Re: rename *MEMALLOC flags (was: Re: [RFC][PATCH 3/4] deadlock prevention core)

[Peter Zijlstra <a.p.zijlstra@chello.nl>]

On Sat, 2006-08-12 at 17:28 +0200, Indan Zupancic wrote:
> On Sat, August 12, 2006 17:06, Peter Zijlstra said:
> > On Sat, 2006-08-12 at 10:41 -0400, Jeff Garzik wrote:
> >> Peter Zijlstra wrote:
> >> > Index: linux-2.6/include/linux/gfp.h
> >> > ===================================================================
> >> > --- linux-2.6.orig/include/linux/gfp.h	2006-08-12 12:56:06.000000000 +0200
> >> > +++ linux-2.6/include/linux/gfp.h	2006-08-12 12:56:09.000000000 +0200
> >> > @@ -46,6 +46,7 @@ struct vm_area_struct;
> >> >  #define __GFP_ZERO	((__force gfp_t)0x8000u)/* Return zeroed page on success */
> >> >  #define __GFP_NOMEMALLOC ((__force gfp_t)0x10000u) /* Don't use emergency reserves */
> >> >  #define __GFP_HARDWALL   ((__force gfp_t)0x20000u) /* Enforce hardwall cpuset memory allocs
> >> */
> >> > +#define __GFP_MEMALLOC  ((__force gfp_t)0x40000u) /* Use emergency reserves */
> >>
> >> This symbol name has nothing to do with its purpose.  The entire area of
> >> code you are modifying could be described as having something to do with
> >> 'memalloc'.
> >>
> >> GFP_EMERGENCY or GFP_USE_RESERVES or somesuch would be a far better
> >> symbol name.
> >>
> >> I recognize that is matches with GFP_NOMEMALLOC, but that doesn't change
> >> the situation anyway.  In fact, a cleanup patch to rename GFP_NOMEMALLOC
> >> would be nice.
> >
> > I'm rather bad at picking names, but here goes:
> >
> > PF_MEMALLOC      -> PF_EMERGALLOC
> > __GFP_NOMEMALLOC -> __GFP_NOEMERGALLOC
> > __GFP_MEMALLOC   -> __GFP_EMERGALLOC
    SOCK_MEMALLOC    -> SOCK_EMERGALLOC
> >
> > Is that suitable and shall I prepare patches? Or do we want more ppl to
> > chime in and have a few more rounds?
> 
> Pardon my ignorance, but if we're doing cleanup anyway, why not use only one flag instead of two?
> Why is __GFP_NOMEMALLOC needed when not setting __GFP_MEMALLOC could mean the same? Or else what
> is the expected behaviour if both flags are set?

__GFP_NOMEMALLOC is most authorative; its use is (afaik) to negate
PF_MEMALLOC.

I agree that having both seems odd, but I haven't spend any significant
time on trying to find a 'nicer' solution.