Windows: ~1 GB RAM per git process, many concurrent (2.56.0.windows.1)

Windows: ~1 GB RAM per git process, many concurrent (2.56.0.windows.1)

[Pierre Bruno]

[-- Attachment #1: Type: text/plain, Size: 827 bytes --]

Hi,

On Windows, git status-type commands use about 1 GB of RAM per
process, and dozens run at once when I use coding agents (OpenCode and
Oh My Pi) in a repository. CPU is near 0%, disk I/O is steady, and the
total is several GB. Since two unrelated tools cause it, I suspect git
or my repo.

Expected: a few short-lived git processes with far less memory.

git version 2.56.0.windows.1 (commit 49d759b698127791a5f3f2759c69b983846711dd,
fsmonitor--daemon enabled), Windows <version>
Repo: <N files / GB>, <N> untracked files
Manual 'git status' with no other tool running: <fast/slow, memory>
Stable 2.55.0: <same/different>
Command line seen in Task Manager: <paste>

Is ~1 GB per process expected here, and is there a recommended config to
reduce it? Full 'git bugreport' attached.

Thanks,
Pierre Bruno

[-- Attachment #2: Screenshot 2026-10-02 130341.png --]
[-- Type: image/png, Size: 117065 bytes --]

[-- Attachment #3: Screenshot 2026-10-02 124754.png --]
[-- Type: image/png, Size: 138173 bytes --]

Re: Windows: ~1 GB RAM per git process, many concurrent (2.56.0.windows.1)

[brian m. carlson]

[-- Attachment #1: Type: text/plain, Size: 1693 bytes --]

On 2026-10-02 at 11:04:00, Pierre Bruno wrote:
> Hi,
> 
> On Windows, git status-type commands use about 1 GB of RAM per
> process, and dozens run at once when I use coding agents (OpenCode and
> Oh My Pi) in a repository. CPU is near 0%, disk I/O is steady, and the
> total is several GB. Since two unrelated tools cause it, I suspect git
> or my repo.
> 
> Expected: a few short-lived git processes with far less memory.
> 
> git version 2.56.0.windows.1 (commit 49d759b698127791a5f3f2759c69b983846711dd,
> fsmonitor--daemon enabled), Windows <version>
> Repo: <N files / GB>, <N> untracked files
> Manual 'git status' with no other tool running: <fast/slow, memory>
> Stable 2.55.0: <same/different>
> Command line seen in Task Manager: <paste>
> 
> Is ~1 GB per process expected here, and is there a recommended config to
> reduce it? Full 'git bugreport' attached.

I think you omitted the attachment, but in any event, I would say that
this is not normally expected for `git status`.  We'd really need to
know what the command line of those processes is for us to know what
they're for; for instance, you may be triggering maintenance on the
repository, in which case packing a large repository could legitimately
use that much memory.  Similarly, if you're using a file system monitor
process, that could consume a large amount of memory in a large
repository.

I don't personally use Windows, so I'm afraid I can't tell you how to
get that information there.  Once you have it, though, it should be
clearer if that's a reasonable amount of memory to be using given the
size of your repository.
-- 
brian m. carlson (they/them)
Toronto, Ontario, CA

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 325 bytes --]

Re: Windows: ~1 GB RAM per git process, many concurrent (2.56.0.windows.1)

[Jeff King]

On Fri, Oct 02, 2026 at 11:04:00AM +0000, Pierre Bruno wrote:

> On Windows, git status-type commands use about 1 GB of RAM per
> process, and dozens run at once when I use coding agents (OpenCode and
> Oh My Pi) in a repository. CPU is near 0%, disk I/O is steady, and the
> total is several GB. Since two unrelated tools cause it, I suspect git
> or my repo.

Is that counting shared, mmap'd memory? Git will mmap the on-disk
packfiles (or on Windows, CreateFileMapping/MapViewOfFile). So if your
repository has a 1GB packfile, that could explain it. And it can lead to
two unintuitive conclusions:

  1. If you have several git-status processes, they're sharing all of
     those pages. So it's still only 1GB of memory use.

  2. Git will make a large map and assume the OS will fault in only the
     pages that are needed. So you may see a large virtual memory size,
     but a lower resident size (these are the terms I'd expect from
     "top" on Linux; I don't know what terms you might see on Windows).

You may also see a large resident size if the OS has faulted in all of
those pages (e.g., due to other commands). Ironically this is a sign
that you _don't_ have a lot of memory pressure (otherwise, the unused
pages would have been dropped). Again, this is coming from the Linux
side of things; I don't know how aggressively (or not) Windows is about
faulting in or releasing pages from mapped files.

-Peff