On 2017/09/08 06:11, Eric Wong wrote:
> cardoso_tiago / hotmail.com wrote:

>> So, after I found that out, I noticed something really
>> strange: Using the `IO#read_nonblock(nread, buffer, exception:
>> false)` reduces the throughput of my solution significantly
>> (it also uses up significantly more memory). When using both
>> `IO#read_nonblock(nread, exception: false)` and
>> `IO#read_nonblock(nread, buffer)`, I get optimal performance.
> 
> It seems you've figured that out in in [ruby-core:82707],
> but I think my earlier note about using thread-local storage
> for short-lived buffers still applies.  Per-socket buffers
> (which you seem to be using) would use more object slots in
> common situations.
> 
>> This might be a bit off-topic. Should I open a new ticket with my findings?
If it's still relevant, then yes, please.

Regards,   Martin.

Unsubscribe: <mailto:ruby-core-request / ruby-lang.org?subject=unsubscribe>
<http://lists.ruby-lang.org/cgi-bin/mailman/options/ruby-core>