[PATCH bpf-next 4/4] riscv, bpf: Mixing bpf2bpf and tailcalls

[Pu Lehui]

On 2023/9/19 19:50, Conor Dooley wrote:
> On Tue, Sep 19, 2023 at 07:23:07PM +0800, Pu Lehui wrote:
>>
>>
>> On 2023/9/19 18:04, Conor Dooley wrote:
>>> On Tue, Sep 19, 2023 at 11:57:11AM +0800, Pu Lehui wrote:
>>>> From: Pu Lehui <pulehui at huawei.com>
>>>>
>>>> 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.
>>>>
>>>> Tests test_bpf.ko and test_verifier have passed, as well as the relative
>>>> testcases of test_progs*.
>>>>
>>>> Signed-off-by: Pu Lehui <pulehui at huawei.com>
>>>
>>> Breaks the build:
>>> ../arch/riscv/net/bpf_jit_comp64.c:846:14: error: use of undeclared identifier 'BPF_TRAMP_F_TAIL_CALL_CTX'
>>>
>>
>> Hi Conor,
>>
>> BPF_TRAMP_F_TAIL_CALL_CTX rely on commit [0], and it has been merged into
>> bpf-next tree.
> 
> I see. I did check the cover to see if there was anything relevant
> there, like a link or base commit, but since there were neither I opted
> to pass on the warning from the patchwork automation we have :) >

Thanks, maybe it should be better to attach it to the cover.

> Thanks & sorry for the noise on this one.
> 
> Thanks,
> Conor.