SharedMemory documentation: System V vs Posix

[nescio007]

BPO	45728
Nosy	@terryjreedy, @pitrou, @applio, @nescio007

^{Note: these values reflect the state of the issue at the time it was migrated and might not reflect the current state.}

Show more details

GitHub fields:

assignee = None
closed_at = None
created_at = <Date 2021-11-05.13:46:16.151>
labels = ['3.11', 'type-feature', '3.9', '3.10', 'docs']
title = 'SharedMemory documentation: System V vs Posix'
updated_at = <Date 2021-11-06.02:30:17.465>
user = 'https://github.com/nescio007'

bugs.python.org fields:

activity = <Date 2021-11-06.02:30:17.465>
actor = 'terry.reedy'
assignee = 'docs@python'
closed = False
closed_date = None
closer = None
components = ['Documentation']
creation = <Date 2021-11-05.13:46:16.151>
creator = 'jkrupp'
dependencies = []
files = []
hgrepos = []
issue_num = 45728
keywords = []
message_count = 2.0
messages = ['405797', '405843']
nosy_count = 5.0
nosy_names = ['terry.reedy', 'pitrou', 'docs@python', 'davin', 'jkrupp']
pr_nums = []
priority = 'normal'
resolution = None
stage = None
status = 'open'
superseder = None
type = 'enhancement'
url = 'https://bugs.python.org/issue45728'
versions = ['Python 3.9', 'Python 3.10', 'Python 3.11']

Linked PRs

• gh-89891: Refer SharedMemory implementation as POSIX #104678
• [3.12] gh-89891: Refer SharedMemory implementation as POSIX (GH-104678) #114873
• [3.11] gh-89891: Refer SharedMemory implementation as POSIX (GH-104678) #114874

[nescio007]

Hi,

the documentation for the shared_memory module states that "shared memory refers to “System V style” shared memory". However, I believe it might be clearer to talk about "POSIX style shared memory" here instead.

Primed by the "System V style"-comment, I had expected that shared memory blocks could be identified through a numeric ID (as used by the system V shared memory APIs shmget, shmat, ...). After looking at the implementation it became clear though, that the newer POSIX-API (shm_open, shm_unlink, ...) is used, which explains why shared memory blocks have a name instead.

Technically, the documentation only uses "System V style" to distinguish it from "distributed shared memory", and further states that "[it] is not necessarily implemented explicitly as such". Yet, this distinction is particularly relevant if memory is to be shared with other (non-python!) processes (although placement of shared_memory in the multiprocessing package might indicate that that is not one of its expected use-cases...)

[terryjreedy]

'System V' is in the first line of the second paragraph of
https://docs.python.org/3/library/multiprocessing.shared_memory.html

Davin, git blame says you wrote this in Feb 2019.

[torokati44]

This note in the docs also confused me (when trying to switch away from the posix_ipc package), and had to check the source that it is indeed "POSIX-style" instead.

See the answer by Terry Lambert here about the differences: https://www.quora.com/What-are-the-major-differences-between-System-V-and-POSIX-Shared-memory-implementations-available-on-Linux

I think that note should not reference "System V" as it is not correct, and only causes confusion.