[wsl2] filesystem performance is much slower than wsl1 in /mnt

[ioweb-gr]

I decided to open this as a separate issue because although it's related to the generic issue of filesystem performance it's directly related to WSL 2 while the other issues are for WSL 1 and it's showing very conflicting results.

• Your Windows build number: (Type ver at a Windows Command Prompt) Version 10.0.18917 Build 18917

• What you're doing and what's happening: I'm testing filesystem write speed in /mnt using dd command. Performing the following tests

WSL2

root@LUCIANO-PC:/home/# dd if=/dev/zero of=/mnt/e/testfile bs=1M count=1000 1000+0 records in 1000+0 records out 1048576000 bytes (1.0 GB, 1000 MiB) copied, 25.939 s, 40.4 MB/s

WSL1

root@LUCIANO-PC:/home/# dd if=/dev/zero of=/mnt/e/testfile bs=1M count=20000 20000+0 records in 20000+0 records out 20971520000 bytes (21 GB, 20 GiB) copied, 47.4897 s, 442 MB/s

On / it's actually the reverse. WSL2 is more than 2 times faster than WSL1.

• What's wrong / what should be happening instead: I would expect the filesystem performance in /mnt to at least be on the same level but it's over 10 times slower.

Another interesting fact is that if I mount the same drive as a cifs share I get 3x performance

WSL 2 (cifs share)

root@LUCIANO-PC:/mnt/sambae# dd if=/dev/zero of=/mnt/sambae/testfile bs=1M count=10000 10000+0 records in 10000+0 records out 10485760000 bytes (10 GB, 9.8 GiB) copied, 84.001 s, 125 MB/s

An update of the current status because it's way too hidden in this thread.

Latest status report: https://github.com/microsoft/WSL/issues/4197#issuecomment-604592340_

[billziss-gh]

On / it's actually the reverse. WSL2 is more than 2 times faster than WSL1.

Can you post the results when doing the same test on a file on / for WSL1 and WSL2?

This is straight I/O on a single file where I would expect to see more or less the same perf between WSL1 and WSL2. My expectation was that WSL2 perf is better than WSL1 only when doing "namespace" operations (i.e. working on lots of small files, listing them, stat'ing them, etc.)

As for the results on /mnt/e they are not very surprising, since I/O has to go through both the Linux and Windows file system stack in likely a new and unoptimized piece of code.

[ioweb-gr]

I see,

I'm assuming you're still refering to / and not /mnt folders about namespace operations. I've repeated the write tests in both environments a few times now and there's no consistent result.

WSL 2

root@LUCIANO-PC:/home/# dd if=/dev/zero of=~/testfile bs=1M count=10000 10000+0 records in 10000+0 records out 10485760000 bytes (10 GB, 9.8 GiB) copied, 7.0873 s, 1.5 GB/s root@LUCIANO-PC:/home/# dd if=/dev/zero of=~/testfile bs=1M count=20000

20000+0 records in 20000+0 records out 20971520000 bytes (21 GB, 20 GiB) copied, 51.3779 s, 408 MB/s root@LUCIANO-PC:/home/# root@LUCIANO-PC:/home/# dd if=/dev/zero of=~/testfile bs=1M count=20000 ^C12360+0 records in 12360+0 records out 12960399360 bytes (13 GB, 12 GiB) copied, 12.1766 s, 1.1 GB/s

WSL1

root@LUCIANO-PC:/home/# dd if=/dev/zero of=~/testfile bs=1M count=10000 10000+0 records in 10000+0 records out 10485760000 bytes (10 GB, 9.8 GiB) copied, 29.7096 s, 353 MB/s root@LUCIANO-PC:/home/#

I'm thinking this has to do something with OS Caching because in task manager even though I can see the file being written, the disk usage write speed is unchanged while in wsl1 I can see it constantly at around 350mb/sec .

As another interesting fact, the vmmem usage while writing the file is increasing rapidly. By the 3rd copy action all my RAM is used up (32GB). It goes back down a while after quitting my shell. So maybe the performance is bottlenecked by this bug as well.

[paultyng]

I'm also experiencing this after an upgrade to WSL2, only slow performance under /mnt, not /, slower than WSL1. I can post numbers as well if necessary. This is on Microsoft Windows [Version 10.0.18936.1000].

[definelicht]

I'm experiencing the same issue, /mnt is extremely slow compared to WSL1 (seconds to run git status on a relatively small repository).

[paultyng]

When I run a git status on a clone of https://github.com/hashicorp/terraform in /mnt I get output like this:

It took 6.97 seconds to enumerate untracked files. 'status -uno'
may speed it up, but you have to be careful not to forget to add
new files yourself (see 'git help status').

[ioweb-gr]

This is an example from the command

pv files.tar.gz | tar -zxf -

The speed is amazingly slow in wsl2 on /mnt

[definelicht]

Do we know what's at the root of this, and whether there are any other workarounds than downgrading?

[benhillis]

@definelicht - Yep we're working on improving the performance. In the meaning working out of your root file system (the ext4 volume) will have MUCH better performance.

[ioweb-gr]

Indeed after the last update I can already see much better performance on WSL2.

root@LUCIANO-PC:~# dd if=/dev/zero of=/mnt/e/testfile bs=1M count=10000 10000+0 records in 10000+0 records out 10485760000 bytes (10 GB, 9.8 GiB) copied, 41.277 s, 254 MB/s

That's on an SSD and it's like 6 times faster than before. Still slower then WSL1 but it's definitely improved.

--correction--- I see much better performance on sequential writes. Untar seems very slow still extracting data at bytes/sec while in wsl1 I can see speeds at kb/sec for the same part

[tuananh]

Indeed after the last update I can already see much better performance on WSL2.

root@LUCIANO-PC:~# dd if=/dev/zero of=/mnt/e/testfile bs=1M count=10000 10000+0 records in 10000+0 records out 10485760000 bytes (10 GB, 9.8 GiB) copied, 41.277 s, 254 MB/s

That's on an SSD and it's like 6 times faster than before. Still slower then WSL1 but it's definitely improved.

--correction--- I see much better performance on sequential writes. Untar seems very slow still extracting data at bytes/sec while in wsl1 I can see speeds at kb/sec for the same part

I got

1000+0 records in
1000+0 records out
1048576000 bytes (1.0 GB, 1000 MiB) copied, 3.57894 s, 293 MB/s

but git status still taking more than 2 seconds on quite small repo

[definelicht]

I think this is a latency rather than a bandwidth issue, surfacing when a large number of files are accessed (e.g., when running git status).

[tuananh]

@definelicht - Yep we're working on improving the performance. In the meaning working out of your root file system (the ext4 volume) will have MUCH better performance.

can you share the cause of it? I'm curious

[gbatalski]

can confirm a very bad performance on /mnt (mvn install takes forever), on /tmp it takes only 2 minutes for the same project

[nono-pxbsd]

After switch configuration from WSL to WSL 2 (Build 18975), without any change on my distro, the server respond time incresed.

Configuration : Debian buster Apache2.X / PHP-FPM 7.3 / MySQL Root : / Web directories : /c/repositories/xxx

WSL 1 : 0.5 sec WSL 2 : 8 sec

With Docker (Ubuntu 18.04) WSL Tech preview i could not test on WSL 1, but WSL 2 takes over 30 sec... I suppose the issue is the same.

[realtica]

This is my performance in a medium project:

GIT_TRACE_PERFORMANCE=1 git status -sb -uno

13:20:04.630972 read-cache.c:2267 performance: 0.017249300 s: read cache .git/index 13:20:07.766270 preload-index.c:146 performance: 3.135180100 s: preload index Refresh index: 100% (3463/3463), done. 13:20:53.531359 read-cache.c:1575 performance: 48.900269600 s: refresh index 13:20:53.981829 read-cache.c:2267 performance: 0.016242500 s: read cache /mnt/e/p/keyed_au/.git/ modules/Files/media/nusoap/index 13:20:54.383135 read-cache.c:1575 performance: 0.401222700 s: refresh index 13:20:54.426399 diff-lib.c:251 performance: 0.000005000 s: diff-files 13:20:54.468000 unpack-trees.c:1554 performance: 0.000087700 s: traverse_trees 13:20:54.468266 unpack-trees.c:465 performance: 0.000164900 s: check_updates 13:20:54.468346 unpack-trees.c:1649 performance: 0.000438000 s: unpack_trees 13:20:54.468368 diff-lib.c:537 performance: 0.001873400 s: diff-index 13:20:54.569793 read-cache.c:2989 performance: 0.014679900 s: write index, changed mask = 2 13:20:54.586550 trace.c:477 performance: 0.736638600 s: git command: /usr/lib/git-core/git status --porcelai n=2 -uno 13:20:55.277184 read-cache.c:2267 performance: 0.031572000 s: read cache /mnt/e/p/keyed_au/.git/ modules/Files/modules/unittest/index 13:20:55.958706 read-cache.c:1575 performance: 0.681302500 s: refresh index 13:20:56.007094 diff-lib.c:251 performance: 0.000003500 s: diff-files 13:20:56.041768 unpack-trees.c:1554 performance: 0.000081200 s: traverse_trees 13:20:56.042014 unpack-trees.c:465 performance: 0.000162100 s: check_updates 13:20:56.042084 unpack-trees.c:1649 performance: 0.000399800 s: unpack_trees 13:20:56.042102 diff-lib.c:537 performance: 0.000454400 s: diff-index 13:20:56.126615 read-cache.c:2989 performance: 0.012095500 s: write index, changed mask = 2 13:20:56.141174 trace.c:477 performance: 1.261971900 s: git command: /usr/lib/git-core/git status --porcelain=2 -uno 13:20:56.148677 diff-lib.c:251 performance: 2.586112000 s: diff-files 13:20:56.190888 unpack-trees.c:1554 performance: 0.000022200 s: traverse_trees 13:20:56.191028 unpack-trees.c:465 performance: 0.000027800 s: check_updates 13:20:56.191106 unpack-trees.c:1649 performance: 0.000639200 s: unpack_trees 13:20:56.191175 diff-lib.c:537 performance: 0.001413100 s: diff-index 13:20:56.215068 trace.c:477 performance: 51.628126000 s: git command: git status -sb -uno

File System exFat, the result is the same is other file system... WSL 2 git=51.62 sec git for windows=0.8 sec

[paslandau]

I'm currently using the normal Docker Desktop with linux containers to run a PHP projekt during local development. The repo is in a normal Windows directory and gets mounted via volume bind in the docker container, so that I can easily manipulate the files through an IDE.

The same setup on a native linux machine has always been multiple times faster and the "problem" (from what I've read) has always been attributed to the Linux VM that is used on Docker Desktop.

Then WSL 2 was announced and I finally came around to set everything up. But when I ran our teststuite for the first time, I was pretty disappointed that the performance was actually much worse than before (40s on the Docker Deskop Linux VM setup vs 100s on the WSL2 setup). Subsequent investigation brought me to this thread.

From what I've read here so far:

Does it even make sense to switch to WSL 2 for local development yet if "bad performance" is your primary reason to do so?

A "no" would be a totally acceptable answer - I just don't want to waste any more time on this if it's simply not ready yet.

[stweedie]

I know that wsl2 is a preview, but is there any hope that accessing \wsl$ from windows or /mnt from wsl will reach acceptable speeds?

[gormulent]

I'd also like to ask about this, doing a git status on something in /mnt/c/.. takes about a minute, something seems wrong ..

[gbatalski]

same with 1903 (18990.1) absolutely useless...

[wusticality]

I am also seeing this, accessing Windows files from the Linux environment is several orders of magnitude slower in WSL2 than in WSL1.

[NBNARADHYA]

Same for me on build 18985

[gkawamoto]

Still happening @ build 18990

[craigloewen-msft]

This experience is being worked on, thank you for posting updates on the build numbers that you are still seeing this issue, but please keep in mind that this can make a fair amount of unnecessary noise for users who are subscribed to this thread, so I'd ask if you could please not do so.

When there are changes in this area, i.e: things that will affect file performance in the mounted drives, we (the WSL team) will be sure to post updates in this thread. Thanks!

[vbsteven]

A possible workaround for slow git could be to use git.exe from WSL2. For me a git status with linux git takes 15 seconds while using git.exe takes 0.3 seconds.

[gkawamoto]

A possible workaround for slow git could be to use git.exe from WSL2. For me a git status with linux git takes 15 seconds while using git.exe takes 0.3 seconds.

Can confirm this behaviour:

[ioweb-gr]

You should see how long npm and composer take to install dependencies :) we can't really alias all commands to the windows counterparts so although it's a workaround I'm hoping we'll see a real solution soon. Currently we are forced back to building things on our linux machines

[er1c]

A possible workaround for slow git could be to use git.exe from WSL2. For me a git status with linux git takes 15 seconds while using git.exe takes 0.3 seconds.

Can confirm this behaviour:

I've been using a little wrapper script like:

#!/bin/bash
# WSL 'git' wrapper, save as /usr/local/bin/git and chmod a+x

REALPATH=`readlink -f ${PWD}`
ARG1=$1
MORE_COMMANDS=(log diff)

if [  "${REALPATH:0:5}" == "/mnt/" ]; then
  # echo "Using windows git.exe"
  if [[ " ${MORE_COMMANDS[@]} " =~ " ${ARG1} " ]]; then
    git.exe "$@" | more
  else
    git.exe "$@"
  fi
else
  # echo "Using linux /usr/bin/git"
  /usr/bin/git "$@"
fi

[danvy]

Same issue on Windows 10 18995 Poor performance of file access to /mnt compared to / Git is 10x slower Cusom prompts using Git are unusable

[wusticality]

I know this goes without saying, but accessing files outside of WSL is absolutely unusable at present. I'm going to have to revert back to WSL1 until this is fixed.

[alshdavid]

Just adding my experience to the list of already well documented experiences. Filesystem access on /mnt/d is unusably slow. Accessing files in ~ is fine, however

[ioweb-gr]

In build 18999 the performance is actually decreased compared to previous builds. For all purposes it's completely unusable. Even tab completion is much slower than before taking over 10 seconds to list the files on an SSD.

[benhillis]

@ioweb-gr - I looked through the commits in that build and nothing jumps out at me that would explain a massive slowdown. Do you have more data you can share?

[ioweb-gr]

Of course, what would you need from me in order to help?

I first noticed it while trying to commit a change to a single file which 20 seconds later hadn't been completed. With windows git it took less than a second for the same operation.

[benhillis]

@ioweb-gr - Is the issue still present?

[ioweb-gr]

let me convert the machine back to wsl2 and get back to you. Since it was taking way too long I reverted it to wsl 1.

[ioweb-gr]

Actually I tried it with another instance which is version 2 and yes the issue persists. While trying tab-completion in the same folder in /mnt it's very slow.

So I'm trying git add compo then pressing tab and waiting to autofil ser.json and it's taking way too long compared to previous version.

ll compo then tab autocompletes it instantly though.

[SvenGroot]

@ioweb-gr I did some research into this, and it appears that this slowdown is caused by the bash-completion script used by git. For git add, this script tried to find all modified files in the repository, for which it needs to "stat" every file, which can be somewhat slow (equivalent to git status).

However, I found this was only really slow if the repository was cloned by Git for Windows. In this case, it seems there's a mismatch between the attributes git expects and the results it gets from stat, so git ends up rehashing every file in your repository. This is not the case if the repository was cloned using git on WSL.

Did you clone your repository using Git for Windows?

[wusticality]

Hey @SvenGroot, I cloned a small repo inside WSL 2 and it takes nearly a minute to do a git status. Other programs are similarly painful if accessing files outside of WSL's environment. Doing a git status on the same repo inside WSL 2 takes about a tenth of a second.

[Luiszaba]

Had to revert back to WSL 1. The issue seems to be global for me.

[ioweb-gr]

Hi @SvenGroot there is high chance it was cloned using git bash for windows since cloning repositories over wsl2 takes ages usually. I'll try in the next days and experiment with different cloning methods and see if the issue persists when it's cloned via wsl.

[SvenGroot]

Please note that for WSL2, we recommend that you put your files in the Linux file system (not under /mnt) whenever possible. It will be much faster (than WSL1 even) and you can still work on them from Windows using \wsl$ or tools such as Visual Studio Code Remote.

We realize that accessing Windows files in WSL2 is slow. We're working on ways to make this better, but not all of those will be ready for the 20h1 release.

[stweedie]

the issue also crosses the boundary when accessing \wsl$ in windows. For instance, I cloned a repo in wsl 2 and am attempting to work on it from sublime. The access is so slow that sublime can't actually index the directory and any plugin also suffers from terrible performance.

Is the 20h1 release the official public release? If fixes for this won't be in for this release, when can we expect them? WSL 2 will be DOA until then for me

[ioweb-gr]

@SvenGroot I understand what you mean however for us it's not always possible to do it like this unfortunately. It very much depends on your workflow for that matter. If I put everything on / I would have to move everything of my projects on my main OS disk which is bad and unsafe. I currently keep them on a separate mirrored drive comprising of 3x2TB SSDs which is mounted on /mnt under wsl but is also portable if needed.

I guess for a lot of people having small ssds on OS there is gonna be a similar workflow so the need to use fast files on /mnt becomes much more important as time passes

[SvenGroot]

If you use the wsl --export and wsl --import commands, it's possible to move your Linux files to a location other than the system drive. This works for both WSL1 and WSL2.

But trust me, we know Windows file system interop performance is important, and we spend a lot of resources trying to improve it.

[ioweb-gr]

Yes I understand that, for my personal reasons I can't export the environment out of my OS drive becauce of automated deployments running here, so unfortunately I cant use that solution. I'm glad your working towards this and I know it's not easy effort. My only concern is that instead of seeing improvement in WSL2 I saw a huge leap backwards when it comes to access /mnt areas so somewhere down the road I think the importance of having this perform as well as possible was overlooked.

In the process I sacrificed VMWare in order to have WSL2 running on Hyper-V because of better compatibility with different functions (for example I couldn't perform git commits in /mnt before in WSL1 as the builds progressed) so I'm just considering this: If I used vmware for a linux environment with mapped network folders I could achieve the same thing WSL2 does for me with better speeds. Comparing the benefits of each solution makes it very hard to choose WSL2 at the moment over WSL1.

[19317362]

I tested on WSL1 and WSL2. The result is WSL1

$ dd if=/dev/zero of=~/testfile bs=1M count=1000 1000+0 records in 1000+0 records out 1048576000 bytes (1.0 GB, 1000 MiB) copied, 0.351729 s, 3.0 GB/s

WSL2 $ dd if=/dev/zero of=~/testfile bs=1M count=1000 1000+0 records in 1000+0 records out 1048576000 bytes (1.0 GB, 1000 MiB) copied, 1.18549 s, 885 MB/s

[er1c]

I've recently been having "I/O" exception issues when trying to do a "git status" in WSL2 - and needing to reboot to restart the VM - I suspect it has to do with running both windows git and linux git at the same time on the same repository. I wish I could create a simple way to repro it - but it's happened at least twice in the last two days. Does anyone know if there is a way to "reboot" WSL2 without a full computer restart?

[ilqvya]

Lol, the new wsl2 is just a VM with network-mounted /mnt folders. So it is expected result to being such slow. I don't know why this WSL2 project exists at all, but it's better to use just Hyper-v or other VM, where i can mount windows folders with SMB or smlz, just like you did it.

On the other hand, wsl1 is VM-free solution, and that's why it's cool and low-latency. You better to improve WSL-kernel to be able to use docker and other heavy soft in wsl-1, without VM-overhead

[zulu12345]

Pls fix this issue. I can't use WSL2 because the /mnt is too slow... A simple "git status" will kill it in a bigger project....

[adam-ludgate]

👍 to get this fixed. Running in /mnt performance has ground to a halt, and I am having to revert back to WSL 1 to continue working