On Wed, Nov 04, 2026 at 08:53:04AM -0700, Keith Busch wrote:
On Wed, Nov 04, 2026 at 04:43:09PM +0200, Jarkko Sakkinen wrote:
quoted
On Wed, Nov 04, 2026 at 10:04:40AM +0100, Christoph Hellwig wrote:
quoted
On Tue, Nov 03, 2026 at 02:29:11PM +0200, Jarkko Sakkinen wrote:
quoted
And essentially what I would want is key_type_dead not to exist.
This version is based on module pinning, and thus not having that
key_put() would keep key type module alive in that case. However, it
turned out this approach will not fly as AF_RXRPC and DNS resolver will
hit the wall given e.g., the remote pinnings.
The approach I'm going for next is to have a type registry instead,
which is independent of the module system (to put story really short).
I think I'm doing this with right process as this really needs a few
trials in order to find patterns that stick.
So what's the plan here. Should we merge this standalone as a fix,
or are you trying to get an entire series in?
If it's useful as a standalone fix, please resend it standalone.
I'll send this as a standalone patch.
I can just apply it as-is unless you plan to make any change from what
was submitted. Just future reference that the intention wasn't clear
when a series' recipients are split up this way.
Yeah, I 100% agree with this.
Br, Jarkko