Bernd Schmidt <[EMAIL PROTECTED]> wrote:
> 1. munmap can now unmap subparts of previously allocated blocks. This
>makes behaviour more consistent with mmu Linux, and allows us to
>simplify and speed up the uClibc malloc implementation.
There's a problem with your alteration to do_munmap()
Bryan Wu wrote:
On Wed, 2007-06-20 at 12:00 +0900, Paul Mundt wrote:
On Fri, Jun 08, 2007 at 03:53:49PM +0200, Bernd Schmidt wrote:
diff --git a/mm/nommu.c b/mm/nommu.c
index 2b16b00..7480a95 100644
--- a/mm/nommu.c
+++ b/mm/nommu.c
[snip]
+ /*
+* Must always set the VM_SPLIT_P
Bernd Schmidt <[EMAIL PROTECTED]> wrote:
> Excess pages will be freed if MAP_SPLIT_PAGES is passed to mmap.
How about MAP_TRIM_EXCESS instead of MAP_SPLIT_PAGES? I think it fits the
function better.
David
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a
Bernd Schmidt <[EMAIL PROTECTED]> wrote:
> + MAP_PRIVATE | MAP_ANONYMOUS
> + | MAP_GROWSDOWN | MAP_SPLIT_PAGES, 0);
The OR-operator should end the first line.
> diff --git a/include/asm-blackfin/mman.h b/include/asm-blackf
On Wed, 2007-06-20 at 12:00 +0900, Paul Mundt wrote:
> On Fri, Jun 08, 2007 at 03:53:49PM +0200, Bernd Schmidt wrote:
> > diff --git a/mm/nommu.c b/mm/nommu.c
> > index 2b16b00..7480a95 100644
> > --- a/mm/nommu.c
> > +++ b/mm/nommu.c
> [snip]
> > + /*
> > +* Must always set the VM_SPLIT_PAGE
On Fri, Jun 08, 2007 at 03:53:49PM +0200, Bernd Schmidt wrote:
> diff --git a/mm/nommu.c b/mm/nommu.c
> index 2b16b00..7480a95 100644
> --- a/mm/nommu.c
> +++ b/mm/nommu.c
[snip]
> + /*
> + * Must always set the VM_SPLIT_PAGES flag for single-page allocations,
> + * to avoid trying to
On Tue, 2007-06-19 at 19:26 -0400, Robin Getz wrote:
> On Fri 8 Jun 2007 09:53, Bernd Schmidt pondered:
> > Here's a patch to move nommu mmap/munmap ever so slightly closer to mmu
> > behaviour. The motivation for this is to be able to deselect uClibc's
> > UCLIBC_UCLINUX_BROKEN_MUNMAP config opti
On Fri 8 Jun 2007 09:53, Bernd Schmidt pondered:
> Here's a patch to move nommu mmap/munmap ever so slightly closer to mmu
> behaviour. The motivation for this is to be able to deselect uClibc's
> UCLIBC_UCLINUX_BROKEN_MUNMAP config option, which speeds up malloc a
> fair bit. I'm interested in c
On 6/11/07, Bernd Schmidt <[EMAIL PROTECTED]> wrote:
Mike Frysinger wrote:
> On 6/9/07, Matt Mackall <[EMAIL PROTECTED]> wrote:
>> On Fri, Jun 08, 2007 at 03:53:49PM +0200, Bernd Schmidt wrote:
>> > 2. It is no longer possible to get blocks smaller than a page through
>> >mmap. This behaviou
Mike Frysinger wrote:
> On 6/9/07, Matt Mackall <[EMAIL PROTECTED]> wrote:
>> On Fri, Jun 08, 2007 at 03:53:49PM +0200, Bernd Schmidt wrote:
>> > 2. It is no longer possible to get blocks smaller than a page through
>> >mmap. This behaviour was used by simplemalloc, which is an insane
>> >
On 6/9/07, Matt Mackall <[EMAIL PROTECTED]> wrote:
On Fri, Jun 08, 2007 at 03:53:49PM +0200, Bernd Schmidt wrote:
> 2. It is no longer possible to get blocks smaller than a page through
>mmap. This behaviour was used by simplemalloc, which is an insane
>way of implementing malloc on nomm
On Sat 9 Jun 2007 15:10, Matt Mackall pondered:
> On Fri, Jun 08, 2007 at 03:53:49PM +0200, Bernd Schmidt wrote:
> > 2. It is no longer possible to get blocks smaller than a page through
> >mmap. This behaviour was used by simplemalloc, which is an insane
> >way of implementing malloc on n
On Fri, Jun 08, 2007 at 03:53:49PM +0200, Bernd Schmidt wrote:
> 2. It is no longer possible to get blocks smaller than a page through
>mmap. This behaviour was used by simplemalloc, which is an insane
>way of implementing malloc on nommu systems and hopefully not used
>by anyone anymo
Here's a patch to move nommu mmap/munmap ever so slightly closer to mmu
behaviour. The motivation for this is to be able to deselect uClibc's
UCLIBC_UCLINUX_BROKEN_MUNMAP config option, which speeds up malloc a
fair bit. I'm interested in comments whether this is a good direction
to go. The patc
14 matches
Mail list logo
