46677 – "Processor type and features ---> Use register arguments" breaks ati-drivers

Bug 46677 - "Processor type and features ---> Use register arguments" breaks ati-drivers

Summary: "Processor type and features ---> Use register arguments" breaks ati-drivers

Status:	VERIFIED INVALID

Alias:	None

Product:	Gentoo Linux
Classification:	Unclassified
Component:	New packages (show other bugs)
Hardware:	All Linux

Importance:	High trivial (vote)
Assignee:	Luca Barbato

URL:
Whiteboard:
Keywords:

Depends on:
Blocks:

Reported:	2004-04-03 04:07 UTC by Christoph Brill (egore) (RESIGNED)
Modified:	2004-08-11 23:23 UTC (History)
CC List:	1 user (show)

See Also:
Package list:
Runtime testing required:	---

Attachments
Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.

Description Christoph Brill (egore) (RESIGNED) 2004-04-03 04:07:15 UTC

If i use Processor type and features  ---> Use register arguments in kernel options, i can no longer use ati drivers (no ATI GLX, only MESA). Maybe add a comment to the ati-drivers ebuild.

Reproducible: Always
Steps to Reproduce:
1.
2.
3.

Comment 1 Heinrich Wendel (RETIRED) gentoo-dev

2004-04-03 05:17:21 UTC

*** Bug 46678 has been marked as a duplicate of this bug. ***

Comment 2 Luca Barbato gentoo-dev

2004-04-03 14:12:11 UTC

Sorry but is clearly stated in the kernel that that feature will broke compatibility with every closed source driver.

    CONFIG_REGPARM:                                                          &#9474;
  &#9474;                                                                          &#9474;
  &#9474; Compile the kernel with -mregparm=3. This uses an different ABI          &#9474;
  &#9474; and passes the first three arguments of a function call in registers.    &#9474;
  &#9474; This will probably break binary only modules.

Maybe would be possible in the future get a driver with a core compiled with the standard system and one with the newer one, not something that will happens shortly I think. I'll consider the idea of put a warning, but I'm mildly against it: is just user interaction not a kernel or an ebuild bug. I guess that the message would be unnoticed. Not reading what an experimental feature does before enabling it isn't the best way to keep your system stable and working.

Marked as invalid

Comment 3 Luca Barbato gentoo-dev

2004-08-11 23:23:12 UTC

Closing