Update #2: See this post for full installation instructions.
Update: Downgrading to the X.org X11 server version 1.16 might not be so troublesome after all. Check out the prototyped procedure I’ve posted.
Well, AMD released a new general Catalyst driver for Linux, but unfortunately the AMD site declares that it supports X.org 6.4 or later but only through 1.16, despite the fact that the Ubuntu fglrx 15.20 driver supports Xorg X11 1.17. Michael Larabel infers that the value 15.101 is less than 15.20, but a commentator seems to point out (rightly?) that the 15.101 value is likely intended to be later than the 15.20 value. Perhaps some information from /var/log/Xorg.0.log will help in determining its proper placement:
[ 12.884] (II) Loading /usr/lib64/xorg/modules/linux/libfglrxdrm.so [ 12.888] (II) Module fglrxdrm: vendor="FireGL - AMD Technologies Inc." [ 12.888] compiled for 1.4.99.906, module version = 15.10.4 [ 12.888] (II) AMD Proprietary Linux Driver Version Identifier:15.10.4 [ 12.888] (II) AMD Proprietary Linux Driver Release Identifier: 15.101.1001 [ 12.888] (II) AMD Proprietary Linux Driver Build Date: Apr 14 2015 15:18:13
It looks like it was built a month and a half ago, and the driver version is 15.10.4, which would make it likely less than 15.20. Given the feature disparity regarding support for Xorg X11 1.17, my guess is that it’s older than the latest Ubuntu driver, as well.
But that’s all purely academic. The point is, the segmentation fault persists when one attempts to install in Fedora 22:
[ 13.405] (--) Chipset Supported AMD Graphics Processor (0x990C) found [ 13.407] (WW) fglrx: No matching Device section for instance (BusID PCI:0@0:1:1) found [ 13.408] (II) fglrx(0): pEnt->device->identifier=0x1b7e740 [ 13.408] (II) fglrx(0): Loading PCS database from /etc/ati/amdpcsdb /etc/ati/amdpcsdb.default [ 13.408] (EE) [ 13.408] (EE) Backtrace: [ 13.412] (EE) 0: /usr/libexec/Xorg (OsLookupColor+0x139) [0x599dd9] [ 13.415] (EE) 1: /lib64/libc.so.6 (__restore_rt+0x0) [0x7fdb1cc5cb1f] [ 13.421] (EE) 2: /usr/lib64/xorg/modules/drivers/fglrx_drv.so (xdl_xs116_atiddxLeaveVT+0x30) [0x7fdb197401e0] [ 13.422] (EE) 3: /usr/libexec/Xorg (xf86DeleteScreen+0x5a) [0x483c6a] [ 13.422] (EE) 4: /usr/libexec/Xorg (InitOutput+0xc93) [0x47d463] [ 13.422] (EE) 5: /usr/libexec/Xorg (remove_fs_handlers+0x22a) [0x43e14a] [ 13.423] (EE) 6: /lib64/libc.so.6 (__libc_start_main+0xf0) [0x7fdb1cc48790] [ 13.423] (EE) 7: /usr/libexec/Xorg (_start+0x29) [0x428729] [ 13.423] (EE) 8: ? (?+0x29) [0x29] [ 13.424] (EE) [ 13.424] (EE) Segmentation fault at address 0x10 [ 13.424] (EE) Fatal server error: [ 13.424] (EE) Caught signal 11 (Segmentation fault). Server aborting
As expected.
So, we can hope that AMD is going to release a driver supporting the X11 1.17 server, or we can try to figure out the pacman package from the previous post. I don’t see many other options for us. I considered the plausibility of downgrading Fedora 22 to the Fedora 21 X11 1.16 package, but my guess is that that will cause more problems than it’s worth (that could be wrong, though).
Bummer.
Engineering Walden 
Ok. So … Sorry for my rudeness, but all I can express from end-user point of view – nVi.. khemmnn, AMD, Fuck You.
Besides, anyone willing to ‘rise up ticket’ in AMD: http://support.amd.com/en-us/contact/email-form ? 🙂
This Driver is functional in Desktop Computer , in the case of laptops, which would be the installation method
I’m not sure I understand – the procedure should be the same for desktops and laptops, though laptops are probably more likely to have AMD graphics cards paired with Intel CPUs, making them susceptible to the issue documented here. Give the procedure a try on your laptop and let me know how it goes.
As an AMD card owner and Fedora user I can only say thanks for share with us your researchs making catalyst work on every Fedora release.
Hey man, glad you could benefit from it! Thanks for the comment.