guest-diag: pin the CE push-buffer window too

The failing wait (uvtw) is uvm_tracker_wait_for_entry for value 3 on a CE
channel whose tracking sem stayed at payload 1 for 14 s: its pushes'
release values 2..3 were never applied because the CE push-buffer pages
(0x120000000 + 0x100000*rank) are outside the GR window and were only
found by the content sweep, late and often wrong (false=3065). Scan both
windows in the guest so the host gets guest-truth translations for the CE
pushes as well.

1 files changed, 39 insertions, 22 deletions

diff --git a/tools/guest-diag/guest_pin_stream.py b/tools/guest-diag/guest_pin_stream.py
index bed4728..ac4f72b 100644
--- a/tools/guest-diag/guest_pin_stream.py
+++ b/tools/guest-diag/guest_pin_stream.py
@@ -30,6 +30,12 @@ REGION = 0x200000
 # ..0x45000) and the per-channel push allocations follow; 512 KB per 2 MB
 # region covers both without scanning holes that hold nothing.
 SPAN = int(os.environ.get("STREAM_SPAN", "0x80000"), 16)
+# CE push-buffer window: one 1 MB allocation per CE channel at
+# 0x120000000 + 0x100000*rank; SPAN2 covers the 1 MB boundary inside a
+# 2 MB region.
+LO2 = int(os.environ.get("STREAM_LO2", "0x120000000"), 16)
+HI2 = int(os.environ.get("STREAM_HI2", "0x120600000"), 16)
+SPAN2 = int(os.environ.get("STREAM_SPAN2", "0x180000"), 16)
 TRACE = ("/sys/kernel/tracing/trace"
          if os.path.exists("/sys/kernel/tracing/trace")
          else "/sys/kernel/debug/tracing/trace")
@@ -108,6 +114,31 @@ def cdt_pid():
     return None
 
 
+def scan_window(fd, lo, hi, span, out):
+    base = lo & ~(REGION - 1)
+    while base < hi:
+        va = max(base, lo) & ~0xFFF
+        end = min(base + span, hi)
+        while va < end:
+            n = min(BATCH, (end - va) // 0x1000)
+            if n <= 0:
+                break
+            os.lseek(fd, (va // 4096) * 8, 0)
+            data = os.read(fd, n * 8)
+            got = len(data) // 8
+            for i in range(got):
+                e = int.from_bytes(data[i * 8:i * 8 + 8], 'little')
+                if not ((e >> 63) & 1):
+                    continue
+                pfn = e & ((1 << 55) - 1)
+                pa = pfn << 12
+                if not pa or (PAMAX and pa >= PAMAX):
+                    continue
+                out[va + i * 0x1000] = pa
+            va += got * 0x1000
+        base += REGION
+
+
 def scan(pid):
     """Present pages in the first SPAN bytes of every REGION, as {va: pa}.
 
@@ -115,32 +146,18 @@ def scan(pid):
     aligned); scanning only the allocation heads keeps the per-loop page
     table work small, which is what makes the stream fast while the client
     is mapping and the guest is loaded.
+
+    Two windows: the GR window (rings/pushbuffers/counter page) and the CE
+    push-buffer window (0x120000000+, one 1 MB allocation per CE channel).
+    The CE window uses a wider span because the per-channel allocations are
+    1 MB apart inside a 2 MB region.
     """
     fd = os.open('/proc/%d/pagemap' % pid, os.O_RDONLY)
     out = {}
     try:
-        base = LO & ~(REGION - 1)
-        while base < HI:
-            va = max(base, LO) & ~0xFFF
-            end = min(base + SPAN, HI)
-            while va < end:
-                n = min(BATCH, (end - va) // 0x1000)
-                if n <= 0:
-                    break
-                os.lseek(fd, (va // 4096) * 8, 0)
-                data = os.read(fd, n * 8)
-                got = len(data) // 8
-                for i in range(got):
-                    e = int.from_bytes(data[i * 8:i * 8 + 8], 'little')
-                    if not ((e >> 63) & 1):
-                        continue
-                    pfn = e & ((1 << 55) - 1)
-                    pa = pfn << 12
-                    if not pa or (PAMAX and pa >= PAMAX):
-                        continue
-                    out[va + i * 0x1000] = pa
-                va += got * 0x1000
-            base += REGION
+        scan_window(fd, LO, HI, SPAN, out)
+        if HI2 > LO2:
+            scan_window(fd, LO2, HI2, SPAN2, out)
     finally:
         os.close(fd)
     return out