netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Simon Horman <horms@kernel.org>
To: Petr Machata <petrm@nvidia.com>
Cc: "David S. Miller" <davem@davemloft.net>,
	Eric Dumazet <edumazet@google.com>,
	Jakub Kicinski <kuba@kernel.org>, Paolo Abeni <pabeni@redhat.com>,
	netdev@vger.kernel.org, Ido Schimmel <idosch@nvidia.com>,
	Amit Cohen <amcohen@nvidia.com>,
	mlxsw@nvidia.com
Subject: Re: [PATCH net-next 09/15] mlxsw: pci: Use only one event queue
Date: Wed, 3 Apr 2024 13:39:21 +0100	[thread overview]
Message-ID: <20240403123921.GE26556@kernel.org> (raw)
In-Reply-To: <23d764f5c032e4c363b98590b746a4b32d2bf900.1712062203.git.petrm@nvidia.com>

On Tue, Apr 02, 2024 at 03:54:22PM +0200, Petr Machata wrote:
> From: Amit Cohen <amcohen@nvidia.com>
> 
> The device supports two event queues. EQ0 is used for command interface
> completion events. EQ1 is used for completion events of RDQ or SDQ.
> 
> Currently, for each EQE (event queue element), we check the queue number
> and handle accordingly. More than that, for each interrupt we schedule
> tasklets for both EQs. This is really ineffective, especially because of
> the fact that EQ0 is used only as part of driver init/fini, when EMADs are
> not available. There is no point to schedule the tasklet for it and check
> each EQE.
> 
> A previous patch changed the code to poll command interface for each use of
> it. It means that now there is no real reason to use EQ0, as we poll the
> command interface.
> 
> Initialize only one event queue and use it as EQ1 (this is determined by
> queue number). Then, for each interrupt we can schedule the tasklet only
> for one queue and we do not have to check the queue number. This
> simplifies the code and should improve performance. Note that polling
> command interface is ok as we use it only as part of driver init/fini.
> 
> Signed-off-by: Amit Cohen <amcohen@nvidia.com>
> Reviewed-by: Ido Schimmel <idosch@nvidia.com>
> Signed-off-by: Petr Machata <petrm@nvidia.com>

Reviewed-by: Simon Horman <horms@kernel.org>


  reply	other threads:[~2024-04-03 12:39 UTC|newest]

Thread overview: 32+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-04-02 13:54 [PATCH net-next 00/15] mlxsw: Preparations for improving performance Petr Machata
2024-04-02 13:54 ` [PATCH net-next 01/15] mlxsw: pci: Move mlxsw_pci_eq_{init, fini}() Petr Machata
2024-04-03 12:36   ` Simon Horman
2024-04-02 13:54 ` [PATCH net-next 02/15] mlxsw: pci: Move mlxsw_pci_cq_{init, fini}() Petr Machata
2024-04-03 12:37   ` Simon Horman
2024-04-02 13:54 ` [PATCH net-next 03/15] mlxsw: pci: Do not setup tasklet from operation Petr Machata
2024-04-03 12:37   ` Simon Horman
2024-04-02 13:54 ` [PATCH net-next 04/15] mlxsw: pci: Arm CQ doorbell regardless of number of completions Petr Machata
2024-04-03 12:37   ` Simon Horman
2024-04-02 13:54 ` [PATCH net-next 05/15] mlxsw: pci: Remove unused counters Petr Machata
2024-04-03 12:38   ` Simon Horman
2024-04-02 13:54 ` [PATCH net-next 06/15] mlxsw: pci: Make style changes in mlxsw_pci_eq_tasklet() Petr Machata
2024-04-03 12:38   ` Simon Horman
2024-04-02 13:54 ` [PATCH net-next 07/15] mlxsw: pci: Poll command interface for each cmd_exec() Petr Machata
2024-04-03 12:38   ` Simon Horman
2024-04-02 13:54 ` [PATCH net-next 08/15] mlxsw: pci: Rename MLXSW_PCI_EQS_COUNT Petr Machata
2024-04-03 12:39   ` Simon Horman
2024-04-02 13:54 ` [PATCH net-next 09/15] mlxsw: pci: Use only one event queue Petr Machata
2024-04-03 12:39   ` Simon Horman [this message]
2024-04-02 13:54 ` [PATCH net-next 10/15] mlxsw: pci: Remove unused wait queue Petr Machata
2024-04-03 12:39   ` Simon Horman
2024-04-02 13:54 ` [PATCH net-next 11/15] mlxsw: pci: Make style change in mlxsw_pci_cq_tasklet() Petr Machata
2024-04-03 12:39   ` Simon Horman
2024-04-02 13:54 ` [PATCH net-next 12/15] mlxsw: pci: Break mlxsw_pci_cq_tasklet() into tasklets per queue type Petr Machata
2024-04-03 12:40   ` Simon Horman
2024-04-02 13:54 ` [PATCH net-next 13/15] mlxsw: pci: Remove mlxsw_pci_sdq_count() Petr Machata
2024-04-03 12:40   ` Simon Horman
2024-04-02 13:54 ` [PATCH net-next 14/15] mlxsw: pci: Remove mlxsw_pci_cq_count() Petr Machata
2024-04-03 12:40   ` Simon Horman
2024-04-02 13:54 ` [PATCH net-next 15/15] mlxsw: pci: Store DQ pointer as part of CQ structure Petr Machata
2024-04-03 12:40   ` Simon Horman
2024-04-04  3:50 ` [PATCH net-next 00/15] mlxsw: Preparations for improving performance patchwork-bot+netdevbpf

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=20240403123921.GE26556@kernel.org \
    --to=horms@kernel.org \
    --cc=amcohen@nvidia.com \
    --cc=davem@davemloft.net \
    --cc=edumazet@google.com \
    --cc=idosch@nvidia.com \
    --cc=kuba@kernel.org \
    --cc=mlxsw@nvidia.com \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    --cc=petrm@nvidia.com \
    /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 NNTP newsgroup(s).