x86: bpf_jit: fix compilation of large bpf programs
Subsystems (5)
PATCH 1/1 x86: bpf_jit: fix compilation of large bpf programs
x86 has variable length encoding. x86 JIT compiler is trying to pick the shortest encoding for given bpf instruction. While doing so the jump targets are changing, so JIT is doing multiple passes over the program. Typical program needs 3 passes. Some very short programs converge with 2 passes. Large programs may need 4 or 5. But specially crafted bpf programs may hit the pass limit and if the program converges on the last iteration the JIT compiler will be producing an image full of 'int 3' insns. Fix this corner case by doing final iteration over bpf program.
---
Daniel wrote the 'Edge hopping nuthouse' test case with 4k jump instructions that managed to trigger this bug.
The test case is nuts and the bug is real.
It's an old bug, but I think worth backporting all the way. Though this fix will apply cleanly only till commit:f3c2af7ba17a ("net: filter: x86: split bpf_jit_compile()") The older kernels should be similar. They have
'for (pass = 0; pass < 10; pass++) {' at the line 153 or so. and all have similar problem as far as I can see.
arch/x86/net/bpf_jit_comp.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-)
arch/x86/net/bpf_jit_comp.c +6 -1
@@ -966,7 +966,12 @@ void bpf_int_jit_compile(struct bpf_prog *prog) } ctx.cleanup_addr = proglen; -for (pass =0 ; pass <10 ; pass++) { +/* JITed image shrinks with every pass and the loop iterates +* until the image stops shrinking. Very large bpf programs +* may converge on the last pass. In such case do one more +* pass to emit the final image +*/ +for (pass =0 ; pass <10 || image; pass++) { proglen =do_jit (prog, addrs, image, oldproglen, &ctx);if (proglen <=0 ) { image =NULL ;
Signature · 1 line
1.7.9.5
Quoted 10 lines · Alexei Starovoitov
x86 has variable length encoding. x86 JIT compiler is trying to pick the shortest encoding for given bpf instruction. While doing so the jump targets are changing, so JIT is doing multiple passes over the program. Typical program needs 3 passes. Some very short programs converge with 2 passes. Large programs may need 4 or 5. But specially crafted bpf programs may hit the pass limit and if the program converges on the last iteration the JIT compiler will be producing an image full of 'int 3' insns. Fix this corner case by doing final iteration over bpf program.
Quoted 3 lines
Fixes 0a14842f5a3c ("net: filter: Just In Time compiler for x86-64")Reported-by Daniel Borkmann <[email protected]>Signed-off-by Alexei Starovoitov <[email protected]>
LGTM, thanks!
David Miller
maintainerFrom: Alexei Starovoitov <[email protected]>
Date: Fri, 22 May 2015 15:42:55 -0700
Quoted 10 lines
x86 has variable length encoding. x86 JIT compiler is trying to pick the shortest encoding for given bpf instruction. While doing so the jump targets are changing, so JIT is doing multiple passes over the program. Typical program needs 3 passes. Some very short programs converge with 2 passes. Large programs may need 4 or 5. But specially crafted bpf programs may hit the pass limit and if the program converges on the last iteration the JIT compiler will be producing an image full of 'int 3' insns. Fix this corner case by doing final iteration over bpf program.
Quoted 3 lines
Fixes 0a14842f5a3c ("net: filter: Just In Time compiler for x86-64")Reported-by Daniel Borkmann <[email protected]>Signed-off-by Alexei Starovoitov <[email protected]>
Applied and queued up for -stable, thanks.
From: Alexei Starovoitov
Quoted 10 lines
Sent: 22 May 2015 23:43
x86 has variable length encoding. x86 JIT compiler is trying to pick the shortest encoding for given bpf instruction. While doing so the jump targets are changing, so JIT is doing multiple passes over the program. Typical program needs 3 passes. Some very short programs converge with 2 passes. Large programs may need 4 or 5. But specially crafted bpf programs may hit the pass limit and if the program converges on the last iteration the JIT compiler will be producing an image full of 'int 3' insns. Fix this corner case by doing final iteration over bpf program.
If the JIT compiler is only changing the encoding of the constants in the x86 instructions (rather than changing the instructions themselves) then there is likely to me an unmeasurable change in the execution time. For instance I don't remember there being a difference in execution time between long and short branches - the only difference is the amount of cache they use.
David
Quoted 6 lines · David Laight
If the JIT compiler is only changing the encoding of the constants in the x86 instructions (rather than changing the instructions themselves) then there is likely to me an unmeasurable change in the execution time. For instance I don't remember there being a difference in execution time between long and short branches - the only difference is the amount of cache they use.
icache is precisely the matter here. In the end, it makes a difference.
You could check this interesting study Ingo did recently :
From: Eric Dumazet
Quoted 10 lines
Sent: 26 May 2015 15:35
Quoted 6 lines · David Laight
If the JIT compiler is only changing the encoding of the constants in the x86 instructions (rather than changing the instructions themselves) then there is likely to me an unmeasurable change in the execution time. For instance I don't remember there being a difference in execution time between long and short branches - the only difference is the amount of cache they use.
Quoted 5 lines
icache is precisely the matter here. In the end, it makes a difference.
You could check this interesting study Ingo did recently :
Yes, interesting, a benchmark that manages to run a lot of code 'cold cache'. Possibly dominated by having a full cache line of code to execute at the beginning of each function.
My guess is that aligning function bodies helps, aligning branch targets (as some old versions of gcc used to do aggressively) is more likely to have a negative effect (increases the icache footprint too much). Unrolling loops further than needed to avoid data stalls needs very careful study.
For JIT, once the obviously short branches have been reduced, further code size reduction might not be worth while.
David
Quoted 1 line · David Laight
Yes, interesting, a benchmark that manages to run a lot of code 'cold cache'.
We have binaries here at Google with 400 or 500 MBytes of text.
Not benchmark, super real workloads you know.
From: Eric Dumazet
Quoted 7 lines
Sent: 26 May 2015 16:30
Quoted 1 line
Yes, interesting, a benchmark that manages to run a lot of code 'cold cache'.
We have binaries here at Google with 400 or 500 MBytes of text.
Not benchmark, super real workloads you know.
Indeed, and a lot of the code is likely to be running 'cold cache'.
I was alluding to the problem where people will benchmark a small function by running in 1000s of times in a tight loop with exactly the same data. Not only is it 'hot cache' but any dynamic branch prediction is 'trained' to the specific data.
David