-----Original Message-----
From: Jakub Kicinski <kuba@kernel.org>
Sent: Wednesday, January 14, 2026 9:55 PM
To: Haiyang Zhang <haiyangz@microsoft.com>
Cc: Haiyang Zhang <redacted>; linux-
hyperv@vger.kernel.org; netdev@vger.kernel.org; KY Srinivasan
[off-list ref]; Wei Liu [off-list ref]; Dexuan Cui
[off-list ref]; Long Li [off-list ref]; Andrew Lunn
[off-list ref]; David S. Miller [off-list ref]; Eric
Dumazet [off-list ref]; Paolo Abeni [off-list ref]; Konstantin
Taranov [off-list ref]; Simon Horman [off-list ref]; Erni
Sri Satya Vennela [off-list ref]; Shradha Gupta
[off-list ref]; Saurabh Sengar
[off-list ref]; Aditya Garg
[off-list ref]; Dipayaan Roy
[off-list ref]; Shiraz Saleem
[off-list ref]; linux-kernel@vger.kernel.org; linux-
rdma@vger.kernel.org; Paul Rosswurm [off-list ref]
Subject: Re: [EXTERNAL] Re: [PATCH V2,net-next, 1/2] net: mana: Add
support for coalesced RX packets on CQE

On Wed, 14 Jan 2026 18:27:50 +0000 Haiyang Zhang wrote:

quoted

quoted

quoted

And, the coalescing can add up to 2 microseconds into one-way

latency.

quoted

quoted

I am asking you how the _device_ (hypervisor?) decides when to

coalesce

quoted

quoted

and when to send a partial CQE (<4 packets in 4 pkt CQE). You are

using

quoted

quoted

the coalescing uAPI, so I'm trying to make sure this is the correct

API.

quoted

quoted

CQE configuration can also be done via ringparam.

When coalescing is enabled, the device waits for packets which can
have the CQE coalesced with previous packet(s). That coalescing process
is finished (and a CQE written to the appropriate CQ) when the CQE is
filled with 4 pkts, or time expired, or other device specific logic is
satisfied.

See, what I'm afraid is happening here is that you are enabling
completion coalescing (how long the device keeps the CQE pending).
Which is _not_ what rx_max_coalesced_frames controls for most NICs.
For most NICs rx_max_coalesced_frames controls IRQ generation logic.

The NIC first buffers up CQEs for typically single digit usecs, and
then once CQE timer exipred and writeback happened it starts an IRQ
coalescing timer. Once the IRQ coalescing timer expires IRQ is
triggered, which schedules NAPI. (broad strokes, obviously many
differences and optimizations exist)

Is my guess correct? Are you controlling CQE coalescing>

Can you control the timeout instead of the frame count?

Our NIC's timeout value cannot be controlled by driver. Also, the
timeout may be changed in future NIC HW.

So, I use the ethtool/rx-frames, which is either 1 or 4 on our
NIC, to switch the CQE coalescing feature on/off.

Thanks,
- Haiyang