git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Patrick Steinhardt <ps@pks.im>
To: Johannes Sixt <j6t@kdbg.org>
Cc: git@vger.kernel.org,
	Johannes Schindelin <johannes.schindelin@gmx.de>,
	Taylor Blau <me@ttaylorr.com>,
	Kristoffer Haugsbakk <kristofferhaugsbakk@fastmail.com>
Subject: Re: [PATCH v2 3/3] compat/mingw: support POSIX semantics for atomic renames
Date: Sun, 27 Oct 2024 18:27:35 +0100	[thread overview]
Message-ID: <Zx54B6FdGprtH0HV@pks.im> (raw)
In-Reply-To: <49619b52-0fea-4179-a829-7ec4a6945055@kdbg.org>

On Sun, Oct 27, 2024 at 05:31:00PM +0100, Johannes Sixt wrote:
> Am 27.10.24 um 16:38 schrieb Patrick Steinhardt:
> > On Sun, Oct 27, 2024 at 02:23:28PM +0100, Johannes Sixt wrote:
> >> Am 24.10.24 um 13:46 schrieb Patrick Steinhardt:
> >>> Windows 10 has introduced the `FILE_RENAME_FLAG_POSIX_SEMANTICS` flag
> >>> that allows us to fix this usecase [1]. When set, it is possible to
> >>> rename a file over a preexisting file even when the target file still
> >>> has handles open. Those handles must have been opened with the
> >>> `FILE_SHARE_DELETE` flag, which we have ensured in the preceding
> >>> commits.
> >>>> Careful readers might have noticed that [1] does not mention the above
> >>> flag, but instead mentions `FILE_RENAME_POSIX_SEMANTICS`. This flag is
> >>> not for use with `SetFileInformationByHandle()` though, which is what we
> >>> use. And while the `FILE_RENAME_FLAG_POSIX_SEMANTICS` flag exists, it is
> >>> not documented on [2] or anywhere else as far as I can tell.
> >>
> >> The Windows 10 SDK defines FILE_RENAME_FLAG_REPLACE_IF_EXISTS and
> >> FILE_RENAME_FLAG_POSIX_SEMANTICS for SetFileInformationByHandle(). That
> >> the documentation lacks "_FLAG_" in the names must be an error in the
> >> documentation.
> >>
> >> I found the mention of FILE_RENAME_POSIX_SEMANTICS quite distracting,
> >> because it is a flag to be used with CreateFileW() and basically only
> >> has to do with case-sensitivity, but nothing with POSIX semantics of
> >> renaming.
> > 
> > I'd still prefer to mention this, because otherwise an astute reader
> > might notice that I'm using a different flag name than what is
> > documented in the docs and figure out that I defined the wrong flag
> > name.
> 
> Ah, I was confused twice here. First, the documentation that you cite[*]
> mentions FILE_RENAME_POSIX_SEMANTICS, but the name does not exist at
> all. There does exist FILE_RENAME_FLAG_POSIX_SEMANTICS. The
> documentation is just wrong. And in my earlier comment I copied the
> inexistent flag name.
> 
> But I meant to cite this flag: FILE_FLAG_POSIX_SEMANTICS (no "RENAME").
> It exists and is for CreateFileW().
> 
> Perhaps you also meant cite the latter one as the flag that "is not for
> use with `SetFileInformationByHandle()`"?
> 
> At any rate, the paragraph as written isn't correct.

I think I'm missing something. That's what the paragraph says:

    Careful readers might have noticed that [1] does not mention the above
    flag, but instead mentions `FILE_RENAME_POSIX_SEMANTICS`. This flag is
    not for use with `SetFileInformationByHandle()` though, which is what we
    use. And while the `FILE_RENAME_FLAG_POSIX_SEMANTICS` flag exists, it is
    not documented on [2] or anywhere else as far as I can tell.

And I'd claim it is correct.

`FILE_RENAME_POSIX_SEMANTICS` exists, this it is not a documentation
error. It is at a lower level than `FILE_RENAME_FLAG_POSIX_SEMANTICS`,
the documentation at [1] refers to "ntifs.h", which is part of the
Windows Driver Kit interfaces. So it is not supposed to be used with
`SetFileInformationByHandle()`, but with `FtlSetInformationFile()` [2],
which _also_ has a separate `FILE_RENAME_INFO` structure that looks the
same as `FILE_RENAME_INFO` defined for `SetFileInformationByHandle()`.
The only difference as far as I can tell is that the flags used for
these structures have slightly different names.

Now I totally get your confusion -- I have been extremely confused by
all of this, as well. It certainly is a documentation error that the
respective `FILE_RENAME_FLAG_POSIX_SEMANTICS` is undocumented, but
having proper docs for this is rather important such that the reader
knows what its behaviour is. So I have no other choice than to link to
the ntifs interfaces, as it documents the actual behaviour, even though
it lives in a different part of the Windows APIs.

Patrick

[1]: https://learn.microsoft.com/en-us/windows-hardware/drivers/ddi/ntifs/ns-ntifs-_file_rename_information
[2]: https://learn.microsoft.com/en-us/windows-hardware/drivers/ddi/fltkernel/nf-fltkernel-fltsetinformationfile

  reply	other threads:[~2024-10-27 17:27 UTC|newest]

Thread overview: 44+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-10-23 15:04 [PATCH 0/3] compat/mingw: implement POSIX-style atomic renames Patrick Steinhardt
2024-10-23 15:04 ` [PATCH 1/3] compat/mingw: share file handles created via `CreateFileW()` Patrick Steinhardt
2024-10-23 16:18   ` Kristoffer Haugsbakk
2024-10-23 17:25     ` Taylor Blau
2024-10-23 17:23   ` Taylor Blau
2024-10-23 17:25     ` Taylor Blau
2024-10-24  6:30     ` Patrick Steinhardt
2024-10-27 13:14     ` Johannes Sixt
2024-10-27 23:46       ` Taylor Blau
2024-10-23 15:05 ` [PATCH 2/3] compat/mingw: allow deletion of most opened files Patrick Steinhardt
2024-10-23 16:17   ` Kristoffer Haugsbakk
2024-10-23 17:30     ` Taylor Blau
2024-10-24  6:30     ` Patrick Steinhardt
2024-10-23 18:07   ` Taylor Blau
2024-10-23 15:05 ` [PATCH 3/3] compat/mingw: support POSIX semantics for atomic renames Patrick Steinhardt
2024-10-23 16:19   ` Kristoffer Haugsbakk
2024-10-24  6:30     ` Patrick Steinhardt
2024-10-24  7:18       ` Kristoffer Haugsbakk
2024-10-24  7:20         ` Patrick Steinhardt
2024-10-23 18:30   ` Taylor Blau
2024-10-23 15:36 ` [PATCH 0/3] compat/mingw: implement POSIX-style " Taylor Blau
2024-10-24 11:46 ` [PATCH v2 " Patrick Steinhardt
2024-10-24 11:46   ` [PATCH v2 1/3] compat/mingw: share file handles created via `CreateFileW()` Patrick Steinhardt
2024-10-24 11:46   ` [PATCH v2 2/3] compat/mingw: allow deletion of most opened files Patrick Steinhardt
2024-10-27 13:17     ` Johannes Sixt
2024-10-27 15:38       ` Patrick Steinhardt
2024-10-27 23:48         ` Taylor Blau
2024-10-27 23:51           ` Taylor Blau
2024-10-24 11:46   ` [PATCH v2 3/3] compat/mingw: support POSIX semantics for atomic renames Patrick Steinhardt
2024-10-27 13:23     ` Johannes Sixt
2024-10-27 15:38       ` Patrick Steinhardt
2024-10-27 16:31         ` Johannes Sixt
2024-10-27 17:27           ` Patrick Steinhardt [this message]
2024-10-27 21:36             ` Johannes Sixt
2024-10-27 23:50               ` Taylor Blau
2024-10-24 16:47   ` [PATCH v2 0/3] compat/mingw: implement POSIX-style " Taylor Blau
2024-10-27 13:27   ` Johannes Sixt
2024-10-27 15:39 ` [PATCH v3 " Patrick Steinhardt
2024-10-27 15:39   ` [PATCH v3 1/3] compat/mingw: share file handles created via `CreateFileW()` Patrick Steinhardt
2024-10-27 15:39   ` [PATCH v3 2/3] compat/mingw: allow deletion of most opened files Patrick Steinhardt
2024-10-27 15:39   ` [PATCH v3 3/3] compat/mingw: support POSIX semantics for atomic renames Patrick Steinhardt
2024-11-06  3:54   ` [PATCH v3 0/3] compat/mingw: implement POSIX-style " Junio C Hamano
2024-11-06  6:44     ` Johannes Sixt
2024-11-06 12:09       ` Junio C Hamano

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=Zx54B6FdGprtH0HV@pks.im \
    --to=ps@pks.im \
    --cc=git@vger.kernel.org \
    --cc=j6t@kdbg.org \
    --cc=johannes.schindelin@gmx.de \
    --cc=kristofferhaugsbakk@fastmail.com \
    --cc=me@ttaylorr.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).