Andrei Vagin [off-list ref]:

On Wed, Jun 24, 2026 at 12:12 AM Askar Safin [off-list ref] wrote:

quoted

Does CRIU actually rely on ability to do SPLICE_F_NONBLOCK vmsplice into
named fifos? Or this is merely a test?

Yes, it does.

I. e. CRIU relies on that named fifo behavior? Okay, I just sent
v2 version of my fixes. The patchset contains fix for named fifos.

Please, test that this fixes that named fifo problem.

I already explained that this isn't just a perfomance degradation, it
actually breaks the pre-dump mechanism in CRIU. vmsplice is invoked from
our parasite code within the context of a user process, where execution
speed is critical. A heavy performance penalty completely invalidates
the pre-dump logic, making the feature useless.

This is very unfortunate. But I still want to remove vmsplice.

At a minimum, we may need to consider a deprecation plan where vmsplice
with SPLICE_F_GIFT triggers a warning for a few releases before these
changes are applied. Alternatively, we could introduce the proposed
behavior alongside a sysctl to fall back to the old behavior and explicitly
state that this fallback path will be completely deprecated in a future kernel
version.

My patches change not only SPLICE_F_GIFT behavior, but also vmsplice
behavior in general.

Let other developers decide what to do (i. e. do nothing, remove
vmsplice now or implement some deprecation scheme).

-- 
Askar Safin