GitHub user PepperJo opened a pull request:
https://github.com/apache/incubator-crail/pull/23
Stagingbuffercache fixes
See commit messages for detail explanation.
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/PepperJo/incubator-crail stagingbuffercachefix
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/incubator-crail/pull/23.patch
To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:
This closes #23
----
commit 671b63554658925e78d1a8c8a11c706ffee8924d
Author: Jonas Pfefferle <pepperjo@...>
Date: 2018-05-02T10:53:47Z
NVMf: staging buffer cache existing entry bug
A buffer might already exist at a certain remote address (resp. LBA) if
a file was deleted and a block was reassigned to a different file. Do not
throw an exception but free the buffer. If we can't free it something
went wrong as there should never be two writers writing to the same
remote address.
https://issues.apache.org/jira/projects/CRAIL/issues/CRAIL-30
Signed-off-by: Jonas Pfefferle <[email protected]>
commit 524ead0f74514e0496c6ecf06a24d88ab334d6e4
Author: Jonas Pfefferle <pepperjo@...>
Date: 2018-05-03T11:38:00Z
NVMf: staging buffer cache clear buffer
Clear buffer position and limit when getting the buffer of the
free queue since position and limit might have been modified before.
https://issues.apache.org/jira/browse/CRAIL-31
Signed-off-by: Jonas Pfefferle <[email protected]>
----
---
