On 01.02.2007 [18:24:39 -0800], Nishanth Aravamudan wrote:
> On 01.02.2007 [18:20:22 -0800], Nishanth Aravamudan wrote:
> > On 01.02.2007 [09:43:01 -0600], Adam Litke wrote:
> > > On Wed, 2007-01-31 at 20:45 -0800, Nishanth Aravamudan wrote:
> <snip>
> > > > + for (i = 0; i < num; i++) {
> > > > + if (seg[i].prot & PROT_WRITE) {
> > > > + for (p = seg[i].vaddr;
> > > > + p <= seg[i].vaddr + seg[i].filesz;
> > > I just realized the above should be (vaddr + filesz + extracopysize).
> > > Otherwise we will throw away the initialized bss data that we've been so
> > > careful to collect earlier.
> >
> > This does complicate things quite a bit, unfortunately, we currently
> > throw away the extracopy information once we've done it. I've modified
> > the code a bit to save this info, and I think I've got it right, but I'd
> > appreciate you and Steve taking a look at the following two patches,
> > which should allow for what you want.
>
> commit 19d3d34954944e84d009cd3282c0abf45a69bdea
> Author: Nishanth Aravamudan <[EMAIL PROTECTED]>
> Date: Thu Feb 1 17:51:47 2007 -0800
>
> elflink: store extracopy information
>
> Add a few fields to the seg_info structure to store the starting address
> and size of the extracopy area for use by the next patch to drop
> pagecache pages. If the extra_vaddr field is NULL, then the pagecache
> dropping code assumes this means there was no extracopy area.
Another note, here, I've got a much more invasive version of this patch
which is mostly just a cleanup once we've decided if we're going to
store extra fields in the seg_info struct. It cleans up do_extracopy to
just store into those fields directly and makes a bit of that code
easier to follow, I think. Maybe it's better to save extra_start and
extra_end in that case (rather than extra_start and an extra_size)? Any
feedback would be appreciated.
Thanks,
Nish
--
Nishanth Aravamudan <[EMAIL PROTECTED]>
IBM Linux Technology Center
-------------------------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier.
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Libhugetlbfs-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/libhugetlbfs-devel
