netdev - Re: [PATCH bpf-next] mm: implement no-MMU variant of vmalloc_user_node

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Mon, 25 Nov 2019 12:38:26 -0800
From:   Andrii Nakryiko <andrii.nakryiko@...il.com>
To:     John Fastabend <john.fastabend@...il.com>
Cc:     Andrii Nakryiko <andriin@...com>, bpf <bpf@...r.kernel.org>,
        Networking <netdev@...r.kernel.org>,
        Alexei Starovoitov <ast@...com>,
        Daniel Borkmann <daniel@...earbox.net>,
        Kernel Team <kernel-team@...com>,
        Johannes Weiner <hannes@...xchg.org>
Subject: Re: [PATCH bpf-next] mm: implement no-MMU variant of vmalloc_user_node_flags

On Mon, Nov 25, 2019 at 12:36 PM John Fastabend
<john.fastabend@...il.com> wrote:
>
> Andrii Nakryiko wrote:
> > To fix build with !CONFIG_MMU, implement it for no-MMU configurations as well.
> >
> > Cc: Johannes Weiner <hannes@...xchg.org>
> > Fixes: fc9702273e2e ("bpf: Add mmap() support for BPF_MAP_TYPE_ARRAY")
> > Reported-by: kbuild test robot <lkp@...el.com>
> > Signed-off-by: Andrii Nakryiko <andriin@...com>
> > ---
> >  mm/nommu.c | 15 +++++++++++++--
> >  1 file changed, 13 insertions(+), 2 deletions(-)
> >
> > diff --git a/mm/nommu.c b/mm/nommu.c
> > index 99b7ec318824..7de592058ab4 100644
> > --- a/mm/nommu.c
> > +++ b/mm/nommu.c
> > @@ -155,11 +155,11 @@ void *__vmalloc_node_flags(unsigned long size, int node, gfp_t flags)
> >       return __vmalloc(size, flags, PAGE_KERNEL);
> >  }
> >
> > -void *vmalloc_user(unsigned long size)
> > +static void *__vmalloc_user_flags(unsigned long size, gfp_t flags)
> >  {
> >       void *ret;
> >
> > -     ret = __vmalloc(size, GFP_KERNEL | __GFP_ZERO, PAGE_KERNEL);
> > +     ret = __vmalloc(size, flags, PAGE_KERNEL);
> >       if (ret) {
> >               struct vm_area_struct *vma;
> >
> > @@ -172,8 +172,19 @@ void *vmalloc_user(unsigned long size)
> >
> >       return ret;
> >  }
> > +
> > +void *vmalloc_user(unsigned long size)
> > +{
> > +     return __vmalloc_user_flags(size, GFP_KERNEL | __GFP_ZERO);
> > +}
> >  EXPORT_SYMBOL(vmalloc_user);
> >
> > +void *vmalloc_user_node_flags(unsigned long size, int node, gfp_t flags)
> > +{
> > +     return __vmalloc_user_flags(size, flags | __GFP_ZERO);
> > +}
> > +EXPORT_SYMBOL(vmalloc_user_node_flags);
> > +
>
> Hi Andrii, my first reaction was that it seemed not ideal to just ignore
> the node value like this but everything I came up with was uglier. I
> guess only user is BPF at the moment so it should be fine.

Yeah, but that's what other node-aware vmalloc() variants do in
nommu.c, so at least it's consistent with other cases. Thanks for
review!

>
> Acked-by: John Fastabend <john.fastabend@...il.com>