Re: don't let mmap allocate down to zero

[linux-os]

On Wed, 26 Jan 2005, Olivier Galibert wrote:

> On Wed, Jan 26, 2005 at 11:38:15AM -0500, linux-os wrote:
> > On Wed, 26 Jan 2005, Rik van Riel wrote:
> >
> > > With some programs the 2.6 kernel can end up allocating memory
> > > at address zero, for a non-MAP_FIXED mmap call!  This causes
> > > problems with some programs and is generally rude to do. This
> > > simple patch fixes the problem in my tests.
> >
> > Does this mean that we can't mmap the screen regen buffer at
> > 0x000b8000 anymore?
>
> No.  Missed the "non-MAP_FIXED" part?  You can always map at 0, you
> just have to ask for it.

Okay.

> > What 'C' standard do you refer to?
>
> Malloc uses mmap to get more memory.  Malloc returning 0 means no
> memory, not "the memory happens to be at 0".  Not that easy to fix in
> the glibc if you want to keep the "segfault on null pointer accesses"
> debugging help too.

malloc is a runtime library. It has its own documented rules.

> Given that the man page itself says that unless you're using MAP_FIXED
> start is only a hint and you should use 0 if you don't care things can
> get real annoying real fast.  Imagine if you want to mmap a <4K file
> and mmap then returns 0, i.e. NULL, as the mapping address as you
> asked.  It's illegal from the point of view of susv3[1] and it's real
> annoying in a C/C++ program.

mmap() can (will) return 0 if you use 0 as the hint and use MAP_FIXED
at 0. That's the reason why one does NOT check for NULL with mmap() but
for MAP_FAILED (which on this system is (void *)-1.

>  OG.
>
> [1]
>  When MAP_FIXED is not set, the implementation uses addr in an
>  implementation-defined manner to arrive at pa. The pa so chosen
>  shall be an area of the address space that the implementation deems
>  suitable for a mapping of len bytes to the file. All implementations
>  interpret an addr value of 0 as granting the implementation complete
>  freedom in selecting pa, subject to constraints described below. A
>  non-zero value of addr is taken to be a suggestion of a process
>  address near which the mapping should be placed. When the
>  implementation selects a value for pa, it never places a mapping at
>  address 0, nor does it replace any extant mapping.



Cheers,
Dick Johnson
Penguin : Linux version 2.6.10 on an i686 machine (5537.79 BogoMips).
 Notice : All mail here is now cached for review by Dictator Bush.
                 98.36% of all statistics are fiction.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/