[bpf-next,0/4] Mixing bpf2bpf and tailcalls for RV64

Message

Pu Lehui Sept. 19, 2023, 3:57 a.m. UTC

In the current RV64 JIT, if we just don't initialize the TCC in subprog,
the TCC can be propagated from the parent process to the subprocess, but
the TCC of the parent process cannot be restored when the subprocess
exits. Since the RV64 TCC is initialized before saving the callee saved
registers into the stack, we cannot use the callee saved register to
pass the TCC, otherwise the original value of the callee saved register
will be destroyed. So we implemented mixing bpf2bpf and tailcalls
similar to x86_64, i.e. using a non-callee saved register to transfer
the TCC between functions, and saving that register to the stack to
protect the TCC value. At the same time, we also consider the scenario
of mixing trampoline.

In addition, some code cleans are also attached to this patchset.

Tests test_bpf.ko and test_verifier have passed, as well as the relative
testcases of test_progs*.

Pu Lehui (4):
  riscv, bpf: Remove redundant ctx->offset initialization
  riscv, bpf: Using kvcalloc to allocate cache buffer
  riscv, bpf: Add RV_TAILCALL_OFFSET macro to format tailcall offset
  riscv, bpf: Mixing bpf2bpf and tailcalls

 arch/riscv/net/bpf_jit.h        |   1 +
 arch/riscv/net/bpf_jit_comp64.c | 104 ++++++++++++++------------------
 arch/riscv/net/bpf_jit_core.c   |   9 +--
 3 files changed, 48 insertions(+), 66 deletions(-)

Comments

Björn Töpel Sept. 26, 2023, 1:30 p.m. UTC | #1

Pu Lehui <[e-mail]> writes:

> In the current RV64 JIT, if we just don't initialize the TCC in subprog,
> the TCC can be propagated from the parent process to the subprocess, but
> the TCC of the parent process cannot be restored when the subprocess
> exits. Since the RV64 TCC is initialized before saving the callee saved
> registers into the stack, we cannot use the callee saved register to
> pass the TCC, otherwise the original value of the callee saved register
> will be destroyed. So we implemented mixing bpf2bpf and tailcalls
> similar to x86_64, i.e. using a non-callee saved register to transfer
> the TCC between functions, and saving that register to the stack to
> protect the TCC value. At the same time, we also consider the scenario
> of mixing trampoline.
>
> In addition, some code cleans are also attached to this patchset.
>
> Tests test_bpf.ko and test_verifier have passed, as well as the relative
> testcases of test_progs*.

Apologies for the review delay. I'm travelling, and will pick it up ASAP
when I'm back.


Björn