Re: [PATCH 00/14] paravirt: cleanup and reorg

[Jürgen Groß]

On 11.09.25 09:48, Peter Zijlstra wrote:

On Thu, Sep 11, 2025 at 08:34:19AM +0200, Juergen Gross wrote:

Some cleanups and reorg of paravirt code and headers:

- The first 2 patches should be not controversial at all, as they
   remove just some no longer needed #include and struct forward
   declarations.

- The 3rd patch is removing CONFIG_PARAVIRT_DEBUG, which IMO has
   no real value, as it just changes a crash to a BUG() (the stack
   trace will basically be the same). As the maintainer of the main
   paravirt user (Xen) I have never seen this crash/BUG() to happen.

- The 4th patch is just a movement of code.

- I don't know for what reason asm/paravirt_api_clock.h was added,
   as all archs supporting it do it exactly in the same way. Patch
   5 is removing it.

- Patches 6-12 are streamlining the paravirt clock interfaces by
   using a common implementation across architectures where possible
   and by moving the related code into common sched code, as this is
   where it should live.

- Patches 13+14 are more like RFC material: patch 13 is doing some
   preparation work to enable patch 14 to move all spinlock related
   paravirt functions into qspinlock.h. If this approach is accepted,
   I'd like to continue with this work by moving most (or all?)
   paravirt functions from paravirt.h into the headers where their
   native counterparts are defined. This is meant to keep the native
   and paravirt function definitions together in one place and
   hopefully to be able to reduce the include hell with paravirt.

Juergen Gross (14):
   x86/paravirt: remove not needed includes of paravirt.h
   x86/paravirt: remove some unneeded struct declarations
   x86/paravirt: remove PARAVIRT_DEBUG config option
   x86/paravirt: move thunk macros to paravirt_types.h
   paravirt: remove asm/paravirt_api_clock.h
   sched: move clock related paravirt code to kernel/sched
   arm/paravirt: use common code for paravirt_steal_clock()
   arm64/paravirt: use common code for paravirt_steal_clock()
   loongarch/paravirt: use common code for paravirt_steal_clock()
   riscv/paravirt: use common code for paravirt_steal_clock()
   x86/paravirt: use common code for paravirt_steal_clock()
   x86/paravirt: move paravirt_sched_clock() related code into tsc.c
   x86/paravirt: allow pv-calls outside paravirt.h
   x86/pvlocks: move paravirt spinlock functions into qspinlock.h

With the note that tip typically likes a capital after the prefix, like:

   x86/paravirt: Remove unneeded includes of paravirt.h

Noted, thanks.

For 1-12:

   Acked-by: Peter Zijlstra (Intel) <pet...@infradead.org>


Now, as to the last two, I'm not sure. Leaking those macros out of PV
isn't particularly nice, then again, not the end of the world either.
Just not sure.

Yes, that's why I didn't continue with all of the other potential movement of
paravirt functions. I want some feedback first. :-)

Its a tradeoff between having functions with / without paravirt in one file
against hiding the paravirt stuff from "normal" readers (not writers, as those
probably need to touch the paravirt variant, too).

BTW, I think the macro leaking isn't the main problem. There are other macros
leaking already.


Juergen

OpenPGP_0xB0DE9DD628BF132F.asc
Description: OpenPGP public key

OpenPGP_signature.asc
Description: OpenPGP digital signature