1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
|
#!/usr/bin/env python3
# Print the distinct pages of the semaphore VAs a process references
# (reuses memprobe.py's scanner; run in the guest as root). The VA window is
# derived from the process's /dev/nvidia* mappings unless lo/hi are given.
# usage: sudo python3 sempages.py <pid> [lo] [hi]
import os
import sys
sys.path.insert(0, '/tmp')
import memprobe as m
pid = int(sys.argv[1])
if len(sys.argv) > 3:
lo = int(sys.argv[2], 16)
hi = int(sys.argv[3], 16)
else:
lo = 1 << 64
hi = 0
with open('/proc/%d/maps' % pid) as f:
for line in f:
parts = line.split()
if len(parts) < 6 or not parts[5].startswith('/dev/nvidia'):
continue
a, b = parts[0].split('-')
lo = min(lo, int(a, 16))
hi = max(hi, int(b, 16))
if hi == 0:
print('no /dev/nvidia mappings')
raise SystemExit(1)
fd = os.open('/proc/%d/mem' % pid, os.O_RDONLY)
# Pin what libcuda actually polls: the semaphore objects referenced by the
# client's trackers (and the semaphore memory they point at). Scanning for
# bare sem-object shapes matches far too many pages once the ring/pushbuffer
# windows are in range; trackers are a structural reference, so pinning is
# limited to pages the client really waits on. Fall back to the shape scan
# only when no tracker resolves.
pages = set()
trackers = m.find_trackers(pid, fd, lo, hi, limit=256)
for _p, _sub, _last, _cur, sem, mem in trackers:
if sem:
pages.add(sem & ~0xfff)
if mem:
pages.add(mem & ~0xfff)
if not pages:
sems = m.sem_objs(pid, fd, lo, hi)
pages = {va & ~0xfff for va in sems}
os.close(fd)
pages = sorted(pages)
print('trackers: %d pages: %d (window 0x%x..0x%x)' %
(len(trackers), len(pages), lo, hi))
for pg in pages:
print('0x%x' % pg)
|
