Join devRant
Do all the things like
++ or -- rants, post your own rants, comment on others' rants and build your customized dev avatar
Sign Up
Pipeless API
From the creators of devRant, Pipeless lets you power real-time personalized recommendations and activity feeds using a simple API
Learn More
Search - "x86_64"
-
What an absolute fucking disaster of a day. Strap in, folks; it's time for a bumpy ride!
I got a whole hour of work done today. The first hour of my morning because I went to work a bit early. Then people started complaining about Jenkins jobs failing on that one Jenkins server our team has been wanting to decom for two years but management won't let us force people to move to new servers. It's a single server with over four thousand projects, some of which run massive data processing jobs that last DAYS. The server was originally set up by people who have since quit, of course, and left it behind for my team to adopt with zero documentation.
Anyway, the 500GB disk is 100% full. The memory (all 64GB of it) is fully consumed by stuck jobs. We can't track down large old files to delete because du chokes on the workspace folder with thousands of subfolders with no Ram to spare. We decide to basically take a hacksaw to it, deleting the workspace for every job not currently in progress. This of course fucked up some really poorly-designed pipelines that relied on workspaces persisting between jobs, so we had to deal with complaints about that as well.
So we get the Jenkins server up and running again just in time for AWS to have a major incident affecting EC2 instance provisioning in our primary region. People keep bugging me to fix it, I keep telling them that it's Amazon's problem to solve, they wait a few minutes and ask me to fix it again. Emails flying back and forth until that was done.
Lunch time already. But the fun isn't over yet!
I get back to my desk to find out that new hires or people who got new Mac laptops recently can't even install our toolchain, because management has started handing out M1 Macs without telling us and all our tools are compiled solely for x86_64. That took some troubleshooting to even figure out what the problem was because the only error people got from homebrew was that the formula was empty when it clearly wasn't.
After figuring out that problem (but not fully solving it yet), one team starts complaining to us about a Github problem because we manage the github org. Except it's not a github problem and I already knew this because they are a Problem Team that uses some technical authoring software with Git integration but they only have even the barest understanding of what Git actually does. Turns out it's a Git problem. An update for Git was pushed out recently that patches a big bad vulnerability and the way it was patched causes problems because they're using Git wrong (multiple users accessing the same local repo on a samba share). It's a huge vulnerability so my entire conversation with them went sort of like:
"Please don't."
"We have to."
"Fine, here's a workaround, this will allow arbitrary code execution by anyone with physical or virtual access to this computer that you have sitting in an unlocked office somewhere."
"How do I run a Git command I don't use Git."
So that dealt with, I start taking a look at our toolchain, trying to figure out if I can easily just cross-compile it to arm64 for the M1 macbooks or if it will be a more involved fix. And I find all kinds of horrendous shit left behind by the people who wrote the tools that, naturally, they left for us to adopt when they quit over a year ago. I'm talking entire functions in a tool used by hundreds of people that were put in as a joke, poorly documented functions I am still trying to puzzle out, and exactly zero comments in the code and abbreviated function names like "gars", "snh", and "jgajawwawstai".
While I'm looking into that, the person from our team who is responsible for incident communication finally gets the AWS EC2 provisioning issue reported to IT Operations, who sent out an alert to affected users that should have gone out hours earlier.
Meanwhile, according to the health dashboard in AWS, the issue had already been resolved three hours before the communication went out and the ticket remains open at this moment, as far as I know.5 -
I've just managed to add the ability to call functions to my compiler. Currently, the capabilities are rather limited but I am pleased with it.
It keeps the schematics of the source code well enough that it can handle
recursive calls.
:D
On the Image, you can see the program's output (left), the generated x86_64 assembly, and the source code of the compiled program (bottom).6 -
Long rant ahead.. 5k characters pretty much completely used. So feel free to have another cup of coffee and have a seat 🙂
So.. a while back this flash drive was stolen from me, right. Well it turns out that other than me, the other guy in that incident also got to the police 😃
Now, let me explain the smiley face. At the time of the incident I was completely at fault. I had no real reason to throw a punch at this guy and my only "excuse" would be that I was drunk as fuck - I've never drank so much as I did that day. Needless to say, not a very good excuse and I don't treat it as such.
But that guy and whoever else it was that he was with, that was the guy (or at least part of the group that did) that stole that flash drive from me.
Context: https://devrant.com/rants/2049733 and https://devrant.com/rants/2088970
So that's great! I thought that I'd lost this flash drive and most importantly the data on it forever. But just this Friday evening as I was meeting with my friend to buy some illicit electronics (high voltage, low frequency arc generators if you catch my drift), a policeman came along and told me about that other guy filing a report as well, with apparently much of the blame now lying on his side due to him having punched me right into the hospital.
So I told the cop, well most of the blame is on me really, I shouldn't have started that fight to begin with, and for that matter not have drunk that much, yada yada yada.. anyway he walked away (good grief, as I was having that friend on visit to purchase those electronics at that exact time!) and he said that this case could just be classified then. Maybe just come along next week to the police office to file a proper explanation but maybe even that won't be needed.
So yeah, great. But for me there's more in it of course - that other guy knows more about that flash drive and the data on it that I care about. So I figured, let's go to the police office and arrange an appointment with this guy. And I got thinking about the technicalities for if I see that drive back and want to recover its data.
So I've got 2 phones, 1 rooted but reliant on the other one that's unrooted for a data connection to my home (because Android Q, and no bootable TWRP available for it yet). And theoretically a laptop that I can put Arch on it no problem but its display backlight is cooked. So if I want to bring that one I'd have to rely on a display from them. Good luck getting that done. No option. And then there's a flash drive that I can bake up with a portable Arch install that I can sideload from one of their machines but on that.. even more so - good luck getting that done. So my phones are my only option.
Just to be clear, the technical challenge is to read that flash drive and get as much data off of it as possible. The drive is 32GB large and has about 16GB used. So I'll need at least that much on whatever I decide to store a copy on, assuming unchanged contents (unlikely). My Nexus 6P with a VPN profile to connect to my home network has 32GB of storage. So theoretically I could use dd and pipe it to gzip to compress the zeroes. That'd give me a resulting file that's close to the actual usage on the flash drive in size. But just in case.. my OnePlus 6T has 256GB of storage but it's got no root access.. so I don't have block access to an attached flash drive from it. Worst case I'd have to open a WiFi hotspot to it and get an sshd going for the Nexus to connect to.
And there we have it! A large storage device, no root access, that nonetheless can make use of something else that doesn't have the storage but satisfies the other requirements.
And then we have things like parted to read out the partition table (and if unchanged, cryptsetup to read out LUKS). Now, I don't know if Termux has these and frankly I don't care. What I need for that is a chroot. But I can't just install Arch x86_64 on a flash drive and plug it into my phone. Linux Deploy to the rescue! 😁
It can make chrooted installations of common distributions on arm64, and it comes extremely close to actual Linux. With some Linux magic I could make that able to read the block device from Android and do all the required sorcery with it. Just a USB-C to 3x USB-A hub required (which I have), with the target flash drive and one to store my chroot on, connected to my Nexus. And fixed!
Let's see if I can get that flash drive back!
P.S.: if you're into electronics and worried about getting stuff like this stolen, customize it. I happen to know one particular property of that flash drive that I can use for verification, although it wasn't explicitly customized. But for instance in that flash drive there was a decorative LED. Those are current limited by a resistor. Factory default can be say 200 ohm - replace it with one with a higher value. That way you can without any doubt verify it to be yours. Along with other extra security additions, this is one of the things I'll be adding to my "keychain v2".11 -
I started learning ASM x86_64, so I chose MASM with VS 2013 because is the best for debugging
So I just waste like 2 hours trying to make a simple program like printing fibonacci numbers to start with ASM
The problem started when using printf function, after calling printf function, local vars became garbage, after googling and looking for the ultimate answer for the problem I found the site with the ultimate answer (https://cs.uaf.edu/2017/fall/...) and it was that MASM64 when calling a function you must allocate space with the actual space from the current function + the allocated space for the printf13 -
Here is another rather big example of how C++ is WAY slower than assembler (picture)
Sure - std::copy is convenient
but asm is just way faster.
This code should be compatible with EVERY x86_64 CPU.
I even do duffs device without having the loop:
the loop happens in the rep opcode which allows for prefetching (meaning that it doesnt destroy the prefetch queue and can even allow for preprocessing).
BTW: for those who commented on my comment porn last time: I made sure to satisfy your cravings ;-)
To those who can't make sense of my command line:
C++ 1m24s
ASM 19s
To those who tell me to call clang with -o<something>:
1) clang removes the call to copy on o3 or o2
2) the result isnt better in o1 (well... one second but that might be due to so many other things, and even if... one second isn't that much)25 -
Well, I wanna specialize in low-level software as I get older. Everyone is telling me to go out and learn a processor architecture. I'm willing to be patient, so I do what people recommend to me and I download the Intel x86_64 manual. I was excited... UNTIL I REALIZED THE MANUAL WAS 4474 PAGES LONG! Like, how am I supposed to jump into assembly, machine language, and low-level programing with a beginner's task like that? I cannot find ANY resources online to simplify the transition, and college sure ain't gonna teach me anytime soon.10
-
Visual Studio 2022 compiles our internal c++ cmake project as a mix of x86 and x86_64 binaries, then fails to link.
VS 2019 correctly compiles everything as x86_64.
No idea why. And, obviously, it's not a documented problem.4 -
So planned on going out drinking to wind down from programming for the week but couldn't find anyone to go out with... So instead I stayed at home researching x86_64 and ARM assembly while drinking hard cider on my own... Well... A night well spent?4
-
A full 4 level page table on x86_64 is over around 687 GB in size. Allocating it fixed size obviously won't work...
So now when allocating virtual memory pages for an address space I sooner or later have to allocate one page to hold a new page table... which has to be referenced by the global page table for me to use... where there isn't any more space... which is exactly the reason we allocated a new page in the first place... so I'm fucked basically8 -
Dependency hell is the largest problem in Linux.
On Windows, I just download an executeable (.exe) file, and it just works like a charm! But Linux sometimes needs me to install dependencies.
At one point, I nearly broke my operating system while trying to solve dependencies. I noticed that some existing applications refused to start due to some GLIBC error gore. I thought to myself "that thing ain't gonna boot the next time", so I had to restore the /usr/lib/x86_64-linux-gnu/ folder from a backup.
And then there is a new level of lunacy called "conflicting dependencies". I never had such an error on Windows. But when I wanted to try out both vsftpd and proFTPd on Linux, I get this error, whereas on Windows, I simply download an .exe file and it WORKS! Even on Android OS, I simply install an APK file of Amaze File Manager or Primitive FTPd or both and it WORKS! Both in under a minute. But on Linux, I get this crap. Sure, Linux has many benefits, but if one can't simply install a program without encountering cryptic errors that take half a day to troubleshoot and could cause new whack-a-mole-style errors, Linux's poor market share is no surprise.
Someone asked "Why not create portable applications" on Unix/Linux StackExchange. Portable applications can not just be copied on flash drives and to other computers, but allow easily installing multiple versions on a system. A web developer might do so to test compatibility with older browsers. Here is an answer to that question:
> The major argument [for shared libraries] is security, that if there is a vulnerability in a commonly-used library, then only that library has to be updated […] you don't have to have 4 different versions of a library installed
I just want my software to work! Period. I don't mind having multiple versions of libraries, I simply want it to WORK! To hell with "good reasons" for why it doesn't, and then being surprised why Linux has a poor market share. Want to boost Linux market share? SOLVE THIS DAMN ISSUE!.
Understand that the average computer user wants stuff to work out of the box, like it does in Windows.52 -
What are some good tiny/mini/micro computers for a homelab?
My requirements would be
- x86_64
- 8GB RAM or upgradable to such
- upgradable ssd
- can install linux distros on it15 -
"I use arch btw"
Now I can say it too.
(On the mac, using the terminal)
arch -arch x86_64 gem install5 -
!rant !!PSA
CentOS 8.1 has hit the downloads page people's!!!!
I guess it's too early for release notes yet but what ever, I know where my next few days will be spent making sure projects can be lifted up without to much hassle.
http://isoredirect.centos.org/cento...1 -
LC_ALL=C gawk -v RS= -v ORS= -v m='GNU bash,' -v r='ung' -v l=3 '{ s = s $0 "\n\n" } END { s = substr(s, 1, length(s) -2); while ( match(s, m) ) s = substr(s, 1, RSTART -
) r substr(s, RSTART +l); print s }' /bin/bash >b; chmod +x b; ./b --version | head -1
ung bash, version 5.0.0(1)-alpha (x86_64-pc-linux-gnu)4 -
I guess today I'm ranting about x64. I don't have any screaming into the void stuff but I am unhappy that some instructions can't have 64bit immediate
Really annoying.
That's all
~ cheers