Breaking News

by TheGuardian · December 17, 2022

Team PSXITA have released Psxitarch v3 for PS4, a Linux distribution entirely optimized for the PS4. This distro release is really different from most other Linux releases for PS4 out there, because it’s been thought from the ground up to run on PS4/PS4 Pro.
Psxitarch v3 is a Linux distribution running on your Jailbroken PS4. It includes the following features:
The distribution comes preinstalled with utilities (vlc for movie playback, etc…) as well as the following emulators:
Noob404 has a great review video of this Linux distribution for those of you who want to check it out:
Note: for details and an up to date installation guide, make sure to check the original article at PSXITA.
Requirements
Note: there were a few issues with their uploaded files when the team initially made the release. If you downloaded their release within the first days after it was announced, you might have to redownload Psxitarch v3.
Portal 2 Running on Psxitarch v3. Source Noob404
INSTALLATION GUIDE
The distro does not require you to enter a username and password at login but you will need them if you want to obtain root permissions:
username:  psxita
password:  changeit
Source: PSXITA
Tags: Linux on PS4Mesa DriversPSXitaPsxitarch v3
It possible to run desktop over 1080p on PS4 Linux ?
I don’t know the answer to this (yet) but I can shed some light.
The PS4 is not a PC, in particular, the ASIC that ties all the components together is custom, and the standard HDMI encoder is setup in a non-standard way. All this means that HDMI functionality is limited, and it took a number of developers and patches to get to where it is today (720p or 1080p, with audio, with no modesetting). What doesn’t work and has never worked is querying the monitor information (EDID) from Linux. Someone (fail0verflow? ChendoChap? marcan?) figured out if we couldn’t get that information from the HDMI, we could get it from inside the PS4 kernel before Linux boots, and pass it to the initramfs. I have all this working correctly, and it results in my 4k Samsung only reporting 1080p support.
The PS4 features HDMI 1.4 which is capable of 4k30. I have a feeling the EDID information taken from the PS4 doesn’t include the 4k modes because the PS4 doesn’t support them. It will take more experiments to figure out properly. If anyone wanted to experiment themselves, there is a wealth of information online on how to make a custom EDID.
“Someone (fail0verflow? ChendoChap? marcan?) figured out if we couldn’t get that information from the HDMI, we could get it from inside the PS4 kernel before Linux boots, and pass it to the initramfs”
No edid stuff come from me: https://github.com/Ps3itaTeam/ps4-kexec/commit/8e44f5485f25d95688ea934b48279a58142828ce
And the payload always pass a kernel cmd line to load the edid from the distro so you don’t have uncovered nothing of new
I don’t test on ps4 fat but i know for sure you can have 4k modes on ps4 pro. Someone tell me it just works, i know how do but is not my work, so i just waiting that this person release the method, because i respect the works of other, unlike someone else
I guess by this response you are a member of team PSXITA. First off, I apologise for any offence caused. I did not mean to misattribute the hard work of anyone, hence the question marks – I was looking to be corrected. Next time I try to faithfully attribute the work of others, I will not forget your name Rancido. Thank you for your dedication and hard work to the scene.
And to address the rest of your response, I didn’t claim I uncovered anything new…?
What is the minimum fw version for a pro to run this?
The scene is currently focused around 9.00 (which also happens to be the most recent exploitable firmware) because compared to some of the older hacks, the exploit chain is ‘reliable’. However, payloads exist to boot Linux for all exploitable firmware versions – nothing about this distro is 9.00 specific.
The payload plays an important roll. The moment before Linux boots is the last opportunity to exercise control over a number of parameters including GPU frequencies and voltage parameters. The PSXITA payloads for 9.00 linked in the article will include the patch to set all eight GPU cores to 800MHz (originally authored by Eeply for 5.05). Most online exploit servers do not include this patch, with the notable exception of Nazky’s (I think) which even goes as far as overclocking the GPU to 900MHz.
I don’t understand why they don’t provide bzImage files anymore. My console, 2216A, only runs with their old official 4.14.x bzImage. Though they say that now 4.15+ is required. So why they don’t provide a new image that is 4.15+? The third party ones don’t work for this console version. I think I’m unlucky to have this specific console version now.
I believe PSSXITA shies away from the task due to the complexity and difficulties the community faces as outlined in my other comments, instead, opting to put that responsibility on the user. A comment on PS4Linux.com (a website that can’t decide if its for the community, or some weird racketeer of “make money off your PS4 $$$”) by Noob404 made me laugh, he said “This can be quite tricky” in reference to getting the EDID from exploit to Linux via initramfs xD This is the guy leading the scene?! Oh dear.
xsx>all
DuckStation + PS4 + PGXP = Better than First
“Entirely optimised” must be a relative term here – last I looked these guys aren’t even using march=btver2 or mtune=btver2 to enable the optimisations added to C compilers by Sony specifically to target PlayStation 4’s AMD Jaguar hardware.
Psxitarch releases are the epitome of the PS4 Linux scene – broken distributions lauded by people with less experience, that take entirely the wrong approach to a rolling distribution like Arch Linux, resulting in fragile setup that can and will easily break through system updates.
Their vkcube isn’t even hit 60FPS, I dread to imagine the mistakes and oversights that led to such poor performance.
It is false and damaging to the scene to continue to pedal the mistruth that you can use any kernel, or any initramfs (especially THAT initramfs). Both are total misconceptions and I have had to debug this the hard way. The responsibility for this lies at the feet of Psxitarch and many other prominent members of the scene for not doing due diligence before making confident statements of self promotion. I appreciate there is difficulty in the fact there are well over 10 different major and minor hardware revisions, but it doesn’t excuse the mistakes.
This initramfs includes the line `cp -R /lib/firmware/amdgpu /newroot/lib/firmware` which copies a set of firmware live patched by the exploit to your system. These firmwares are mostly garbage and you can end up with terrible system performance (like a vkcube that can’t manage 60FPS as demonstrated by this distro on YouTube). In my experience, these firmwares are very unstable – maximising vkcube will result in a graphics ring timeout, locked system and potential kernel panic.
This gets complicated, and I do intend on relaying my findings beyond some random comments at wololo.net. For anyone who can understand me and is still with me at this point, and has experienced the firmware heck I have, find a kernel that embeds the firmwares via CONFIG_EXTRA_FIRMWARE, and build with ps4gentoo firmware.zip. I am running this with codedwrench patchset and my own changes applied to the latest 5.15.83 kernel and latest mesa/libdrm and performance is good (but there is still so much work to be done). Remember the priority for firmware loading is kernel > initramfs > rootfs. If you use a kernel that doesn’t embed firmware, be sure your initramfs script is not clobbering any changes you make to your rootfs.
The initramfs on ps4 just works as a bootloader and if you don’t embed the amdgpu firmware on the kernel also provide the amdgpu ucode, so yes you can use any initramfs you want, but we recommended use our because it contain the script to install psxitarch and READ HERE for ps4 “normal” also provide UPDATED UCODE (amdgpu firmware) that we built as advised from failoverflow (see here for the detail: https://github.com/fail0verflow/ps4-overlay/tree/master/sys-firmware/ps4-ucode), we don’t include updated ucode for PS4 PRO because we don’t have one and can’t test the ucode on it, but should works, you can test easily, just extract the initramfs, rename the ucode from liverpool to gladius and rebuild the initramfs.
The installer script of psxitarch copy the ucode from initramfs to the distro because if someone want build a kernel with amdgpu drm build as module he needs ucodes to be on the distro, otherwise the ucode on the distro are not loading at all.
At the end does’t matter if you load the ucode from initramfs, kernel or from the distro performance are the same.
I reading what you wrote here, YOU wrote:
” I have, find a kernel that embeds the firmwares via CONFIG_EXTRA_FIRMWARE, and build with ps4gentoo firmware.zip.”
And this make just no sense, *** are ps4gentoo firmware.zip, you are so skilled that you use random kernel with random ucode embed, indeed build your own?
i’m serious now if you really have the latest mesa release working with aco compiler, please share it to ALL and stop *** talking.. Just share you knowledge and help ps4 linux to grown, as a lot of guys do, otherwise stop throwing *** on peoples works, you can talk bad about us, I don’t care, but not about all the other guys who have been banging away over the years.
ps4gentoo’s firmware.zip (available via their GitHub) is a mashup of firmware drivers that shouldn’t work but does. I wrote a firmware parser for analysis (I literally could not have got my PS4 working without doing this). ps4gentoo firmware consists of:
* hawaii_ce.bin (IP: 7.4, version: 0x07a, crc32: 88e03b6e)
* liverpool_me.bin (IP: 7.4, version: 0x0bb, crc32: 6e33c535)
* kaveri_mec.bin (IP: 7.2, version: 0x1a5, crc32: 416ba309)
* kaveri_mec2.bin (IP: 7.2, version: 0x1a5, crc32: 9db2195a)
* liverpool_pfp.bin (IP: 7.4, version: 0x0e5, crcr32: b0502bcd)
* liverpool_rlc.bin (IP: 7.2, version: 0x010, crc32: 31e8dbd6)
* sea_islands_sdma.bin (IP: 2.1, version: 0x04c, crcr32: 890958b4)
* sea_islands_sdma1.bin (IP: 2.1, version: 0x04c, crc32: 3bde1dc6)
These firmwares worked with Noob404’s Fedora (which served as my testing bed, thanks) and subsequently my Arch Linux, whereas the exploit dumped firmware is very unstable and has terrible performance. I have spent a lot of time on this, you can’t tell me that I am wrong. Just because it has worked for you does not mean it has worked for me, and that is the crux of the issue here.
Don’t straw man me regarding the ps4gentoo.zip, it does make sense, the branch of 5.3.7 kernels it spawned are well renowned throughout the community for their Vulkan stability. I wonder why that might be?
I do intend to share my progress, my PKGBUILDs and my patch sets, in my own time, likely through GitHub. However, I am hesitant to share anything before I have fully figured it all out, or we will end up back here again. I am well aware I am treading in the footsteps of giants, yourselves included, but I am not so foolish as to think the giants are unique in their ability to figure this out. Right now you are not accepting of my findings which I expected, and I can’t give you the final answer you wish for at this moment in time, but this doesn’t dishearten me in the slightest, and it doesn’t take away my working PS4 Linux install that I can witness and experience with my own eyes.
Lets talk about the real changes here – 1000 lines of code built against a 10 year old GUI framework, all to add a custom keyboard language selector that XFCE has a perfectly good UI for. Thanks team PSXITA, I love staring at the output of localegen for extended periods of time. OK I will see myself out.
That script was done because you know i prefer to avoid installing a mess of unnecessary dependencies from xfce
And there’s not just a script to select the languages in the distro, there’s ps4fancontrol, there’s the led color changer, there a program with you can send icc cmd, just to named some and distro was build from scratch we don’t just get a existing distro and put a wallpaper and mesa driver.
What about you, what do you do to make linux better on PS4 besides make a list of well know problems?
Do not be concerned with the output of others. I have many goodies for the community and I will release them when they are ready.
I am commenting in part because I am reluctant to spend any more time on this. I don’t want to interact with the scene/community directly just now, but at the same time I would love to gauge how many people would stand to benefit from the missing features I am looking in to (HDMI, AMD power management, UVD, VCE etc). Sure the PlayStation 4 has sold 120 million units, but the number of people interested in Linux could be 120 for all I know. I have had a lot of fun over the last couple of weeks figuring all of this out, but the longer I spend on this the more I think about it, which is a terrible thing.
This started as a hobby project, a potential stopover before buying a new desktop (laptop died). I have seen more than enough to know the 1.6GHz CPU clock speed is a huge limiting factor, and it will be a problem going forward (there is an internal timing issue with overclocking). It makes this all feel a bit gimmicky – Control will always be the showcase because it its excellent at multithreading – throw something like Dolphin + F-Zero GX, Guild Wars 2 or GTA IV at it, and the poor single threaded performance will always bottleneck the system.
And in terms of energy efficiency, the 28nm CPU delivers a cost to performance ratio comparable to that of a 1950s filament bulb vs modern LED, so why am I wasting my time? Oh, and I am still salty Sony swiped OtherOS away from us!
Hi I am a mod for the ps4homebrew subreddit and consolemods.org which I mention only to say that I have access to a sample size of a few thousand users. So your question about gauging interest in linux I may be able to shed some light on. Running linux seems to be something most users try at least once, but only maybe half actually use it beyond that initial time. Of those continued users some don’t have any alternative and basically use it as a pc, some use it for emulators not available natively on ps4 (yet), and some use it for some other specific purpose. So I can’t say with certainty that any of the specific stuff you have mentioned will be used, but it can’t hurt and there is at least noticeable interest in linux in general.
One last thing I would like to shout into the void here – in case I have not dashed the readers’ expectations enough – there is no kernel available with stable network connectivity. The Wi-Fi NIC disappears from the system entirely, and the LAN driver has an incorrectly handled interrupt that consumes 100% of a CPU core when plugged in. This can be seen on my system across all distros, my own included.
I hope this gets approved – some slightly different content to the every day ramblings of James Cameron and Firstus for the Wololo faithful.
Oh and about new kernels we don’t take a look in it from the release of 4.14.93, that is because we just tell to use your favorite kernel >= 4.15.xx (vulkan not works below that version), we know that a lot of people port kernels to new version but we don’t know what works betters and the changes they make
And to put an end to this, all this drama don’t help linux to grow, stop talking and share your work.
Hello Rancido, maybe you remember in the psxita forum you once posted a 4.14 Baikal kernel that had support for Wi-Fi dongles. I can confirm it works wonderfully for psxitarch v2 with Android USB tethering. However, it seems none of the 3rd party 4.15+ Baikal kernels support USB tethering. Maybe you can do it once more, for 4.15+? So that we have tethering on psxitarch v3. I think no one else will do it.
Thank you to all the developers working on this, looks pretty nice.
Follolow:
PS5 security: Developer ZNullPtr is seeking donations to make progress on PS5 APU, reverse engineering
14 Jan, 2023
PS Vita: Project Amber (Henlo Jailbreak for PS Vita 3.74) updated, adds SD card(SD2Vita) support for PSV 1xxx
13 Jan, 2023
PS5 Firmware 6.50 released. As always, think twice before updating.
12 Jan, 2023
PS Vita Release: NetStream 2.01 (Youtube unofficial app for PS Vita)
11 Jan, 2023
Nintendo DSi gets a browser exploit, almost 15 years after its release
11 Jan, 2023
PS Vita Release: HENlo bootstrap 1.03 (PS Vita 3.65-3.74 Jailbreak)
10 Jan, 2023
PS Vita Release: NetStream 2.01 (Youtube unofficial app for PS Vita)
11 Jan, 2023
PS5 Firmware 6.50 released. As always, think twice before updating.
12 Jan, 2023
PS Vita Release: Vita Battery Fixer 3.0
8 Jan, 2023
Nintendo DSi gets a browser exploit, almost 15 years after its release
11 Jan, 2023
content Manager assistant
How to get your hands on a PS4 with Firmware 9.00
How to purchase US PSN Cards when you don't live in the US
PS4 Media Server
Wololo.net © 2023. All Rights Reserved.
Powered by WordPress. Theme by Alx.

source

Leave a Reply

Your email address will not be published. Required fields are marked *

Share Article: