Re: [PATCH] mingw: enable atomic O_APPEND

From: Junio C Hamano <gitster@pobox.com>
Date: Aug 13, 2018, 21:55 UTC
Message-ID: <xmqqo9e62aos.fsf@gitster-ct.c.googlers.com>
In-Reply-To: <87eff2rmgt.fsf@evledraar.gmail.com>

Ævar Arnfjörð Bjarmason <avarab@gmail.com> writes:

> O_APPEND in POSIX is race-free only up to PIPE_MAX bytes written at a
> time, which is e.g. 2^12 by default on linux, after that all bets are
> off and the kernel is free to interleave different write calls.

We are only interested in the use of race-safe append in the tracing, so we should be OK in practice. but then ...

> So there's still a use for a race_safe_append_open() wrapper function,
> to O_APPEND and do the PIPE_MAX assertion. Otherwise you're calling a
> "safe" function which isn't safe at all anymore.

... having an assert that would never trigger is not all that bad ;-)

$ cat FOOTERMessages come from the public archive at lore.kernel.org/git, fetched every hour. The front page is chosen and written each morning by an AI editor and can be wrong; the threads themselves are the record. About and API. For agents: an MCP server at https://gitlist.dev/mcp, and any thread, story or person page as Markdown by adding .md to its URL (or sending Accept: text/markdown). Details in /llms.txt.