Re: [PATCH 3/5] pagemap: Make pagemap_read enforce reading in multiples of 8

[Matt Mackall <mpm@selenic.com>]

On Thu, 2008-06-05 at 12:05 -0400, Thomas Tuttle wrote:
> On Thu, Jun 5, 2008 at 12:03 PM, Matt Mackall <mpm@selenic.com> wrote:
> >
> > On Thu, 2008-06-05 at 11:06 -0400, Thomas Tuttle wrote:
> >> Since kpagecount and kpageflags require reads in multiples of 8, and
> >> it simplifies add_to_pagemap significantly, I added the same
> >> requirement to /proc/pid/pagemap.
> >
> > I'm generally fine with this. Another approach that's perhaps more
> > friendly is when someone tries to do a 24-byte read, do a 16-byte read,
> > leaving the file pointer aligned. Not sure if that's completely kosher
> > though.
> 
> This doesn't require that they read exactly 8 bytes, just that they
> read some multiple of 8 bytes.  (They can read 8, 16, 24, etc..., just
> not something like 12.)

Yes, I got that, I'm just mathematically impaired this morning. Read
that as "when someone tries to do a *20* byte read, do a 16-byte read.."
In other words, round down to the nearest multiple of 8.

-- 
Mathematics is the supreme nostalgia of our time.