|
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] Re: [PATCH v4 03/42] xen/riscv: introduce csr_volatile_read64()
On 18.11.2026 10:50, Baptiste Le Duc wrote:
>> On RV32 a 64-bit CSR is accessed as two halves, <csr> and <csr>H. Reading
>> the halves with two separate instructions isn't always safe: for a CSR
>> which hardware increments, a carry from the low half into the high one can
>> happen in between, so a plain pair of reads can produce a value the CSR
>> never held. Therefore the high half is re-read and the sequence retried if
>> it changed in the meantime.
>>
>> Only CSRs which change under Xen's feet need that, hence the name. The
>> majority can be read with a plain pair of csr_read()s, so no helper for
>> those is introduced here as there would be no user for it.
>>
>> Use it for CSR_TIME, which is exactly such a counter, and widen cycles_t to
>> uint64_t. Otherwise get_cycles() would still truncate the time counter to
>> 32 bits on RV32.
>>
>> No functional change on RV64.
>>
>> Fixes: a541ddadec0a ("xen/riscv: introduce time.h")
>> Signed-off-by: Oleksii Kurochko <oleksii.kurochko@xxxxxxxxx>
>> ---
>> Changes in v4:
>> - Take the CSR name without the CSR_ prefix, so that the macro parameter
>> is used as an operand of ## only (MISRA C Rule 20.12).
>> - Drop the (void) of the ...H constant on RV64.
>> - Initialise hi_ in its declaration.
>
> Jan gave his Reviewed-by on v3. It is gone in v4. Dropping it may be
> fine given the macro rework, but please say so in the changelog :)
>
>> [ ... 45 lines skipped ... ]
>> #else
>> #define csr_write64(name, val) csr_write(CSR_ ## name, val)
>> +
>> +#define csr_volatile_read64(name) csr_read(CSR_ ## name)
>
> Nit: On v3 Jan suggested "#define csr_volatile_read64 csr_read64", and you
> answered you would then move csr_read64() into this patch. In v4
> csr_read64() still only comes with "xen/riscv: implement vCPU context
> switching".
>
> On RV64 I think it would read better like that:
> #define csr_read64(name) csr_read(CSR_ ## name)
> #define csr_volatile_read64(name) csr_read64(name)
#define csr_read64(name) csr_read(CSR_ ## name)
#define csr_volatile_read64 csr_read64
i.e. matching what I suggested earlier. Otherwise what is passed into
csr_volatile_read64() would be macro-expanded before further
expanding csr_read64(), thus prepending CSR_ only to the result of
that expansion. (In the common case there'll be no difference, but
things can go wrong if some unexpected new #define appears.)
Jan
|
![]() |
Lists.xenproject.org is hosted with RackSpace, monitoring our |