Re: [git head] Should X86_PAT really default to yes?

On Friday, May 02, 2008 12:37 pm Pallipadi, Venkatesh wrote:
-----Original Message-----
From: Frans Pop [mailto:elendil@xxxxxxxxx]
Sent: Friday, May 02, 2008 12:22 PM
To: linux-kernel@xxxxxxxxxxxxxxx
Cc: Pallipadi, Venkatesh; Ingo Molnar
Subject: [git head] Should X86_PAT really default to yes?

With X86_PAT enabled, when X is started I get about 40 lines
(with varying
addresses) like:
kernel: Xorg:3358 /dev/mem expected mapping type write-back for
807bf000-81000000, got uncached-minus

These messages? They're coming from the kernel it looks like, from the
map_devmem routine in pat.c. I'm not sure they're accurate though; for PCI
regions /dev/mem is *supposed* to map with UC- and not WB, so maybe this
function needs to be updated?

And when X exits I get a bunch of lines like:
kernel: Xorg:3349 freeing invalid memtype 80020000-8002a000

I also noticed artifacts (a band of about 2 cm high across the
screen) after
X goes to black but before the switch to VT1.

This is just a transient issue during VT switch or server exit though, right?
X functionality isn't affected, and your VTs work fine? If so, it might not
be a PAT issue but just a different memory layout or something (and therefore
it would really just be a cosmetic bug in the X driver).

I really think PAT should be on by default; if you're running into real
functional or performance problems we'd better get them fixed rather than
disabling PAT...

Thanks,
Jesse

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

Relevant Pages

  • [GIT PULL] x86 fixes
    ... I wanted to wait one more day with the PAT bits ... ... Also note the kernel might malfunction if you disable ... This parameter disables that behavior, ... extern int kernel_map_sync_memtype(u64 base, unsigned long size, ...
    (Linux-Kernel)
  • mouse problem I believe is in the kernel
    ... And because of the restarts of GPM and X not fixing it, ... suspect it is a kernel issue. ... switch back, GPM no longer works. ... When I unplug the mouse and/or keyboard from Windows 98 and plug them ...
    (comp.os.linux.development.system)
  • Re: how do threads work?
    ... >> kernel for a little help, in the form of regular timer signals. ... it is signals. ... The kernel doesn't care if you switch stack causing such ... one CPU bound thrad and one I/O bound thread. ...
    (comp.os.linux.development.system)
  • eth0 works with Linksys hub but not Netgear switch?!?
    ... network works fine when it's plugged into a Linksys ... EW5HUB hub, but not when it's plugged into a Netgear FS608v2 switch ... kernel: ip_tables: 2000-2002 Netfilter core team ...
    (comp.os.linux.networking)
  • Re: MSDOS emulation in MS V7 ?
    ... you can switch to 32-bit protected mode and from there switch to ... It would work exactly same way it does with a 32-bit kernel except that ... you're right DOSEMU must be using a software CPU emulator on 64-bit Linux.. ...
    (comp.os.msdos.programmer)