https://github.com/torvalds/linux
Revision 3ad33b2436b545cbe8b28e53f3710432cad457ab authored by Lee Schermerhorn on 15 November 2007, 00:59:10 UTC, committed by Linus Torvalds on 15 November 2007, 02:45:38 UTC
We hit the BUG_ON() in mm/rmap.c:vma_address() when trying to migrate via
mbind(MPOL_MF_MOVE) a non-anon region that spans multiple vmas.  For
anon-regions, we just fail to migrate any pages beyond the 1st vma in the
range.

This occurs because do_mbind() collects a list of pages to migrate by
calling check_range().  check_range() walks the task's mm, spanning vmas as
necessary, to collect the migratable pages into a list.  Then, do_mbind()
calls migrate_pages() passing the list of pages, a function to allocate new
pages based on vma policy [new_vma_page()], and a pointer to the first vma
of the range.

For each page in the list, new_vma_page() calls page_address_in_vma()
passing the page and the vma [first in range] to obtain the address to get
for alloc_page_vma().  The page address is needed to get interleaving
policy correct.  If the pages in the list come from multiple vmas,
eventually, new_page_address() will pass that page to page_address_in_vma()
with the incorrect vma.  For !PageAnon pages, this will result in a bug
check in rmap.c:vma_address().  For anon pages, vma_address() will just
return EFAULT and fail the migration.

This patch modifies new_vma_page() to check the return value from
page_address_in_vma().  If the return value is EFAULT, new_vma_page()
searchs forward via vm_next for the vma that maps the page--i.e., that does
not return EFAULT.  This assumes that the pages in the list handed to
migrate_pages() is in address order.  This is currently case.  The patch
documents this assumption in a new comment block for new_vma_page().

If new_vma_page() cannot locate the vma mapping the page in a forward
search in the mm, it will pass a NULL vma to alloc_page_vma().  This will
result in the allocation using the task policy, if any, else system default
policy.  This situation is unlikely, but the patch documents this behavior
with a comment.

Note, this patch results in restarting from the first vma in a multi-vma
range each time new_vma_page() is called.  If this is not acceptable, we
can make the vma argument a pointer, both in new_vma_page() and it's caller
unmap_and_move() so that the value held by the loop in migrate_pages()
always passes down the last vma in which a page was found.  This will
require changes to all new_page_t functions passed to migrate_pages().  Is
this necessary?

For this patch to work, we can't bug check in vma_address() for pages
outside the argument vma.  This patch removes the BUG_ON().  All other
callers [besides new_vma_page()] already check the return status.

Tested on x86_64, 4 node NUMA platform.

Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Acked-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
1 parent e1a1c99
Raw File
Tip revision: 3ad33b2436b545cbe8b28e53f3710432cad457ab authored by Lee Schermerhorn on 15 November 2007, 00:59:10 UTC
Migration: find correct vma in new_vma_page()
Tip revision: 3ad33b2
ver_linux
#!/bin/sh
# Before running this script please ensure that your PATH is
# typical as you use for compilation/istallation. I use
# /bin /sbin /usr/bin /usr/sbin /usr/local/bin, but it may
# differ on your system.
#
PATH=/sbin:/usr/sbin:/bin:/usr/bin:/usr/local/sbin:$PATH
echo 'If some fields are empty or look unusual you may have an old version.'
echo 'Compare to the current minimal requirements in Documentation/Changes.'
echo ' '

uname -a
echo ' '

gcc --version 2>&1| head -n 1 | grep -v gcc | awk \
'NR==1{print "Gnu C                 ", $1}'

gcc --version 2>&1| grep gcc | awk \
'NR==1{print "Gnu C                 ", $3}'

make --version 2>&1 | awk -F, '{print $1}' | awk \
      '/GNU Make/{print "Gnu make              ",$NF}'

echo "binutils               $(ld -v | egrep -o '[0-9]+\.[0-9\.]+')"

echo -n "util-linux             "
fdformat --version | awk '{print $NF}' | sed -e s/^util-linux-// -e s/\)$//

echo -n "mount                  "
mount --version | awk '{print $NF}' | sed -e s/^mount-// -e s/\)$//

depmod -V  2>&1 | awk 'NR==1 {print "module-init-tools     ",$NF}'

tune2fs 2>&1 | grep "^tune2fs" | sed 's/,//' |  awk \
'NR==1 {print "e2fsprogs             ", $2}'

fsck.jfs -V 2>&1 | grep version | sed 's/,//' |  awk \
'NR==1 {print "jfsutils              ", $3}'

reiserfsck -V 2>&1 | grep ^reiserfsck | awk \
'NR==1{print "reiserfsprogs         ", $2}'

fsck.reiser4 -V 2>&1 | grep ^fsck.reiser4 | awk \
'NR==1{print "reiser4progs          ", $2}'

xfs_db -V 2>&1 | grep version | awk \
'NR==1{print "xfsprogs              ", $3}'

pccardctl -V 2>&1| grep pcmciautils | awk '{print "pcmciautils           ", $2}'

cardmgr -V 2>&1| grep version | awk \
'NR==1{print "pcmcia-cs             ", $3}'

quota -V 2>&1 | grep version | awk \
'NR==1{print "quota-tools           ", $NF}'

pppd --version 2>&1| grep version | awk \
'NR==1{print "PPP                   ", $3}'

isdnctrl 2>&1 | grep version | awk \
'NR==1{print "isdn4k-utils          ", $NF}'

showmount --version 2>&1 | grep nfs-utils | awk \
'NR==1{print "nfs-utils             ", $NF}'

echo -n "Linux C Library        "
sed -n -e '/^.*\/libc-\([^/]*\)\.so$/{s//\1/;p;q}' < /proc/self/maps

ldd -v > /dev/null 2>&1 && ldd -v || ldd --version |head -n 1 | awk \
'NR==1{print "Dynamic linker (ldd)  ", $NF}'

ls -l /usr/lib/lib{g,stdc}++.so  2>/dev/null | awk -F. \
       '{print "Linux C++ Library      " $4"."$5"."$6}'

ps --version 2>&1 | grep version | awk \
'NR==1{print "Procps                ", $NF}'

ifconfig --version 2>&1 | grep tools | awk \
'NR==1{print "Net-tools             ", $NF}'

# Kbd needs 'loadkeys -h',
loadkeys -h 2>&1 | awk \
'(NR==1 && ($3 !~ /option/)) {print "Kbd                   ", $3}'

# while console-tools needs 'loadkeys -V'.
loadkeys -V 2>&1 | awk \
'(NR==1 && ($2 ~ /console-tools/)) {print "Console-tools         ", $3}'

oprofiled --version 2>&1 | awk \
'(NR==1 && ($2 == "oprofile")) {print "oprofile              ", $3}'

expr --v 2>&1 | awk 'NR==1{print "Sh-utils              ", $NF}'

udevinfo -V 2>&1 | grep version | awk '{print "udev                  ", $3}'

iwconfig --version 2>&1 | awk \
'(NR==1 && ($3 == "version")) {print "wireless-tools        ",$4}'

if [ -e /proc/modules ]; then
    X=`cat /proc/modules | sed -e "s/ .*$//"`
    echo "Modules Loaded         "$X
fi
back to top