_BRK_INS_BYTE(x) \
arch/arm64/net/bpf_jit_comp.c +30 -9
index 7ae3354..7108895 100644
@@ -19,12 +19,13 @@
 #define pr_fmt(fmt) "bpf_jit: " fmt
 
 #include <linux/filter.h>
-#include <linux/moduleloader.h>
 #include <linux/printk.h>
 #include <linux/skbuff.h>
 #include <linux/slab.h>
+
 #include <asm/byteorder.h>
 #include <asm/cacheflush.h>
+#include <asm/debug-monitors.h>
 
 #include "bpf_jit.h"
 
@@ -119,6 +120,14 @@ static inline int bpf2a64_offset(int bpf_to, int bpf_from,
 	return to - from;
 }
 
+static void jit_fill_hole(void *area, unsigned int size)
+{
+	u32 *ptr;
+	/* We are guaranteed to have aligned memory. */
+	for (ptr = area; size >= sizeof(u32); size -= sizeof(u32))
+		*ptr++ = cpu_to_le32(AARCH64_BREAK_FAULT);
+}
+
 static inline int epilogue_offset(const struct jit_ctx *ctx)
 {
 	int to = ctx->offset[ctx->prog->len - 1];
@@ -613,8 +622,10 @@ void bpf_jit_compile(struct bpf_prog *prog)
 
 void bpf_int_jit_compile(struct bpf_prog *prog)
 {
+	struct bpf_binary_header *header;
 	struct jit_ctx ctx;
 	int image_size;
+	u8 *image_ptr;
 
 	if (!bpf_jit_enable)
 		return;
@@ -636,23 +647,25 @@ void bpf_int_jit_compile(struct bpf_prog *prog)
 		goto out;
 
 	build_prologue(&ctx);
-
 	build_epilogue(&ctx);
 
 	/* Now we know the actual image size. */
 	image_size = sizeof(u32) * ctx.idx;
-	ctx.image = module_alloc(image_size);
-	if (unlikely(ctx.image == NULL))
+	header = bpf_jit_binary_alloc(image_size, &image_ptr,
+				      sizeof(u32), jit_fill_hole);
+	if (header == NULL)
 		goto out;
 
 	/* 2. Now, the actual pass. */
 
+	ctx.image = (u32 *)image_ptr;
 	ctx.idx = 0;
+
 	build_prologue(&ctx);
 
 	ctx.body_offset = ctx.idx;
 	if (build_body(&ctx)) {
-		module_free(NULL, ctx.image);
+		bpf_jit_binary_free(header);
 		goto out;
 	}
 
@@ -663,17 +676,25 @@ void bpf_int_jit_compile(struct bpf_prog *prog)
 		bpf_jit_dump(prog->len, image_size, 2, ctx.image);
 
 	bpf_flush_icache(ctx.image, ctx.image + ctx.idx);
+
+	set_memory_ro((unsigned long)header, header->pages);
 	prog->bpf_func = (void *)ctx.image;
 	prog->jited = 1;
-
 out:
 	kfree(ctx.offset);
 }
 
 void bpf_jit_free(struct bpf_prog *prog)
 {
-	if (prog->jited)
-		module_free(NULL, prog->bpf_func);
+	unsigned long addr = (unsigned long)prog->bpf_func & PAGE_MASK;
+	struct bpf_binary_header *header = (void *)addr;
+
+	if (!prog->jited)
+		goto free_filter;
+
+	set_memory_rw(addr, header->pages);
+	bpf_jit_binary_free(header);
 
-	kfree(prog);
+free_filter:
+	bpf_prog_unlock_free(prog);
 }
Signature · 1 line
1.9.3
Acked-by on v4 1/1
#1 lore raw
Quoted 41 lines · Daniel Borkmann

This is the ARM64 variant for 314beb9bcab ("x86: bpf_jit_comp: secure bpf jit against spraying attacks").

Thanks to commit 11d91a770f1f ("arm64: Add CONFIG_DEBUG_SET_MODULE_RONX support") which added necessary infrastructure, we can now implement RO marking of eBPF generated JIT image pages and randomize start offset for the JIT code, so that it does not reside directly on a page boundary anymore. Likewise, the holes are filled with illegal instructions: here we use BRK #0x100 (opcode 0xd4202000) to trigger a fault in the kernel (unallocated BRKs would trigger a fault through do_debug_exception). This seems more reliable as we don't have a guaranteed undefined instruction space on ARM64.

This is basically the ARM64 variant of what we already have in ARM via commit 55309dd3d4cd ("net: bpf: arm: address randomize and write protect JIT code"). Moreover, this commit also presents a merge resolution due to conflicts with commit 60a3b2253c41 ("net: bpf: make eBPF interpreter images read-only") as we don't use kfree() in bpf_jit_free() anymore to release the locked bpf_prog structure, but instead bpf_prog_unlock_free() through a different allocator.

JIT tested on aarch64 with BPF test suite.

Reference: http://mainisusuallyafunction.blogspot.com/2012/11/attacking-hardened-linux-systems-with.html

Cc Zi Shen Lim <[email protected]>
Cc Will Deacon <[email protected]>
Cc Catalin Marinas <[email protected]>
Cc David S. Miller <[email protected]>
Cc Alexei Starovoitov <[email protected]>

---
v3->v4:
- Inclusion into debug-monitors.h as suggested by Will
v2->v3:
- Use cpu_to_le32() as suggested by Zi/Will
v1->v2:
- Use brk insn as suggested by Catalin
Note:
- This patch depends on net-next being merged to mainline due to the mentioned merge conflict.

Quoted 3 lines
 arch/arm64/include/asm/debug-monitors.h |  8 +++++++
 arch/arm64/net/bpf_jit_comp.c           | 39 +++++++++++++++++++++++++--------
 2 files changed, 38 insertions(+), 9 deletions(-)

Thanks, Daniel, this looks great:

We can merge this at -rc1.

Will

Reviewed-by
#2 lore raw
Quoted 42 lines · Will Deacon
Quoted 41 lines · Daniel Borkmann

This is the ARM64 variant for 314beb9bcab ("x86: bpf_jit_comp: secure bpf jit against spraying attacks").

Thanks to commit 11d91a770f1f ("arm64: Add CONFIG_DEBUG_SET_MODULE_RONX support") which added necessary infrastructure, we can now implement RO marking of eBPF generated JIT image pages and randomize start offset for the JIT code, so that it does not reside directly on a page boundary anymore. Likewise, the holes are filled with illegal instructions: here we use BRK #0x100 (opcode 0xd4202000) to trigger a fault in the kernel (unallocated BRKs would trigger a fault through do_debug_exception). This seems more reliable as we don't have a guaranteed undefined instruction space on ARM64.

This is basically the ARM64 variant of what we already have in ARM via commit 55309dd3d4cd ("net: bpf: arm: address randomize and write protect JIT code"). Moreover, this commit also presents a merge resolution due to conflicts with commit 60a3b2253c41 ("net: bpf: make eBPF interpreter images read-only") as we don't use kfree() in bpf_jit_free() anymore to release the locked bpf_prog structure, but instead bpf_prog_unlock_free() through a different allocator.

JIT tested on aarch64 with BPF test suite.

Reference: http://mainisusuallyafunction.blogspot.com/2012/11/attacking-hardened-linux-systems-with.html

Cc Zi Shen Lim <[email protected]>
Cc Will Deacon <[email protected]>
Cc Catalin Marinas <[email protected]>
Cc David S. Miller <[email protected]>
Cc Alexei Starovoitov <[email protected]>

---
v3->v4:
- Inclusion into debug-monitors.h as suggested by Will
v2->v3:
- Use cpu_to_le32() as suggested by Zi/Will
v1->v2:
- Use brk insn as suggested by Catalin
Note:
- This patch depends on net-next being merged to mainline due to the mentioned merge conflict.

Quoted 8 lines
Quoted 3 lines
 arch/arm64/include/asm/debug-monitors.h |  8 +++++++
 arch/arm64/net/bpf_jit_comp.c           | 39 +++++++++++++++++++++++++--------
 2 files changed, 38 insertions(+), 9 deletions(-)

Thanks, Daniel, this looks great:

Quoted 3 lines

We can merge this at -rc1.

Will

#3 lore raw
In reply to Daniel Borkmann

Hi Catalin, Will,

On Tue, Sep 16, 2014 at 12:48 AM, Daniel Borkmann <[email protected]> wrote:
[...]

arch/arm64/net/bpf_jit_comp.c static inline int bpf2a64_offset(int bpf_to, int bpf_from, · line 129

Quoted 7 lines
arch/arm64/net/bpf_jit_comp.c +7 -0
+static void jit_fill_hole(void *area, unsigned int size)
+{
+       u32 *ptr;
+       /* We are guaranteed to have aligned memory. */
+       for (ptr = area; size >= sizeof(u32); size -= sizeof(u32))
+               *ptr++ = cpu_to_le32(AARCH64_BREAK_FAULT);
+}

[...]

Out of curiosity, I looked at objdump of the above code.

0000000000000088 <jit_fill_hole>:

      88:       71000c3f        cmp     w1, #0x3
      8c:       54000149        b.ls    b4 <jit_fill_hole+0x2c>
      90:       51001022        sub     w2, w1, #0x4
      94:       927e7442        and     x2, x2, #0xfffffffc
      98:       91001042        add     x2, x2, #0x4
      9c:       8b020002        add     x2, x0, x2
      a0:       52840001        mov     w1, #0x2000
 // #8192  <-- loops here
      a4:       72ba8401        movk    w1, #0xd420, lsl #16
      a8:       b8004401        str     w1, [x0],#4  <-- is there an
optimization such that we loop here?
      ac:       eb02001f        cmp     x0, x2
      b0:       54ffff81        b.ne    a0 <jit_fill_hole+0x18>
      b4:       d65f03c0        ret

I'm wondering if there's any optimizations that'll generate code that loops at 0xa8 instead of 0xa0. w1 only needs to loaded with the constant once, but here we're reloading it on every iteration of the loop.

Thanks,
z