Re: About commit "io: change inX() to have their own IO barrier overrides"

[John Garry <john.garry@huawei.com>]

On 06/03/2020 07:54, Arnd Bergmann wrote:
> On Fri, Mar 6, 2020 at 4:44 AM Sinan Kaya <okaya@kernel.org> wrote:
>>
>> On 3/3/2020 11:40 AM, Arnd Bergmann wrote:
>>>> -          ret = read##bw(PCI_IOBASE + addr);
>>>> +          __io_pbr();
>>>> +          ret = __raw_read##bw(PCI_IOBASE + addr);
>>>> +          __io_pbr();
>>> __io_par();
>>>
>>
>> Why do we need to change read##bw above?
>>
>> read##bw already provides strong ordering guarantees across multiple
>> architectures.
> 
> The exact semantics of inl() and readl() are slightly different, so they
> have distinct sets of barriers in the asm-generic/io.h implementation.
> 
> For instance, the arm64 architectures defines in_par() as '__iormb(v)',
> but defines __io_ar() as a  '__rmb()'. Similarly, riscv defines them
> as "fence i,ior" and "fence i,r".
> 
> You could argue that the definitions are wrong (I have not checked the
> history of the definitions), but as long as the inb() in asm-generic/io.h
> uses those, the implementation in lib/logic_pio.c uses the same ones
> to make the two behave the same way.
> 

So the change would look like:

-- a/lib/logic_pio.c
+++ b/lib/logic_pio.c
@@ -229,13 +229,21 @@ unsigned long 
logic_pio_trans_cpuaddr(resource_size_t addr)
  }

  #if defined(CONFIG_INDIRECT_PIO) && defined(PCI_IOBASE)
+
+#define logic_in_to_cpu_b(x) (x)
+#define logic_in_to_cpu_w(x) __le16_to_cpu(x)
+#define logic_in_to_cpu_l(x) __le32_to_cpu(x)
+
  #define BUILD_LOGIC_IO(bw, type)                                      \
  type logic_in##bw(unsigned long addr)                                 \
  {                                                                     \
         type ret = (type)~0;                                           \
                                                                        \
         if (addr < MMIO_UPPER_LIMIT) {                                 \
-               ret = read##bw(PCI_IOBASE + addr);                     \
+               void __iomem *_addr = PCI_IOBASE + addr;               \
+               __io_pbr();                                            \
+               ret = logic_in_to_cpu_##bw(__raw_read##bw(_addr));     \
+               __io_par(ret);                                         \
         } else if (addr >= MMIO_UPPER_LIMIT && addr < IO_SPACE_LIMIT) {\
                 struct logic_pio_hwaddr *entry = find_io_rang

We could prob combine the le_to_cpu and __raw_read into a single macro.

John