pythongh-141770: Annotate anonymous mmap usage if "-X dev" is used (pythongh-142079)

corona10 · web-flow · commit c4ccaf4b1051 · 2025-12-08T14:47:19.000Z
diff --git a/Doc/whatsnew/3.15.rst b/Doc/whatsnew/3.15.rst
@@ -1236,6 +1236,12 @@ Build changes
   modules that are missing or packaged separately.
   (Contributed by Stan Ulbrych and Petr Viktorin in :gh:`139707`.)
 
+* Annotating anonymous mmap usage is now supported if Linux kernel supports
+  :manpage:`PR_SET_VMA_ANON_NAME <PR_SET_VMA(2const)>` (Linux 5.17 or newer).
+  Annotations are visible in ``/proc/<pid>/maps`` if the kernel supports the feature
+  and :option:`-X dev <-X>` is passed to the Python or Python is built in :ref:`debug mode <debug-build>`.
+  (Contributed by Donghee Na in :gh:`141770`)
+
 
 Porting to Python 3.15
 ======================
diff --git a/Include/internal/pycore_mmap.h b/Include/internal/pycore_mmap.h
@@ -0,0 +1,45 @@
+#ifndef Py_INTERNAL_MMAP_H
+#define Py_INTERNAL_MMAP_H
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#ifndef Py_BUILD_CORE
+#  error "this header requires Py_BUILD_CORE define"
+#endif
+
+#include "pycore_pystate.h"
+
+#if defined(HAVE_PR_SET_VMA_ANON_NAME) && defined(__linux__)
+#  include <linux/prctl.h>
+#  include <sys/prctl.h>
+#endif
+
+#if defined(HAVE_PR_SET_VMA_ANON_NAME) && defined(__linux__)
+static inline void
+_PyAnnotateMemoryMap(void *addr, size_t size, const char *name)
+{
+#ifndef Py_DEBUG
+    if (!_Py_GetConfig()->dev_mode) {
+        return;
+    }
+#endif
+    assert(strlen(name) < 80);
+    int old_errno = errno;
+    prctl(PR_SET_VMA, PR_SET_VMA_ANON_NAME, (unsigned long)addr, size, name);
+    /* Ignore errno from prctl */
+    /* See: https://bugzilla.redhat.com/show_bug.cgi?id=2302746 */
+    errno = old_errno;
+}
+#else
+static inline void
+_PyAnnotateMemoryMap(void *Py_UNUSED(addr), size_t Py_UNUSED(size), const char *Py_UNUSED(name))
+{
+}
+#endif
+
+#ifdef __cplusplus
+}
+#endif
+#endif  // !Py_INTERNAL_MMAP_H
diff --git a/Makefile.pre.in b/Makefile.pre.in
@@ -1378,6 +1378,7 @@ PYTHON_HEADERS= \
 		$(srcdir)/Include/internal/pycore_long.h \
 		$(srcdir)/Include/internal/pycore_memoryobject.h \
 		$(srcdir)/Include/internal/pycore_mimalloc.h \
+		$(srcdir)/Include/internal/pycore_mmap.h \
 		$(srcdir)/Include/internal/pycore_modsupport.h \
 		$(srcdir)/Include/internal/pycore_moduleobject.h \
 		$(srcdir)/Include/internal/pycore_namespace.h \
diff --git a/Misc/NEWS.d/next/Core_and_Builtins/2025-11-29-18-14-28.gh-issue-141770.JURnvg.rst b/Misc/NEWS.d/next/Core_and_Builtins/2025-11-29-18-14-28.gh-issue-141770.JURnvg.rst
@@ -0,0 +1,2 @@
+Annotate anonymous mmap usage only when supported by the
+Linux kernel and if ``-X dev`` is used or Python is built in debug mode. Patch by Donghee Na.
diff --git a/Modules/_ctypes/malloc_closure.c b/Modules/_ctypes/malloc_closure.c
@@ -14,6 +14,7 @@
 #  endif
 #endif
 #include "ctypes.h"
+#include "pycore_mmap.h"          // _PyAnnotateMemoryMap()
 
 /* BLOCKSIZE can be adjusted.  Larger blocksize will take a larger memory
    overhead, but allocate less blocks from the system.  It may be that some
@@ -74,14 +75,16 @@ static void more_core(void)
     if (item == NULL)
         return;
 #else
+    size_t mem_size = count * sizeof(ITEM);
     item = (ITEM *)mmap(NULL,
-                        count * sizeof(ITEM),
+                        mem_size,
                         PROT_READ | PROT_WRITE | PROT_EXEC,
                         MAP_PRIVATE | MAP_ANONYMOUS,
                         -1,
                         0);
     if (item == (void *)MAP_FAILED)
         return;
+    _PyAnnotateMemoryMap(item, mem_size, "cpython:ctypes");
 #endif
 
 #ifdef MALLOC_CLOSURE_DEBUG
diff --git a/Modules/mmapmodule.c b/Modules/mmapmodule.c
@@ -26,6 +26,7 @@
 #include "pycore_abstract.h"      // _Py_convert_optional_to_ssize_t()
 #include "pycore_bytesobject.h"   // _PyBytes_Find()
 #include "pycore_fileutils.h"     // _Py_stat_struct
+#include "pycore_mmap.h"          // _PyAnnotateMemoryMap()
 #include "pycore_weakref.h"       // FT_CLEAR_WEAKREFS()
 
 #include <stddef.h>               // offsetof()
@@ -1951,6 +1952,7 @@ new_mmap_object(PyTypeObject *type, PyObject *args, PyObject *kwdict)
         PyErr_SetFromErrno(PyExc_OSError);
         return NULL;
     }
+    _PyAnnotateMemoryMap(m_obj->data, map_size, "cpython:mmap");
     m_obj->access = (access_mode)access;
     return (PyObject *)m_obj;
 }
diff --git a/Objects/obmalloc.c b/Objects/obmalloc.c
@@ -2,6 +2,7 @@
 
 #include "Python.h"
 #include "pycore_interp.h"        // _PyInterpreterState_HasFeature
+#include "pycore_mmap.h"          // _PyAnnotateMemoryMap()
 #include "pycore_object.h"        // _PyDebugAllocatorStats() definition
 #include "pycore_obmalloc.h"
 #include "pycore_obmalloc_init.h"
@@ -467,6 +468,7 @@ _PyMem_ArenaAlloc(void *Py_UNUSED(ctx), size_t size)
     if (ptr == MAP_FAILED)
         return NULL;
     assert(ptr != NULL);
+    _PyAnnotateMemoryMap(ptr, size, "cpython:pymalloc");
     return ptr;
 #else
     return malloc(size);
diff --git a/PCbuild/pythoncore.vcxproj b/PCbuild/pythoncore.vcxproj
@@ -277,6 +277,7 @@
     <ClInclude Include="..\Include\internal\pycore_llist.h" />
     <ClInclude Include="..\Include\internal\pycore_lock.h" />
     <ClInclude Include="..\Include\internal\pycore_long.h" />
+    <ClInclude Include="..\Include\internal\pycore_mmap.h" />
     <ClInclude Include="..\Include\internal\pycore_modsupport.h" />
     <ClInclude Include="..\Include\internal\pycore_moduleobject.h" />
     <ClInclude Include="..\Include\internal\pycore_namespace.h" />
diff --git a/PCbuild/pythoncore.vcxproj.filters b/PCbuild/pythoncore.vcxproj.filters
@@ -735,6 +735,9 @@
     <ClInclude Include="..\Include\internal\pycore_long.h">
       <Filter>Include\internal</Filter>
     </ClInclude>
+    <ClInclude Include="..\Include\internal\pycore_mmap.h">
+      <Filter>Include\internal</Filter>
+    </ClInclude>
     <ClInclude Include="..\Include\internal\pycore_modsupport.h">
       <Filter>Include\internal</Filter>
     </ClInclude>
diff --git a/Python/jit.c b/Python/jit.c
@@ -16,6 +16,7 @@
 #include "pycore_intrinsics.h"
 #include "pycore_list.h"
 #include "pycore_long.h"
+#include "pycore_mmap.h"
 #include "pycore_opcode_metadata.h"
 #include "pycore_opcode_utils.h"
 #include "pycore_optimizer.h"
@@ -75,6 +76,9 @@ jit_alloc(size_t size)
     int prot = PROT_READ | PROT_WRITE;
     unsigned char *memory = mmap(NULL, size, prot, flags, -1, 0);
     int failed = memory == MAP_FAILED;
+    if (!failed) {
+        _PyAnnotateMemoryMap(memory, size, "cpython:jit");
+    }
 #endif
     if (failed) {
         jit_error("unable to allocate memory");
diff --git a/Python/perf_jit_trampoline.c b/Python/perf_jit_trampoline.c
@@ -61,6 +61,7 @@
 #include "pycore_ceval.h"         // _PyPerf_Callbacks
 #include "pycore_frame.h"
 #include "pycore_interp.h"
+#include "pycore_mmap.h"          // _PyAnnotateMemoryMap()
 #include "pycore_runtime.h"       // _PyRuntime
 
 #ifdef PY_HAVE_PERF_TRAMPOLINE
@@ -1085,6 +1086,7 @@ static void* perf_map_jit_init(void) {
         close(fd);
         return NULL;  // Memory mapping failed
     }
+    _PyAnnotateMemoryMap(perf_jit_map_state.mapped_buffer, page_size, "cpython:perf_jit_trampoline");
 #endif
 
     perf_jit_map_state.mapped_size = page_size;
diff --git a/Python/perf_trampoline.c b/Python/perf_trampoline.c
@@ -132,6 +132,7 @@ any DWARF information available for them).
 #include "Python.h"
 #include "pycore_ceval.h"         // _PyPerf_Callbacks
 #include "pycore_interpframe.h"   // _PyFrame_GetCode()
+#include "pycore_mmap.h"          // _PyAnnotateMemoryMap()
 #include "pycore_runtime.h"       // _PyRuntime
 
 
@@ -290,6 +291,7 @@ new_code_arena(void)
         perf_status = PERF_STATUS_FAILED;
         return -1;
     }
+    _PyAnnotateMemoryMap(memory, mem_size, "cpython:perf_trampoline");
     void *start = &_Py_trampoline_func_start;
     void *end = &_Py_trampoline_func_end;
     size_t code_size = end - start;
diff --git a/configure b/configure
diff --git a/configure.ac b/configure.ac
@@ -5583,6 +5583,13 @@ AC_CHECK_DECLS([UT_NAMESIZE],
               [],
               [@%:@include <utmp.h>])
 
+AC_CHECK_DECLS([PR_SET_VMA_ANON_NAME],
+              [AC_DEFINE([HAVE_PR_SET_VMA_ANON_NAME], [1],
+                         [Define if you have the 'PR_SET_VMA_ANON_NAME' constant.])],
+              [],
+              [@%:@include <linux/prctl.h>
+               @%:@include <sys/prctl.h>])
+
 # check for openpty, login_tty, and forkpty
 
 AC_CHECK_FUNCS([openpty], [],
diff --git a/pyconfig.h.in b/pyconfig.h.in
@@ -228,6 +228,10 @@
 /* Define to 1 if you have the <db.h> header file. */
 #undef HAVE_DB_H
 
+/* Define to 1 if you have the declaration of 'PR_SET_VMA_ANON_NAME', and to 0
+   if you don't. */
+#undef HAVE_DECL_PR_SET_VMA_ANON_NAME
+
 /* Define to 1 if you have the declaration of 'RTLD_DEEPBIND', and to 0 if you
    don't. */
 #undef HAVE_DECL_RTLD_DEEPBIND
@@ -996,6 +1000,9 @@
 /* Define if your compiler supports function prototype */
 #undef HAVE_PROTOTYPES
 
+/* Define if you have the 'PR_SET_VMA_ANON_NAME' constant. */
+#undef HAVE_PR_SET_VMA_ANON_NAME
+
 /* Define to 1 if you have the 'pthread_condattr_setclock' function. */
 #undef HAVE_PTHREAD_CONDATTR_SETCLOCK
 

Original file line number	Diff line number	Diff line change
`@@ -0,0 +1,2 @@`
	`1`	`+Annotate anonymous mmap usage only when supported by the`
	`2`	+Linux kernel and if ``-X dev`` is used or Python is built in debug mode. Patch by Donghee Na.
Original file line number	Diff line number	Diff line change
`@@ -26,6 +26,7 @@`
`26`	`26`	`#include "pycore_abstract.h" // _Py_convert_optional_to_ssize_t()`
`27`	`27`	`#include "pycore_bytesobject.h" // _PyBytes_Find()`
`28`	`28`	`#include "pycore_fileutils.h" // _Py_stat_struct`
	`29`	`+#include "pycore_mmap.h" // _PyAnnotateMemoryMap()`
`29`	`30`	`#include "pycore_weakref.h" // FT_CLEAR_WEAKREFS()`
`30`	`31`
`31`	`32`	`#include <stddef.h> // offsetof()`
`@@ -1951,6 +1952,7 @@ new_mmap_object(PyTypeObject type, PyObject args, PyObject *kwdict)`
`1951`	`1952`	`PyErr_SetFromErrno(PyExc_OSError);`
`1952`	`1953`	`return NULL;`
`1953`	`1954`	`}`
	`1955`	`+ _PyAnnotateMemoryMap(m_obj->data, map_size, "cpython:mmap");`
`1954`	`1956`	`m_obj->access = (access_mode)access;`
`1955`	`1957`	`return (PyObject *)m_obj;`
`1956`	`1958`	`}`