On 25/09/2023 12:44, Zheng Yejian wrote:
> On 2023/9/25 18:02, Christian Loehle wrote:
>> buffer_size_kb no longer shows the requested amount, but the one that
>> is actually used internally for the ring buffer.
>>
>> commit 6d98a0f2ac3c ("tracing: Set actual size after ring buffer resize")
>> changed the sysfs behavior such that value read will always show the
>> actual size, while previously it showed the size that was requested
>> through the sysfs interface, even if it was rounded up to fulfill
>> the request.
>> So the documentation can state that more precisely now.
>>
>> Signed-off-by: Christian Loehle <christian.loehle@xxxxxxx>
>> ---
>> Documentation/trace/ftrace.rst | 2 +-
>> 1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/Documentation/trace/ftrace.rst b/Documentation/trace/ftrace.rst
>> index 23572f6697c0..2e066b3b6edc 100644
>> --- a/Documentation/trace/ftrace.rst
>> +++ b/Documentation/trace/ftrace.rst
>> @@ -191,7 +191,7 @@ of ftrace. Here is a list of some of the key files:
>> A few extra pages may be allocated to accommodate buffer management
>> meta-data. If the last page allocated has room for more bytes
>> than requested, the rest of the page will be used,
>> - making the actual allocation bigger than requested or shown.
>> + making the actual allocation bigger than requested.
>
> Hi, the actual allocation should still be bigger than shown due to the
> loss of accuracy when doing unit conversion from bytes to kilobytes (see
> tracing_entries_read()).
>
Right, the sysfs obviously only allows for KB aligned setting, but you're right.
If set on the cmdline non-KB multiples are possible and accuracy is lost.
Nevermind then.
> --
>
> Thanks,
> Zheng Yejian
>
>> ( Note, the size may not be a multiple of the page size
>> due to buffer management meta-data. )
>>
>
