News for hackers. Stuff that matters.
Re:63 Cores Blocked by Seven Instructions by pbsd on Monday October 21, 2019 @09:54AM (#21312237)
For each input source file, cl.exe creates at least 7 temporary files (with suffixes "gl", "sy", "ex", "in", "db", "md", "lk"). The churn of creating and deleting those, coupled with the slowness of performing checkpointing on a huge empty drive, seem to be the root cause here.

This appears somewhat related to this bug report: https://developercommunity.visualstudio.com/content/problem/...

Marking the temporary files as FILE_ATTRIBUTE_TEMPORARY could improve things, without having to go into significant Windows kernel changes.

Re:63 Cores Blocked by Seven Instructions by NullPrefix on Monday October 21, 2019 @11:41AM (#21313542)
> Moving %TEMP% to a RAM drive made the builds noticeably faster and also left the system responsive, even during two concurrent 20-way parallel builds (40 compile jobs in parallel, no loss of system responsiveness).

This matters more than anything else. You can play all you want with FILE_ATTRIBUTE_TEMPORARY or whatever else markings but the OS will just not care about them enough.

Re:63 Cores Blocked by Seven Instructions by MuffinFlavored on Monday October 21, 2019 @01:02PM (#21314580)
Does creating a RAM drive in Windows still require third-party software/drivers or is it natively supported by Microsoft these days? I haven't tried to do it since Windows 7 I believe.
Re:63 Cores Blocked by Seven Instructions by pdimitar on Monday October 21, 2019 @02:30PM (#21315586)
AFAIK, it still requires 3rd party software.

However, having a Samsung SSD and enabling RAPID mode in Samsung Magician (SSD management software) which effectively uses an invisible RAM disk accelerated my games' startup times by a factor of at least 3x. Some games even start 5x times faster.

I do recognise that games have a much different workload than compiler jobs of course; but invisibly utilising a RAM disk might help a lot regardless.

Re:63 Cores Blocked by Seven Instructions by the8472 on Monday October 21, 2019 @05:54PM (#21317885)
> still require third-party software/drivers

Back in the 3.1 and 9x days it didn't require 3rd party drivers, thanks to ramdrive.sys

Re:63 Cores Blocked by Seven Instructions by NullPrefix on Monday October 21, 2019 @02:30PM (#21315595)
That whole Linux integration into Windows marketing campaign could have you believe that Linux things just work on Windows now and you can just "mount -t tmpfs none /path/to/my/mount -o size=64G".

I doubt that is going to happen. File system mounts do not run on marketing fuel.

Re:63 Cores Blocked by Seven Instructions by jrockway on Monday October 21, 2019 @04:51PM (#21317209)
The latest variant of WSL uses a VM. So that will work, but you won't be seeing that mount inside Windows.
Re:63 Cores Blocked by Seven Instructions by Dylan16807 on Monday October 21, 2019 @06:03PM (#21317968)
Well, you can access it over the \\wsl$ shared drive. But that share can only do about 4k IOps.
Re:63 Cores Blocked by Seven Instructions by anon73044 on Monday October 21, 2019 @07:14PM (#21318638)
I've often wondered about the viability of putting all my build agent's folders in a ram disk to speed up build times.

Granted, IT won't be thrilled that I'll need ~4-8x the RAM, but the devs would certainly love the speed.

....now if only I could make fastlane take less than 2 hours....

Re:63 Cores Blocked by Seven Instructions by BeeOnRope on Monday October 21, 2019 @08:18PM (#21319057)
As long as you have the RAM available and your OS has reasonable file system caching behavior - you should already get that effect.

On Linux, I can run a build repeatedly that touches several GB of files and there is basically zero IO during the build, because everything is in the page cache.

There is some IO at some point after the build as the dirty files (e.g., .o files) get written out, but if if you delete them fast enough, even that doesn't happen.

Re:63 Cores Blocked by Seven Instructions by zamadatix on Monday October 21, 2019 @03:17PM (#21316108)
Isn't FILE_ATTRIBUTE_TEMPORARY the same thing as putting it in the RAM drive (minus needing a special mount and what part of the disk gets written when you run out of RAM)?
Re:63 Cores Blocked by Seven Instructions by snagglegaggle on Monday October 21, 2019 @10:53AM (#21312928)
> Marking the temporary files as FILE_ATTRIBUTE_TEMPORARY could improve things, without having to go into significant Windows kernel changes.

Having used Cygwin and MinGW (and less so WSL) NTFS is probably the main factor here. Especially when using Cygwin program compilation is very slow not just due to process creation but file access, potentially on disk.

I could see checkpointing contributing, but having used backups/file history on Windows server, you will see CPU use irregularities that I think are similar in cause to this but you do not see CPU use irregularities as bad as described in the article.

The pathological behavior of NTFS with many files is easy to prove once you encounter it.[0] This would be exposed to the kernel as well and is likely holding up NtfsCheckpointVolume. At least in my experience the problem goes deeper, and how NTFS structures are handled also contributes; for example, trying to enumerate and copy files in certain ways is extremely slow even if the files are easily enumerable.

You can say that if disabling checkpointing gets rid of noticeable slowdown it is the right thing to do, but there are people who will rightly not want to disable it.

[0] https://stackoverflow.com/questions/197162/ntfs-performance-...

Re:63 Cores Blocked by Seven Instructions by pault on Monday October 21, 2019 @02:57PM (#21315896)
Time the difference between an npm install, with all its thousands of tiny files, on NTFS and ext4. It's excruciating.
Re:63 Cores Blocked by Seven Instructions by cpach on Monday October 21, 2019 @04:25PM (#21316896)
Just copying Emacs for Windows from one folder to another takes surprisingly long time :)
Re:63 Cores Blocked by Seven Instructions by Filligree on Monday October 21, 2019 @04:45PM (#21317130)
I've definitely noticed this behaviour. It's excruciating.

There's a ZFS driver for Windows now. I haven't tried it yet, but I think I will. Can't be any worse.