2024年9月27日 12:28，Boqun Feng <boqun.feng@xxxxxxxxx> wrote：
>
> On Fri, Sep 27, 2024 at 09:37:50AM +0800, Boqun Feng wrote:
>>
>>
>> On Fri, Sep 27, 2024, at 9:30 AM, Mathieu Desnoyers wrote:
>>> On 2024-09-27 02:01, Boqun Feng wrote:
>>>> #define ADDRESS_EQ(var, expr) \
>>>> ({ \
>>>> bool _____cmp_res = (unsigned long)(var) == (unsigned long)(expr); \
>>>> \
>>>> OPTIMIZER_HIDE_VAR(var); \
>>>> _____cmp_res; \
>>>> })
>>>
>>> If the goal is to ensure gcc uses the register populated by the
>>> second, I'm afraid it does not work. AFAIU, "hiding" the dependency
>>> chain does not prevent the SSA GVN optimization from combining the
>
> Note it's not hiding the dependency, rather the equality,
>
>>> registers as being one and choosing one arbitrary source. "hiding"
>
> after OPTIMIZER_HIDE_VAR(var), compiler doesn't know whether 'var' is
> equal to 'expr' anymore, because OPTIMIZER_HIDE_VAR(var) uses "=r"(var)
> to indicate the output is overwritten. So when 'var' is referred later,
> compiler cannot use the register for a 'expr' value or any other
> register that has the same value, because 'var' may have a different
> value from the compiler's POV.
>
>>> the dependency chain before or after the comparison won't help here.
>>>
>>> int fct_hide_var_compare(void)
>>> {
>>> int *a, *b;
>>>
>>> do {
>>> a = READ_ONCE(p);
>>> asm volatile ("" : : : "memory");
>>> b = READ_ONCE(p);
>>> } while (!ADDRESS_EQ(a, b));
>>
>> Note that ADDRESS_EQ() only hide first parameter, so this should be ADDRESS_EQ(b, a).
>>
>
> I replaced ADDRESS_EQ(a, b) with ADDRESS_EQ(b, a), and the compile
> result shows it can prevent the issue:
>
> gcc 14.2 x86-64:
>
> fct_hide_var_compare:
> .L2:
> mov rcx, QWORD PTR p[rip]
> mov rdx, QWORD PTR p[rip]
> mov rax, rdx
> cmp rcx, rdx
> jne .L2
> mov eax, DWORD PTR [rax]
> ret
>
> gcc 14.2.0 ARM64:
>
> fct_hide_var_compare:
> adrp x2, p
> add x2, x2, :lo12:p
> .L2:
> ldr x3, [x2]
> ldr x1, [x2]
> mov x0, x1
> cmp x3, x1
> bne .L2
> ldr w0, [x0]
> ret
>
> Link to godbolt:
>
> https://godbolt.org/z/a7jsfzjxY
Checking the assembly generated by different compilers for the kernel on the local machine will yield more accurate results. Some optimizations are restricted by the kernel. Therefore, if you use Godbolt, ensure that the compiler arguments match those used for the kernel.
>
> Regards,
> Boqun
>
>> Regards,
>> Boqun
>>
>>> return *b;
>>> }
>>>
>>> gcc 14.2 x86-64:
>>>
>>> fct_hide_var_compare:
>>> mov rax,QWORD PTR [rip+0x0] # 67 <fct_hide_var_compare+0x7>
>>> mov rdx,QWORD PTR [rip+0x0] # 6e <fct_hide_var_compare+0xe>
>>> cmp rax,rdx
>>> jne 60 <fct_hide_var_compare>
>>> mov eax,DWORD PTR [rax]
>>> ret
>>> main:
>>> xor eax,eax
>>> ret
>>>
>>> gcc 14.2.0 ARM64:
>>>
>>> fct_hide_var_compare:
>>> adrp x0, .LANCHOR0
>>> add x0, x0, :lo12:.LANCHOR0
>>> .L12:
>>> ldr x1, [x0]
>>> ldr x2, [x0]
>>> cmp x1, x2
>>> bne .L12
>>> ldr w0, [x1]
>>> ret
>>> p:
>>> .zero 8
>>>
>>>
>>> --
>>> Mathieu Desnoyers
>>> EfficiOS Inc.
>>> https://www.efficios.com
