[LU-8166] sanity test 101g assumes 4k page size Created: 19/May/16 Updated: 27/Jun/18 Resolved: 27/Jun/18 |
|
| Status: | Closed |
| Project: | Lustre |
| Component/s: | None |
| Affects Version/s: | None |
| Fix Version/s: | None |
| Type: | Bug | Priority: | Minor |
| Reporter: | Oleg Drokin | Assignee: | James Nunez (Inactive) |
| Resolution: | Duplicate | Votes: | 0 |
| Labels: | None | ||
| Issue Links: |
|
||||||||
| Severity: | 3 | ||||||||
| Rank (Obsolete): | 9223372036854775807 | ||||||||
| Description |
|
There's a hardcoded assumption that 16M is 4096 pages which is wrong on some architectures. Andreas advises that we can actually calculate proper pages with: local page_count=$((16 * 1048576 / $(page_size)))
[ "$mp" -eq "$page_count" ] ||
and also in a couple mor espots in that same test. |
| Comments |
| Comment by Oleg Drokin [ 20/May/16 ] |
|
See the discussion here: http://review.whamcloud.com/#/c/20282/2/lustre/tests/sanity.sh,unified |
| Comment by Gerrit Updater [ 29/Jun/16 ] |
|
James Nunez (james.a.nunez@intel.com) uploaded a new patch: http://review.whamcloud.com/21084 |
| Comment by James Nunez (Inactive) [ 27/Jun/18 ] |
|
sanity test 101g page size parameters were generalized with the patch for LU-8139 at https://review.whamcloud.com/25335. |