AcoSail/kernel_optimize_test
8
0
Fork 0
forked from luck/tmp_suning_uos_patched
Code Pull Requests Activity

sparc64: Perf should save/restore fault info

Browse Source 
There have been several reports of random processes being killed with
a bus error or segfault during userspace stack walking in perf.  One
of the root causes of this problem is an asynchronous modification to
thread_info fault_address and fault_code, which stems from a perf
counter interrupt arriving during kernel processing of a "benign"
fault, such as a TSB miss. Since perf_callchain_user() invokes
copy_from_user() to read user stacks, a fault is not only possible,
but probable. Validity checks on the stack address merely cover up the
problem and reduce its frequency.

The solution here is to save and restore fault_address and fault_code
in perf_callchain_user() so that the benign fault handler is not
disturbed by a perf interrupt.

Signed-off-by: Rob Gardner <rob.gardner@oracle.com>
Signed-off-by: Dave Aldridge <david.j.aldridge@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This commit is contained in:
Rob Gardner 2015-12-22 21:16:07 -07:00 committed by David S. Miller
parent 3f74306ac8
commit 833526941f
1 changed files with 4 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
arch/sparc/kernel/perf_event.c
View File

@ -1828,6 +1828,8 @@ static void perf_callchain_user_32(struct perf_callchain_entry *entry,
void
perf_callchain_user(struct perf_callchain_entry *entry, struct pt_regs *regs)
{
u64 saved_fault_address = current_thread_info()->fault_address;
u8 saved_fault_code = get_thread_fault_code();
mm_segment_t old_fs;
perf_callchain_store(entry, regs->tpc);
@ -1850,4 +1852,6 @@ perf_callchain_user(struct perf_callchain_entry *entry, struct pt_regs *regs)
pagefault_enable();
set_fs(old_fs);
set_thread_fault_code(saved_fault_code);
current_thread_info()->fault_address = saved_fault_address;
}