Re: [PATCH v4 bpf 0/4] vmalloc: bpf: introduce VM_ALLOW_HUGE_VMAP

From: Song Liu <song@kernel.org>
To: Luis Chamberlain <mcgrof@kernel.org>
Cc: bpf <bpf@vger.kernel.org>, Linux-MM <linux-mm@kvack.org>,
	open list <linux-kernel@vger.kernel.org>,
	Alexei Starovoitov <ast@kernel.org>,
	Daniel Borkmann <daniel@iogearbox.net>,
	Kernel Team <kernel-team@fb.com>,
	Andrew Morton <akpm@linux-foundation.org>,
	"Edgecombe, Rick P" <rick.p.edgecombe@intel.com>,
	Christoph Hellwig <hch@infradead.org>,
	imbrenda@linux.ibm.com
Subject: Re: [PATCH v4 bpf 0/4] vmalloc: bpf: introduce VM_ALLOW_HUGE_VMAP
Date: Fri, 15 Apr 2022 18:34:16 -0700	[thread overview]
Message-ID: <CAPhsuW6LY8g6SZpW0CBOe22FUd2e3oSq4RW1GDxZoPmHYBqwvw@mail.gmail.com> (raw)
In-Reply-To: <YlnCBqNWxSm3M3xB@bombadil.infradead.org>

On Fri, Apr 15, 2022 at 12:05 PM Luis Chamberlain <mcgrof@kernel.org> wrote:
>
> On Fri, Apr 15, 2022 at 09:44:09AM -0700, Song Liu wrote:
> > Changes v3 => v4:
> > 1. Fix __weak module_alloc_huge; remove unused vmalloc_huge; rename
> >    __vmalloc_huge => vmalloc_huge. (Christoph Hellwig)
> > 2. Use vzalloc (as it was before vmalloc_no_huge) and clean up comments in
> >    kvm_s390_pv_alloc_vm.
> >
> > Changes v2 => v3:
> > 1. Use __vmalloc_huge in alloc_large_system_hash.
> > 2. Use EXPORT_SYMBOL_GPL for new functions. (Christoph Hellwig)
> > 3. Add more description about the issues and changes.(Christoph Hellwig,
> >    Rick Edgecombe).
> >
> > Changes v1 => v2:
> > 1. Add vmalloc_huge(). (Christoph Hellwig)
> > 2. Add module_alloc_huge(). (Christoph Hellwig)
> > 3. Add Fixes tag and Link tag. (Thorsten Leemhuis)
> >
> > Enabling HAVE_ARCH_HUGE_VMALLOC on x86_64 and use it for bpf_prog_pack has
> > caused some issues [1], as many users of vmalloc are not yet ready to
> > handle huge pages. To enable a more smooth transition to use huge page
> > backed vmalloc memory, this set replaces VM_NO_HUGE_VMAP flag with an new
> > opt-in flag, VM_ALLOW_HUGE_VMAP. More discussions about this topic can be
> > found at [2].
> >
> > Patch 1 removes VM_NO_HUGE_VMAP and adds VM_ALLOW_HUGE_VMAP.
> > Patch 2 uses VM_ALLOW_HUGE_VMAP in bpf_prog_pack.
> >
> > [1] https://lore.kernel.org/lkml/20220204185742.271030-1-song@kernel.org/
> > [2] https://lore.kernel.org/linux-mm/20220330225642.1163897-1-song@kernel.org/
>
> Looks good except for that I think this should just wait for v5.19. The
> fixes are so large I can't see why this needs to be rushed in other than
> the first assumptions of the optimizations had some flaws addressed here.

We need these changes to fix issues like [3]. Note that there might
still be some
undiscovered issues with huge page backed vmalloc memory on powerpc, which
had HAVE_ARCH_HUGE_VMALLOC enabled since the 5.15 kernel. As we
agreed, the new opt-in flag is a safer approach here. We probably should have
1/4 and 2/4 back ported to stable. Therefore, I think shipping this
set now would
give us a more reliable 5.18 release.

Does this make sense?

Thanks,
Song

[3] https://lore.kernel.org/lkml/14444103-d51b-0fb3-ee63-c3f182f0b546@molgen.mpg.de/