[COMMITTED 2.34 0/2] Improve aligned_alloc and calloc test coverage

[COMMITTED 2.34 0/2] Improve aligned_alloc and calloc test coverage

[Miguel Martin]

From: Miguel Martín <mmartinv@redhat.com>

Backports to improve aligned_alloc and calloc test coverage in glibc 2.34:

DJ Delorie (1):
  aligned_alloc: conform to C17

Joe Simmons-Talbott (1):
  malloc: Improve aligned_alloc and calloc test coverage.

 malloc/Makefile                   | 18 ++++++-
 malloc/malloc-debug.c             |  9 +++-
 malloc/malloc.c                   | 26 ++++++++--
 malloc/tst-aligned-alloc-random.c | 43 +++++++++++++++++
 malloc/tst-aligned-alloc-static.c |  1 +
 malloc/tst-aligned-alloc.c        | 80 +++++++++++++++++++++++++++++++
 malloc/tst-aligned_alloc-lib.c    | 72 ++++++++++++++++++++++++++++
 malloc/tst-malloc-random.c        | 20 ++++++++
 malloc/tst-malloc.c               |  4 ++
 manual/memory.texi                |  2 +-
 10 files changed, 268 insertions(+), 7 deletions(-)
 create mode 100644 malloc/tst-aligned-alloc-random.c
 create mode 100644 malloc/tst-aligned-alloc-static.c
 create mode 100644 malloc/tst-aligned-alloc.c
 create mode 100644 malloc/tst-aligned_alloc-lib.c
 create mode 100644 malloc/tst-malloc-random.c

--
2.45.1

[COMMITTED 2.34 1/2] aligned_alloc: conform to C17

[Miguel Martin]

From: DJ Delorie <dj@redhat.com>

This patch adds the strict checking for power-of-two alignments
in aligned_alloc(), and updates the manual accordingly.

Reviewed-by: Carlos O'Donell <carlos@redhat.com>
(cherry picked from commit d1417176a35d27ffb8da0ffb1e33154163b6eeb2)
---
 malloc/Makefile                   |  4 +-
 malloc/malloc-debug.c             |  9 +++-
 malloc/malloc.c                   | 26 ++++++++--
 malloc/tst-aligned-alloc-static.c |  1 +
 malloc/tst-aligned-alloc.c        | 80 +++++++++++++++++++++++++++++++
 manual/memory.texi                |  2 +-
 6 files changed, 116 insertions(+), 6 deletions(-)
 create mode 100644 malloc/tst-aligned-alloc-static.c
 create mode 100644 malloc/tst-aligned-alloc.c

diff --git a/malloc/Makefile b/malloc/Makefile
index 9b70831d38..b02c49a687 100644
--- a/malloc/Makefile
+++ b/malloc/Makefile
@@ -43,10 +43,12 @@ tests := mallocbug tst-malloc tst-valloc tst-calloc tst-obstack \
 	 tst-tcfree1 tst-tcfree2 tst-tcfree3 \
 	 tst-safe-linking \
 	 tst-mallocalign1 \
+	 tst-aligned-alloc
 
 tests-static := \
 	 tst-interpose-static-nothread \
-	 tst-interpose-static-thread
+	 tst-interpose-static-thread \
+	 tst-aligned-alloc-static
 
 # Test for the malloc_set_state symbol removed in glibc 2.25.
 ifeq ($(have-GLIBC_2.23)$(build-shared),yesyes)
diff --git a/malloc/malloc-debug.c b/malloc/malloc-debug.c
index 3d7e6d44fd..95ad42b68e 100644
--- a/malloc/malloc-debug.c
+++ b/malloc/malloc-debug.c
@@ -300,7 +300,14 @@ __debug_memalign (size_t alignment, size_t bytes)
   return _debug_mid_memalign (alignment, bytes, RETURN_ADDRESS (0));
 }
 strong_alias (__debug_memalign, memalign)
-strong_alias (__debug_memalign, aligned_alloc)
+static void *
+__debug_aligned_alloc (size_t alignment, size_t bytes)
+{
+  if (!powerof2 (alignment) || alignment == 0)
+    return NULL;
+  return _debug_mid_memalign (alignment, bytes, RETURN_ADDRESS (0));
+}
+strong_alias (__debug_aligned_alloc, aligned_alloc)
 
 static void *
 __debug_pvalloc (size_t bytes)
diff --git a/malloc/malloc.c b/malloc/malloc.c
index d31e985ecc..5ce3223777 100644
--- a/malloc/malloc.c
+++ b/malloc/malloc.c
@@ -3407,6 +3407,29 @@ __libc_memalign (size_t alignment, size_t bytes)
   void *address = RETURN_ADDRESS (0);
   return _mid_memalign (alignment, bytes, address);
 }
+libc_hidden_def (__libc_memalign)
+
+/* For ISO C17.  */
+void *
+weak_function
+aligned_alloc (size_t alignment, size_t bytes)
+{
+  if (!__malloc_initialized)
+    ptmalloc_init ();
+
+/* Similar to memalign, but starting with ISO C17 the standard
+   requires an error for alignments that are not supported by the
+   implementation.  Valid alignments for the current implementation
+   are non-negative powers of two.  */
+  if (!powerof2 (alignment) || alignment == 0)
+    {
+      __set_errno (EINVAL);
+      return 0;
+    }
+
+  void *address = RETURN_ADDRESS (0);
+  return _mid_memalign (alignment, bytes, address);
+}
 
 static void *
 _mid_memalign (size_t alignment, size_t bytes, void *address)
@@ -3465,9 +3488,6 @@ _mid_memalign (size_t alignment, size_t bytes, void *address)
           ar_ptr == arena_for_chunk (mem2chunk (p)));
   return tag_new_usable (p);
 }
-/* For ISO C11.  */
-weak_alias (__libc_memalign, aligned_alloc)
-libc_hidden_def (__libc_memalign)
 
 void *
 __libc_valloc (size_t bytes)
diff --git a/malloc/tst-aligned-alloc-static.c b/malloc/tst-aligned-alloc-static.c
new file mode 100644
index 0000000000..d504473094
--- /dev/null
+++ b/malloc/tst-aligned-alloc-static.c
@@ -0,0 +1 @@
+#include "tst-aligned-alloc.c"
diff --git a/malloc/tst-aligned-alloc.c b/malloc/tst-aligned-alloc.c
new file mode 100644
index 0000000000..8bd6527147
--- /dev/null
+++ b/malloc/tst-aligned-alloc.c
@@ -0,0 +1,80 @@
+/* Test for C17 alignment requirements.
+   Copyright (C) 2023 Free Software Foundation, Inc.
+   This file is part of the GNU C Library.
+
+   The GNU C Library is free software; you can redistribute it and/or
+   modify it under the terms of the GNU Lesser General Public
+   License as published by the Free Software Foundation; either
+   version 2.1 of the License, or (at your option) any later version.
+
+   The GNU C Library is distributed in the hope that it will be useful,
+   but WITHOUT ANY WARRANTY; without even the implied warranty of
+   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+   Lesser General Public License for more details.
+
+   You should have received a copy of the GNU Lesser General Public
+   License along with the GNU C Library; if not, see
+   <https://www.gnu.org/licenses/>.  */
+
+#include <errno.h>
+#include <malloc.h>
+#include <stdint.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <libc-diag.h>
+#include <support/check.h>
+
+static int
+do_test (void)
+{
+  void *p1;
+  void *p2;
+  void *p3;
+  void *p4;
+  void *p5;
+
+  errno = 0;
+
+  /* The implementation supports alignments that are non-negative powers of 2.
+     We test 5 distinct conditions here:
+     - A non-negative power of 2 alignment e.g. 64.
+     - A degenerate zero power of 2 alignment e.g. 1.
+     - A non-power-of-2 alignment e.g. 65.
+     - A zero alignment.
+     - A corner case SIZE_MAX / 2 + 1 alignment.
+  */
+
+  p1 = aligned_alloc (64, 64);
+
+  if (p1 == NULL)
+    FAIL_EXIT1 ("aligned_alloc(64, 64) failed");
+
+  p2 = aligned_alloc (1, 64);
+
+  if (p2 == NULL)
+    FAIL_EXIT1 ("aligned_alloc(1, 64) failed");
+
+  p3 = aligned_alloc (65, 64);
+
+  if (p3 != NULL)
+    FAIL_EXIT1 ("aligned_alloc(65, 64) did not fail");
+
+  p4 = aligned_alloc (0, 64);
+
+  if (p4 != NULL)
+    FAIL_EXIT1 ("aligned_alloc(0, 64) did not fail");
+
+  /* This is an alignment like 0x80000000...UL */
+  p5 = aligned_alloc (SIZE_MAX / 2 + 1, 64);
+
+  if (p5 != NULL)
+    FAIL_EXIT1 ("aligned_alloc(SIZE_MAX/2+1, 64) did not fail");
+
+  free (p1);
+  free (p2);
+  return 0;
+}
+
+#define TEST_FUNCTION do_test ()
+#include "../test-skeleton.c"
diff --git a/manual/memory.texi b/manual/memory.texi
index 0b2b9c8795..5c16f7aee6 100644
--- a/manual/memory.texi
+++ b/manual/memory.texi
@@ -995,7 +995,7 @@ power of two than that, use @code{aligned_alloc} or @code{posix_memalign}.
 @c Alias to memalign.
 The @code{aligned_alloc} function allocates a block of @var{size} bytes whose
 address is a multiple of @var{alignment}.  The @var{alignment} must be a
-power of two and @var{size} must be a multiple of @var{alignment}.
+power of two.
 
 The @code{aligned_alloc} function returns a null pointer on error and sets
 @code{errno} to one of the following values:
-- 
2.45.1

[COMMITTED 2.34 2/2] malloc: Improve aligned_alloc and calloc test coverage.

[Miguel Martin]

From: Joe Simmons-Talbott <josimmon@redhat.com>

Add a DSO (malloc/tst-aligned_alloc-lib.so) that can be used during
testing to interpose malloc with a call that randomly uses either
aligned_alloc, __libc_malloc, or __libc_calloc in the place of malloc.
Use LD_PRELOAD with the DSO to mirror malloc/tst-malloc.c testing as an
example in malloc/tst-malloc-random.c.  Add malloc/tst-aligned-alloc-random.c
as another example that does a number of malloc calls with randomly sized,
but limited to 0xffff, requests.

The intention is to be able to utilize existing malloc testing to ensure
that similar allocation APIs are also exposed to the same rigors.

Reviewed-by: DJ Delorie <dj@redhat.com>
(cherry picked from commit 3395157ff2b0657d70c36169156f67440205c8bf)
---
 malloc/Makefile                   | 16 ++++++-
 malloc/tst-aligned-alloc-random.c | 43 ++++++++++++++++++
 malloc/tst-aligned_alloc-lib.c    | 72 +++++++++++++++++++++++++++++++
 malloc/tst-malloc-random.c        | 20 +++++++++
 malloc/tst-malloc.c               |  4 ++
 5 files changed, 153 insertions(+), 2 deletions(-)
 create mode 100644 malloc/tst-aligned-alloc-random.c
 create mode 100644 malloc/tst-aligned_alloc-lib.c
 create mode 100644 malloc/tst-malloc-random.c

diff --git a/malloc/Makefile b/malloc/Makefile
index b02c49a687..5bd73a32b5 100644
--- a/malloc/Makefile
+++ b/malloc/Makefile
@@ -39,11 +39,13 @@ tests := mallocbug tst-malloc tst-valloc tst-calloc tst-obstack \
 	 tst-malloc-tcache-leak \
 	 tst-malloc_info tst-mallinfo2 \
 	 tst-malloc-too-large \
+	 tst-malloc-random \
 	 tst-malloc-stats-cancellation \
 	 tst-tcfree1 tst-tcfree2 tst-tcfree3 \
 	 tst-safe-linking \
 	 tst-mallocalign1 \
-	 tst-aligned-alloc
+	 tst-aligned-alloc \
+	 tst-aligned-alloc-random
 
 tests-static := \
 	 tst-interpose-static-nothread \
@@ -123,7 +125,8 @@ extra-libs = libmemusage libc_malloc_debug
 extra-libs-others = $(extra-libs)
 
 # Helper objects for some tests.
-extra-tests-objs += \
+extra-test-objs += \
+  tst-aligned_alloc-lib.so \
   tst-interpose-aux-nothread.o \
   tst-interpose-aux-thread.o \
 
@@ -131,6 +134,9 @@ test-extras = \
   tst-interpose-aux-nothread \
   tst-interpose-aux-thread \
 
+modules-names = \
+  tst-aligned_alloc-lib
+
 libmemusage-routines = memusage
 libmemusage-inhibit-o = $(filter-out .os,$(object-suffixes))
 
@@ -329,3 +335,9 @@ tst-mallocstate-malloc-check-ENV = LD_PRELOAD=$(objpfx)libc_malloc_debug.so
 # libc_malloc_debug.so.
 $(objpfx)tst-mallocstate: $(objpfx)libc_malloc_debug.so
 $(objpfx)tst-mallocstate-malloc-check: $(objpfx)libc_malloc_debug.so
+
+$(objpfx)tst-aligned-alloc-random.out: $(objpfx)tst-aligned_alloc-lib.so
+$(objpfx)tst-malloc-random.out: $(objpfx)tst-aligned_alloc-lib.so
+
+tst-aligned-alloc-random-ENV = LD_PRELOAD=$(objpfx)tst-aligned_alloc-lib.so
+tst-malloc-random-ENV = LD_PRELOAD=$(objpfx)tst-aligned_alloc-lib.so
diff --git a/malloc/tst-aligned-alloc-random.c b/malloc/tst-aligned-alloc-random.c
new file mode 100644
index 0000000000..f2825ce38f
--- /dev/null
+++ b/malloc/tst-aligned-alloc-random.c
@@ -0,0 +1,43 @@
+/* Test for randomized malloc that calls aligned_alloc
+   Copyright (C) 2024 Free Software Foundation, Inc.
+   This file is part of the GNU C Library.
+
+   The GNU C Library is free software; you can redistribute it and/or
+   modify it under the terms of the GNU Lesser General Public
+   License as published by the Free Software Foundation; either
+   version 2.1 of the License, or (at your option) any later version.
+
+   The GNU C Library is distributed in the hope that it will be useful,
+   but WITHOUT ANY WARRANTY; without even the implied warranty of
+   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+   Lesser General Public License for more details.
+
+   You should have received a copy of the GNU Lesser General Public
+   License along with the GNU C Library; if not, see
+   <https://www.gnu.org/licenses/>.  */
+
+#include <stdlib.h>
+#include <support/check.h>
+#include <time.h>
+
+static int
+do_test (void)
+{
+  void *p1;
+  int i;
+
+  srandom (time (NULL));
+
+  for (i = 0; i < 1024; i++)
+  {
+    size_t size = random () & 0xffff;
+
+    p1 = malloc (size);
+    TEST_VERIFY (p1 != NULL);
+  }
+
+  return 0;
+}
+
+
+#include <support/test-driver.c>
diff --git a/malloc/tst-aligned_alloc-lib.c b/malloc/tst-aligned_alloc-lib.c
new file mode 100644
index 0000000000..0205df5acf
--- /dev/null
+++ b/malloc/tst-aligned_alloc-lib.c
@@ -0,0 +1,72 @@
+/* Module used for improved aligned_alloc testing.
+   Copyright (C) 2024 Free Software Foundation, Inc.
+   Copyright The GNU Toolchain Authors.
+   This file is part of the GNU C Library.
+
+   The GNU C Library is free software; you can redistribute it and/or
+   modify it under the terms of the GNU Lesser General Public License as
+   published by the Free Software Foundation; either version 2.1 of the
+   License, or (at your option) any later version.
+
+   The GNU C Library is distributed in the hope that it will be useful,
+   but WITHOUT ANY WARRANTY; without even the implied warranty of
+   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+   Lesser General Public License for more details.
+
+   You should have received a copy of the GNU Lesser General Public
+   License along with the GNU C Library; see the file COPYING.LIB.  If
+   not, see <https://www.gnu.org/licenses/>.  */
+
+#include <array_length.h>
+#include <libc-symbols.h>
+#include <stdlib.h>
+
+extern void *__libc_malloc (size_t size);
+extern void *__libc_calloc (size_t n, size_t size);
+
+int aligned_alloc_count = 0;
+int libc_malloc_count = 0;
+int libc_calloc_count = 0;
+
+/* Get a random alignment value.  Biased towards the smaller values.  Must be
+   a power of 2. */
+static size_t get_random_alignment (void)
+{
+  size_t aligns[] = {
+    1, 2, 4, 8, 16, 32, 64, 128, 256, 512, 1024, 2048, 4096, 8192, 16384
+  };
+
+  return aligns[random () % array_length (aligns)];
+}
+
+static void *get_random_alloc (size_t size)
+{
+  void *retval;
+  size_t align;
+
+  switch (random() % 3)
+  {
+    case 1:
+      align = get_random_alignment ();
+      retval = aligned_alloc (align, size);
+      aligned_alloc_count++;
+      break;
+    case 2:
+      retval = __libc_calloc (1, size);
+      libc_calloc_count++;
+      break;
+    default:
+      retval = __libc_malloc (size);
+      libc_malloc_count++;
+      break;
+  }
+
+  return retval;
+}
+
+
+void * __random_malloc (size_t size)
+{
+  return get_random_alloc (size);
+}
+strong_alias (__random_malloc, malloc)
diff --git a/malloc/tst-malloc-random.c b/malloc/tst-malloc-random.c
new file mode 100644
index 0000000000..762b70c918
--- /dev/null
+++ b/malloc/tst-malloc-random.c
@@ -0,0 +1,20 @@
+/* Test malloc with random calls to aligned_alloc and calloc.
+
+   Copyright (C) 2024 Free Software Foundation, Inc.
+   This file is part of the GNU C Library.
+
+   The GNU C Library is free software; you can redistribute it and/or
+   modify it under the terms of the GNU Lesser General Public
+   License as published by the Free Software Foundation; either
+   version 2.1 of the License, or (at your option) any later version.
+
+   The GNU C Library is distributed in the hope that it will be useful,
+   but WITHOUT ANY WARRANTY; without even the implied warranty of
+   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+   Lesser General Public License for more details.
+
+   You should have received a copy of the GNU Lesser General Public
+   License along with the GNU C Library; if not, see
+   <https://www.gnu.org/licenses/>.  */
+
+#include "tst-malloc.c"
diff --git a/malloc/tst-malloc.c b/malloc/tst-malloc.c
index a587cbded6..98be3ee254 100644
--- a/malloc/tst-malloc.c
+++ b/malloc/tst-malloc.c
@@ -19,7 +19,9 @@
 #include <errno.h>
 #include <malloc.h>
 #include <stdio.h>
+#include <stdlib.h>
 #include <libc-diag.h>
+#include <time.h>
 
 static int errors = 0;
 
@@ -36,6 +38,8 @@ do_test (void)
   void *p, *q;
   int save;
 
+  srandom (time (NULL));
+
   errno = 0;
 
   DIAG_PUSH_NEEDS_COMMENT;
-- 
2.45.1

Re: [COMMITTED 2.34 2/2] malloc: Improve aligned_alloc and calloc test coverage.

[Florian Weimer]

* Miguel Martin via Libc-stable:

> From: Joe Simmons-Talbott <josimmon@redhat.com>
>
> Add a DSO (malloc/tst-aligned_alloc-lib.so) that can be used during
> testing to interpose malloc with a call that randomly uses either
> aligned_alloc, __libc_malloc, or __libc_calloc in the place of malloc.
> Use LD_PRELOAD with the DSO to mirror malloc/tst-malloc.c testing as an
> example in malloc/tst-malloc-random.c.  Add malloc/tst-aligned-alloc-random.c
> as another example that does a number of malloc calls with randomly sized,
> but limited to 0xffff, requests.
>
> The intention is to be able to utilize existing malloc testing to ensure
> that similar allocation APIs are also exposed to the same rigors.
>
> Reviewed-by: DJ Delorie <dj@redhat.com>
> (cherry picked from commit 3395157ff2b0657d70c36169156f67440205c8bf)

In general, we don't backport such mere test additions to upstream
release branches, especially if they do not relate to known previous
regressions or might cover potential regressions in changes we backport
later.  One reason is that we like to maintain monoticity in the sense
that if a commit has been backported to 2.X, we also want it backported
to 2.X+1 (if it wasn't already committed during 2.X+1 development).  For
earlier branches such as 2.34, that requires fixing a lot of other
branches, and the branch count is steadily growing over time.  This
effort is better spent elsewhere.

Our downstream procedures do not require that the branch is backported
to the upstream base branch.

Thanks,
Florian

Re: [COMMITTED 2.34 2/2] malloc: Improve aligned_alloc and calloc test coverage.

[Miguel Martín]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

On Tue, 2024-05-21 at 12:42 +0200, Florian Weimer wrote:
> * Miguel Martin via Libc-stable:
> 
> > From: Joe Simmons-Talbott <josimmon@redhat.com>
> > 
> > Add a DSO (malloc/tst-aligned_alloc-lib.so) that can be used during
> > testing to interpose malloc with a call that randomly uses either
> > aligned_alloc, __libc_malloc, or __libc_calloc in the place of
> > malloc.
> > Use LD_PRELOAD with the DSO to mirror malloc/tst-malloc.c testing
> > as an
> > example in malloc/tst-malloc-random.c.  Add malloc/tst-aligned-
> > alloc-random.c
> > as another example that does a number of malloc calls with randomly
> > sized,
> > but limited to 0xffff, requests.
> > 
> > The intention is to be able to utilize existing malloc testing to
> > ensure
> > that similar allocation APIs are also exposed to the same rigors.
> > 
> > Reviewed-by: DJ Delorie <dj@redhat.com>
> > (cherry picked from commit
> > 3395157ff2b0657d70c36169156f67440205c8bf)
> 
> In general, we don't backport such mere test additions to upstream
> release branches, especially if they do not relate to known previous
> regressions or might cover potential regressions in changes we
> backport
> later.  One reason is that we like to maintain monoticity in the
> sense
> that if a commit has been backported to 2.X, we also want it
> backported
> to 2.X+1 (if it wasn't already committed during 2.X+1 development). 
> For
> earlier branches such as 2.34, that requires fixing a lot of other
> branches, and the branch count is steadily growing over time.  This
> effort is better spent elsewhere.
> 
> Our downstream procedures do not require that the branch is
> backported
> to the upstream base branch.
> 
> Thanks,
> Florian
> 

Ok, it make sense.

Then, we only need to backport d141717 to glibc/2.39/master so
all the commits are in upstream 2.39 before backporting them downstream
to 2.34, right?
-----BEGIN PGP SIGNATURE-----
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=ZHmN
-----END PGP SIGNATURE-----

Re: [COMMITTED 2.34 2/2] malloc: Improve aligned_alloc and calloc test coverage.

[Adhemerval Zanella Netto]

On 21/05/24 10:00, Miguel Martín via Libc-stable wrote:
> On Tue, 2024-05-21 at 12:42 +0200, Florian Weimer wrote:
>> * Miguel Martin via Libc-stable:
> 
>>> From: Joe Simmons-Talbott <josimmon@redhat.com>
>>>
>>> Add a DSO (malloc/tst-aligned_alloc-lib.so) that can be used during
>>> testing to interpose malloc with a call that randomly uses either
>>> aligned_alloc, __libc_malloc, or __libc_calloc in the place of
>>> malloc.
>>> Use LD_PRELOAD with the DSO to mirror malloc/tst-malloc.c testing
>>> as an
>>> example in malloc/tst-malloc-random.c.  Add malloc/tst-aligned-
>>> alloc-random.c
>>> as another example that does a number of malloc calls with randomly
>>> sized,
>>> but limited to 0xffff, requests.
>>>
>>> The intention is to be able to utilize existing malloc testing to
>>> ensure
>>> that similar allocation APIs are also exposed to the same rigors.
>>>
>>> Reviewed-by: DJ Delorie <dj@redhat.com>
>>> (cherry picked from commit
>>> 3395157ff2b0657d70c36169156f67440205c8bf)
> 
>> In general, we don't backport such mere test additions to upstream
>> release branches, especially if they do not relate to known previous
>> regressions or might cover potential regressions in changes we
>> backport
>> later.  One reason is that we like to maintain monoticity in the
>> sense
>> that if a commit has been backported to 2.X, we also want it
>> backported
>> to 2.X+1 (if it wasn't already committed during 2.X+1 development). 
>> For
>> earlier branches such as 2.34, that requires fixing a lot of other
>> branches, and the branch count is steadily growing over time.  This
>> effort is better spent elsewhere.
> 
>> Our downstream procedures do not require that the branch is
>> backported
>> to the upstream base branch.
> 
>> Thanks,
>> Florian
> 
> 
> Ok, it make sense.
> 
> Then, we only need to backport d141717 to glibc/2.39/master so
> all the commits are in upstream 2.39 before backporting them downstream
> to 2.34, right?
> 

Also, do you plan to backport it to all other affected releases branches?
Because it does not make much sense to have it only for 2.34, 2.39, and
master.

Re: [COMMITTED 2.34 2/2] malloc: Improve aligned_alloc and calloc test coverage.

[Miguel Martín]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

On Tue, 2024-05-21 at 10:06 -0300, Adhemerval Zanella Netto wrote:
> 
> 
> On 21/05/24 10:00, Miguel Martín via Libc-stable wrote:
> > On Tue, 2024-05-21 at 12:42 +0200, Florian Weimer wrote:
> > > * Miguel Martin via Libc-stable:
> > 
> > > > From: Joe Simmons-Talbott <josimmon@redhat.com>
> > > > 
> > > > Add a DSO (malloc/tst-aligned_alloc-lib.so) that can be used
> > > > during
> > > > testing to interpose malloc with a call that randomly uses
> > > > either
> > > > aligned_alloc, __libc_malloc, or __libc_calloc in the place of
> > > > malloc.
> > > > Use LD_PRELOAD with the DSO to mirror malloc/tst-malloc.c
> > > > testing
> > > > as an
> > > > example in malloc/tst-malloc-random.c.  Add malloc/tst-aligned-
> > > > alloc-random.c
> > > > as another example that does a number of malloc calls with
> > > > randomly
> > > > sized,
> > > > but limited to 0xffff, requests.
> > > > 
> > > > The intention is to be able to utilize existing malloc testing
> > > > to
> > > > ensure
> > > > that similar allocation APIs are also exposed to the same
> > > > rigors.
> > > > 
> > > > Reviewed-by: DJ Delorie <dj@redhat.com>
> > > > (cherry picked from commit
> > > > 3395157ff2b0657d70c36169156f67440205c8bf)
> > 
> > > In general, we don't backport such mere test additions to
> > > upstream
> > > release branches, especially if they do not relate to known
> > > previous
> > > regressions or might cover potential regressions in changes we
> > > backport
> > > later.  One reason is that we like to maintain monoticity in the
> > > sense
> > > that if a commit has been backported to 2.X, we also want it
> > > backported
> > > to 2.X+1 (if it wasn't already committed during 2.X+1
> > > development). 
> > > For
> > > earlier branches such as 2.34, that requires fixing a lot of
> > > other
> > > branches, and the branch count is steadily growing over time. 
> > > This
> > > effort is better spent elsewhere.
> > 
> > > Our downstream procedures do not require that the branch is
> > > backported
> > > to the upstream base branch.
> > 
> > > Thanks,
> > > Florian
> > 
> > 
> > Ok, it make sense.
> > 
> > Then, we only need to backport d141717 to glibc/2.39/master so
> > all the commits are in upstream 2.39 before backporting them
> > downstream
> > to 2.34, right?
> > 
> 
> Also, do you plan to backport it to all other affected releases
> branches?
> Because it does not make much sense to have it only for 2.34, 2.39,
> and
> master.
> 

No, AFAIU I only need to backport it to 2.39, the 2.34 backport will be
done downstream.
-----BEGIN PGP SIGNATURE-----
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=FzBj
-----END PGP SIGNATURE-----

Re: [COMMITTED 2.34 2/2] malloc: Improve aligned_alloc and calloc test coverage.

[Sam James]

Adhemerval Zanella Netto via Libc-stable <libc-stable@sourceware.org> writes:

> On 21/05/24 10:00, Miguel Martín via Libc-stable wrote:
>> On Tue, 2024-05-21 at 12:42 +0200, Florian Weimer wrote:
>>> * Miguel Martin via Libc-stable:
>> 
>>>> From: Joe Simmons-Talbott <josimmon@redhat.com>
>>>>
>>>> Add a DSO (malloc/tst-aligned_alloc-lib.so) that can be used during
>>>> testing to interpose malloc with a call that randomly uses either
>>>> aligned_alloc, __libc_malloc, or __libc_calloc in the place of
>>>> malloc.
>>>> Use LD_PRELOAD with the DSO to mirror malloc/tst-malloc.c testing
>>>> as an
>>>> example in malloc/tst-malloc-random.c.  Add malloc/tst-aligned-
>>>> alloc-random.c
>>>> as another example that does a number of malloc calls with randomly
>>>> sized,
>>>> but limited to 0xffff, requests.
>>>>
>>>> The intention is to be able to utilize existing malloc testing to
>>>> ensure
>>>> that similar allocation APIs are also exposed to the same rigors.
>>>>
>>>> Reviewed-by: DJ Delorie <dj@redhat.com>
>>>> (cherry picked from commit
>>>> 3395157ff2b0657d70c36169156f67440205c8bf)
>> 
>>> In general, we don't backport such mere test additions to upstream
>>> release branches, especially if they do not relate to known previous
>>> regressions or might cover potential regressions in changes we
>>> backport
>>> later.  One reason is that we like to maintain monoticity in the
>>> sense
>>> that if a commit has been backported to 2.X, we also want it
>>> backported
>>> to 2.X+1 (if it wasn't already committed during 2.X+1 development). 
>>> For
>>> earlier branches such as 2.34, that requires fixing a lot of other
>>> branches, and the branch count is steadily growing over time.  This
>>> effort is better spent elsewhere.
>> 
>>> Our downstream procedures do not require that the branch is
>>> backported
>>> to the upstream base branch.
>> 
>>> Thanks,
>>> Florian
>> 
>> 
>> Ok, it make sense.
>> 
>> Then, we only need to backport d141717 to glibc/2.39/master so
>> all the commits are in upstream 2.39 before backporting them downstream
>> to 2.34, right?
>> 
>
> Also, do you plan to backport it to all other affected releases branches?
> Because it does not make much sense to have it only for 2.34, 2.39, and
> master.

Yes, as Florian said, it should be monotonic.

(I don't know if we have that documented, though, and maybe this
discussion shows it needs improvement.)

Re: [COMMITTED 2.34 2/2] malloc: Improve aligned_alloc and calloc test coverage.

[Florian Weimer]

* Miguel Martín:

> Then, we only need to backport d141717 to glibc/2.39/master so
> all the commits are in upstream 2.39 before backporting them downstream
> to 2.34, right?

We don't need to backport this to 2.39, either.

Thanks,
Florian

Re: [COMMITTED 2.34 0/2] Improve aligned_alloc and calloc test coverage

[Florian Weimer]

* Miguel Martin via Libc-stable:

> From: Miguel Martín <mmartinv@redhat.com>
>
> Backports to improve aligned_alloc and calloc test coverage in glibc 2.34:
>
> DJ Delorie (1):
>   aligned_alloc: conform to C17
>
> Joe Simmons-Talbott (1):
>   malloc: Improve aligned_alloc and calloc test coverage.

As far as I can see, none of these patches have actually been committed
(pushed to the official repository).

Thanks,
Florian

Re: [COMMITTED 2.34 0/2] Improve aligned_alloc and calloc test coverage

[Miguel Martín]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

On Tue, 2024-05-21 at 12:43 +0200, Florian Weimer wrote:
> * Miguel Martin via Libc-stable:
> 
> > From: Miguel Martín <mmartinv@redhat.com>
> > 
> > Backports to improve aligned_alloc and calloc test coverage in
> > glibc 2.34:
> > 
> > DJ Delorie (1):
> >   aligned_alloc: conform to C17
> > 
> > Joe Simmons-Talbott (1):
> >   malloc: Improve aligned_alloc and calloc test coverage.
> 
> As far as I can see, none of these patches have actually been
> committed
> (pushed to the official repository).
> 
> Thanks,
> Florian
> 

Both are in master and the first one has been already backported to
release/2.39/master:

git branch --contains d1417176a35d27ffb8da0ffb1e33154163b6eeb2
  master
  release/2.39/master

git show -s d1417176a35d27ffb8da0ffb1e33154163b6eeb2
commit d1417176a35d27ffb8da0ffb1e33154163b6eeb2
Author: DJ Delorie <dj@redhat.com>
Date:   Tue Mar 21 00:46:43 2023 -0400

    aligned_alloc: conform to C17
    
    This patch adds the strict checking for power-of-two alignments
    in aligned_alloc(), and updates the manual accordingly.
    
    Reviewed-by: Carlos O'Donell <carlos@redhat.com>

git branch --contains 3395157ff2b0657d70c36169156f67440205c8bf
  master

git show -s 3395157ff2b0657d70c36169156f67440205c8bf
commit 3395157ff2b0657d70c36169156f67440205c8bf
Author: Joe Simmons-Talbott <josimmon@redhat.com>
Date:   Tue May 14 14:36:50 2024 +0000

    malloc: Improve aligned_alloc and calloc test coverage.
    
    Add a DSO (malloc/tst-aligned_alloc-lib.so) that can be used during
    testing to interpose malloc with a call that randomly uses either
    aligned_alloc, __libc_malloc, or __libc_calloc in the place of    
malloc.
    Use LD_PRELOAD with the DSO to mirror malloc/tst-malloc.c testing 
as an
    example in malloc/tst-malloc-random.c.  Add malloc/tst-aligned-   
alloc-random.c
    as another example that does a number of malloc calls with randomly
sized,
    but limited to 0xffff, requests.
    
    The intention is to be able to utilize existing malloc testing to 
ensure
    that similar allocation APIs are also exposed to the same rigors.
    
    Reviewed-by: DJ Delorie <dj@redhat.com>



-----BEGIN PGP SIGNATURE-----
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=DXW+
-----END PGP SIGNATURE-----

Re: [COMMITTED 2.34 0/2] Improve aligned_alloc and calloc test coverage

[Miguel Martín]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

On Tue, 2024-05-21 at 12:43 +0200, Florian Weimer wrote:
> * Miguel Martin via Libc-stable:
> 
> > From: Miguel Martín <mmartinv@redhat.com>
> > 
> > Backports to improve aligned_alloc and calloc test coverage in
> > glibc 2.34:
> > 
> > DJ Delorie (1):
> >   aligned_alloc: conform to C17
> > 
> > Joe Simmons-Talbott (1):
> >   malloc: Improve aligned_alloc and calloc test coverage.
> 
> As far as I can see, none of these patches have actually been
> committed
> (pushed to the official repository).
> 
> Thanks,
> Florian
> 

After talking to Arjun I think I know what you actually mean and the
answer is no, none of them have been pushed to the upstream repos, the
subject prefix should be PATCH, not COMMITTED. Sorry for the confusion.

-----BEGIN PGP SIGNATURE-----

iQIzBAEBCAAdFiEERBNxlsW9fRmqDQAeOZwMQZdKxwAFAmZMmNUACgkQOZwMQZdK
xwAqIRAA3rVYQTnHa1vld0Zcz4AsMyTJIy5BZX16yvqWt5vIdw+L9Vfg38XJJldT
3hgkeJ8dVZcyWXiDvquR00KieYagSm10PXOcCX38TmFPHHyOivI5icxaLDbb5ph/
u9xzq0jAgIIe5B5xO2nEFKPYKoK0APH5Kjk4orzJS+YY47ZXQgFDTNwVVF5yLZBa
NvCg40ruNPI3Zz5MPtDQu3KUEiYsxbezf5SyrvQeuCEs71YCh7PCmUboGOcxExJ4
NYvdxdGPqyxOQ36SpiN24fANThjXkuCdvYo9q95kfbkwmm60iV2sH8wvoVtSaZx6
pNDjTkvjMkpZkXPytrZGS78lhSv1YntWNimidKUmKw9zOU9nnOzD3YXnDI8eDx+b
5GUaG4BsgK9STX4Dd5KbQJzMdOKE4kf36xxQ8nO7zATKUei2EHgjnIf7C7E97amG
mUiScQVEC5i6eOa5bBxLU1wjK39wXh8RRt01xKrj538BRFagGNsbAhF+kI9DMH3e
mZl67npQ3U+8xn99bAcAM+M5XQNBdF/6EF2WDpgry+qUj81G+eixpbq9fAOtloAd
VxCdridJ1Yh4TNNBBSGKMjNXHDmAQRRTViwCwJu9jG/RHCkhNLRVxumv4rz5GLcV
mXEhYPpWcqeaKmrWe+FUrmPw6DYm8NcdHr1x2P/pBvP3QhbXifw=
=XGlw
-----END PGP SIGNATURE-----