]>
Commit | Line | Data |
---|---|---|
c58f753d | 1 | 2008-06-04 Brian Behlendorf <behlendorf1@llnl.gov> |
2 | ||
3 | * : Tag spl-0.3.2 | |
4 | ||
5 | * : Extensive improvements to the build system to detect kernel | |
6 | API changes so we can flexibly build with a wider range of kernel | |
7 | versions. The code has now been testing with the 2.6.18-32chaos | |
8 | and 2.6.25.3-18.fc9 kernels, however we should also be compatible | |
9 | with other kernels in the range of 2.6.18-2.6.25. | |
10 | ||
57d86234 | 11 | 2008-05-25 Brian Behlendorf <behlendorf1@llnl.gov> |
12 | ||
13 | * configure.ac, autoconf/* : Initial pass at resolving | |
14 | API changes introduced by kernels newer than 2.6.18. | |
15 | ||
65a045da | 16 | 2008-05-21 Brian Behlendorf <behlendorf1@llnl.gov> |
17 | ||
18 | * : Tag spl-0.3.1 | |
19 | ||
20 | * : License headers including URCL added for release. | |
21 | ||
b2585b36 | 22 | 2008-05-21 Brian Behlendorf <behlendorf1@llnl.gov> |
23 | ||
24 | * : Tag spl-0.3.0 | |
25 | ||
26 | * configure.ac: Improved autotools support. | |
27 | ||
9ab1ac14 | 28 | 2008-04-26 Brian Behlendorf <behlendorf1@llnl.gov> |
29 | ||
30 | * include/sys/mutex.h : Implemented a close approximation | |
31 | of adaptive mutexes. These changes however required me to | |
32 | export a new symbol from the kernel proper 'task_curr()' | |
33 | which means we are now dependant on a patched kernel. | |
34 | ||
3561541c | 35 | 2008-04-24 Brian Behlendorf <behlendorf1@llnl.gov> |
36 | ||
37 | * : Tag spl-0.2.1 | |
38 | ||
39 | * modules/spl/spl-proc.c : Add /proc/sys/spl/version. | |
40 | ||
1bac409f | 41 | 2008-04-24 Herb Wartens <wartens2@llnl.gov> |
42 | ||
43 | * include/sys/kmem.h : Make sure that when calling __vmem_alloc | |
44 | that we do not have __GFP_ZERO set. Once the memory is allocated | |
45 | then zero out the memory if __GFP_ZERO is passed to | |
46 | __vmem_alloc. | |
47 | ||
55152ebb | 48 | 2008-04-16 Herb Wartens <wartens2@llnl.gov> |
49 | ||
50 | * modules/spl/spl-kmem.c : Make sure to disable interrupts | |
51 | when necessary to avoid deadlocks. We were seeing the deadlock | |
52 | when calling kmem_cache_generic_constructor() and then an interrupt | |
53 | forced us to end up calling kmem_cache_generic_destructor() | |
54 | which caused our deadlock. | |
55 | ||
f1ca4da6 | 56 | 2008-02-26 Brian Behlendorf <behlendorf1@llnl.gov> |
57 | ||
58 | : Initial commit of the solaris porting layer (spl). Included | |
59 | in addition to the source is an initial autoconf / configure | |
60 | style build system. |