Re: [uml-devel] [PATCH v2] delete unnecessary bootmem struct page array

[Real Name <enjoymindful@gmail.com>]

On Sat, Jun 14, 2014 at 11:44:04AM +0200, Richard Weinberger wrote:
> Hi!
> 
> Am 03.06.2014 07:30, schrieb Real Name:
> > From: Honggang Li <enjoymindful@gmail.com>
> > 
> > The patch based on linux-next-2014-06-02.
> > 
> > The old init_maps function does two things:
> > 1) allocates and initializes one struct page array for bootmem
> > 2) count the number of total pages
> > 
> > After removed the source code related to the unnecessary array, the name 
> > 'init_maps' is some kind of improper named, as it just count the number of
> > total page numbers. So, I renamed the function as 'mem_total_pages'.
> > 
> > I tested the patch through repeat reboot the uml kernel many times.
> > [real@name linux-next]$ make ARCH=um defconfig
> > [real@name linux-next]$ make ARCH=um linux
> > [real@name linux-next]$ file linux
> > linux: ELF 64-bit LSB executable, x86-64, version 1 (SYSV), dynamically linked (uses shared libs), for GNU/Linux 2.6.32, not stripped
> > [real@name linux-next]$ ./linux ubda=/home/real/linux-next/Fedora20-AMD64-root_fs mem=256m && sync && echo 1
> > [real@name linux-next]$ ./linux ubda=/home/real/linux-next/Fedora20-AMD64-root_fs mem=256m && sync && echo 2
> > (repeat reboot the uml kernel many times..)
> 
> Can you please include in the changelog the commit sha1 which made the old init_maps() obsolete?
> I had a look at the pre-git linux tree, looks like init_maps() wasn't touched for more than 10 years.

hi, richard

what is the pre-git linux tree? I searched it with google, but failed.

The v2.6.12 kernel is the oldest one available from the linux-next git tree. And it has duplicated
struct page arrays. So, any suggestion how to find the commit you wanted?

linux-2.4.20 + uml-patch-2.4.20-8 works on *old* redhat-9 virtual machine. And it seems has duplicated
struct page array too.

The first struct page array
------------------------
linux-2.4.20/arch/um/kernel/physmem.c
157 int init_maps(unsigned long len)
158 {
159         struct page *p, *map;
160         int i, n;
161 
162         n = len >> PAGE_SHIFT;
163         len = n * sizeof(struct page);
164 
165         if(kmalloc_ok){
166                 map = kmalloc(len, GFP_KERNEL);
167                 if(map == NULL) map = vmalloc(len);
168         }
169         else map = alloc_bootmem_low_pages(len);
170 
171         if(map == NULL)
172                 return(-ENOMEM);
173 
174         for(i = 0; i < n; i++){
175                 p = &map[i];
176                 set_page_count(p, 0);
177                 SetPageReserved(p);
178                 INIT_LIST_HEAD(&p->list);
179         }
180 
181         mem_map = map;
182         max_mapnr = n;
183         return(0);
184 }


The second struct page array
-----------------------
mm/memory.c
73 mem_map_t * mem_map; // global define

mm/page_alloc.c
839 void __init free_area_init(unsigned long *zones_size)
840 {
841         free_area_init_core(0, &contig_page_data, &mem_map, zones_size, 0, 0, 0);
842 }

--------
mm/page_alloc.c
685 void __init free_area_init_core(int nid, pg_data_t *pgdat, struct page **gmap,
686         unsigned long *zones_size, unsigned long zone_start_paddr,
687         unsigned long *zholes_size, struct page *lmem_map)
688 {
........
716         map_size = (totalpages + 1)*sizeof(struct page);
717         if (lmem_map == (struct page *)0) {
718                 lmem_map = (struct page *) alloc_bootmem_node(pgdat, map_size);
719                 lmem_map = (struct page *)(PAGE_OFFSET +
720                         MAP_ALIGN((unsigned long)lmem_map - PAGE_OFFSET));
721         }
722         *gmap = pgdat->node_mem_map = lmem_map;


> 
> Thanks,
> //richard
> 
> > Honggang Li (1):
> >   delete unnecessary bootmem struct page array
> > 
> >  arch/um/include/shared/mem_user.h |  2 +-
> >  arch/um/kernel/physmem.c          | 32 ++++++--------------------------
> >  arch/um/kernel/um_arch.c          |  7 +------
> >  3 files changed, 8 insertions(+), 33 deletions(-)
> > 

------------------------------------------------------------------------------
HPCC Systems Open Source Big Data Platform from LexisNexis Risk Solutions
Find What Matters Most in Your Big Data with HPCC Systems
Open Source. Fast. Scalable. Simple. Ideal for Dirty Data.
Leverages Graph Analysis for Fast Processing & Easy Data Exploration
http://p.sf.net/sfu/hpccsystems
_______________________________________________
User-mode-linux-devel mailing list
User-mode-linux-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel