netdev - Re: [PATCH net-next 2/2] net/mlx4: Revert "mlx4: set maximal number of default RSS queues"

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Thu, 2 Jan 2014 11:35:23 +0200
From:	Or Gerlitz <ogerlitz@...lanox.com>
To:	Yuval Mintz <yuvalmin@...adcom.com>,
	Or Gerlitz <or.gerlitz@...il.com>
CC:	Amir Vadai <amirv@...lanox.com>,
	"David S. Miller" <davem@...emloft.net>,
	"netdev@...r.kernel.org" <netdev@...r.kernel.org>,
	Eugenia Emantayev <eugenia@...lanox.com>,
	Ido Shamay <idos@...lanox.com>
Subject: Re: [PATCH net-next 2/2] net/mlx4: Revert "mlx4: set maximal number
 of default RSS queues"

On 02/01/2014 08:04, Yuval Mintz wrote:
>>> [...] If you believe this is a better default (or some relaxation will be, e.g.,  16
>> instead of 8),  why not set it as default for ALL multi-queue networking
>> drivers?
>>
>> Going back to your original commit 16917b87a "net-next: Add
>> netif_get_num_default_rss_queues" I am still not clear why we want
>>
>> 1. why we want a common default to all MQ devices?
> Although networking benefits from multiple Interrupt vectors
> (enabling more rings, better performance, etc.), bounding this
> number only to the number of cpus is unreasonable as it strains
> system resources; e.g., consider a 40-cpu server - we might wish
> to have 40 vectors per device, but that means that connecting
> several devices to the same server might cause other functions
> to fail probe as they will no longer be able to acquire interrupt
> vectors of their own.

Modern servers which have tens of CPUs typically have thousands of MSI-X 
vectors which means you should be easily able to plug four cards into a 
server with 64 cores which will consume 256 out of the 1-4K vectors out 
there. Anyway, let me continue your approach - how about raising the 
default hard limit to 16 or having it as the number of cores @ the numa 
node where the card is plugged?

Or.


>
> Since networking has an API allowing the user to manually set the
> number of channels, the default is upper-bounded.
>
>> 2. why this default has to be hard coded and not derived e.g from the
>> number of cores or alike attribute of the system?
> This is not entirely correct; The default number is derived from
> the number of online cpus - it's only upper bounded by some
> hard-coded value.
>
> Cheers,
> Yuval

--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html