bpo-37471: mmap module adding FreeBSD specific flag into the constants #14513
Conversation
There was a problem hiding this comment.
Would it be possible to document the constant in https://docs.python.org/dev/library/mmap.html ?
It seems like other MAP_xxx constants are not documented yet and so should be documented as well.
|
They are (shortly) in the flags section. Ok for adding the new constant too. |
|
The minimum doc should be a list of constants, like: https://docs.python.org/dev/library/os.html#os.O_RDONLY |
|
Fair enough ! ought to be before madvise's I think. |
Doc/library/mmap.rst
Outdated
| @@ -81,7 +81,11 @@ To map anonymous memory, -1 should be passed as the fileno along with the length | |||
| private copy-on-write mapping, so changes to the contents of the mmap | |||
| object will be private to this process, and :const:`MAP_SHARED` creates a | |||
| mapping that's shared with all other processes mapping the same areas of | |||
| the file. The default value is :const:`MAP_SHARED`. | |||
| the file. :const:`MAP_ALIGNED_SUPER` (FreeBSD only) allows to map to | |||
| pages of 2MB (x86) or so, it is enabled by default, at boot time, | |||
There was a problem hiding this comment.
| pages of 2MB (x86) or so, it is enabled by default, at boot time, | |
| pages of 2 MiB (x86) or so, it is enabled by default, at boot time, |
Is it x86 (32-bit) or x86-64 (64-bit)?
There was a problem hiding this comment.
Write maybe (x86, x86-64) in this case... But I'm not sure if Python documentation should go too far in the description, since FreeBSD may decide to add support for MAP_ALIGNED_SUPER for other platforms.
There was a problem hiding this comment.
It supports more than x86 already in fact but you re right I ll try to find more concise wording
| @@ -0,0 +1 @@ | |||
| Adding new 'MAP_ALIGNED_SUPER' FreeBSD constant. | |||
There was a problem hiding this comment.
| Adding new 'MAP_ALIGNED_SUPER' FreeBSD constant. | |
| Adding new :data:`mmap.MAP_ALIGNED_SUPER` FreeBSD constant. |
| MAP_PRIVATE | ||
| MAP_DENYWRITE | ||
| MAP_EXECUTABLE | ||
| MAP_ALIGNED_SUPER |
There was a problem hiding this comment.
You have to document that MAP_ALIGNED_SUPER is new, example:
.. versionadded:: 3.9
Add :data:`MAP_ALIGNED_SUPER` constant.
Doc/library/mmap.rst
Outdated
| the file. :const:`MAP_ALIGNED_SUPER` (FreeBSD only) allows to map to | ||
| pages of 2MB (x86) or so, it is enabled by default, at boot time, | ||
| in most of architectures and exists since 2008. | ||
| :ref:`MAP_* constants <mmap-constants>`. |
There was a problem hiding this comment.
"MAP_* constants" is not really a sentence. Maybe move it after the default, and write something like:
The default value is :const:`MAP_SHARED`.
See :ref:`MAP_* constants <mmap-constants>`.
Doc/library/mmap.rst
Outdated
| @@ -341,3 +359,6 @@ MADV_* Constants | |||
| Availability: Systems with the madvise() system call. | |||
|
|
|||
| .. versionadded:: 3.8 | |||
|
|
|||
| .. versionadded:: 3.9 | |||
| Add :data:`MAP_ALIGNED_SUPER` constant. | |||
There was a problem hiding this comment.
Can you move this to the your new MAP_* Constants section?
| @@ -0,0 +1 @@ | |||
| Adding new :data:`mmap.MAP_CONCEAL` OpenBSD constant. | |||
There was a problem hiding this comment.
Can you try to merge the 2 changelog entries? Put content of the 2nd entry in the first one, remove the second file.
| MAP_DENYWRITE | ||
| MAP_EXECUTABLE | ||
| MAP_ALIGNED_SUPER | ||
| MAP_CONCEAL |
There was a problem hiding this comment.
You have to document that MAP_CONCEAL is new.
Doc/library/mmap.rst
Outdated
| the file. The default value is :const:`MAP_SHARED`. | ||
| the file. :const:`MAP_ALIGNED_SUPER` (FreeBSD only) to use | ||
| super pages feature, enabled by default in most of architectures | ||
| and exists since 2008. |
There was a problem hiding this comment.
I don't understand this sentence :-/ I suggest:
:const:`MAP_ALIGNED_SUPER` (FreeBSD only) uses super pages feature,
it is enabled by default in most of architectures and exists since 2008.
https://bugs.python.org/issue37471