diff options
Diffstat (limited to 'tools/guest-diag/guest_pin_stream.py')
| -rw-r--r-- | tools/guest-diag/guest_pin_stream.py | 55 |
1 files changed, 53 insertions, 2 deletions
diff --git a/tools/guest-diag/guest_pin_stream.py b/tools/guest-diag/guest_pin_stream.py index bb253a2..175ed23 100644 --- a/tools/guest-diag/guest_pin_stream.py +++ b/tools/guest-diag/guest_pin_stream.py @@ -1544,6 +1544,18 @@ def scan_vmas(fd, pid, out): parts = line.split() if len(parts) < 5: continue + # Skip mappings with no read permission: a CUDA device buffer + # is a PROT_NONE reservation (e.g. `---p 00000000 00:00 0`), + # potentially hundreds of MB. Its pages are never present in + # the pagemap, so scanning it produces no pairs - but it + # consumed the VMA_MAX_PAGES budget and the scan never reached + # the client's present staging buffers that sit ABOVE the + # device reservation (observed: a client whose device range + # started below its staging buffer got no pins at all and its + # H2D copies failed with rc=-2 forever). + perms = parts[1] if len(parts) >= 2 else '' + if 'r' not in perms: + continue # Anonymous mappings (no path field) are the client's UVM sysmem # buffers when the driver maps them without a /dev/zero backing: # the second client's cudaMalloc'd data buffers show up exactly @@ -1760,8 +1772,21 @@ def main(): st_d = os.stat("/tmp/guest_want_vas.txt") if st_d.st_mtime != _want_state.get("mtime"): with open("/tmp/guest_want_vas.txt") as fw: - wvas = [int(t, 16) for t in fw.read().split() - if t.startswith("0x")] + # r582: tolerate malformed historical lines. The + # r581 boot's scanner died here: a transport-escaped + # append merged VAs into one token containing + # literal backslash-n ('0x20400f000\\n0x121000000\\n + # ...') and int(t, 16) raised ValueError out of + # main(), stopping ALL new pins (the observed + # pool-pin latency). Extract every 0x token instead; + # a bad token is skipped, never fatal. + wvas = [] + for t in re.findall(r"0x[0-9a-fA-F]+", + fw.read()): + try: + wvas.append(int(t, 16)) + except ValueError: + pass _want_state["mtime"] = st_d.st_mtime for wv in wvas[-64:]: if wv in _want_state["done"]: @@ -1794,6 +1819,32 @@ def main(): except Exception: ok = False got = [] + # r586: a demanded VA outside the CE + # pushbuffer window (the GRFILL staging pages, + # high client VAs) is a plain client mapping + # page: resolve it through the client's own + # pagemap. Without this the demand for those + # VAs was a silent no-op (resolve_va_by_content + # rejects anything outside PB_CE_LO..PB_CE_HI) + # and the fills kept dropping them. + if not ok and not got and pid: + try: + pfd = os.open('/proc/%d/pagemap' % pid, + os.O_RDONLY) + try: + sub = {} + vpg = wv & ~0xFFF + scan_range(pfd, vpg, vpg + 0x1000, + sub) + finally: + os.close(pfd) + if sub: + for v, p in sub.items(): + got.append('0x%x,0x%x,1\n' + % (v, p)) + ok = True + except OSError: + pass if ok and got: lines.extend(got) _want_state["done"].add(wv) |