All of lore.kernel.org
 help / color / mirror / Atom feed
* [Bug 110956] List of 19.20-812932 release mistakes
@ 2019-06-21 19:41 bugzilla-daemon
  2019-06-21 23:19 ` bugzilla-daemon
                   ` (40 more replies)
  0 siblings, 41 replies; 42+ messages in thread
From: bugzilla-daemon @ 2019-06-21 19:41 UTC (permalink / raw)
  To: dri-devel


[-- Attachment #1.1: Type: text/plain, Size: 3267 bytes --]

https://bugs.freedesktop.org/show_bug.cgi?id=110956

            Bug ID: 110956
           Summary: List of 19.20-812932 release mistakes
           Product: DRI
           Version: unspecified
          Hardware: x86-64 (AMD64)
                OS: Linux (All)
            Status: NEW
          Severity: normal
          Priority: medium
         Component: DRM/AMDgpu-pro
          Assignee: dri-devel@lists.freedesktop.org
          Reporter: ashark@linuxcomp.ru

I was repacking amdgpu-pro (ubuntu archive) for Arch linux, and while doing
this, I have noticed many probable mistakes. Here is my list.

- wsa-amdgpu package has empty copyright file
- Release page says that it supports ubuntu x86_64, but actually it can be
installed on ubuntu x86. Is this an error?
- In clinfo package description Homepage link is broken (404 page)
- In alternative dependencies there are such occurances: libva1-amdgpu,
libva2-amdgpu, libvdpau-1-amdgpu. But these are not provided neither by ubuntu
repos, nor in bundled archive. So is it an error?
- Provided libdrm packages have MIT licence. But are they built completely from
open source? In the changelog I can see that it is some
amd-mainline-hybrid-master20190125. And actually if I omit these packages, then
proprietay libgl driver and clinfo utility crashes. I want to avoid needing of
installation of these libdrm packages (libdrm-amdgpu-amdgpu1,
libdrm-amdgpu-common, libdrm2-amdgpu). Is that possible?
- amdgpu-pro(-hwe) and amdgpu-pro-lib32 depend on amdgpu(-hwe), but actually
they should depend on amdgpu-lib(-hwe) (just like open variant packages).
Because of that even when running installer with --no-dkms, it still is in
packages list (because amdgpu(-hwe) depends on it). Is it intensional ar just a
mistake?
- In libgl1-amdgpu-mesa-dri in postinst script in Support I+A hybrid graphics
there is such condition:
if [ -f ... ] && [ "str1" != "str2" ]; then
You just compare two different strings? I guess you wanted to compare folder
contents. But now that condition will always be false.
in if [ "${f%%/*}"... there is extra percent symbol, you should remove it. Btw,
in rpm variants scriptlets, it is already fixed.
- In the documentation at readthedocs there is missing information about Open
Vulkan component
- There is actually no PX package in bundled archive, so maybe remove it
completely from installer script and from the documentation?
- Release page says that you need to install lunar_sdk for vulkan to work. Is
it really true? And even then, it says that you need version 1.1.106.0, but in
vulkan-amdgpu-pro inside json file we have 1.1.108. The documentation is
outdated?
- roct-amdgpu-pro and roct-amdgpu-pro-dev have MIT licence. Are they actually
open souce components? If yes, then why do they called with -pro suffix?
- Can you remove hardcoded declining of installation on unsupported ubuntu
release? Just leave a warning (as you let it for non-ubuntu systems), but let
the users to decide. Otherwise, they need to edit release version or modify
debian packages internals (I even did a special script for this and published
in this bugtracking system).

-- 
You are receiving this mail because:
You are the assignee for the bug.

[-- Attachment #1.2: Type: text/html, Size: 4590 bytes --]

[-- Attachment #2: Type: text/plain, Size: 159 bytes --]

_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

^ permalink raw reply	[flat|nested] 42+ messages in thread
end of thread, other threads:[~2019-11-19  8:00 UTC | newest]

Thread overview: 42+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2019-06-21 19:41 [Bug 110956] List of 19.20-812932 release mistakes bugzilla-daemon
2019-06-21 23:19 ` bugzilla-daemon
2019-06-22  4:53 ` bugzilla-daemon
2019-06-22 15:41 ` bugzilla-daemon
2019-06-22 15:42 ` bugzilla-daemon
2019-06-22 15:45 ` bugzilla-daemon
2019-06-22 15:46 ` bugzilla-daemon
2019-06-22 15:48 ` bugzilla-daemon
2019-06-22 15:49 ` bugzilla-daemon
2019-06-22 15:49 ` bugzilla-daemon
2019-06-22 15:52 ` bugzilla-daemon
2019-06-22 15:54 ` bugzilla-daemon
2019-06-22 16:11 ` bugzilla-daemon
2019-06-22 16:13 ` bugzilla-daemon
2019-06-22 16:13 ` bugzilla-daemon
2019-06-22 16:37 ` bugzilla-daemon
2019-06-22 16:48 ` bugzilla-daemon
2019-06-22 17:57 ` bugzilla-daemon
2019-06-22 17:57 ` bugzilla-daemon
2019-06-22 17:57 ` bugzilla-daemon
2019-06-22 17:57 ` bugzilla-daemon
2019-06-22 17:57 ` bugzilla-daemon
2019-06-22 17:57 ` bugzilla-daemon
2019-06-22 17:57 ` bugzilla-daemon
2019-06-22 17:57 ` bugzilla-daemon
2019-06-22 17:57 ` bugzilla-daemon
2019-06-22 17:57 ` bugzilla-daemon
2019-06-24  8:59 ` bugzilla-daemon
2019-07-29 20:05 ` bugzilla-daemon
2019-07-29 20:08 ` bugzilla-daemon
2019-07-29 20:15 ` bugzilla-daemon
2019-07-29 20:24 ` bugzilla-daemon
2019-07-29 20:28 ` bugzilla-daemon
2019-07-31 15:24 ` bugzilla-daemon
2019-11-19  8:00 ` bugzilla-daemon
2019-11-19  8:00 ` bugzilla-daemon
2019-11-19  8:00 ` bugzilla-daemon
2019-11-19  8:00 ` bugzilla-daemon
2019-11-19  8:00 ` bugzilla-daemon
2019-11-19  8:00 ` bugzilla-daemon
2019-11-19  8:00 ` bugzilla-daemon
2019-11-19  8:00 ` bugzilla-daemon

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.