Hi,
On 9.10.24 16:13, Andrey Zhadchenko wrote:
On 9/30/24 19:32, Alexander Atanasov wrote:
md_page is always present in memory. In that case
md_page->page could be mapped once and we would not need to perform
kmap_atomic/kunmap_atomic for each access.
https://virtuozzo.atlassian.net/browse/VSTOR-91659
Suggested-by: Denis V. Lunev <d...@openvz.org>
Signed-off-by: Alexander Atanasov <alexander.atana...@virtuozzo.com>
---
drivers/md/dm-ploop-bat.c | 18 +++++++-----------
drivers/md/dm-ploop-cmd.c | 30 +++++++++++-------------------
drivers/md/dm-ploop-map.c | 35 +++++++++++++----------------------
drivers/md/dm-ploop.h | 14 +++++++-------
4 files changed, 38 insertions(+), 59 deletions(-)
Do we still need kmap/kunmap in the following places?
1. ploop_delta_check_header()
I think it was already mapped during
ploop_read_delta_metadata()->ploop_prealloc_md_pages(md_root, 0,
1)->ploop_alloc_md_page()
2. ploop_resize() - should also be mapped already
No, we do not. I will update and resend.
_______________________________________________
Devel mailing list
Devel@openvz.org
https://lists.openvz.org/mailman/listinfo/devel
