Arch Linux User Repository

Currently fails to build, downgrading binutils to 2.39 on Arch worked ($ downgrade binutils). Issue is related to a change with libbfd.a that binutils made in 2.40.

Related GitHub issue: https://github.com/gnif/LookingGlass/issues/1056

Fails to build as of today due to ZSTD, related GitHub issue: https://github.com/gnif/LookingGlass/issues/1056

@xiretza: So far I've not. It only happened twice over the whole lifespan thus far afaik so I haven't bothered. I don't know about others.

That said, checking the artifact now, the checksum still seems to be the same. I downloaded the artifact fresh (twice to confirm) and it still matches the checksum of PKGBUILD 2:B6-2. It seems it hasn't changed at the moment. Can you recheck on your end?

Looks like it changed again, now it's 25d720c7dd2711c86ebde73c6f4db990b60d4015d075ca4b33b60e16937bf700... Has anyone complained to upstream about this yet?

I have compared the website artifact against the sourcecode and I've been unable to find any differences (maybe the artifact VERSION file changed but that isn't relevant/dangerous here) so I have updated the checksum. The update should be pushed to the AUR shortly.

@untwisted; that is possible if the upstream artifact has changed. This should generally never just randomly happen and if it does, it can be an indication of a malicious actor having been able to either change the artifact or intercept yours somehow.

Now I highly doubt that has actually happened, this seems way to niche of an item to target so that means upstream probably repackaged it for some reason on another. That said, I will not just blindly update the checksum without verifying the new artifact but I'm going to have to do that later.

Looks like the sha512sum is incorrect? Installation was failing so I downloaded the package from the Looking Glass site and ran a sha512 manually and got 8b7a4b6d33380299512bef103331635c9e9855ae0712cbe4d9ccbec2c2499c77318502aa3e0259451ef32c414ef5c8fbc103513bcd924570097dd866934cbe10.

Modifying the PKGBUILD to this SHA value enabled the build to work.

diff --git a/PKGBUILD b/PKGBUILD
index 299d696..b8a340d 100644
--- a/PKGBUILD
+++ b/PKGBUILD
@@ -6,8 +6,8 @@ pkgname=("${pkgbase}"
          "${pkgbase}-host"
          "obs-plugin-${pkgbase}")
 epoch=2
-pkgver=B5.0.1
-pkgrel=3
+pkgver=B6
+pkgrel=1
 pkgdesc="An extremely low latency KVMFR (KVM FrameRelay) implementation for guests with VGA PCI Passthrough"
 url="https://looking-glass.io/"
 arch=('x86_64')
@@ -15,7 +15,7 @@ license=('GPL2')
 makedepends=('cmake' 'fontconfig' 'spice-protocol' 'wayland-protocols'
              'libxss' 'libxi' 'libxpresent' 'obs-studio')
 source=("looking-glass-${pkgver}.tar.gz::https://looking-glass.io/artifact/${pkgver}/source")
-sha512sums=('2600722fba5a9b794556391a636df09d284dc5460c5e9b4718bc2c044e6ece68890893c18c30780245207c0e92194f9691269bdcd590f118a616417d1e74bbd6')
+sha512sums=('402729dc51a6e8995a64384de6e9549861a0db6a87c9627e5832ea548b4ab6b49f538cf12ae510caa360c39806671785ce7560aaf07e7c46532fe0271884f2cb')

 _lgdir="${pkgbase}-${pkgver}"

The DKMS module reports error exit 10, for linux-xanmod-edge kernel, but works with my other kernels. Is there a way to fix this?

Oh, my bad. When I first tried to install obs-plugin-looking-glass there was a conflict warning somewhere so I thought that installing looking-glass had already installed all 4 packages by itself. Thanks for the info.