Re: [PATCH v5 02/22] liveupdate: luo_core: integrate with KHO

[Pasha Tatashin]

On Fri, Nov 14, 2025 at 6:30 AM Mike Rapoport <rppt@xxxxxxxxxx> wrote:
>
> On Fri, Nov 07, 2025 at 04:03:00PM -0500, Pasha Tatashin wrote:
> > Integrate the LUO with the KHO framework to enable passing LUO state
> > across a kexec reboot.
> >
> > When LUO is transitioned to a "prepared" state, it tells KHO to
> > finalize, so all memory segments that were added to KHO preservation
> > list are getting preserved. After "Prepared" state no new segments
> > can be preserved. If LUO is canceled, it also tells KHO to cancel the
> > serialization, and therefore, later LUO can go back into the prepared
> > state.
> >
> > This patch introduces the following changes:
> > - During the KHO finalization phase allocate FDT blob.
> > - Populate this FDT with a LUO compatibility string ("luo-v1").
> >
> > LUO now depends on `CONFIG_KEXEC_HANDOVER`. The core state transition
> > logic (`luo_do_*_calls`) remains unimplemented in this patch.
> >
> > Signed-off-by: Pasha Tatashin <pasha.tatashin@xxxxxxxxxx>
> > ---
> >  include/linux/liveupdate.h         |   6 +
> >  include/linux/liveupdate/abi/luo.h |  54 +++++++
> >  kernel/liveupdate/luo_core.c       | 243 ++++++++++++++++++++++++++++-
> >  kernel/liveupdate/luo_internal.h   |  17 ++
> >  mm/mm_init.c                       |   4 +
> >  5 files changed, 323 insertions(+), 1 deletion(-)
> >  create mode 100644 include/linux/liveupdate/abi/luo.h
> >  create mode 100644 kernel/liveupdate/luo_internal.h
> >
> > diff --git a/include/linux/liveupdate.h b/include/linux/liveupdate.h
> > index 730b76625fec..0be8804fc42a 100644
> > --- a/include/linux/liveupdate.h
> > +++ b/include/linux/liveupdate.h
> > @@ -13,6 +13,8 @@
> >
> >  #ifdef CONFIG_LIVEUPDATE
> >
> > +void __init liveupdate_init(void);
> > +
> >  /* Return true if live update orchestrator is enabled */
> >  bool liveupdate_enabled(void);
> >
> > @@ -21,6 +23,10 @@ int liveupdate_reboot(void);
> >
> >  #else /* CONFIG_LIVEUPDATE */
> >
> > +static inline void liveupdate_init(void)
> > +{
> > +}
>
> The common practice is to place brackets at the same line with function
> declaration.

Sure.

>
> ...
>
> > +static int __init luo_early_startup(void)
> > +{
> > +     phys_addr_t fdt_phys;
> > +     int err, ln_size;
> > +     const void *ptr;
> > +
> > +     if (!kho_is_enabled()) {
> > +             if (liveupdate_enabled())
> > +                     pr_warn("Disabling liveupdate because KHO is disabled\n");
> > +             luo_global.enabled = false;
> > +             return 0;
> > +     }
> > +
> > +     /* Retrieve LUO subtree, and verify its format. */
> > +     err = kho_retrieve_subtree(LUO_FDT_KHO_ENTRY_NAME, &fdt_phys);
> > +     if (err) {
> > +             if (err != -ENOENT) {
> > +                     pr_err("failed to retrieve FDT '%s' from KHO: %pe\n",
> > +                            LUO_FDT_KHO_ENTRY_NAME, ERR_PTR(err));
> > +                     return err;
> > +             }
> > +
> > +             return 0;
> > +     }
> > +
> > +     luo_global.fdt_in = __va(fdt_phys);
>
> phys_to_virt is clearer, isn't it?

Sure

>
> > +     err = fdt_node_check_compatible(luo_global.fdt_in, 0,
> > +                                     LUO_FDT_COMPATIBLE);
>
> ...
>
> > +void __init liveupdate_init(void)
> > +{
> > +     int err;
> > +
> > +     err = luo_early_startup();
> > +     if (err) {
> > +             pr_err("The incoming tree failed to initialize properly [%pe], disabling live update\n",
> > +                    ERR_PTR(err));
> > +             luo_global.enabled = false;
> > +     }
> > +}
> > +
> > +/* Called during boot to create LUO fdt tree */
>
>                          ^ create outgoing

OK

>
> > +static int __init luo_late_startup(void)
> > +{
> > +     int err;
> > +
> > +     if (!liveupdate_enabled())
> > +             return 0;
> > +
> > +     err = luo_fdt_setup();
> > +     if (err)
> > +             luo_global.enabled = false;
> > +
> > +     return err;
> > +}
> > +late_initcall(luo_late_startup);
>
> It would be nice to have a comment explaining why late_initcall() is fine
> and why there's no need to initialize the outgoing fdt earlier.

I will add a comment; basically it is fine because the outgoing data
structures are only used after we enter userspace.

>
> > +/**
> > + * luo_alloc_preserve - Allocate, zero, and preserve memory.
>
> I think this and the "free" counterparts would be useful for any KHO users,
> even those that don't need LUO.

I will move them to KHO.

>
> > + * @size: The number of bytes to allocate.
> > + *
> > + * Allocates a physically contiguous block of zeroed pages that is large
> > + * enough to hold @size bytes. The allocated memory is then registered with
> > + * KHO for preservation across a kexec.
> > + *
> > + * Note: The actual allocated size will be rounded up to the nearest
> > + * power-of-two page boundary.
> > + *
> > + * @return A virtual pointer to the allocated and preserved memory on success,
> > + * or an ERR_PTR() encoded error on failure.
> > + */
> > +void *luo_alloc_preserve(size_t size)
> > +{
> > +     struct folio *folio;
> > +     int order, ret;
> > +
> > +     if (!size)
> > +             return ERR_PTR(-EINVAL);
> > +
> > +     order = get_order(size);
> > +     if (order > MAX_PAGE_ORDER)
> > +             return ERR_PTR(-E2BIG);
>
> High order allocations would likely fail or at least cause a heavy reclaim.
> For now it seems that we won't be needing really large contiguous chunks so
> maybe limiting this to PAGE_ALLOC_COSTLY_ORDER?

Let's use MAX_PAGE_ORDER for now, my concern is that
PAGE_ALLOC_COSTLY_ORDER too fragile to make it part of ABI. If
allocation fails, the user will have to deal with it, as we return a
proper error code.

> Later if we'd need higher order allocations we can try to allocate with
> __GFP_NORETRY or __GFP_RETRY_MAYFAIL with a fallback to vmalloc.
>
> > +
> > +     folio = folio_alloc(GFP_KERNEL | __GFP_ZERO, order);
> > +     if (!folio)
> > +             return ERR_PTR(-ENOMEM);
> > +
> > +     ret = kho_preserve_folio(folio);
> > +     if (ret) {
> > +             folio_put(folio);
> > +             return ERR_PTR(ret);
> > +     }
> > +
> > +     return folio_address(folio);
> > +}
> > +
>
> --
> Sincerely yours,
> Mike.