<feed xmlns='http://www.w3.org/2005/Atom'>
<title>delta/ocaml.git/Changes, branch 4.05</title>
<subtitle>github.com: ocaml/ocaml
</subtitle>
<link rel='alternate' type='text/html' href='http://91.123.203.49/cgit/delta/ocaml.git/'/>
<entry>
<title>Changes for 1330</title>
<updated>2022-12-21T17:38:30+00:00</updated>
<author>
<name>David Allsopp</name>
<email>david.allsopp@metastack.com</email>
</author>
<published>2022-12-17T12:27:15+00:00</published>
<link rel='alternate' type='text/html' href='http://91.123.203.49/cgit/delta/ocaml.git/commit/?id=b94a2820d0a039e88a9affd2ac577ebe56afc5f9'/>
<id>b94a2820d0a039e88a9affd2ac577ebe56afc5f9</id>
<content type='text'>
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
</pre>
</div>
</content>
</entry>
<entry>
<title>Update Changes</title>
<updated>2022-01-17T11:15:41+00:00</updated>
<author>
<name>Xavier Leroy</name>
<email>xavierleroy@users.noreply.github.com</email>
</author>
<published>2021-03-05T18:14:07+00:00</published>
<link rel='alternate' type='text/html' href='http://91.123.203.49/cgit/delta/ocaml.git/commit/?id=0d2c1fd77b4715c70d0ffc4348f1bc2d3eaacbf7'/>
<id>0d2c1fd77b4715c70d0ffc4348f1bc2d3eaacbf7</id>
<content type='text'>
Updated separately to allow the previous patch to be used for multiple
releases.
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Updated separately to allow the previous patch to be used for multiple
releases.
</pre>
</div>
</content>
</entry>
<entry>
<title>Define __USE_MINGW_ANSI_STDIO=0 for mingw-w64 (#9939)</title>
<updated>2020-11-27T16:27:19+00:00</updated>
<author>
<name>David Allsopp</name>
<email>david.allsopp@metastack.com</email>
</author>
<published>2020-11-27T09:29:35+00:00</published>
<link rel='alternate' type='text/html' href='http://91.123.203.49/cgit/delta/ocaml.git/commit/?id=c684ef1cb76d67a59dbf395040303b26fc5b7931'/>
<id>c684ef1cb76d67a59dbf395040303b26fc5b7931</id>
<content type='text'>
The latest mingw-w64 _mingw.h header now automatically sets __USE_MINGW_ANSI_STDIO=1 for C99 and later. mingw-w64 always defines snprintf but the inline definition when __USE_MINGW_ANSI_STDIO=0 temporarily #undef's any snprintf macro so works in harmony with ours. The other case does not do this so we get declaration errors.

Fixes: #9938 .
(cherry picked from commit a03b6035ebe4682950018e0e1337bab18771fd81)
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
The latest mingw-w64 _mingw.h header now automatically sets __USE_MINGW_ANSI_STDIO=1 for C99 and later. mingw-w64 always defines snprintf but the inline definition when __USE_MINGW_ANSI_STDIO=0 temporarily #undef's any snprintf macro so works in harmony with ours. The other case does not do this so we get declaration errors.

Fixes: #9938 .
(cherry picked from commit a03b6035ebe4682950018e0e1337bab18771fd81)
</pre>
</div>
</content>
</entry>
<entry>
<title>Fix integer overflows when unmarshaling a bigarray</title>
<updated>2019-07-22T13:54:04+00:00</updated>
<author>
<name>Stephane Glondu</name>
<email>steph@glondu.net</email>
</author>
<published>2019-01-25T13:34:23+00:00</published>
<link rel='alternate' type='text/html' href='http://91.123.203.49/cgit/delta/ocaml.git/commit/?id=c6ca3afc78b75d7748e4e09e56c6b020418be06e'/>
<id>c6ca3afc78b75d7748e4e09e56c6b020418be06e</id>
<content type='text'>
Malicious or corrupted marshaled data can result in a bigarray
with impossibly large dimensions that cause overflow when computing
the in-memory size of the bigarray.  Disaster ensues when the data
is read in a too small memory area.  This commit checks for overflows
when computing the in-memory size of the bigarray.

This patch is based on one by Xavier Leroy and has been modified to
use caml_ba_multov instead of caml_umul_overflow which is unavailable
in OCaml 4.05.0.

The original commit hash is 85162eee9d4072fa9c2f498f03cd94e357033eec.

Origin: https://github.com/ocaml/ocaml/pull/1718
Bug: https://github.com/ocaml/ocaml/issues/7765
Bug-Debian: https://bugs.debian.org/895472
Bug-CVE: CVE-2018-9838
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Malicious or corrupted marshaled data can result in a bigarray
with impossibly large dimensions that cause overflow when computing
the in-memory size of the bigarray.  Disaster ensues when the data
is read in a too small memory area.  This commit checks for overflows
when computing the in-memory size of the bigarray.

This patch is based on one by Xavier Leroy and has been modified to
use caml_ba_multov instead of caml_umul_overflow which is unavailable
in OCaml 4.05.0.

The original commit hash is 85162eee9d4072fa9c2f498f03cd94e357033eec.

Origin: https://github.com/ocaml/ocaml/pull/1718
Bug: https://github.com/ocaml/ocaml/issues/7765
Bug-Debian: https://bugs.debian.org/895472
Bug-CVE: CVE-2018-9838
</pre>
</div>
</content>
</entry>
<entry>
<title>debugger: avoid off-by-one overflow in debugger socket path (#1753)</title>
<updated>2019-07-19T14:39:37+00:00</updated>
<author>
<name>Anil Madhavapeddy</name>
<email>anil@recoil.org</email>
</author>
<published>2018-05-04T16:54:11+00:00</published>
<link rel='alternate' type='text/html' href='http://91.123.203.49/cgit/delta/ocaml.git/commit/?id=b79a3a37ded5e2041fa80f5fbd2ab283f2388730'/>
<id>b79a3a37ded5e2041fa80f5fbd2ab283f2388730</id>
<content type='text'>
Modern gcc have a `-Werror=stringop-truncation` warning that tells
us that the strncpy used to copy the CAML_DEBUG_SOCKET value from
the environment misses out the space for a terminating NUL.

Rather than silently truncate the path, this patch raises an
exception if the path is too long, and also fixes the off-by-one
in the strncpy invocation.

with feedback from Stephen Dolan

(cherry picked from commit 698f64825efd3c3f1344c69c2c75cf309bd9c428)
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Modern gcc have a `-Werror=stringop-truncation` warning that tells
us that the strncpy used to copy the CAML_DEBUG_SOCKET value from
the environment misses out the space for a terminating NUL.

Rather than silently truncate the path, this patch raises an
exception if the path is too long, and also fixes the off-by-one
in the strncpy invocation.

with feedback from Stephen Dolan

(cherry picked from commit 698f64825efd3c3f1344c69c2c75cf309bd9c428)
</pre>
</div>
</content>
</entry>
<entry>
<title>byterun: do not alias function arguments to sigprocmask</title>
<updated>2019-07-19T14:10:55+00:00</updated>
<author>
<name>Anil Madhavapeddy</name>
<email>anil@recoil.org</email>
</author>
<published>2018-05-02T08:44:10+00:00</published>
<link rel='alternate' type='text/html' href='http://91.123.203.49/cgit/delta/ocaml.git/commit/?id=b35e3d48bc392f764e1edc99ce6a1b1a68358ed4'/>
<id>b35e3d48bc392f764e1edc99ce6a1b1a68358ed4</id>
<content type='text'>
Recent gcc versions (e.g. as found in Fedora 28) check that
restrict-qualified parameters do not alias each other in
function calls. One such example that triggers a compile
error is sigprocmask in the bytecode runtime.

The error this changeset fixes is:

```
signals.c: In function 'caml_execute_signal':
signals.c:152:33: error: passing argument 3 to restrict-qualified parameter aliases with argument 2 [-Werror=restrict]
   sigprocmask(SIG_BLOCK, &amp;sigs, &amp;sigs);
```

(cherry picked from commit ebcc2f8594138a5ba6b020083fc42d57a914433f)
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Recent gcc versions (e.g. as found in Fedora 28) check that
restrict-qualified parameters do not alias each other in
function calls. One such example that triggers a compile
error is sigprocmask in the bytecode runtime.

The error this changeset fixes is:

```
signals.c: In function 'caml_execute_signal':
signals.c:152:33: error: passing argument 3 to restrict-qualified parameter aliases with argument 2 [-Werror=restrict]
   sigprocmask(SIG_BLOCK, &amp;sigs, &amp;sigs);
```

(cherry picked from commit ebcc2f8594138a5ba6b020083fc42d57a914433f)
</pre>
</div>
</content>
</entry>
<entry>
<title>Revised configuration test for int64 alignment (#1371)</title>
<updated>2017-09-27T14:00:53+00:00</updated>
<author>
<name>Xavier Leroy</name>
<email>xavier.leroy@inria.fr</email>
</author>
<published>2017-09-27T14:00:53+00:00</published>
<link rel='alternate' type='text/html' href='http://91.123.203.49/cgit/delta/ocaml.git/commit/?id=23eb28c41ca175284b165ae4da58749afd37fcc6'/>
<id>23eb28c41ca175284b165ae4da58749afd37fcc6</id>
<content type='text'>
MacOS High Sierra balks at the redefinition of `int64_t`.
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
MacOS High Sierra balks at the redefinition of `int64_t`.
</pre>
</div>
</content>
</entry>
<entry>
<title>Revert "Merge pull request #1272 from garrigue/fix_mpr7601"</title>
<updated>2017-09-06T12:43:01+00:00</updated>
<author>
<name>Jacques Garrigue</name>
<email>garrigue@math.nagoya-u.ac.jp</email>
</author>
<published>2017-09-06T12:43:01+00:00</published>
<link rel='alternate' type='text/html' href='http://91.123.203.49/cgit/delta/ocaml.git/commit/?id=9d15fa031fb77db8682325460a0cec3f8ae7dd35'/>
<id>9d15fa031fb77db8682325460a0cec3f8ae7dd35</id>
<content type='text'>
This reverts commit a38350d5c02d021657c18a36b56d79ef669f7054, reversing
changes made to af92b8137538385734417d628dd2905232900791.
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
This reverts commit a38350d5c02d021657c18a36b56d79ef669f7054, reversing
changes made to af92b8137538385734417d628dd2905232900791.
</pre>
</div>
</content>
</entry>
<entry>
<title>Fix MPR#7601</title>
<updated>2017-08-29T00:39:52+00:00</updated>
<author>
<name>Jacques Garrigue</name>
<email>garrigue@math.nagoya-u.ac.jp</email>
</author>
<published>2017-08-29T00:39:52+00:00</published>
<link rel='alternate' type='text/html' href='http://91.123.203.49/cgit/delta/ocaml.git/commit/?id=34abd62a638a4e4b4a744eab455fa2ed83941317'/>
<id>34abd62a638a4e4b4a744eab455fa2ed83941317</id>
<content type='text'>
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
</pre>
</div>
</content>
</entry>
<entry>
<title>Take PLT-clobbered registers into account at Ialloc (#1304)</title>
<updated>2017-08-28T17:17:27+00:00</updated>
<author>
<name>Mark Shinwell</name>
<email>mshinwell@gmail.com</email>
</author>
<published>2017-08-28T17:09:57+00:00</published>
<link rel='alternate' type='text/html' href='http://91.123.203.49/cgit/delta/ocaml.git/commit/?id=af92b8137538385734417d628dd2905232900791'/>
<id>af92b8137538385734417d628dd2905232900791</id>
<content type='text'>
Mark PLT-clobbered registers as destroyed across the Ialloc instruction.

Currently only x86-64 is affected, in PIC mode only, and only with the glibc dynamic loader.

Cherry-pick of [trunk 70c585d]
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Mark PLT-clobbered registers as destroyed across the Ialloc instruction.

Currently only x86-64 is affected, in PIC mode only, and only with the glibc dynamic loader.

Cherry-pick of [trunk 70c585d]
</pre>
</div>
</content>
</entry>
</feed>
