fix #744 aka #753: implement memory USS / PSS on Linux

5 files changed, 48 insertions, 7 deletions

diff --git a/HISTORY.rst b/HISTORY.rst
index 27b628b9..96af5586 100644
--- a/HISTORY.rst
+++ b/HISTORY.rst
@@ -6,6 +6,7 @@ Bug tracker at https://github.com/giampaolo/psutil/issues
 **Enhancements**
 
 - #732: Process.environ().  (patch by Frank Benkstein)
+- #753: [Linux] Process memory USS and PSS.
 
 **Bug fixes**
 
diff --git a/README.rst b/README.rst
index 81fe9bb3..cebf841c 100644
--- a/README.rst
+++ b/README.rst
@@ -239,7 +239,7 @@ Process management
     >>> p.memory_info()
     pmem(rss=7471104, vms=68513792)
     >>> p.memory_info_ex()
-    extmem(rss=9662464, vms=49192960, shared=3612672, text=2564096, lib=0, data=5754880, dirty=0)
+    pextmem(rss=10915840, vms=67608576, shared=3313664, text=2310144, lib=0, data=7262208, dirty=0, uss=9830400, pss=1216512)
     >>> p.memory_maps()
     [pmmap_grouped(path='/lib/x8664-linux-gnu/libutil-2.15.so', rss=32768, size=2125824, pss=32768, shared_clean=0, shared_dirty=0, private_clean=20480, private_dirty=12288, referenced=32768, anonymous=12288, swap=0),
      pmmap_grouped(path='/lib/x8664-linux-gnu/libc-2.15.so', rss=3821568, size=3842048, pss=3821568, shared_clean=0, shared_dirty=0, private_clean=0, private_dirty=3821568, referenced=3575808, anonymous=3821568, swap=0),
diff --git a/docs/index.rst b/docs/index.rst
index 1c564783..aa1bd36b 100644
--- a/docs/index.rst
+++ b/docs/index.rst
@@ -1006,9 +1006,9 @@ Process class
      +--------+---------+-------+---------+--------------------+
      | dirty  |         |       |         | nonpaged_pool      |
      +--------+---------+-------+---------+--------------------+
-     |        |         |       |         | pagefile           |
+     | uss    |         |       |         | pagefile           |
      +--------+---------+-------+---------+--------------------+
-     |        |         |       |         | peak_pagefile      |
+     | pss    |         |       |         | peak_pagefile      |
      +--------+---------+-------+---------+--------------------+
      |        |         |       |         | private            |
      +--------+---------+-------+---------+--------------------+
@@ -1022,6 +1022,8 @@ Process class
      >>> p.memory_info_ex()
      pextmem(rss=15491072, vms=84025344, shared=5206016, text=2555904, lib=0, data=9891840, dirty=0)
 
+     .. versionchanged:: 3.5.0 added `uss` and `pss` fields on Linux.
+
   .. method:: memory_percent()
 
      Compare physical system memory to process resident memory (RSS) and
diff --git a/psutil/_pslinux.py b/psutil/_pslinux.py
index ee613564..04c95cf3 100644
--- a/psutil/_pslinux.py
+++ b/psutil/_pslinux.py
@@ -56,6 +56,7 @@ __extra__all__ = [
 
 # --- constants
 
+HAS_SMAPS = os.path.exists('/proc/%s/smaps' % os.getpid())
 HAS_PRLIMIT = hasattr(cext, "linux_prlimit")
 
 # RLIMIT_* constants, not guaranteed to be present on all kernels
@@ -217,7 +218,7 @@ svmem = namedtuple(
     'svmem', ['total', 'available', 'percent', 'used', 'free',
               'active', 'inactive', 'buffers', 'cached'])
 
-pextmem = namedtuple('pextmem', 'rss vms shared text lib data dirty')
+pextmem = namedtuple('pextmem', 'rss vms shared text lib data dirty uss pss')
 
 pmmap_grouped = namedtuple(
     'pmmap_grouped', ['path', 'rss', 'size', 'pss', 'shared_clean',
@@ -972,7 +973,9 @@ class Process(object):
                                 int(vms) * PAGESIZE)
 
     @wrap_exceptions
-    def memory_info_ex(self):
+    def memory_info_ex(self,
+                       _private_re=re.compile(b"Private.*:\s+(\d+)"),
+                       _shared_re=re.compile(b"Shared.*:\s+(\d+)")):
         #  ============================================================
         # | FIELD  | DESCRIPTION                         | AKA  | TOP  |
         #  ============================================================
@@ -983,13 +986,30 @@ class Process(object):
         # | lib    | library (unused in Linux 2.6)       | lrs  |      |
         # | data   | data + stack                        | drs  | DATA |
         # | dirty  | dirty pages (unused in Linux 2.6)   | dt   |      |
+        # | -----------------------------------------------------------
+        # | uss    | unique set size ("real memory")     |      |      |
+        # | pss    | proportional set size               |      |      |
         #  ============================================================
         with open_binary("%s/%s/statm" % (self._procfs_path, self.pid)) as f:
             vms, rss, shared, text, lib, data, dirty = \
                 [int(x) * PAGESIZE for x in f.readline().split()[:7]]
-        return pextmem(rss, vms, shared, text, lib, data, dirty)
+        if HAS_SMAPS:
+            # Note: using two regexes is faster than reading the file
+            # line by line.
+            # XXX: on Python 3 the 2 regexes are 30% slower than on
+            # Python 2 though. Figure out why.
+            with open_binary("%s/%s/smaps" % (self._procfs_path, self.pid),
+                             buffering=BIGGER_FILE_BUFFERING) as f:
+                smaps_data = f.read()
+            uss = sum(map(int, _private_re.findall(smaps_data))) * 1024
+            pss = sum(map(int, _shared_re.findall(smaps_data))) * 1024
+        else:
+            # usually means we're on kernel < 2.6.14 or CONFIG_MMU kernel
+            # configuration option is not enabled.
+            uss = pss = 0
+        return pextmem(rss, vms, shared, text, lib, data, dirty, uss, pss)
 
-    if os.path.exists('/proc/%s/smaps' % os.getpid()):
+    if HAS_SMAPS:
 
         @wrap_exceptions
         def memory_maps(self):
diff --git a/test/_linux.py b/test/_linux.py
index d36afc53..637ad1f7 100644
--- a/test/_linux.py
+++ b/test/_linux.py
@@ -599,6 +599,24 @@ class LinuxSpecificTestCase(unittest.TestCase):
             self.assertEqual(psutil.Process().exe(), "/home/foo")
             self.assertEqual(psutil.Process().cwd(), "/home/foo")
 
+    def test_uss_pss_mem_against_mem_maps(self):
+        src = textwrap.dedent("""
+            import time
+            with open("%s", "w") as f:
+                time.sleep(10)
+            """ % TESTFN)
+        sproc = pyrun(src)
+        self.addCleanup(reap_children)
+        call_until(lambda: os.listdir('.'), "'%s' not in ret" % TESTFN)
+        p = psutil.Process(sproc.pid)
+        time.sleep(.1)
+        memex = p.memory_info_ex()
+        maps = p.memory_maps(grouped=False)
+        self.assertEqual(
+            memex.uss, sum([x.private_dirty + x.private_clean for x in maps]))
+        self.assertEqual(
+            memex.pss, sum([x.shared_dirty + x.shared_clean for x in maps]))
+
 
 def main():
     test_suite = unittest.TestSuite()