Re: [arch-general] Issue building Electron 1.8
Hi Jonathon, That might be; will probably ping the Electron team for getting a bit of help on this. Any other suggestions? Possibly from Electron official packagers. Thanks in advance, Giovanni Santini Il giorno gio 11 apr 2019 alle ore 21:58 Jonathon Fernyhough < jonat...@manjaro.org> ha scritto: > On 11/04/2019 13:08, Giovanni Santini via arch-general wrote: > > I am trying to build Electron 1.8.X for an AUR package which requires it. > > Given the last commit to the 1.8.x branch was in November 2018 [1] I > suspect it probably doesn't support LLVM 8. > > [1] > > https://github.com/electron/electron/commit/f1a1197e69b5c386ce8a9b6279c2c211f9a5fb9f > >
[arch-general] Issue building Electron 1.8
Hi everyone, I am trying to build Electron 1.8.X for an AUR package which requires it. In order to build it, I checked out the `electron` package, switched to the commit of the latest 1.8 and built the package. Everything goes quite well, except for the final linking step, which gives an error as in here: https://pastebin.com/WScSCUP5 I am not really sure why this happens and would like to have some help understanding what is wrong; as soon as I/we can narrow down the issue, I can work on it and/or report it to the Electron devs. :) Thanks in advance, Giovanni Santini
Re: [arch-general] Tips for a stable GNOME Shell?
Hello again,
Thanks to your suggestions, I decided to
1. Replace GDM with LightDM
2. Install LXDE as lightweight DE
The installation was easy and I am really glad of the RAM performance
obtained (I suppose GDM was the real resource hog).
These is only one small desire that is still unexpressed for me: I both
have mutter and openbox locally installed and I would like to use only
mutter (looks much better) when using LXDE.
The problem is that I have literally no idea of how to set up its keys;
from its man, I should change them from the Control Center, that is not
present in LXDE (it's GNOME stuff) and I am not sure of which GNOME
settings services I should execute to make it work. If anyone has any
tips, please speak ahead!
Thanks again to all, anyways :)
--
Giovanni Santini
My blog: http://giovannisantini.tk
My code: https://git{hub,lab}.com/ItachiSan
My GPG: 2FADEBF5
Re: [arch-general] Tips for a stable GNOME Shell?
Il 08/02/2018 00:36, strupo ha scritto:
> On Wed, 7 Feb 2018 21:28:26 +0100
> Giovanni Santini via arch-general wrote:
>
>> Thanks for the suggestion regarding the Plasma DE, I never heard of it
>> before.
>
> You may know it better by the name KDE.
>
Oh. Thanks to point it out.
Am I so ignorant in the matter? Doh.
--
Giovanni Santini
My blog: http://giovannisantini.tk
My code: https://git{hub,lab}.com/ItachiSan
My GPG: 2FADEBF5
Re: [arch-general] Tips for a stable GNOME Shell?
Good evening,
Thank you so much for your input; it is really appreciated.
For you information, I am using a (kinda old) Acer E1-570G, which has
4GB of RAM and a i5 3337 CPU.
Thanks for the suggestion regarding the Plasma DE, I never heard of it
before; I will try to give it a shot, considering XFCE (which was
suggested) as possible fallback.
I am also interested in Michael's setup; except for ZRam, did you use
anything else?
Thanks again to everyone!
--
Giovanni Santini
My blog: http://giovannisantini.tk
My code: https://git{hub,lab}.com/ItachiSan
My GPG: 2FADEBF5
[arch-general] Tips for a stable GNOME Shell?
Good evening,
I am writing here since I do believe people here might have found
solutions already to my problems.
Sadly, I am the problem, as I love GNOME Shell (ops).
Jokes aside, I love its interface and behaviour; although, it is really
hard to use it on a real-context basis for me.
What it happens is that if I execute RAM-consuming applications, GNOME
Shell behaves really badly, swapping a lot with memory.
The usual scenario is me trying to send some e-mails, while I have
Visual Studio Code and Firefox for some coding; usually, this leads to
huge slowdown, up to making the system unusable.
This doesn't happen when using a GNOME-friendly i3 session, executing by
far many more RAM-consuming applications (such as running Franz with
multiple services, Telegram Desktop and others).
I got some good boosts from the following actions:
- Disabling almost all the Shell extensions, except for my 'essential' ones.
- Using a X11 session instead of Wayland
- Tweaking swap and VFS parameters (there is a web article referenced in
the ArchWiki which is really good)
So I have two questions:
1. Am I nuts? Did I do something really bad to my GNOME Shell without
knowing that? How could I repair my setup?
2. If this is it (GNOME Shell is TOO heavy), is there any lightweight DE
that offers something similar? I would need at least the search within
apps and files for sure.
Thanks in advance for replies and sorry for such a long message.
--
Giovanni Santini
My blog: http://giovannisantini.tk
My code: https://git{hub,lab}.com/ItachiSan
My GPG: 2FADEBF5
Re: [arch-general] Thunderbird doesn't show e-mails
Il 29/10/2017 16:10, Ryan Petris via arch-general ha scritto:
> On 10/29/2017 09:34 AM, Guus Snijders via arch-general wrote:
>>
>> Just a wild guess, but you probably have a display filter active in
>> Thunderbird...
>>
>> My first thought was about the protocol (pop3 vs imap), but since the mails
>> are actually there in the mbox file, it'sprobably just a filter.
>>
>>
>> Mvg, Guus Snijders
>
> There's an "Unread" button above the email list on the left, maybe
> that's pressed?
>
Il 29/10/2017 20:30, matthew dyer via arch-general ha scritto:
>
>
> Hi,
>
> Under view, menu, there should be an opsion to show read messages. HTH.
>
> Matthew
>
>
>
> Sent from Mail for Windows 10
>
Thank you for all the suggestions, however it is not so trivial.
I tried all your options and nope, nothing worked.
However, enabling quick filter display some messages! Not really the
best though.
--
Giovanni Santini
My blog: http://giovannisantini.tk
My code: https://git{hub,lab}.com/ItachiSan
My GPG: 2FADEBF5
[arch-general] Thunderbird doesn't show e-mails
Hello everyone,
I have a weird issue.
My uni e-mail uses Office 365 as e-mail provider; I've added it in
Thunderbird, but I can see only unread e-mails. All the read one are not
appearing in my inbox.
However, if I `cat` the INBOX file, I can see the e-mails and if I mark
them as unread from somewhere else I can see them.
Also, adding the e-mail address on Thunderbird on Windows doesn't have
the same problem.
Any idea to further investgation?
--
Giovanni Santini
My blog: http://giovannisantini.tk
My code: https://git{hub,lab}.com/ItachiSan
My GPG: 2FADEBF5
Re: [arch-general] Detect broken DHCP setup
Il 06/09/2017 01:09, Leonid Isaev via arch-general ha scritto:
>
> What does it mean a valid DHCP setup? By reconnection you mean that your
> client
> re-request a lease from the server? Also, dbus has nothing to do with dhcp
> settings...
I know DBus has nothing to do with DHCP; what I meant is that
NetworkManager shows in its DBus interface when the DHCP configuration
is not valid, so people can reset it.
And almost, what I would do is to restart the connection.
>
> In any case, my advice is to get rid of NetworkManager as well as systemd-*
> tools. If you want a robust dhcp setup on a simple client with a single
> network
> card, use dhcpcd (no need even for netctl) because it provides link status
> detection. But don't use dhcpcd@.service provided with the package, instead
> replace it with:
> ...
> -<
> The crucial part is "-Bt 0" which makes dhcpcd wait forever for a lease (read
> the manpage for other options you might need, for example, in my setup I
> constrain the demon to only deal with ipv4). My (compatible with read-only
> root
> filesystem) /etc/dhcpcd.conf is:
> ...
> these are mostly default settings. Maybe you need to add "nomtu" in case your
> ISP does something idiotic with this setting (mine does :)). Oh, and hardcode
> the DNS settings in /etc/resolv.conf, so a broken dhcp server has no control
> over them.
>
Thanks a lot for the configuration files and the suggestions! If there's
no better solution I can go for them! :)
I still noticed that systemd exposes a DBus network interface
(*org.freedesktop.network1*) which should have proper information but I
found very little documentation online (if none) regarding it...
--
Giovanni Santini
My blog: http://giovannisantini.tk
My code: https://git{hub,lab}.com/ItachiSan
My GPG: 2FADEBF5
[arch-general] Detect broken DHCP setup
Good morning,
tl;dr: I was wondering what could be the best way to check a DHCP setup.
Briefly: our landlord provides us Wifi through a telephonic company
service. However, the routers placed in the building after some time
lose the DHCP settings.
While I can fix it on my computer with a reconnection, such a method is
not applicable to my Raspberrys.
On one I have Raspbian, on the other ArchLinuxARM. While I found a
possible solution for the Raspbian, for ALARM I have some doubt.
On that device I am using *systemd-networkd + systemd-resolved* for the
network setup. However, I saw no real method to check if the DHCP
configuration is valid (while NetworkManager provides an element through
DBus inspection).
Does anyone has some knowledge about it?
To be fair, I am asking as I am sure the 'ping Google DNSs and if it is
screwed restart the networking, use this as a cron job' will likely
work, but I was interested in a more elegant way to approach it (like
the NetworkManaged DBus introspection).
Thanks in advance to all!
--
Giovanni Santini
My blog: http://giovannisantini.tk
My code: https://git{hub,lab}.com/ItachiSan
My GPG: 2FADEBF5
Re: [arch-general] git undetectable tag replacement? (Was: Sébastien Luttringer and Tobias Powalowski)
Il 03/07/2017 04:14, Eli Schwartz via arch-general ha scritto:
>
> So I was under the impression that git tags encode the tagname in the
> actual blob, and I didn't see how that attack (rooted in the basic
> nature of a branch as a lightweight, mutable, *pushable* pointer to a
> commit) was supposed to work unless of course it was talking about a
> lightweight tag (which is not really meant for public/permanent use)...
>
> Having actually tested this out, I find myself quite bewildered.
>
> Because, git *does* encode the tagname in the blob, like I thought.
> And... you *can* simply copy .git/refs/tags/tagname to create a fake
> tag, and then you see something quite bewildering:
>
> ```
> [eschwartz@arch ~/git]$ git clone https://github.com/systemd/systemd
> [...]
> [eschwartz@arch ~/git]$ cd systemd
> [eschwartz@arch ~/git/systemd]$ echo "$(git show-ref -s v233)" >
> .git/refs/tags/v232
> [eschwartz@arch ~/git/systemd]$ git tag -v v232
> object d60c527009133a1ed3d69c14b8c837c790e78d10
> type commit
> tag v233
> tagger Lennart Poettering 1488405946 +0100
>
> systemd 233
> gpg: Signature made Wed 01 Mar 2017 05:05:46 PM EST
> gpg:using RSA key B63B21879C3485B0
> gpg: Good signature from "Lennart Poettering "
> [unknown]
> gpg: aka "Lennart Poettering "
> [unknown]
> gpg: aka "Lennart Poettering (Red Hat)
> " [unknown]
> gpg: aka "Lennart Poettering (Sourceforge.net)
> " [unknown]
> gpg: WARNING: This key is not certified with a trusted signature!
> gpg: There is no indication that the signature belongs to the
> owner.
> Primary key fingerprint: 63CD A1E5 D3FC 22B9 98D2 0DD6 327F 2695 1A01 5CC4
> Subkey fingerprint: 16B1 C4EE C0BC 021A C777 F681 B63B 2187 9C34 85B0
> ```
>
> I simply do not understand why git doesn't error when it sees that a tag
> ref named v232 points to a tag blob named v233. The information is all
> there, and it seems trivial to add a check.
>
> Oh well, one more reason to use commit hashes rather than pkgver tags.
> (The other reason is, because upstreams themselves are quite capable of
> modifying tags and re-signing them without waiting for some attacker to
> play with the ref pointers.)
>
> Anyway, I learned something today.
>
I think I can reply to this. Just skip if not interesed to last part.
"""Thanks""" to a lab of a class I had the previous semester, I can
describe you exactly how Git blobs are... There are:
- normal *blob*s: standard files compressed using Deflate, have only a
small special header at the top
- *tree*s: a specially encoded file, pointing to blobs or other
*tree*s... call them *blobbed directories*
- *commits*: another deflated file, keeping the commit message and a
reference to a *tree*.
To be 100% I made one just now to check and I can say that we have
- *tags*: a plain file referring to a commit object
You see the point: you just renamed a text file pointing others. Unless
you change the commit binary file (doing this won't pass the gpg check)
or fake the whole .git blobs, it will point to the v233 commit and all
the stuff below it.
If you are instead saying that Git should notice that the file in the
refs folder is named improperly, just `man git` and see what *git* is :P :D
tl;dr: Too much an hassle; destroying a Git repo is easy, faking it is not.
P.S. Thanks to that class, we made some Java classes to handle Git
blobs, check them if you want: https://gitlab.com/ItachiSan/lab01/
--
Giovanni Santini
My blog: http://giovannisantini.tk
My code: https://git{hub,lab}.com/ItachiSan
My GPG: 2FADEBF5
signature.asc
Description: OpenPGP digital signature
Re: [arch-general] Broken devtools?
Il 22/06/2017 23:15, Bruno Pagani ha scritto:
>
> devtools scripts use the first mirror from your system mirror list. Can
> you be more specific about the issue you’re encountering? It’s likely
> that your first mirror does not work for instance, in which case
> devtools fails.
>
> Bruno
>
Not really sure of what it happened, before it wasn't working...
With no changes, it started to work again.
I do suppose there was some problem due to the latest pacman update (I
do suppose I've updated pacman and used the chroot before rebooting).
Seems now it is working. It was a weird issue, thanks again.
--
Giovanni Santini
My blog: http://giovannisantini.tk
My code: https://git{hub,lab}.com/ItachiSan
My GPG: 2FADEBF5
[arch-general] Broken devtools?
Good afternoon,
I was trying to build a package with `devtools` *extra-x86_64-build*,
but is doesn't use any mirror.
I've noticed also that adding a mirror to the root chroot mirrorlist has
no effect and gets removed at the first build attempt.
--
Giovanni Santini
My blog: http://giovannisantini.tk
My code: https://git{hub,lab}.com/ItachiSan
My GPG: 2FADEBF5
Re: [arch-general] ReadyDLNA/MiniDLNA doesn't work behind wireless
Il 05/06/2017 23:11, Mike Cloaked via arch-general ha scritto:
> On Mon, Jun 5, 2017 at 4:12 PM, ITwrx.org wrote:
>> i just installed minidlna and got it working from vlc over wireless. it
>> didn't work at first. "systemctl status minindlna" showed that it wasn't
>> able to identify the network interface i provided. i have no idea
>> why(old method in minidlna?). i commented that line in config back out
>> and made sure minidlna could read my testing media directory and it
>> worked at that point. i used nmap -sS -sU -T4 -A -v 192.168.1.1 to see
>> if port 8200 was open. it was.
>>
>> so, if "systemctl status minidlna" or "journalctl -b" shows no problems
>> for minidlna then maybe test with nmap from both wired and wireless and
>> see if there is a difference. If there is, then you have network config
>> issue. the arch wiki page for minidlna has a section about wireless in
>> the troubleshooting section i notice.
>>
>
> Although I don't run minidlna over the wireless interface on the server I
> use, and only use the wired interface, I presume that the key here may be
> as you suggest that the lines in /etc/minidlna.conf near the start of the
> file that are relevant, which in my case are:
>
> # port for HTTP (descriptions, SOAP, media transfer) traffic
> port=8200
>
> # network interfaces to serve, comma delimited
> #network_interface=eth0
> network_interface=eth0
>
> would need to include the correctly specified interface(s) for the server
> running minidlna for the OP. I wonder what the latter line actually is for
> that file in the problem machine?
>
Good morning,
Thanks to everyone for your support.
Seems the problem was different, as my router started acting really weird.
I just replaced it and MiniDLNA 'n stuff work flawlessly.
So, always replace defective routers. :P
Thanks again.
--
Giovanni Santini
My blog: http://giovannisantini.tk
My code: https://git{hub,lab}.com/ItachiSan
My GPG: 2FADEBF5
Re: [arch-general] ReadyDLNA/MiniDLNA doesn't work behind wireless
Il 04/06/2017 18:08, ITwrx.org ha scritto:
> this computer is the one you successfully accessed the minidlna web page
> from? or a different computer connected via ethernet? i'm just trying to
> verify that you have successfully accessed the webpage for minidlna from
> a wireless device/computer.
>
> if yes, then you could use nmap or a nmap gui (nmap-qt4 ?) to scan the
> host that is supposed to be serving minidlna and make sure that port is
> actually open/visible. IOW, just start ruling things out. If that port
> is open/accessible from a remote machine then try different apps on the
> clients to make sure it's not just buggy clients. you might also try
> different minidlna servers. i know when i tried to use minidlna the
> servers and client apps were very picky and some were very flakey too. i
> ended up ditching minidlna(even though i had it working most of the
> time) and just using samba just because i didn't like how non-robust it
> all was.
>
The computer I was referring to is the server I am trying to access.
I've accessed the webpage from my laptop.
There is a strange thing indeed: the server has actually no firewall,
however the 1900 UDP port, which should be open for SSDP seems closed.
Here it is the result of some investigation
---
# On my laptop
$ nmap -sU -p 1900 santini_server
Starting Nmap 7.40 ( https://nmap.org ) at 2017-06-05 11:32 ora legale
Europa occidentale
Nmap scan report for santini_server (192.168.0.109)
Host is up (0.027s latency).
PORT STATE SERVICE
1900/udp closed upnp
MAC Address: 00:1E:2A:43:47:3E (Netgear)
Nmap done: 1 IP address (1 host up) scanned in 12.43 seconds
# On the server
$ sudo iptables --list
Chain INPUT (policy ACCEPT)
target prot opt source destination
Chain FORWARD (policy ACCEPT)
target prot opt source destination
Chain OUTPUT (policy ACCEPT)
target prot opt source destination
---
No idea why MiniDLNA does not check the port.
--
Giovanni Santini
My blog: http://giovannisantini.tk
My code: https://git{hub,lab}.com/ItachiSan
My GPG: 2FADEBF5
Re: [arch-general] ReadyDLNA/MiniDLNA doesn't work behind wireless
Il 04/06/2017 17:24, ihad ha scritto:
> Hi,
>
> ...
>
> Just a wild guess: is your wireless network a separate subnet? If so,
> there's your problem. AFAIK MiniDLNA uses SSDP[1] to announce itself,
> using a site-local multicast-IP that isn't routed. You need to bridge
> your wireless network into the wired one, meaning it has to be the same
> subnet.
>
Hey,
no, both of them are configured through DHCP and they both use the same
subnet (mask 255.255.255.0, base 192.168.0.0).
--
Giovanni Santini
My blog: http://giovannisantini.tk
My code: https://git{hub,lab}.com/ItachiSan
My GPG: 2FADEBF5
Re: [arch-general] ReadyDLNA/MiniDLNA doesn't work behind wireless
Il 04/06/2017 16:10, ITwrx.org ha scritto:
> The status page is properly accessible from where? A x86 computer that's
> on your wired network? or from a device connected to network via wireless?
MiniDLNA publishes automatically a web page on port 8200 for checking
its status.
My computer is an old x64-capable server.
It is connected to my network through a wireless USB adapter.
--
Giovanni Santini
My blog: http://giovannisantini.tk
My code: https://git{hub,lab}.com/ItachiSan
My GPG: 2FADEBF5
[arch-general] ReadyDLNA/MiniDLNA doesn't work behind wireless
Good afternoon,
as said in the subject I am having some issues running MiniDLNA on my
home server.
The service fires up properly (the status web page is properly
accessible), however the server is not visible when I scan for it in my
DLNA-enanched applications, on all the platforms (my Android phone and
tablet + my computer).
In the past it was working when my home server was connected through
Ethernet, but I had to make a really bad setup and I would prefer to
keep it on wireless.
Did anyone have a problem similar to mine?
--
Giovanni Santini
My blog: http://giovannisantini.tk
My code: https://git{hub,lab}.com/ItachiSan
My GPG: 2FADEBF5
Re: [arch-general] Where do you host your personal repository?
Il 15/05/2017 20:56, Jelle van der Waa ha scritto:
>
> I use a private server, OBS has some major issues (it's crammed, doesn't
> have all the packages due to licensing).
>
Luckily, now Arch:Community is present and works pretty well.
Had only to add ffmpeg to my repo.
>
> What kind of integration are you looking for?
>
Something like:
- Updating my master repo, which contains all the packages as git modules
- Using CI (Travis I suppose) having the package built and deployed to
the releases of the same service
Il 15/05/2017 22:02, SanskritFritz ha scritto:
> I also plan to create a custom repo somewhere, so I have the same question.
> In the meantime you can check out this list, maybe you get some ideas:
> https://wiki.archlinux.org/index.php/Unofficial_user_repositories
If you want some suggestion about using OpenBuildService, just tell me. ;)
--
Giovanni Santini
My blog: http://giovannisantini.tk
My code: https://git{hub,lab}.com/ItachiSan
My GPG: 2FADEBF5
[arch-general] Where do you host your personal repository?
Good morning,
It has been a few time since I am hosting packages on OpenSUSE OBS (Open
Build Service). I find myself kinda good, but I have to tackle some
problems it makes that I don't really like.
So I wanted to ask to you where do you host your packages, if you have a
private server or you take advantage of already present services.
I was also wondering if some integration with Gitlab/Github could be
possible. That would be cool.
--
Giovanni Santini
My blog: http://giovannisantini.tk
My code: https://git{hub,lab}.com/ItachiSan
My GPG: 2FADEBF5
Re: [arch-general] Best practices for creating an Arch Docker image
Il 28/04/2017 07:23, nfnty via arch-general ha scritto:
> I've been maintaining minimal and hardened Arch images and container
> configurations for several years now that are updated regularly [1].
>
> All my images are based on the image `nfnty/arch-mini` [2] which has
> been made to be as minimal and automated as possible. It is built from
> scratch using a bootstrap archive that is built inside another container
> using the image `nfnty/arch-bootstrap` [3]. The bootstrap archive can be
> built outside of the container using the same script [4] as the image does.
>
> You can find many of my images with automated builds on Docker Hub [5].
>
> //nfnty
>
>
> [1] https://github.com/nfnty/dockerfiles
> [2]
> https://github.com/nfnty/dockerfiles/tree/master/images/arch-mini/latest
> [3]
> https://github.com/nfnty/dockerfiles/tree/master/images/arch-bootstrap/latest
>
> [4]
> https://github.com/nfnty/dockerfiles/blob/master/images/arch-bootstrap/latest/scripts/build.sh
>
> [5] https://hub.docker.com/u/nfnty
>
>
Thank you a lot for your feedback. :)
Hope to see the official image soon on Docker Hub. Cheers!
--
Giovanni Santini
My blog: http://giovannisantini.tk
My code: https://git{hub,lab}.com/ItachiSan
My GPG: 2FADEBF5
[arch-general] Best practices for creating an Arch Docker image
Good evening to everybody,
I got interested in Docker lately and I've decided to create an
ArchLinux image for it.
There's one suggested from the ArchWiki (*base/archlinux*) but I wanted
to learn from scratch.
So, I've then some questions:
1. As the root filesystem, I've made a repacked version of the bootstrap
tarball. Even though it is not so clean, it works and it is easy peasy
to do. Should I go still for a `pacstrap`?
2. Theorically, one step of the Dockerfile should be installing the
whole 'base' group, which includes also the kernel, which is not really
needed in a container.
So this questions splits up in 2 parts:
- which packages I can ignore of the 'base' group?
- which packages present in the bootstrap OS can be removed?
This is because I think the Docker image should contain only the least
number of packages of an Arch system; ideally, *pacman* and the needed
core utils.
3. I'm having a GPGME error with the i686 tarball... Upgrading GPGME
breaks pacman, upgrading pacman does the same. If someone is interested
in helping me, I would be glad to share the Dockerfile.
A first working code is here: https://github.com/ItachiSan/dockerfiles
--
Giovanni Santini
My blog: http://giovannisantini.tk
My code: https://git{hub,lab}.com/ItachiSan
My GPG: 2FADEBF5
[arch-general] Bash broken: no PATH and completion
Good morning,
I want to share one of my recent issues with bash.
Everything was working nicely up to some time ago, but recently I'm
having some issues.
The major two things I've noticed are the following:
1. bash doesn't load */etc/profile*, thus not loading all the scripts in
*/etc/profile.d/*
2. bash-completion does not load properly all completion; for example,
the completion for *optirun* is present but it is not loaded.
Does anybody else have issues with it lately?
I didn't notice info about bash config changes.
--
Giovanni Santini
My blog: http://giovannisantini.tk
My code: https://git{hub,lab}.com/ItachiSan
My GPG: 2FADEBF5
[arch-general] Fwd: Re: GRUB doesn't boot Windows behind Secure Boot
Il 07/12/2016 17:25, Giovanni Santini ha scritto:
> Il 29/11/2016 10:49, David Phillips ha scritto:
>>
>> I'm not familiar with GRUB anymore, but it sounds like perhaps the file
>> bootmgfw.efi has not been signed with a key that is trusted by your board's
>> SecureBoot feature. Hence, booting with SB enabled causes a load failure
>> while booting without SB fixes the problem.
>>
>
Good afternoon,
as here I had no major suggestion about what to do, I decided to ask
help in the official mailing list as you can see in [1].
After some tests, we found a small bug in grub that prevented the
chainload of EFI executables.
This fixes my problem. However, it is not present in the official
package yet; I suppose that opening a bug report online should be enough.
[1] http://lists.gnu.org/archive/html/help-grub/2016-12/msg1.html
[2]
http://git.savannah.gnu.org/cgit/grub.git/commit/?id=ce95549cc54b5d6f494608a7c390dba3aab4fba7
--
Giovanni Santini
My blog: http://giovannisantini.tk
My code: https://git{hub,lab}.com/ItachiSan
My GPG: 2FADEBF5
Re: [arch-general] GRUB doesn't boot Windows behind Secure Boot
Il 29/11/2016 10:49, David Phillips ha scritto:
>
> I'm not familiar with GRUB anymore, but it sounds like perhaps the file
> bootmgfw.efi has not been signed with a key that is trusted by your board's
> SecureBoot feature. Hence, booting with SB enabled causes a load failure
> while booting without SB fixes the problem.
>
Hello there,
as said previously, I would have posted here after having back a proper
setup. I had to fight over a week to set up the EFI partition and
Windows Boot Manager properly, but I made it somehow.
So, that's what I did:
- I've resetted the UEFI firmware, so that everything was clean
- I've installed GRUB again and I setted up Preloader and HashTool as
stated at [1]
- With HashTool, I've enrolled the Grub EFI binary and also the proper
Windows EFI binaries.
I still face the same error; additionally, also chainloading HashTool
from GRUB gives me errors (with Secure Boot, from here SB, on, tried
only with SB on as it is useless with SB off).
As before, turning off SB allows GRUB to chainload Windows flawlessly.
Still, I would like to keep SB on.
I'm adding also my grub.cfg file at [2].
Hope to have some feedback soon!
Regards
[1] https://wiki.archlinux.org/index.php/Secure_Boot#Set_up_PreLoader
[2] http://paste.ubuntu.com/23594049/
--
Giovanni Santini
My blog: http://giovannisantini.tk
My code: https://git{hub,lab}.com/ItachiSan
My GPG: 2FADEBF5
Re: [arch-general] GRUB doesn't boot Windows behind Secure Boot
Il 29/11/2016 10:49, David Phillips ha scritto:
>
> I'm not familiar with GRUB anymore, but it sounds like perhaps the file
> bootmgfw.efi has not been signed with a key that is trusted by your board's
> SecureBoot feature. Hence, booting with SB enabled causes a load failure
> while booting without SB fixes the problem.
>
> Like I said, not familiar with GRUB, so not sure if that file is grub's
> problem or not. Just a pointer to something for you to investigate.
>
> Thanks,
> David
>
Thank you for your reply David.
This is strange, as Secure Boot works flawlessly booting directly
Windows Boot Manager (that is also the file mentioned above!).
So maybe GRUB doesn't like it anyways... I will think about it in next
days, as I also broke up Windows Boot Manager (resizing EFI partition is
a nope for BCD settings, seems).
Will ping here after some progress.
--
Giovanni Santini
My blog: http://giovannisantini.tk
My code: https://git{hub,lab}.com/ItachiSan
My GPG: 2FADEBF5
[arch-general] GRUB doesn't boot Windows behind Secure Boot
Good afternoon,
I have been using GRUB for a long time, but I'm having a strange issue.
My setup consists of ArchLinux as main OS and Windows 10 and Ubuntu
12.04 as secondary OSes.
Turning up Secure Boot in my firmware options results in such an error
when chainloading Windows:
/EndEntire
file path:
/ACPI(yadda)/PCI(yadda)/Sata(0,0,0)/HD(yaddayadda)/File(\EFI\Microsoft\Boot)/File(bootmgfw.efi)/EndEntire
error: cannot load image.
The strange thing is that disabling Secure Boot make it works.
Not really sure what I'm missing here.
--
Giovanni Santini
My blog: http://giovannisantini.tk
My code: https://git{hub,lab}.com/ItachiSan
My GPG: 2FADEBF5
Re: [arch-general] RetroArch refuses to go back into windowed mode after fullscreen
You should check if there is some difference between the two kconfig. (Sorry for top post, GMail from Android doesn't allow me an easy reply if not top posting) --- Giovanni Santini My blog: http://giovannisantini.tk My code: https://github.com/ItachiSan My code, again: https://gitlab.com/u/ItachiSan My Twitter: https://twitter.com/santini__gio My Facebook: https://www.facebook.com/giovanni.santini My Google+: https://plus.google.com/+GiovanniSantini/ My GPG: 2FADEBF5 Il 09 giu 2016 11:08 PM, "Diego Viola via arch-general" < arch-general@archlinux.org> ha scritto: > On Thu, Jun 9, 2016 at 2:46 PM, Diego Viola wrote: > > On Thu, Jun 9, 2016 at 11:19 AM, Diego Viola > wrote: > >> Hello, > >> > >> I've been having this poblem for a while now, and I reported the > >> problem here, because I think its' a bug in the Intel driver. > >> > >> https://bugs.freedesktop.org/show_bug.cgi?id=93844 > >> > >> The problem happens when I use retroarch and I play a game, I press F > >> to go into fullscreen mode, but when I press F again, the video is > >> paused and retroarch hangs. > >> > >> This happens independently of the core I'm using in RA. > >> > >> At the moment, the problem is present with the 4.6.1-2-ARCH kernel, > >> but if I compile 4.6.1 myself, the problem does not occur. > >> > >> Is there anything the package maintainers can do about this? Or this > >> is something only the Intel developers can tackle? > >> > >> My system is archlinux x86_64, Intel i915 driver, Dual-Core CPU > >> E5500 @ 2.80GHz desktop machine, 2GB of RAM. I was able to reproduce > >> this on my ThinkPad T450 as well, so I don't think the hardware is > >> that relevant. > >> > >> Thanks, > >> Diego > > > > Tobias, Thomas, any ideas on this? > > > > I've tried compiling 4.6.1 myself and then RetroArch doesn't hang > > after toggling windowed-mode from fullscreen. > > > > It hangs with the stock 4.6.1-ARCH kernel though. > > > > Diego > > It turns out with a clean (make defconfig/menuconfig) of 4.6.2 it > doesn't hang, when I used Arch's /proc/config.gz it hangs. > > By "it hangs" I mean when I switch from fullscreen to windowed-mode on > RetroArch. > > Diego >
