71b936e93aa6de6fa10bd4b5d93a266604409072 |
|
25-Apr-2009 |
subrata_modak <subrata_modak> |
The attached patch fixes this bug. This is the bug of the kernel. The following patch has fixed the bug. commit af76aba00fdcfb21535c9f9872245d14097a4561 Author: Rusty Russell <rusty@rustcorp.com.au> Date: Mon Mar 30 22:05:11 2009 -0600 cpumask: fix seq_bitmap_*() functions. 1) seq_bitmap_list() should take a const. 2) All the seq_bitmap should use cpumask_bits(). Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> But after applying this patch, Case 5 and 6 still failed. It is because the max size of CPUSET in the glibc is 1024, but the max size of your kernel is 4096, when we invoke the sched_getaffinity, the kernel will check whether the size of mask passed into the function is >= the size of the mask in the kernel. If not, the function returns -1. So it is not the bug of the tesecase. This patch contains two lib files written by Paul Jackson who is ex-maintainer of cpuset. These two lib files are used for both this bug fix and the testcases that will be added in future. Signed-Off-By: Miao Xie <miaox@cn.fujitsu.com>.
/external/ltp/testcases/kernel/controllers/cpuset/cpuset_lib/cpuset.h
|