[PATCH bpf-next v2] bpf: support doubleword alignment in bpf_jit_binary_alloc

[PATCH bpf-next v2] bpf: support doubleword alignment in bpf_jit_binary_alloc

[Ilya Leoshkevich]

Currently passing alignment greater than 4 to bpf_jit_binary_alloc does
not work: in such cases it silently aligns only to 4 bytes.

On s390, in order to load a constant from memory in a large (>512k) BPF
program, one must use lgrl instruction, whose memory operand must be
aligned on an 8-byte boundary.

This patch makes it possible to request 8-byte alignment from
bpf_jit_binary_alloc, and also makes it issue a warning when an
unsupported alignment is requested.

Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
---

v1 -> v2: Simply bump alignment to 8, don't try to be too generic.

 include/linux/filter.h | 6 ++++--
 kernel/bpf/core.c      | 4 ++++
 2 files changed, 8 insertions(+), 2 deletions(-)

diff --git a/include/linux/filter.h b/include/linux/filter.h
index 7a6f8f6f1da4..ad80e9c6111c 100644
--- a/include/linux/filter.h
+++ b/include/linux/filter.h
@@ -515,10 +515,12 @@ struct sock_fprog_kern {
 	struct sock_filter	*filter;
 };
 
+/* Some arches need doubleword alignment for their instructions and/or data */
+#define BPF_IMAGE_ALIGNMENT 8
+
 struct bpf_binary_header {
 	u32 pages;
-	/* Some arches need word alignment for their instructions */
-	u8 image[] __aligned(4);
+	u8 image[] __aligned(BPF_IMAGE_ALIGNMENT);
 };
 
 struct bpf_prog {
diff --git a/kernel/bpf/core.c b/kernel/bpf/core.c
index c1fde0303280..99693f3c4e99 100644
--- a/kernel/bpf/core.c
+++ b/kernel/bpf/core.c
@@ -31,6 +31,7 @@
 #include <linux/rcupdate.h>
 #include <linux/perf_event.h>
 #include <linux/extable.h>
+#include <linux/log2.h>
 #include <asm/unaligned.h>
 
 /* Registers */
@@ -815,6 +816,9 @@ bpf_jit_binary_alloc(unsigned int proglen, u8 **image_ptr,
 	struct bpf_binary_header *hdr;
 	u32 size, hole, start, pages;
 
+	WARN_ON_ONCE(!is_power_of_2(alignment) ||
+		     alignment > BPF_IMAGE_ALIGNMENT);
+
 	/* Most of BPF filters are really small, but if some of them
 	 * fill a page, allow at least 128 extra bytes to insert a
 	 * random section of illegal instructions.
-- 
2.23.0

Re: [PATCH bpf-next v2] bpf: support doubleword alignment in bpf_jit_binary_alloc

[Daniel Borkmann]

On 11/15/19 1:37 PM, Ilya Leoshkevich wrote:
> Currently passing alignment greater than 4 to bpf_jit_binary_alloc does
> not work: in such cases it silently aligns only to 4 bytes.
> 
> On s390, in order to load a constant from memory in a large (>512k) BPF
> program, one must use lgrl instruction, whose memory operand must be
> aligned on an 8-byte boundary.
> 
> This patch makes it possible to request 8-byte alignment from
> bpf_jit_binary_alloc, and also makes it issue a warning when an
> unsupported alignment is requested.
> 
> Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>

Applied, thanks!