Whamcloud - gitweb
LU-11825 clio: Remove pio feature 60/34460/5
authorPatrick Farrell <paf@cray.com>
Sun, 23 Dec 2018 21:04:08 +0000 (16:04 -0500)
committerOleg Drokin <green@whamcloud.com>
Mon, 8 Apr 2019 06:32:03 +0000 (06:32 +0000)
commit81fe463c5ac1e8d5b51b680c1ddcd490bfc3e623
tree4345de1980f8daf778a8e65c9fc7189893f5c303
parentc4eed4899ac799d6f7ffd3504add9d80eca71990
LU-11825 clio: Remove pio feature

The pio feature was an interesting concept, but never
achieved real use in production, primarily because it was
actually slower than normal i/o for some common workloads,
so it could not be on by default.

It has had a trivial deadlock in it (with truncate) since
release, and this has never been reported by a customer.

The attemp to rewrite readahead to use ptasks was never
fully successful, and is being superseded by a different
implementation.

Given all this, there is no reason to keep the extra
complexity in the already complicated clio code.

Lustre-change: https://review.whamcloud.com/33912
Lustre-commit: 2b0a34fe43bf4fce5560af61a45e5393c96070a9

Signed-off-by: Patrick Farrell <paf@cray.com>
Change-Id: Ie558db946af94189e35e985e8894706c14ed5f39
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Wang Shilong <wshilong@whamcloud.com>
Signed-off-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/34460
Tested-by: Jenkins
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Patrick Farrell <pfarrell@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
14 files changed:
lustre/include/cl_object.h
lustre/include/obd_support.h
lustre/llite/file.c
lustre/llite/llite_internal.h
lustre/llite/lproc_llite.c
lustre/llite/rw26.c
lustre/llite/vvp_internal.h
lustre/llite/vvp_io.c
lustre/lov/lov_io.c
lustre/obdclass/cl_io.c
lustre/obdclass/cl_object.c
lustre/osc/osc_io.c
lustre/osc/osc_lock.c
lustre/tests/sanity.sh