Skip to content

Instantly share code, notes, and snippets.

View slp's full-sized avatar

Sergio López slp

243 followers · 1 following
View GitHub Profile
@slp
slp / 0001-Update-vendor-of-containers-storage.patch
Created July 18, 2022 12:03
From 96d52b6e40060c104cd4e08c2fab73ad1a899fba Mon Sep 17 00:00:00 2001
From: Sergio Lopez <[email protected]>
Date: Mon, 18 Jul 2022 13:51:24 +0200
Subject: [PATCH 1/4] Update vendor of containers/storage
Backport 39880517a19ffde1ca2fca5f29cb051309e56773 from the main
branch.
Signed-off-by: Sergio Lopez <[email protected]>
---
@slp
slp / OpenFaaS-libkrun-Fedora33.md
Created February 25, 2021 13:53
Installing OpenFaaS with libkrun-based isolation on Fedora 33

Installing OpenFaaS with libkrun-based isolation on Fedora 33

Setting up the system

Revert to cgroup v1

The version of containerd that OpenFaaS will download during the installation doesn't support cgroup v2, so we need to configure systemd to avoid using the Unified cgroup Hierarchy, and reboot the system:

@slp
slp / flatpak_evergreen_firefox.md
Created June 30, 2020 07:54
Quick recipe for an evergreen Firefox using Flatpak 
flatpak --user install flathub org.mozilla.firefox
flatpak run org.mozilla.firefox

# Personalize the browser, log in to the places you
# want to visit with your own account, and quit it.

# Save a copy of the app's data directory to use it
# as a template.
cp -a $HOME/.var/app/org.mozilla.firefox \
@slp
slp / keybase.md
Created May 29, 2020 11:15

Keybase proof

I hereby claim:

  • I am slp on github.
  • I am slpnix (https://keybase.io/slpnix) on keybase.
  • I have a public key ASB2otsGKzZEFFPxwpuoQ0ZXOVqZ6DMEAMpnb_fNVDtQ8Ao

To claim this, I am signing this object:

@slp
slp / linux_microvm_v3.config
Created August 29, 2019 15:32
#
# Automatically generated file; DO NOT EDIT.
# Linux/x86 5.3.0-rc3 Kernel Configuration
#
#
# Compiler: gcc (GCC) 9.1.1 20190503 (Red Hat 9.1.1-1)
#
CONFIG_CC_IS_GCC=y
CONFIG_GCC_VERSION=90101
@slp
slp / diagnose_boot_issue.md
Created August 29, 2017 06:47

This morning, my laptop got stuck for a minute during boot in the typical waiting for... message coming from systemd.

Of course, the easiest and funniest approach here would be blaming systemd and writing an angry tweet about how it sucks, but let's see if we can do better.

First, once the system booted, let's find out which services failed:

[localhost ~]$ systemctl --failed
  UNIT                        LOAD   ACTIVE SUB    DESCRIPTION                                 
● systemd-udev-settle.service loaded failed failed udev Wait for Complete Device Initialization
@slp
slp / peek_cns_using_sysutils.md
Created August 24, 2017 09:25
Peeking into a container's namespace using system utils 
[root@rhel74-docker ~]# docker run -ti --rm alpine
Unable to find image 'alpine:latest' locally
Trying to pull repository registry.access.redhat.com/alpine ... 
Trying to pull repository docker.io/library/alpine ... 
latest: Pulling from docker.io/library/alpine
88286f41530e: Pull complete 
Digest: sha256:1072e499f3f655a032e88542330cf75b02e7bdf673278f701d7ba61629ee3ebe
/ # ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN qlen 1
@slp
slp / read_latency_test_KVM
Last active February 26, 2017 07:45
Running on KVM (4.4.0-21-generic/qemu-2.5.0).
Test computes cost of write() and read() moving 16384 x 4K blocks.
Filesystem is tmpfs for Linux and GNU Hurd (translator), memfs for Magenta.
Magenta (014c20cb2b99f1ec80c8d244c243bf17cf1f134f)
==================================================
elapsed time: 900.43 ms cost per call: 54.96 us
@slp
slp / read_latency_test_RPi3
Last active February 22, 2017 07:32
Running on Raspberry Pi 3.
Test computes cost of write() and read() moving 16384 x 4K blocks.
Filesystem is tmpfs for Linux and memfs for Magenta.
Magenta (3e34cf26a478ca07dd620e741e2140a9a6b7fd8b)
==================================================
elapsed time: 1987.18 ms cost per call: 121.29 us
@slp
slp / Notes on Google's Magenta.md
Last active February 27, 2017 13:09

This document is a draft. This information may be incomplete and/or inaccurate

About

After reading an article on Hacker News about Google's Fuchsia OS, I decided to take a look at its microkernel, Magenta.

I always had an interest on microkernels, and over the years I've read a lot of code from Mach variants, L4 family, MINIX 3, HelenOS, and so on. I even maintain a MkLinux (OSFMach+Linux) repo to preserve its code (https://github.com/slp/mkunity) and made minor contributions to GNU Mach and GNU Hurd.

I'm also curious about Google's approach with Magenta because writing a pure microkernel in the era of SSDs and 10 Gbps interfaces, is one the most challenging tasks you can find. To be able to extract the full potential of such devices, you need to both find the shortest possible path from user application to device (that is, keeping the lowest CPU induced latency) and being able to parallelize the IO load without compromises.