Re: Limit on max_entries of CPU_ARRAY_MAP

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



We'll have to see how it plays out performance wise, but we're doing
some tracking on a per IP per port basis across a /24. We'd originally
tried to use one map with a key that's a combination of IP and port,
but had to switch to map in map with IP being outer key and port being
inner (or could go vice versa).

--Zvi

On Tue, Jan 30, 2018 at 5:21 PM, Alexei Starovoitov
<alexei.starovoitov@xxxxxxxxx> wrote:
> On Mon, Jan 29, 2018 at 11:00:27AM -0800, Zvi Effron wrote:
>> Would it make sense to increase that limit for 64-bit systems? All of
>> the comments on why that limit exists that I saw mentioned that
>> userspace wouldn't be able to access all of the elements if it were
>> bigger. But on a 64-bit system, shouldn't userspace be able to access
>> more than 4GB?
>
> What is the _real_ use case for more than 4Gbyte maps?
>
> So far I've seen only one such case that had use to map-in-map
> as a workaround to grow the total map size to ~20Gbyte,
> but it eventually it was scaled down because at such sizes
> performance was bad.
>



[Index of Archives]     [Linux Networking Development]     [Fedora Linux Users]     [Linux SCTP]     [DCCP]     [Gimp]     [Yosemite Campsites]

  Powered by Linux