Re: [PATCH v2] mm: Fix checking unmapped holes for mbind

Re: [PATCH v2] mm: Fix checking unmapped holes for mbind

[Andrew Morton]

(cc linux-man@vger.kernel.org)

On Tue, 29 Oct 2019 17:56:06 +0800 "Li Xinhai" <lixinhai.lxh@gmail.com> wrote:

> queue_pages_range() will check for unmapped holes besides queue pages for
> migration. The rules for checking unmapped holes are:
> 1 Unmapped holes at any part of the specified range should be reported as
>   EFAULT if mbind() for none MPOL_DEFAULT cases;
> 2 Unmapped holes at any part of the specified range should be ignored if
>   mbind() for MPOL_DEFAULT case;
> Note that the second rule is the current implementation, but it seems
> conflicts the Linux API definition.

Can you quote the part of the API definition which you're looking at?

My mbind(2) manpage says

ERRORS
       EFAULT Part or all of the memory range specified by nodemask and maxn-
              ode points outside your accessible address space.  Or, there was
              an unmapped hole in the specified memory range specified by addr
              and len.

(I assume the first sentence meant to say "specified by addr and len")

I agree with your interpretation, but there's no mention here that
MPOL_DEFAULT is treated differently and I don't see why it should be.


More broadly, I worry that it's too late to change this - existing
applications might fail if we change the implementation in the proposed
fashion.  So perhaps what we should do here is to change the manpage to
match reality?

Is the current behavior causing you any problems in a real-world use
case?

Re: [PATCH v2] mm: Fix checking unmapped holes for mbind

[Li Xinhai]

On 2019-10-31 at 12:08 Andrew Morton wrote:
>(cc linux-man@vger.kernel.org)
>
>On Tue, 29 Oct 2019 17:56:06 +0800 "Li Xinhai" <lixinhai.lxh@gmail.com> wrote:
>
>> queue_pages_range() will check for unmapped holes besides queue pages for
>> migration. The rules for checking unmapped holes are:
>> 1 Unmapped holes at any part of the specified range should be reported as
>>   EFAULT if mbind() for none MPOL_DEFAULT cases;
>> 2 Unmapped holes at any part of the specified range should be ignored if
>>   mbind() for MPOL_DEFAULT case;
>> Note that the second rule is the current implementation, but it seems
>> conflicts the Linux API definition.
>
>Can you quote the part of the API definition which you're looking at?
>
>My mbind(2) manpage says
>
>ERRORS
>       EFAULT Part or all of the memory range specified by nodemask and maxn-
>              ode points outside your accessible address space.  Or, there was
>              an unmapped hole in the specified memory range specified by addr
>              and len.
>
>(I assume the first sentence meant to say "specified by addr and len")
> 

this part: 
"Or, there was an unmapped hole in the specified memory range specified by addr 
and len."
is concerned by my patch.

>I agree with your interpretation, but there's no mention here that
>MPOL_DEFAULT is treated differently and I don't see why it should be.
> 
The first rule match the manpage, but the current mempolicy implementation only 
reports EFAULT if holes are within range, or at the head side of range. No EFAULT 
reported if hole at tail side of range. I suppose the first rule has to be fixed.

The seconde rule, when MPOL_DEFAULT is used, was summarized by me according 
to mempolicy implementation. Actually, this rule does not follow manpage and exsits 
for long days. In my understanding, this rule is reasonable (in code,  the internal flag 
MPOL_MF_DISCONTIG_OK is used for that purpose, there is comments for reason) 
and we'd better keep it.

>
>More broadly, I worry that it's too late to change this - existing
>applications might fail if we change the implementation in the proposed
>fashion.  So perhaps what we should do here is to change the manpage to
>match reality?
> 
I prefer add description in manpage for the second rule, so no change to our code. 
Only fix for first rule.

>Is the current behavior causing you any problems in a real-world use
>case? 
I was using mbind() with MPOL_DEFAULT(or MPOL_BIND) to reset a range of address 
(which maybe contiguous or not in the whole range) to the default policy (to a specific 
node), and observed this issue. If mbind() call for each mapping one by one, we don't see the 
issue.

- Xinhai

Re: [PATCH v2] mm: Fix checking unmapped holes for mbind

[Michal Hocko]

On Wed 30-10-19 21:08:36, Andrew Morton wrote:
> (cc linux-man@vger.kernel.org)
> 
> On Tue, 29 Oct 2019 17:56:06 +0800 "Li Xinhai" <lixinhai.lxh@gmail.com> wrote:
> 
> > queue_pages_range() will check for unmapped holes besides queue pages for
> > migration. The rules for checking unmapped holes are:
> > 1 Unmapped holes at any part of the specified range should be reported as
> >   EFAULT if mbind() for none MPOL_DEFAULT cases;
> > 2 Unmapped holes at any part of the specified range should be ignored if
> >   mbind() for MPOL_DEFAULT case;
> > Note that the second rule is the current implementation, but it seems
> > conflicts the Linux API definition.
> 
> Can you quote the part of the API definition which you're looking at?
> 
> My mbind(2) manpage says
> 
> ERRORS
>        EFAULT Part or all of the memory range specified by nodemask and maxn-
>               ode points outside your accessible address space.  Or, there was
>               an unmapped hole in the specified memory range specified by addr
>               and len.
> 
> (I assume the first sentence meant to say "specified by addr and len")

My understanding is that this really refers to area pointed to by nodemask.
Btw. why there is any special casing around the unmapped holes with the
address space range? This looks like an antipattern to other address
space operations to me. E.g. munmap simply unmaps all existing vmas in
the given range, mprotect, madvise etc. behave the same.

So my question is, do we want to remove that weird restriction and
simply act on all existing VMAs in the range? The only situation this
could regress would be if somebody used mbind to probe for existing VMAs
and that sounds a more than sensible to me. Or am I missing anything?
-- 
Michal Hocko
SUSE Labs

Re: [PATCH v2] mm: Fix checking unmapped holes for mbind

[Li Xinhai]

On 2019-10-31 at 19:26 Michal Hocko wrote:
>On Wed 30-10-19 21:08:36, Andrew Morton wrote:
>> (cc linux-man@vger.kernel.org)
>>
>> On Tue, 29 Oct 2019 17:56:06 +0800 "Li Xinhai" <lixinhai.lxh@gmail.com> wrote:
>>
>> > queue_pages_range() will check for unmapped holes besides queue pages for
>> > migration. The rules for checking unmapped holes are:
>> > 1 Unmapped holes at any part of the specified range should be reported as
>> >   EFAULT if mbind() for none MPOL_DEFAULT cases;
>> > 2 Unmapped holes at any part of the specified range should be ignored if
>> >   mbind() for MPOL_DEFAULT case;
>> > Note that the second rule is the current implementation, but it seems
>> > conflicts the Linux API definition.
>>
>> Can you quote the part of the API definition which you're looking at?
>>
>> My mbind(2) manpage says
>>
>> ERRORS
>>        EFAULT Part or all of the memory range specified by nodemask and maxn-
>>               ode points outside your accessible address space.  Or, there was
>>               an unmapped hole in the specified memory range specified by addr
>>               and len.
>>
>> (I assume the first sentence meant to say "specified by addr and len")
>
>My understanding is that this really refers to area pointed to by nodemask.
>Btw. why there is any special casing around the unmapped holes with the
>address space range? This looks like an antipattern to other address
>space operations to me. E.g. munmap simply unmaps all existing vmas in
>the given range, mprotect, madvise etc. behave the same.
>
>So my question is, do we want to remove that weird restriction and
>simply act on all existing VMAs in the range? The only situation this
>could regress would be if somebody used mbind to probe for existing VMAs
>and that sounds a more than sensible to me. Or am I missing anything?
>--
>Michal Hocko
>SUSE Labs 

yes, mbind() care about the unmapped holes for non MPOL_DEFAULT cases, but 
other operations don't care those holes. It seems no clues about why that 
restriction was decided.

At present, if it is hard to decide to remove this restriction, we may keep the 
current behavior. New patch posted for this purpose,  
"[PATCH v3] mm: Fix checking unmapped holes for mbind".
Thanks.

Xinhai