Julius Plenz

I don’t have a 32 bit system at hand to try this out. Could you recompile (`make clean all`) using the referenced commit and see if this fixes the issue?...

Okay, so it’s a first step. I’ll try compiling some 32 bit binary when I have the time. Meanwhile, as per the README, have you tried `nocache -n 2` (or,...

A more generic solution would be to introduce a method to either include (“whitelist”) or exclude (“blacklist”) certain glob patterns. The problem here is that the only way you can...

True; but the functionality of `nocache` is achieved by the wrapper shell script setting the `LD_PRELOAD` env variable. The initializer of `nocache.so` is only called from the specified executable and...

Use `fnmatch(3)` without `FNM_PATHNAME`. ``` $ cat fnmatch.c #include int main(int argc, char *argv[]) { return fnmatch("foo/*", argv[1], 0); } $ gcc -Wall -o fnmatch fnmatch.c $ ./fnmatch foo/bar/baz &&...

I’ve added some comments to your commit https://github.com/beroal/nocache/commit/c3956d384d04837dc33dc1756dd7e73754aae919 > Do you think that OS uses the last suggestion instead of joining suggestions from all processes? The reality is a bit...

I left some comments on https://github.com/beroal/nocache/commit/1e6061c9879b21f1d22607cc5d783f5abdf20a3f again. > Then the Linux kernel is kind of stupid. Yes, and you’re welcome to improve it. The code is in `mm/fadvise.c`. Beware though...

Hi, > fadvise() works just fine without prior fsync() - buffer pages are flushed out in nice batches, and still marked for reuse. The `posix_fadvise()` man page says regarding `POSIX_FADV_DONTNEED`...

Addendum: Let me quote [Tobi Oetiker on this](http://insights.oetiker.ch/linux/fadvise/): > As you can see we are calling fdatasync right before calling `posix_fadvise`, this makes sure that all data associated with the...

It’s a nice idea, but I wonder if it helps in this use case (lots of very small files). Do you propose to leave the fd’s open? In that case...