diff options
author | Vitaly Kuznetsov <vkuznets@redhat.com> | 2015-03-18 12:29:23 -0700 |
---|---|---|
committer | Greg Kroah-Hartman <gregkh@linuxfoundation.org> | 2015-03-25 11:53:54 +0100 |
commit | 5abbbb75d733793b1637441691de95b7665cee85 (patch) | |
tree | d63dd5bba8ebd7b3102fb77affe60a952b953c1b /drivers/hv | |
parent | f3f6eb80872becdbce07f7d2670d3f8ae5119752 (diff) |
Drivers: hv: hv_balloon: don't lose memory when onlining order is not natural
Memory blocks can be onlined in random order. When this order is not natural
some memory pages are not onlined because of the redundant check in
hv_online_page().
Here is a real world scenario:
1) Host tries to hot-add the following (process_hot_add):
pg_start=rg_start=0x48000, pfn_cnt=111616, rg_size=262144
2) This results in adding 4 memory blocks:
[ 109.057866] init_memory_mapping: [mem 0x48000000-0x4fffffff]
[ 114.102698] init_memory_mapping: [mem 0x50000000-0x57ffffff]
[ 119.168039] init_memory_mapping: [mem 0x58000000-0x5fffffff]
[ 124.233053] init_memory_mapping: [mem 0x60000000-0x67ffffff]
The last one is incomplete but we have special has->covered_end_pfn counter to
avoid onlining non-backed frames and hv_bring_pgs_online() function to bring
them online later on.
3) Now we have 4 offline memory blocks: /sys/devices/system/memory/memory9-12
$ for f in /sys/devices/system/memory/memory*/state; do echo $f `cat $f`; done | grep -v onlin
/sys/devices/system/memory/memory10/state offline
/sys/devices/system/memory/memory11/state offline
/sys/devices/system/memory/memory12/state offline
/sys/devices/system/memory/memory9/state offline
4) We bring them online in non-natural order:
$grep MemTotal /proc/meminfo
MemTotal: 966348 kB
$echo online > /sys/devices/system/memory/memory12/state && grep MemTotal /proc/meminfo
MemTotal: 1019596 kB
$echo online > /sys/devices/system/memory/memory11/state && grep MemTotal /proc/meminfo
MemTotal: 1150668 kB
$echo online > /sys/devices/system/memory/memory9/state && grep MemTotal /proc/meminfo
MemTotal: 1150668 kB
As you can see memory9 block gives us zero additional memory. We can also
observe a huge discrepancy between host- and guest-reported memory sizes.
The root cause of the issue is the redundant pg >= covered_start_pfn check (and
covered_start_pfn advancing) in hv_online_page(). When upper memory block in
being onlined before the lower one (memory12 and memory11 in the above case) we
advance the covered_start_pfn pointer and all memory9 pages do not pass the
check. If the assumption that host always gives us requests in sequential order
and pg_start always equals rg_start when the first request for the new HA
region is received (that's the case in my testing) is correct than we can get
rid of covered_start_pfn and pg >= start_pfn check in hv_online_page() is
sufficient.
The current char-next branch is broken and this patch fixes
the bug.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Diffstat (limited to 'drivers/hv')
-rw-r--r-- | drivers/hv/hv_balloon.c | 14 |
1 files changed, 4 insertions, 10 deletions
diff --git a/drivers/hv/hv_balloon.c b/drivers/hv/hv_balloon.c index f1f17c5ad8d8..014256a2ef88 100644 --- a/drivers/hv/hv_balloon.c +++ b/drivers/hv/hv_balloon.c @@ -428,14 +428,13 @@ struct dm_info_msg { * currently hot added. We hot add in multiples of 128M * chunks; it is possible that we may not be able to bring * online all the pages in the region. The range - * covered_start_pfn : covered_end_pfn defines the pages that can + * covered_end_pfn defines the pages that can * be brough online. */ struct hv_hotadd_state { struct list_head list; unsigned long start_pfn; - unsigned long covered_start_pfn; unsigned long covered_end_pfn; unsigned long ha_end_pfn; unsigned long end_pfn; @@ -679,8 +678,7 @@ static void hv_online_page(struct page *pg) list_for_each(cur, &dm_device.ha_region_list) { has = list_entry(cur, struct hv_hotadd_state, list); - cur_start_pgp = (unsigned long) - pfn_to_page(has->covered_start_pfn); + cur_start_pgp = (unsigned long)pfn_to_page(has->start_pfn); cur_end_pgp = (unsigned long)pfn_to_page(has->covered_end_pfn); if (((unsigned long)pg >= cur_start_pgp) && @@ -692,7 +690,6 @@ static void hv_online_page(struct page *pg) __online_page_set_limits(pg); __online_page_increment_counters(pg); __online_page_free(pg); - has->covered_start_pfn++; } } } @@ -736,10 +733,9 @@ static bool pfn_covered(unsigned long start_pfn, unsigned long pfn_cnt) * is, update it. */ - if (has->covered_end_pfn != start_pfn) { + if (has->covered_end_pfn != start_pfn) has->covered_end_pfn = start_pfn; - has->covered_start_pfn = start_pfn; - } + return true; } @@ -784,7 +780,6 @@ static unsigned long handle_pg_range(unsigned long pg_start, pgs_ol = pfn_cnt; hv_bring_pgs_online(start_pfn, pgs_ol); has->covered_end_pfn += pgs_ol; - has->covered_start_pfn += pgs_ol; pfn_cnt -= pgs_ol; } @@ -845,7 +840,6 @@ static unsigned long process_hot_add(unsigned long pg_start, list_add_tail(&ha_region->list, &dm_device.ha_region_list); ha_region->start_pfn = rg_start; ha_region->ha_end_pfn = rg_start; - ha_region->covered_start_pfn = pg_start; ha_region->covered_end_pfn = pg_start; ha_region->end_pfn = rg_start + rg_size; } |