| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Previously the heap coalesce function looked at the "free" flag
on the next block to determine if two blocks could be merged
together. The coalesce function is ran with all cores synchronized
so no one should be modifying the heap, but there is still a
subtle race condition.
A user task could be in the process of freeing a block, marking it
"free" but not actually inserting it onto the queues yet. The
coalesce will combine this block together but then when the user
task resumes it continues to try to insert it onto the queue.
This causes both the larger merged block and the smaller block to
be on a free queue and memory corruption when both blocks are used.
This is resolved by having a new flag indicating that the block
is known to be part of the "coalesce" group and using this in
addition to the "free" flag. Only blocks which are actually present
on the free queues are considered to be part of the "coalesce"
group.
Change-Id: If33d15d731d32e07e01104244ebc65daf2295878
Reviewed-on: http://gfw160.austin.ibm.com:8080/gerrit/4520
Reviewed-by: Douglas R. Gilbert <dgilbert@us.ibm.com>
Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
Tested-by: Jenkins Server
Reviewed-by: A. Patrick Williams III <iawillia@us.ibm.com>
|
| |
|
|
|
|
|
|
| |
Change-Id: I25ec156fe3c9088d2de2254017407dbadd2fac31
Reviewed-on: http://gfw160.austin.ibm.com:8080/gerrit/4255
Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
Tested-by: Jenkins Server
Reviewed-by: A. Patrick Williams III <iawillia@us.ibm.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The coalesce was causing time issues in VPO.
Reduce from an O(n^2) to O(n) algorithm.
Results from one particular execution:
- Old - 91649004 cycles for 7471 chunks
- New - 02668146 cycles for 7676 chunks
<3% cycle time of original algorithm.
Change-Id: I7145c6b430dccdb3f08d186a1ee5ea2f86aa3f81
Reviewed-on: http://gfw160.austin.ibm.com:8080/gerrit/2942
Tested-by: Jenkins Server
Reviewed-by: Van H. Lee <vanlee@us.ibm.com>
Reviewed-by: Douglas R. Gilbert <dgilbert@us.ibm.com>
Reviewed-by: Thi N. Tran <thi@us.ibm.com>
Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
Reviewed-by: A. Patrick Williams III <iawillia@us.ibm.com>
|
| |
|
|
|
|
|
| |
Change-Id: Iaac392b9f4287ba888e454532c4061d6a14c6e5c
Reviewed-on: http://gfw160.austin.ibm.com:8080/gerrit/593
Tested-by: Jenkins Server
Reviewed-by: Douglas R. Gilbert <dgilbert@us.ibm.com>
|
| |
|
|
|
|
|
|
| |
Change-Id: Iada5e8b69c7919d2b59febd861450abeb7c45287
Reviewed-on: http://gfw160.austin.ibm.com:8080/gerrit/451
Tested-by: Jenkins Server
Reviewed-by: MATTHEW S. BARTH <msbarth@us.ibm.com>
Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
|
| |
|
|
|
|
|
|
| |
Change-Id: Ibe725a43e6366d9113ec99df1cc6aafa7bbb770e
Reviewed-on: http://gfw160.austin.ibm.com:8080/gerrit/431
Tested-by: Jenkins Server
Reviewed-by: A. Patrick Williams III <iawillia@us.ibm.com>
Reviewed-by: Douglas R. Gilbert <dgilbert@us.ibm.com>
|
| |
|
|
|
|
|
|
|
|
|
|
| |
redundancies and also to have a single place to update the memory
map if needed.
See Task 3507.
Change-Id: I8f2d632983abe6d6798784e975cd93057018594b
Reviewed-on: http://gfw160.austin.ibm.com:8080/gerrit/330
Tested-by: Jenkins Server
Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
|
| |
|
|
|
|
|
| |
Change-Id: I205f2409e56032cfc0aaf01d7e26d357f0b86373
Reviewed-on: http://gfw160.austin.ibm.com:8080/gerrit/277
Reviewed-by: A. Patrick Williams III <iawillia@us.ibm.com>
Tested-by: Jenkins Server
|
| |
|
|
|
|
|
|
|
|
|
| |
- Segment Manager
- Base / Device Segments
- Block for Base image.
Change-Id: Ic0c058e5c5b210ec1c48d30f6ed9f9837d74a3c8
Reviewed-on: http://gfw160.austin.ibm.com:8080/gerrit/193
Tested-by: Jenkins Server
Reviewed-by: MATTHEW S. BARTH <msbarth@us.ibm.com>
|
| | |
|
| | |
|
| |
|
