#
c14aafed |
| 18-Mar-2015 |
Navdeep Parhar <np@FreeBSD.org> |
Merge r278538 through r280226.
|
#
59fa1525 |
| 17-Mar-2015 |
Baptiste Daroussin <bapt@FreeBSD.org> |
Merge from head
|
#
ad7ea69d |
| 16-Mar-2015 |
Nathan Whitehorn <nwhitehorn@FreeBSD.org> |
Convert PTE eviction lock from an RW lock to a RM lock. It is held for writing approximately never (< 0.00000001% under heavy VM load, and it can go for months without ever being acquired in normal o
Convert PTE eviction lock from an RW lock to a RM lock. It is held for writing approximately never (< 0.00000001% under heavy VM load, and it can go for months without ever being acquired in normal operation). This provides a 10% (2-minute) improvement in wall clock time for make -j32 buildworld on a 4-core 32-thread POWER8.
show more ...
|
#
7426d572 |
| 15-Mar-2015 |
Baptiste Daroussin <bapt@FreeBSD.org> |
Merge from head
|
#
e0125cfd |
| 14-Mar-2015 |
Dimitry Andric <dim@FreeBSD.org> |
Merge ^/head r279893 through r279984.
|
#
1cd30eb6 |
| 13-Mar-2015 |
Nathan Whitehorn <nwhitehorn@FreeBSD.org> |
Deallocate any leftover page table entries in the LPAR at boot. This prevents contamination from a previous kernel (e.g. after shutdown -r).
|
#
fa1e92b6 |
| 04-Mar-2015 |
Baptiste Daroussin <bapt@FreeBSD.org> |
Merge from head
|
#
072aeeb6 |
| 02-Mar-2015 |
Navdeep Parhar <np@FreeBSD.org> |
Merge r278538 through r279514.
|
#
0d56a8cb |
| 26-Feb-2015 |
Dimitry Andric <dim@FreeBSD.org> |
Merge ^/head r279163 through r279308.
|
#
827cc9b9 |
| 24-Feb-2015 |
Nathan Whitehorn <nwhitehorn@FreeBSD.org> |
New pmap implementation for 64-bit PowerPC processors. The main focus of this change is to improve concurrency: - Drop global state stored in the shadow overflow page table (and all other global st
New pmap implementation for 64-bit PowerPC processors. The main focus of this change is to improve concurrency: - Drop global state stored in the shadow overflow page table (and all other global state) - Remove all global locks - Use per-PTE lock bits to allow parallel page insertion - Reconstruct state when requested for evicted PTEs instead of buffering it during overflow
This drops total wall time for make buildworld on a 32-thread POWER8 system by a factor of two and system time by a factor of three, providing performance 20% better than similarly clocked Core i7 Xeons per-core. Performance on smaller SMP systems, where PMAP lock contention was not as much of an issue, is nearly unchanged.
Tested on: POWER8, POWER5+, G5 UP, G5 SMP (64-bit and 32-bit kernels) Merged from: user/nwhitehorn/ppc64-pmap-rework Looked over by: jhibbits, andreast MFC after: 3 months Relnotes: yes Sponsored by: FreeBSD Foundation
show more ...
|
#
d8fee543 |
| 12-Feb-2015 |
Baptiste Daroussin <bapt@FreeBSD.org> |
Merge from HEAD
|
#
53f2fbca |
| 11-Feb-2015 |
Glen Barber <gjb@FreeBSD.org> |
MFH: r278202,r278205-r278590
Sponsored by: The FreeBSD Foundation
|
#
93466fc6 |
| 10-Feb-2015 |
Dimitry Andric <dim@FreeBSD.org> |
Merge ^/head r278351 through r278498.
|
#
3b336798 |
| 09-Feb-2015 |
Nathan Whitehorn <nwhitehorn@FreeBSD.org> |
Mark invalid page table entries correctly for PMAP as well as for the hypervisor. This prevents an infinite loop where processes with evicted pages would page fault forever when PMAP decided the evic
Mark invalid page table entries correctly for PMAP as well as for the hypervisor. This prevents an infinite loop where processes with evicted pages would page fault forever when PMAP decided the evicted pages on which the process was faulting was actually present and did not need to be restored.
Found while building LLVM with make -j32.
Sponsored by: FreeBSD Foundation
show more ...
|
#
07e765cd |
| 09-Feb-2015 |
Nathan Whitehorn <nwhitehorn@FreeBSD.org> |
Fix typo in PTE insertion overflow handling: use the page we're actually returning, not the one we just looked at.
|
#
a840d2f5 |
| 08-Feb-2015 |
Nathan Whitehorn <nwhitehorn@FreeBSD.org> |
Add some error checking on the supplied page size list. This makes sure that we (a) get the correct large page size to provide to pmap and (b) we can alert the user if running under incorrectly-confi
Add some error checking on the supplied page size list. This makes sure that we (a) get the correct large page size to provide to pmap and (b) we can alert the user if running under incorrectly-configured PowerKVM on POWER7 and POWER8 systems.
MFC after: 1 week
show more ...
|
Revision tags: release/10.1.0, release/9.3.0 |
|
#
3b8f0845 |
| 28-Apr-2014 |
Simon J. Gerraty <sjg@FreeBSD.org> |
Merge head
|
#
84e51a1b |
| 23-Apr-2014 |
Alan Somers <asomers@FreeBSD.org> |
IFC @264767
|
#
5748b897 |
| 19-Feb-2014 |
Martin Matuska <mm@FreeBSD.org> |
Merge head up to r262222 (last merge was incomplete).
|
#
485ac45a |
| 04-Feb-2014 |
Peter Grehan <grehan@FreeBSD.org> |
MFC @ r259205 in preparation for some SVM updates. (for real this time)
|
Revision tags: release/10.0.0 |
|
#
4af7c894 |
| 23-Nov-2013 |
Andreas Tobler <andreast@FreeBSD.org> |
Limit the large page size to 16MB for now.
MFC after: 1 week
|
#
50d3286d |
| 11-Nov-2013 |
Gleb Smirnoff <glebius@FreeBSD.org> |
Merge head r232040 through r258006.
|
#
12e7de22 |
| 31-Oct-2013 |
Andreas Tobler <andreast@FreeBSD.org> |
Ouch, do the shift properly. An uin64_t needs the right postfix when shifting.
|
#
064bee34 |
| 30-Oct-2013 |
Peter Grehan <grehan@FreeBSD.org> |
MFC @ r256071
This is just prior to the bhyve_npt_pmap import so will allow just the change to be merged for easier debug.
|
#
0bfd163f |
| 18-Oct-2013 |
Gleb Smirnoff <glebius@FreeBSD.org> |
Merge head r233826 through r256722.
|