Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

osd: bluestore: more fixes #7130

Merged
merged 32 commits into from Jan 11, 2016
Merged

osd: bluestore: more fixes #7130

merged 32 commits into from Jan 11, 2016

Conversation

liewegas
Copy link
Member

@liewegas liewegas commented Jan 6, 2016

No description provided.

@@ -464,6 +487,37 @@ int BlockDevice::read(uint64_t off, uint64_t len, bufferlist *pbl,
return r < 0 ? r : 0;
}

int BlockDevice::read_buffered(uint64_t off, uint64_t len, char *buf)
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

what's the difference with BlockDevice::read(off, len, pbl, NULL, true)?

Copy link
Member Author

@liewegas liewegas Jan 7, 2016 via email

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@liewegas liewegas force-pushed the wip-bluestore branch 2 times, most recently from 2763dde to 9fde446 Compare January 8, 2016 14:44
Signed-off-by: Sage Weil <sage@redhat.com>
Signed-off-by: Sage Weil <sage@redhat.com>
Signed-off-by: Sage Weil <sage@redhat.com>
Doing it in the aio thread ensures that there isn't a use-after-free.
Note that this is only bluefs written files.

Signed-off-by: Sage Weil <sage@redhat.com>
Signed-off-by: Sage Weil <sage@redhat.com>
Signed-off-by: Sage Weil <sage@redhat.com>
Signed-off-by: Sage Weil <sage@redhat.com>
Signed-off-by: Sage Weil <sage@redhat.com>
Signed-off-by: Sage Weil <sage@redhat.com>
Wait on fsync and _flush_log only.

Signed-off-by: Sage Weil <sage@redhat.com>
Signed-off-by: Sage Weil <sage@redhat.com>
Signed-off-by: Sage Weil <sage@redhat.com>
Signed-off-by: Sage Weil <sage@redhat.com>
fdatasync(2) does a cache flush on the device, which we want to avoid if
we didn't actually write anything.

Signed-off-by: Sage Weil <sage@redhat.com>
Signed-off-by: Sage Weil <sage@redhat.com>
Signed-off-by: Sage Weil <sage@redhat.com>
Signed-off-by: Sage Weil <sage@redhat.com>
Signed-off-by: Sage Weil <sage@redhat.com>
Signed-off-by: Sage Weil <sage@redhat.com>
Signed-off-by: Sage Weil <sage@redhat.com>
Signed-off-by: Sage Weil <sage@redhat.com>
Signed-off-by: Sage Weil <sage@redhat.com>
These are on my todo list elsewhere.

Signed-off-by: Sage Weil <sage@redhat.com>
Signed-off-by: Sage Weil <sage@redhat.com>
The object may have trailing zeros.

Signed-off-by: Sage Weil <sage@redhat.com>
Signed-off-by: Sage Weil <sage@redhat.com>
Signed-off-by: Sage Weil <sage@redhat.com>
Signed-off-by: Sage Weil <sage@redhat.com>
If we're past EOF, we need to zero the rest of the block.

Signed-off-by: Sage Weil <sage@redhat.com>
Prefill the objectstore by fragmenting the freespace.  Max free allocations
are bluestore_debug_prefragment_max bytes (default 1M).  Prefill defaults
to 0, of course.  One would normally just set this to .8 to get something
resembling an aged disk.

Signed-off-by: Sage Weil <sage@redhat.com>
- drop the MIN cruft
- unreserve what we didn't get

Signed-off-by: Sage Weil <sage@redhat.com>
liewegas added a commit that referenced this pull request Jan 11, 2016
@liewegas liewegas merged commit 9e9603e into ceph:master Jan 11, 2016
@liewegas liewegas deleted the wip-bluestore branch January 11, 2016 13:59
@ghost ghost changed the title bluestore: more fixes osd: bluestore: more fixes Feb 10, 2016
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
2 participants