Sitsofe Wheeler

This sounds like a dupe of #731 ...

@chamarthy Great report! Can you try the [`experimental_verify`](https://fio.readthedocs.io/en/latest/fio_doc.html#cmdoption-arg-experimental-verify) option?

@chamarthy any news?

OK the results reported by @bcran were kind of what we would expect. The non-experimental verify actually extends a data structure with I/Os to verify whereas experimental just regenerates what...

(Pinging @axboe on this one)

(@axboe ping)

@JonKohler I assume you are saying it needs to be clearer that you have to set `sqthread_poll` to a value (i.e. `sqthread_poll=1). If so I'd argue the documentation is unclear......

@111alan This sounds like #875 ... Do things get better if you fully write to your files and then do I/O to them? If so the issue is that unlike...

@111alan I mean: 1. Create the file write to it **entirely** once: ``` fio --thread --filename=testfile --rw=write --size=10G` --name=writeonce --bs=1M --end_fsync=1 ``` 2. Ensure that fio uses the fully written...

A quick note on this one: I recently added an option to create PDB information (when a clang build is done). As I'm not a Windows user I don't know...