Re: [PATCH v5 3/3] xen: add helpers to allocate unpopulated memory
From: Roger Pau Monné
Date: Fri Sep 04 2020 - 04:42:48 EST
On Fri, Sep 04, 2020 at 09:00:18AM +0200, Jürgen Groß wrote:
> On 03.09.20 18:38, Roger Pau Monné wrote:
> > On Thu, Sep 03, 2020 at 05:30:07PM +0200, Jürgen Groß wrote:
> > > On 01.09.20 10:33, Roger Pau Monne wrote:
> > > > To be used in order to create foreign mappings. This is based on the
> > > > ZONE_DEVICE facility which is used by persistent memory devices in
> > > > order to create struct pages and kernel virtual mappings for the IOMEM
> > > > areas of such devices. Note that on kernels without support for
> > > > ZONE_DEVICE Xen will fallback to use ballooned pages in order to
> > > > create foreign mappings.
> > > >
> > > > The newly added helpers use the same parameters as the existing
> > > > {alloc/free}_xenballooned_pages functions, which allows for in-place
> > > > replacement of the callers. Once a memory region has been added to be
> > > > used as scratch mapping space it will no longer be released, and pages
> > > > returned are kept in a linked list. This allows to have a buffer of
> > > > pages and prevents resorting to frequent additions and removals of
> > > > regions.
> > > >
> > > > If enabled (because ZONE_DEVICE is supported) the usage of the new
> > > > functionality untangles Xen balloon and RAM hotplug from the usage of
> > > > unpopulated physical memory ranges to map foreign pages, which is the
> > > > correct thing to do in order to avoid mappings of foreign pages depend
> > > > on memory hotplug.
> > > >
> > > > Note the driver is currently not enabled on Arm platforms because it
> > > > would interfere with the identity mapping required on some platforms.
> > > >
> > > > Signed-off-by: Roger Pau Monné <roger.pau@xxxxxxxxxx>
> > >
> > > Sorry, I just got a build error for x86 32-bit build:
> > >
> > > WARNING: unmet direct dependencies detected for ZONE_DEVICE
> > > Depends on [n]: MEMORY_HOTPLUG [=n] && MEMORY_HOTREMOVE [=n] &&
> > > SPARSEMEM_VMEMMAP [=n] && ARCH_HAS_PTE_DEVMAP [=n]
> > > Selected by [y]:
> > > - XEN_UNPOPULATED_ALLOC [=y] && XEN [=y] && X86 [=y]
> > > GEN Makefile
> > > CC kernel/bounds.s
> > > CALL /home/gross/korg/src/scripts/atomic/check-atomics.sh
> > > UPD include/generated/bounds.h
> > > CC arch/x86/kernel/asm-offsets.s
> > > In file included from /home/gross/korg/src/include/linux/mmzone.h:19:0,
> > > from /home/gross/korg/src/include/linux/gfp.h:6,
> > > from /home/gross/korg/src/include/linux/slab.h:15,
> > > from /home/gross/korg/src/include/linux/crypto.h:19,
> > > from /home/gross/korg/src/arch/x86/kernel/asm-offsets.c:9:
> > > /home/gross/korg/src/include/linux/page-flags-layout.h:95:2: error: #error
> > > "Not enough bits in page flags"
> > > #error "Not enough bits in page flags"
> > > ^~~~~
> > > make[2]: *** [/home/gross/korg/src/scripts/Makefile.build:114:
> > > arch/x86/kernel/asm-offsets.s] Error 1
> > > make[1]: *** [/home/gross/korg/src/Makefile:1175: prepare0] Error 2
> > > make[1]: Leaving directory '/home/gross/korg/x8632'
> > > make: *** [Makefile:185: __sub-make] Error 2
> >
> > Sorry for this. I've tested a 32bit build but I think it was before
> > the last Kconfig changes. I'm a little unsure how to solve this, as
> > ZONE_DEVICE doesn't select the required options for it to run, but
> > rather depends on them to be available.
> >
> > You can trigger something similar on x86-64 by doing:
> >
> > $ make ARCH=x86_64 xen.config
> > Using .config as base
> > Merging ./kernel/configs/xen.config
> > Merging ./arch/x86/configs/xen.config
> > #
> > # merged configuration written to .config (needs make)
> > #
> > scripts/kconfig/conf --olddefconfig Kconfig
> >
> > WARNING: unmet direct dependencies detected for ZONE_DEVICE
> > Depends on [n]: MEMORY_HOTPLUG [=y] && MEMORY_HOTREMOVE [=n] && SPARSEMEM_VMEMMAP [=y] && ARCH_HAS_PTE_DEVMAP [=y]
> > Selected by [y]:
> > - XEN_UNPOPULATED_ALLOC [=y] && XEN [=y] && X86_64 [=y]
> > #
> > # configuration written to .config
> > #
> >
> > I think the only solution is to have XEN_UNPOPULATED_ALLOC depend on
> > ZONE_DEVICE rather than select it?
>
> Yes, I think so.
>
> I've folded that in and now build is fine.
Thanks, I assume no further action is needed on my side.
Roger.