Re: [PATCH net RESEND] net/mlx5: Use unique names for software steering caches
From: Christian Borntraeger
Date: Tue Oct 27 2026 - 08:47:57 EST
Am 15.09.26 um 13:13 schrieb Leon Romanovsky:
On Mon, Sep 14, 2026 at 10:58:43AM +0200, Christian Borntraeger wrote:Ping.
Am 06.08.26 um 13:17 schrieb Leon Romanovsky:
On Tue, Aug 04, 2026 at 02:57:56PM -0700, Jakub Kicinski wrote:Any updates on this?
On Tue, 4 Aug 2026 11:04:55 +0300 Leon Romanovsky wrote:
On Mon, Aug 03, 2026 at 06:26:14PM -0700, Jakub Kicinski wrote:
On Fri, 31 Jul 2026 00:02:00 +0300 Tariq Toukan wrote:
Prefix each cache name with the device name to make it unique.
Why do we need a per-device kmem_cache?
Isn't the right fix to make this driver-level state?
This kmem_cache is required only for software managed steering mode, which is
configured per device by mlx5_fs_mode_set(). In general, resources
should be allocated only when they are actually needed.
Of course you have to wrap the init in a lock and count users.
Trivial stuff.
I don't see why we should abandon the simple, architecturally
correct approach of allocating and managing resources where they
belong in favor of a more complex global scheme involving locking
and reference counting.
No, we are waiting for Jakub to explain why should we complicate simple
and working solution.
So in the end this is something for you to sort out with Jakub.
I can just say that we see this issue in our s390 CI without the patch
so the patch as is is
Tested-by: Christian Borntraeger <borntraeger@xxxxxxxxxxxxx>
I can offer to test alternative patches if any.