Re: [PATCH 1/2] RDMA/nldev: add resource summary max values for usage rate display

[Leon Romanovsky <leon@kernel.org>]

On Tue, May 12, 2026 at 03:38:59PM +0800, Tao Cui wrote:
> 
> Hi，Leon
> 
> Thanks for the review. You're right that a percentage alone is not
> very helpful to users.
> 
> 在 2026/5/11 18:12, Leon Romanovsky 写道:
> > On Thu, Apr 23, 2026 at 02:13:51PM +0800, Tao Cui wrote:
> >> Add RDMA_NLDEV_ATTR_RES_SUMMARY_ENTRY_MAX netlink attribute to expose
> >> device resource limits (max_qp, max_cq, max_mr, max_pd, max_srq) in
> >> the resource summary alongside the existing current count. This allows
> >> userspace tools like iproute2's rdma to display resource usage rates.
> > 
> > While I'm fine with the overall idea, I think we should spend more time  
> > determining the proper display format for this information. Once we agree  
> > on how it should be presented, that output should be included in the commit  
> > message.
> > 
> > Presenting utilization as a percentage seems too imprecise, and users would  
> > likely prefer to see the maximum value instead.
> > 
> > Thanks
> 
> I originally chose the percentage format for its intuitiveness.
> The expected output was described in the corresponding iproute2
> patch [1], but I should have included it in this kernel commit
> message as well.
> 
> [1] https://lore.kernel.org/all/20260423064711.360024-1-cuitao@kylinos.cn/
> 
> Here are the display formats I considered:
> 1) qp 123 (0.0%)            -- percentage only, loses the actual max
> 2) qp 123 (max 131072)      -- clear but verbose
> 3) qp 123/131072            -- curr/max, concise and common in Linux tools
> 4) qp 123/131072 (0.0%)     -- all info, but too crowded for one line
> 
> After comparing them, I think format 3 is the best fit. It is concise,
> follows conventions used by other Linux tools (e.g. df, free), and
> users can easily estimate the percentage from the curr/max values.
> 
>   Before: 0: mlx5_0: qp 123  cq 45  mr 200  pd 10
>   After:  0: mlx5_0: qp 123/131072  cq 45/65536  mr 200/1000000  pd 10/32768
> 
> In JSON output, both "curr" and "max" fields will be provided so that
> scripts can compute percentages if needed.
> 
> The kernel side change (exposing the max value) remains the same. I'll
> update the commit messages to include the expected output format and
> resubmit.
> 
> Does this look acceptable?

Yes, I agree that option number 3 appears to be the best choice.

Thanks