From: "Mattias Rönnblom" <hofors@lysator.liu.se>
To: Bruce Richardson <bruce.richardson@intel.com>
Cc: "Mattias Rönnblom" <mattias.ronnblom@ericsson.com>,
dev@dpdk.org, "Morten Brørup" <mb@smartsharesystems.com>,
"Stephen Hemminger" <stephen@networkplumber.org>
Subject: Re: [RFC v3 1/6] eal: add static per-lcore memory allocation facility
Date: Tue, 20 Feb 2024 17:26:17 +0100 [thread overview]
Message-ID: <67d41a52-5e45-4906-91f1-57a5906e614c@lysator.liu.se> (raw)
In-Reply-To: <ZdSPb9SLnH9WsoWW@bricha3-mobl1.ger.corp.intel.com>
On 2024-02-20 12:39, Bruce Richardson wrote:
> On Tue, Feb 20, 2024 at 11:47:14AM +0100, Mattias Rönnblom wrote:
>> On 2024-02-20 10:11, Bruce Richardson wrote:
>>> On Tue, Feb 20, 2024 at 09:49:03AM +0100, Mattias Rönnblom wrote:
>>>> Introduce DPDK per-lcore id variables, or lcore variables for short.
>>>>
>>>> An lcore variable has one value for every current and future lcore
>>>> id-equipped thread.
>>>>
>>>> The primary <rte_lcore_var.h> use case is for statically allocating
>>>> small chunks of often-used data, which is related logically, but where
>>>> there are performance benefits to reap from having updates being local
>>>> to an lcore.
>>>>
>>>> Lcore variables are similar to thread-local storage (TLS, e.g., C11
>>>> _Thread_local), but decoupling the values' life time with that of the
>>>> threads.
>
> <snip>
>
>>>> +/*
>>>> + * Avoid using offset zero, since it would result in a NULL-value
>>>> + * "handle" (offset) pointer, which in principle and per the API
>>>> + * definition shouldn't be an issue, but may confuse some tools and
>>>> + * users.
>>>> + */
>>>> +#define INITIAL_OFFSET 1
>>>> +
>>>> +char rte_lcore_var[RTE_MAX_LCORE][RTE_MAX_LCORE_VAR] __rte_cache_aligned;
>>>> +
>>>
>>> While I like the idea of improved handling for per-core variables, my main
>>> concern with this set is this definition here, which adds yet another
>>> dependency on the compile-time defined RTE_MAX_LCORE value.
>>>
>>
>> lcore variables replaces one RTE_MAX_LCORE-dependent pattern with another.
>>
>> You could even argue the dependency on RTE_MAX_LCORE is reduced with lcore
>> variables, if you look at where/in how many places in the code base this
>> macro is being used. Centralizing per-lcore data management may also provide
>> some opportunity in the future for extending the API to cope with some more
>> dynamic RTE_MAX_LCORE variant. Not without ABI breakage of course, but we
>> are not ever going to change anything related to RTE_MAX_LCORE without
>> breaking the ABI, since this constant is everywhere, including compiled into
>> the application itself.
>>
>
> Yep, that is true if it's widely used.
>
>>> I believe we already have an issue with this #define where it's impossible
>>> to come up with a single value that works for all, or nearly all cases. The
>>> current default is still 128, yet DPDK needs to support systems where the
>>> number of cores is well into the hundreds, requiring workarounds of core
>>> mappings or customized builds of DPDK. Upping the value fixes those issues
>>> at the cost to memory footprint explosion for smaller systems.
>>>
>>
>> I agree this is an issue.
>>
>> RTE_MAX_LCORE also need to be sized to accommodate not only all cores used,
>> but the sum of all EAL threads and registered non-EAL threads.
>>
>> So, there is no reliable way to discover what RTE_MAX_LCORE is on a
>> particular piece of hardware, since the actual number of lcore ids needed is
>> up to the application.
>>
>> Why is the default set so low? Linux has MAX_CPUS, which serves the same
>> purpose, which is set to 4096 by default, if I recall correctly. Shouldn't
>> we at least be able to increase it to 256?
>
> The default is so low because of the mempool caches. These are an array of
> buffer pointers with 512 (IIRC) entries per core up to RTE_MAX_LCORE.
>
>>
>>> I'm therefore nervous about putting more dependencies on this value, when I
>>> feel we should be moving away from its use, to allow more runtime
>>> configurability of cores.
>>>
>>
>> What more specifically do you have in mind?
>>
>
> I don't think having a dynamically scaling RTE_MAX_LCORE is feasible, but
> what I would like to see is a runtime specified value. For example, you
> could run DPDK with EAL parameter "--max-lcores=1024" for large systems or
> "--max-lcores=32" for small ones. That would then be used at init-time to
> scale all internal datastructures appropriately.
>
Sounds reasonably to me, especially if you would take gradual approach.
By gradual I mean something like adding a function
rte_lcore_max_possible(), or something like that, returning the EAL
init-specified value. DPDK libraries/PMDs could then gradually be made
aware and taking advantage of knowing that lcore ids will always be
below a certain threshold, usually significantly lower than RTE_MAX_LCORE.
The only change required for lcore variables would be that the FOREACH
macro would use the run-time-max value, rather than RTE_MAX_LCORE, which
in turn would leave all the higher-numbered lcore id buffers
untouched/unmapped.
The set of possible lcore ids could also be expressed as a bitset, if
you have machine with a huge amount of cores, running many small DPDK
instances.
> /Bruce
>
> <snip for brevity>
next prev parent reply other threads:[~2024-02-20 16:26 UTC|newest]
Thread overview: 72+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-02-08 18:16 [RFC 0/5] Lcore variables Mattias Rönnblom
2024-02-08 18:16 ` [RFC 1/5] eal: add static per-lcore memory allocation facility Mattias Rönnblom
2024-02-09 8:25 ` Morten Brørup
2024-02-09 11:46 ` Mattias Rönnblom
2024-02-09 13:04 ` Morten Brørup
2024-02-19 7:49 ` Mattias Rönnblom
2024-02-19 11:10 ` Morten Brørup
2024-02-19 14:31 ` Mattias Rönnblom
2024-02-19 15:04 ` Morten Brørup
2024-02-19 9:40 ` [RFC v2 0/5] Lcore variables Mattias Rönnblom
2024-02-19 9:40 ` [RFC v2 1/5] eal: add static per-lcore memory allocation facility Mattias Rönnblom
2024-02-20 8:49 ` [RFC v3 0/6] Lcore variables Mattias Rönnblom
2024-02-20 8:49 ` [RFC v3 1/6] eal: add static per-lcore memory allocation facility Mattias Rönnblom
2024-02-20 9:11 ` Bruce Richardson
2024-02-20 10:47 ` Mattias Rönnblom
2024-02-20 11:39 ` Bruce Richardson
2024-02-20 13:37 ` Morten Brørup
2024-02-20 16:26 ` Mattias Rönnblom [this message]
2024-02-21 9:43 ` Jerin Jacob
2024-02-21 10:31 ` Morten Brørup
2024-02-21 14:26 ` Mattias Rönnblom
2024-02-22 9:22 ` Morten Brørup
2024-02-23 10:12 ` Mattias Rönnblom
2024-02-25 15:03 ` [RFC v4 0/6] Lcore variables Mattias Rönnblom
2024-02-25 15:03 ` [RFC v4 1/6] eal: add static per-lcore memory allocation facility Mattias Rönnblom
2024-02-27 9:58 ` Morten Brørup
2024-02-27 13:44 ` Mattias Rönnblom
2024-02-27 15:05 ` Morten Brørup
2024-02-27 16:27 ` Mattias Rönnblom
2024-02-27 16:51 ` Morten Brørup
2024-02-28 10:09 ` [RFC v5 0/6] Lcore variables Mattias Rönnblom
2024-02-28 10:09 ` [RFC v5 1/6] eal: add static per-lcore memory allocation facility Mattias Rönnblom
2024-03-19 12:52 ` Konstantin Ananyev
2024-03-20 10:24 ` Mattias Rönnblom
2024-03-20 14:18 ` Konstantin Ananyev
2024-05-06 8:27 ` [RFC v6 0/6] Lcore variables Mattias Rönnblom
2024-05-06 8:27 ` [RFC v6 1/6] eal: add static per-lcore memory allocation facility Mattias Rönnblom
2024-05-06 8:27 ` [RFC v6 2/6] eal: add lcore variable test suite Mattias Rönnblom
2024-05-06 8:27 ` [RFC v6 3/6] random: keep PRNG state in lcore variable Mattias Rönnblom
2024-05-06 8:27 ` [RFC v6 4/6] power: keep per-lcore " Mattias Rönnblom
2024-05-06 8:27 ` [RFC v6 5/6] service: " Mattias Rönnblom
2024-05-06 8:27 ` [RFC v6 6/6] eal: keep per-lcore power intrinsics " Mattias Rönnblom
2024-02-28 10:09 ` [RFC v5 2/6] eal: add lcore variable test suite Mattias Rönnblom
2024-02-28 10:09 ` [RFC v5 3/6] random: keep PRNG state in lcore variable Mattias Rönnblom
2024-02-28 10:09 ` [RFC v5 4/6] power: keep per-lcore " Mattias Rönnblom
2024-02-28 10:09 ` [RFC v5 5/6] service: " Mattias Rönnblom
2024-02-28 10:09 ` [RFC v5 6/6] eal: keep per-lcore power intrinsics " Mattias Rönnblom
2024-02-25 15:03 ` [RFC v4 2/6] eal: add lcore variable test suite Mattias Rönnblom
2024-02-25 15:03 ` [RFC v4 3/6] random: keep PRNG state in lcore variable Mattias Rönnblom
2024-02-25 15:03 ` [RFC v4 4/6] power: keep per-lcore " Mattias Rönnblom
2024-02-25 15:03 ` [RFC v4 5/6] service: " Mattias Rönnblom
2024-02-25 16:28 ` Mattias Rönnblom
2024-02-25 15:03 ` [RFC v4 6/6] eal: keep per-lcore power intrinsics " Mattias Rönnblom
2024-02-20 8:49 ` [RFC v3 2/6] eal: add lcore variable test suite Mattias Rönnblom
2024-02-20 8:49 ` [RFC v3 3/6] random: keep PRNG state in lcore variable Mattias Rönnblom
2024-02-20 15:31 ` Morten Brørup
2024-02-20 8:49 ` [RFC v3 4/6] power: keep per-lcore " Mattias Rönnblom
2024-02-20 8:49 ` [RFC v3 5/6] service: " Mattias Rönnblom
2024-02-22 9:42 ` Morten Brørup
2024-02-23 10:19 ` Mattias Rönnblom
2024-02-20 8:49 ` [RFC v3 6/6] eal: keep per-lcore power intrinsics " Mattias Rönnblom
2024-02-19 9:40 ` [RFC v2 2/5] eal: add lcore variable test suite Mattias Rönnblom
2024-02-19 9:40 ` [RFC v2 3/5] random: keep PRNG state in lcore variable Mattias Rönnblom
2024-02-19 11:22 ` Morten Brørup
2024-02-19 14:04 ` Mattias Rönnblom
2024-02-19 15:10 ` Morten Brørup
2024-02-19 9:40 ` [RFC v2 4/5] power: keep per-lcore " Mattias Rönnblom
2024-02-19 9:40 ` [RFC v2 5/5] service: " Mattias Rönnblom
2024-02-08 18:16 ` [RFC 2/5] eal: add lcore variable test suite Mattias Rönnblom
2024-02-08 18:16 ` [RFC 3/5] random: keep PRNG state in lcore variable Mattias Rönnblom
2024-02-08 18:16 ` [RFC 4/5] power: keep per-lcore " Mattias Rönnblom
2024-02-08 18:16 ` [RFC 5/5] service: " Mattias Rönnblom
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=67d41a52-5e45-4906-91f1-57a5906e614c@lysator.liu.se \
--to=hofors@lysator.liu.se \
--cc=bruce.richardson@intel.com \
--cc=dev@dpdk.org \
--cc=mattias.ronnblom@ericsson.com \
--cc=mb@smartsharesystems.com \
--cc=stephen@networkplumber.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).