Re: vmalloc with GFP_NOFS

[Mike Rapoport <rppt@linux.vnet.ibm.com>]

On Wed, May 09, 2018 at 08:13:51AM -0700, Darrick J. Wong wrote:
> On Wed, May 09, 2018 at 03:42:22PM +0200, Michal Hocko wrote:
> > On Tue 24-04-18 13:25:42, Michal Hocko wrote:
> > [...]
> > > > As a suggestion, could you take
> > > > documentation about how to convert to the memalloc_nofs_{save,restore}
> > > > scope api (which I think you've written about e-mails at length
> > > > before), and put that into a file in Documentation/core-api?
> > > 
> > > I can.
> > 
> > Does something like the below sound reasonable/helpful?
> > ---
> > =================================
> > GFP masks used from FS/IO context
> > =================================
> > 
> > :Date: Mapy, 2018
> > :Author: Michal Hocko <mhocko@kernel.org>
> > 
> > Introduction
> > ============
> > 
> > FS resp. IO submitting code paths have to be careful when allocating
> 
> Not sure what 'FS resp. IO' means here -- 'FS and IO' ?
> 
> (Or is this one of those things where this looks like plain English text
> but in reality it's some sort of markup that I'm not so familiar with?)
> 
> Confused because I've seen 'resp.' used as shorthand for
> 'responsible'...
> 
> > memory to prevent from potential recursion deadlocks caused by direct
> > memory reclaim calling back into the FS/IO path and block on already
> > held resources (e.g. locks). Traditional way to avoid this problem
> 
> 'The traditional way to avoid this deadlock problem...'
> 
> > is to clear __GFP_FS resp. __GFP_IO (note the later implies clearing
> > the first as well) in the gfp mask when calling an allocator. GFP_NOFS
> > resp. GFP_NOIO can be used as shortcut.
> > 
> > This has been the traditional way to avoid deadlocks since ages. It
> 
> I think this sentence is a little redundant with the previous sentence,
> you could chop it out and join this paragraph to the one before it.
> 
> > turned out though that above approach has led to abuses when the restricted
> > gfp mask is used "just in case" without a deeper consideration which leads
> > to problems because an excessive use of GFP_NOFS/GFP_NOIO can lead to
> > memory over-reclaim or other memory reclaim issues.
> > 
> > New API
> > =======
> > 
> > Since 4.12 we do have a generic scope API for both NOFS and NOIO context
> > ``memalloc_nofs_save``, ``memalloc_nofs_restore`` resp. ``memalloc_noio_save``,
> > ``memalloc_noio_restore`` which allow to mark a scope to be a critical
> > section from the memory reclaim recursion into FS/IO POV. Any allocation
> > from that scope will inherently drop __GFP_FS resp. __GFP_IO from the given
> > mask so no memory allocation can recurse back in the FS/IO.
> > 
> > FS/IO code then simply calls the appropriate save function right at
> > the layer where a lock taken from the reclaim context (e.g. shrinker)
> > is taken and the corresponding restore function when the lock is

Seems like the second "is taken" got there by mistake

> > released. All that ideally along with an explanation what is the reclaim
> > context for easier maintenance.
> > 
> > What about __vmalloc(GFP_NOFS)
> > ==============================
> > 
> > vmalloc doesn't support GFP_NOFS semantic because there are hardcoded
> > GFP_KERNEL allocations deep inside the allocator which are quit non-trivial
> 
> ...which are quite non-trivial...
> 
> > to fix up. That means that calling ``vmalloc`` with GFP_NOFS/GFP_NOIO is
> > almost always a bug. The good news is that the NOFS/NOIO semantic can be
> > achieved by the scope api.
> > 
> > In the ideal world, upper layers should already mark dangerous contexts
> > and so no special care is required and vmalloc should be called without
> > any problems. Sometimes if the context is not really clear or there are
> > layering violations then the recommended way around that is to wrap ``vmalloc``
> > by the scope API with a comment explaining the problem.
> 
> Otherwise looks ok to me based on my understanding of how all this is
> supposed to work...
> 
> Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
> 
> --D
> 
> > -- 
> > Michal Hocko
> > SUSE Labs
> 

-- 
Sincerely yours,
Mike.