[RFC] riscv/entry: issue about a0/orig_a0 register and ENOSYS

[Alexandre Ghiti]

On 16/10/2024 14:23, Celeste Liu wrote:
> On 2024-10-16 20:00, Alexandre Ghiti wrote:
>> Hi Celeste,
>>
>> Thank you for looking into this and really sorry about the late response.
>>
>> On 17/09/2024 06:09, Celeste Liu wrote:
>>> Before PTRACE_GET_SYSCALL_INFO was implemented in v5.3, the only way to
>>> get syscall arguments was to get user_regs_struct via PTRACE_GETREGSET.
>>> On some architectures where a register is used as both the first
>>> argument and the return value and thus will be changed at some stage of
>>> the syscall process, something like orig_a0 is provided to save the
>>> original argument register value. But RISC-V doesn't export orig_a0 in
>>> user_regs_struct (This ABI was designed at e2c0cdfba7f6 ("RISC-V:
>>> User-facing API").) so userspace application like strace will get the
>>> right or wrong result depends on the operation order in do_trap_ecall_u()
>>> function.
>>>
>>> This requires we put the ENOSYS process after syscall_enter_from_user_mode()
>>> or syscall_handler()[1]. Unfortunately, the generic entry API
>>> syscall_enter_from_user_mode() requires we
>>>
>>> *  process ENOSYS before syscall_enter_from_user_mode()
>>
>> Where does this requirement come from?
>>
>>
>>> *  or only set a0 to ENOSYS when the return value of
>>>      syscall_enter_from_user_mode() != -1
>>>
>>> Again, if we choose the latter way to avoid conflict with the first
>>> issue, we will meet the third problem: strace depends on that kernel
>>> will return ENOSYS when syscall nr is -1 to implement their syscall
>>> tampering feature[2].
>>
>> IIUC, seccomp and others in syscall_enter_from_user_mode() could return -1 and then we could not differentiate with the syscall number being -1.
>>
>> But could we imagine, to distinguish between an error and the syscall number being -1, checking again the syscall number after we call syscall_enter_from_user_mode()? If the syscall number is -1, then we set ENOSYS otherwise we don't do anything (a bit like what you did in 52449c17bdd1 ("riscv: entry: set a0 = -ENOSYS only when syscall != -1")).
>>
>> Let me know if I completely misunderstood here!
> Yeah. I found this a bit later after I post this RFC. I include it in a update reply,
> copy here as well:
>
>> But from another angle, syscall number is in a7 register, so we can call the
>> get_syscall_nr() after calling the syscall_enter_from_user_mode() to bypass the
>> information lost of the return value of the syscall_enter_from_user_mode(). But
>> in this way, the syscall number in the syscall_enter_from_user_mode() return
>> value is useless, and we can remove it directly.
> So if we get syscall number from a7 register again, the syscall number part of
> the return value of syscall_enter_from_user_mode() is useless completely.
> I think it's better to drop it so the later new architecture developer will not
> run into the same issue. (Actually, the syscall number returned by
> syscall_enter_from_user_mode() is also the result of get_syscall_nr() at the end
> of it.) But it will affect other architecture's code so I think there still need
> some discussions.
>
> Or if you think it's better to post a patch and then discuss in patch thread
> directly, I'm glad to do this.


Great that we have a solution that does not need to change the ABI :)

I think we should start by implementing a fix for riscv only that 
implements the get_syscall_nr() after syscall_enter_from_user_mode() so 
that we can merge that in 6.12-rcX.

And after that, you could come with the nicer solution you propose.

Do you think you can send a patch for the quick fix soon? In the 
meantime, I'm adding the strace testsuite to my CI to make sure it works 
and we don't break it again :)

Thanks!

Alex


>
>> Thanks again for the thorough explanation,
>>
>> Alex
>>
>>
>>> Actually, we tried the both ways in 52449c17bdd1 ("riscv: entry: set
>>> a0 = -ENOSYS only when syscall != -1") and 61119394631f ("riscv: entry:
>>> always initialize regs->a0 to -ENOSYS") before.
>>>
>>> Naturally, there is a solution:
>>>
>>> 1. Just add orig_a0 in user_regs_struct and let strace use it as
>>>      loongarch does. So only two problems, which can be resolved without
>>>      conflict, are left here.
>>>
>>> The conflicts are the direct result of the limitation of generic entry
>>> API, so we have another two solutions:
>>>
>>> 2. Give up the generic entry API, and switch back to the
>>>      architecture-specific standardalone implementation.
>>> 3. Redesign the generic entry API: the problem was caused by
>>>      syscall_enter_from_user_mode() using the value -1 (which is unused
>>>      normally) of syscall nr to inform syscall was reject by seccomp/bpf.
>>>
>>> In theory, the Solution 1 is best:
>>>
>>> *  a0 was used for two purposes in ABI, so using two variables to store
>>>      it is natural.
>>> *  Userspace can implement features without depending on the internal
>>>      behavior of the kernel.
>>>
>>> Unfortunately, it's difficult to implement based on the current code.
>>> The RISC-V defined struct pt_regs as below:
>>>
>>>           struct pt_regs {
>>>                   unsigned long epc;
>>>           ...
>>>                   unsigned long t6;
>>>                   /* Supervisor/Machine CSRs */
>>>                   unsigned long status;
>>>                   unsigned long badaddr;
>>>                   unsigned long cause;
>>>                   /* a0 value before the syscall */
>>>                   unsigned long orig_a0;
>>>           };
>>>
>>> And user_regs_struct needs to be a prefix of struct pt_regs, so if we
>>> want to include orig_a0 in user_regs_struct, we will need to include
>>> Supervisor/Machine CSRs as well. It's not a big problem. Since
>>> struct pt_regs is the internal ABI of the kernel, we can reorder it.
>>> Unfortunately, struct user_regs_struct is defined as below:
>>>
>>>           struct user_regs_struct {
>>>                   unsigned long pc;
>>>           ...
>>>                   unsigned long t6;
>>>           };
>>>
>>> It doesn't contain something like reserved[] as padding to leave the
>>> space to add more registers from struct pt_regs!
>>> The loongarch do the right thing as below:
>>>
>>>           struct user_pt_regs {
>>>                   /* Main processor registers. */
>>>                   unsigned long regs[32];
>>>           ...
>>>                   unsigned long reserved[10];
>>>           } __attribute__((aligned(8)));
>>>
>>> RISC-V can't include orig_a0 in user_regs_struct without breaking UABI.
>>>
>>> Need a discussion to decide to use which solution, or is there any
>>> other better solution?
>>>
>>> [1]: https://github.com/strace/strace/issues/315
>>> [2]: https://lore.kernel.org/linux-riscv/20240627071422.GA2626@altlinux.org/
>>>
>>>
>>> _______________________________________________
>>> linux-riscv mailing list
>>> linux-riscv at lists.infradead.org
>>> http://lists.infradead.org/mailman/listinfo/linux-riscv