with an PAT-enabled kernel, when using uvesafb or vesafb, these drivers will
create uncached-minus PAT entries for the framebuffer memory because they use
ioremap() (not the *_cache or *_wc variants). When the framebuffer memory
intersects with the video RAM used by Xorg, the complete video RAM will be
mapped uncached-minus what results in a serve performance penalty.
Here are the correct MTRR entries created by uvesafb:
schlicht@netbook:~$ cat /proc/mtrr
reg00: base=0x000000000 ( 0MB), size= 2048MB, count=1: write-back
reg01: base=0x06ff00000 ( 1791MB), size= 1MB, count=1: uncachable
reg02: base=0x070000000 ( 1792MB), size= 256MB, count=1: uncachable
reg03: base=0x0d0000000 ( 3328MB), size= 16MB, count=1: write-combining
And here are the problematic PAT entries:
schlicht@netbook:~$ sudo cat /sys/kernel/debug/x86/pat_memtype_list
PAT memtype list:
write-back @ 0x0-0x1000
uncached-minus @ 0x6fedd000-0x6fee3000
uncached-minus @ 0x6fee2000-0x6fee3000
uncached-minus @ 0x6fee2000-0x6fee3000
uncached-minus @ 0x6fee2000-0x6fee3000
uncached-minus @ 0x6fee2000-0x6fee3000
uncached-minus @ 0x6fee2000-0x6fee3000
uncached-minus @ 0x6fee2000-0x6fee3000
uncached-minus @ 0x6fee2000-0x6fee3000
uncached-minus @ 0x6fee3000-0x6fee4000
uncached-minus @ 0x6fee3000-0x6fee4000
uncached-minus @ 0x6fee3000-0x6fee4000
uncached-minus @ 0xd0000000-0xe0000000 <-- created by xserver-xorg
uncached-minus @ 0xd0000000-0xd1194000 <-- created by uvesafb
uncached-minus @ 0xf4000000-0xf4009000
uncached-minus @ 0xf4200000-0xf4400000
uncached-minus @ 0xf5000000-0xf5010000
uncached-minus @ 0xf5100000-0xf5104000
uncached-minus @ 0xf5400000-0xf5404000
uncached-minus @ 0xf5404000-0xf5405000
uncached-minus @ 0xf5404000-0xf5405000
uncached-minus @ 0xfed00000-0xfed01000
Therefore I created the attached patch for uvesafb which uses ioremap_wc() to
create the correct PAT entries, as shown below:
schlicht@netbook:~$ sudo cat /sys/kernel/debug/x86/pat_memtype_list
PAT memtype list:
write-back @ 0x0-0x1000
uncached-minus @ 0x6fedd000-0x6fee3000
uncached-minus @ 0x6fee2000-0x6fee3000
uncached-minus @ 0x6fee2000-0x6fee3000
uncached-minus @ 0x6fee2000-0x6fee3000
uncached-minus @ 0x6fee2000-0x6fee3000
uncached-minus @ 0x6fee2000-0x6fee3000
uncached-minus @ 0x6fee2000-0x6fee3000
uncached-minus @ 0x6fee2000-0x6fee3000
uncached-minus @ 0x6fee3000-0x6fee4000
uncached-minus @ 0x6fee3000-0x6fee4000
uncached-minus @ 0x6fee3000-0x6fee4000
write-combining @ 0xd0000000-0xe0000000
write-combining @ 0xd0000000-0xd1194000
uncached-minus @ 0xf4000000-0xf4009000
uncached-minus @ 0xf4200000-0xf4400000
uncached-minus @ 0xf5000000-0xf5010000
uncached-minus @ 0xf5100000-0xf5104000
uncached-minus @ 0xf5400000-0xf5404000
uncached-minus @ 0xf5404000-0xf5405000
uncached-minus @ 0xf5404000-0xf5405000
uncached-minus @ 0xfed00000-0xfed01000
This results in a performance gain, objectively measurable with e.g.
x11perf -comppixwin10 -comppixwin100 -comppixwin500:
1: x11perf_xaa.log
2: x11perf_xaa_patched.log
1 2 Operation
-------- ---------------- -----------------
124000.0 202000.0 ( 1.63) Composite 10x10 from pixmap to window
3340.0 24400.0 ( 7.31) Composite 100x100 from pixmap to window
131.0 1150.0 ( 8.78) Composite 500x500 from pixmap to window
You can see the serve performance gain when composing larger pixmaps to window.
The patches replace the ioremap() function with the variant matching the mtrr-
parameter. To create "write-back" PAT entries, the ioremap_cache() function
must be called after creating the MTRR entries, and the ioremap_cache() region
must completely fit into the MTRR region, this is why the MTRR region size is
now rounded up to the next power-of-two.
Signed-off-by: Thomas Schlichter <thomas.schlichter@web.de>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
Netlink message processing in the kernel is synchronous these days,
capabilities can be checked directly in security_netlink_recv() from
the current process.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Reviewed-by: James Morris <jmorris@namei.org>
[chrisw: update to include pohmelfs and uvesafb]
Signed-off-by: Chris Wright <chrisw@sous-sol.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
This is more kernel-ish, saves some space, and also allows us to
expand the ops without breaking all the callers who are happy for the
new members to be NULL.
The few places which defined their own param types are changed to the
new scheme (more which crept in recently fixed in following patches).
Since we're touching them anyway, we change get() and set() to take a
const struct kernel_param (which they really are). This causes some
harmless warnings until we fix them (in following patches).
To reduce churn, module_param_call creates the ops struct so the callers
don't have to change (and casts the functions to reduce warnings).
The modern version which takes an ops struct is called module_param_cb.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Reviewed-by: Takashi Iwai <tiwai@suse.de>
Tested-by: Phil Carmody <ext-phil.2.carmody@nokia.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Ville Syrjala <syrjala@sci.fi>
Cc: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Cc: Alessandro Rubini <rubini@ipvvis.unipv.it>
Cc: Michal Januszewski <spock@gentoo.org>
Cc: Trond Myklebust <Trond.Myklebust@netapp.com>
Cc: "J. Bruce Fields" <bfields@fieldses.org>
Cc: Neil Brown <neilb@suse.de>
Cc: linux-kernel@vger.kernel.org
Cc: linux-input@vger.kernel.org
Cc: linux-fbdev-devel@lists.sourceforge.net
Cc: linux-nfs@vger.kernel.org
Cc: netdev@vger.kernel.org
percpu.h is included by sched.h and module.h and thus ends up being
included when building most .c files. percpu.h includes slab.h which
in turn includes gfp.h making everything defined by the two files
universally available and complicating inclusion dependencies.
percpu.h -> slab.h dependency is about to be removed. Prepare for
this change by updating users of gfp and slab facilities include those
headers directly instead of assuming availability. As this conversion
needs to touch large number of source files, the following script is
used as the basis of conversion.
http://userweb.kernel.org/~tj/misc/slabh-sweep.py
The script does the followings.
* Scan files for gfp and slab usages and update includes such that
only the necessary includes are there. ie. if only gfp is used,
gfp.h, if slab is used, slab.h.
* When the script inserts a new include, it looks at the include
blocks and try to put the new include such that its order conforms
to its surrounding. It's put in the include block which contains
core kernel includes, in the same order that the rest are ordered -
alphabetical, Christmas tree, rev-Xmas-tree or at the end if there
doesn't seem to be any matching order.
* If the script can't find a place to put a new include (mostly
because the file doesn't have fitting include block), it prints out
an error message indicating which .h file needs to be added to the
file.
The conversion was done in the following steps.
1. The initial automatic conversion of all .c files updated slightly
over 4000 files, deleting around 700 includes and adding ~480 gfp.h
and ~3000 slab.h inclusions. The script emitted errors for ~400
files.
2. Each error was manually checked. Some didn't need the inclusion,
some needed manual addition while adding it to implementation .h or
embedding .c file was more appropriate for others. This step added
inclusions to around 150 files.
3. The script was run again and the output was compared to the edits
from #2 to make sure no file was left behind.
4. Several build tests were done and a couple of problems were fixed.
e.g. lib/decompress_*.c used malloc/free() wrappers around slab
APIs requiring slab.h to be added manually.
5. The script was run on all .h files but without automatically
editing them as sprinkling gfp.h and slab.h inclusions around .h
files could easily lead to inclusion dependency hell. Most gfp.h
inclusion directives were ignored as stuff from gfp.h was usually
wildly available and often used in preprocessor macros. Each
slab.h inclusion directive was examined and added manually as
necessary.
6. percpu.h was updated not to include slab.h.
7. Build test were done on the following configurations and failures
were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my
distributed build env didn't work with gcov compiles) and a few
more options had to be turned off depending on archs to make things
build (like ipr on powerpc/64 which failed due to missing writeq).
* x86 and x86_64 UP and SMP allmodconfig and a custom test config.
* powerpc and powerpc64 SMP allmodconfig
* sparc and sparc64 SMP allmodconfig
* ia64 SMP allmodconfig
* s390 SMP allmodconfig
* alpha SMP allmodconfig
* um on x86_64 SMP allmodconfig
8. percpu.h modifications were reverted so that it could be applied as
a separate patch and serve as bisection point.
Given the fact that I had only a couple of failures from tests on step
6, I'm fairly confident about the coverage of this conversion patch.
If there is a breakage, it's likely to be something in one of the arch
headers which should be easily discoverable easily on most builds of
the specific arch.
Signed-off-by: Tejun Heo <tj@kernel.org>
Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
Remove fb_save_state() and fb_restore_state operations from frame buffer layer.
They are used only in two drivers:
1. savagefb - and cause bug #11248
2. uvesafb
Usage of these operations is misunderstood in both drivers so kill these
operations, fix the bug #11248 and avoid confusion in the future.
Tested on Savage 3D/MV card and the patch fixes the bug #11248.
The frame buffer layer uses these funtions during switch between graphics
and text mode of the console, but these drivers saves state before
switching of the frame buffer (in the fb_open) and after releasing it (in
the fb_release). This defeats the purpose of these operations.
Addresses http://bugzilla.kernel.org/show_bug.cgi?id=11248
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Reported-by: Jochen Hein <jochen@jochen.org>
Tested-by: Jochen Hein <jochen@jochen.org>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Michal Januszewski <spock@gentoo.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Acked-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Acked-by: Evgeniy Polyakov <zbr@ioremap.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
The connector documentation states that the argument to the callback
function is always a pointer to a struct cn_msg, but rather than encode it
in the API itself, it uses a void pointer everywhere. This doesn't make
much sense to encode the pointer in documentation as it prevents proper C
type checking from occurring and can easily allow people to use the wrong
pointer type. So convert the argument type to an explicit struct cn_msg
pointer.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
1) Now module_param(..., invbool, ...) requires a bool, and similarly
module_param(..., bool, ...) allows it, change pmi_setpal to a bool.
2) #define param_get_scroll to NULL, since it can never be called (perm
argument to module_param_named is 0).
3) Return -EINVAL from param_set_scroll if the value is bad, so it's
reported.
Note that I don't think the old fb_get_options() is required for new
drivers: the parameters automatically work as uvesafb.XXX=... anyway.
Acked-by: Michał Januszewski <spock@gentoo.org>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
uvesafb incorrectly sets the length of the color fields to 6 bits for
PSEUDOCOLOR modes, even though 8 bits are always used per pixel. Fix this
by setting the length to 8.
The switch of the DAC width from the default 6 bits to 8 bits is retained
and tracked internally in the driver, but never exposed to userspace.
Signed-off-by: Michal Januszewski <spock@gentoo.org>
Acked-by: Krzysztof Helt <krzysztof.h1@poczta.fm>
Cc: <syrjala@sci.fi>
Cc: Geert Uytterhoeven <geert.uytterhoeven@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
If the vbemode option is used, uvesafb calls fb_get_mode() without first
setting the resolution in info->var. This results in a division by zero
in fb_get_mode(), as evidenced e.g. in [1]. Fix this by ensuring the
info->var structure is populated before fb_get_mode() is called.
[1] http://bugzilla.kernel.org/show_bug.cgi?id=11661#c37
Signed-off-by: Michal Januszewski <spock@gentoo.org>
Cc: Krzysztof Helt <krzysztof.h1@poczta.fm>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
Acked-by: Michal Januszewski <michalj@gmail.com>
Cc: Krzysztof Helt <krzysztof.h1@poczta.fm>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
GFP_KERNEL is legal here - we don't need to use gfp_any().
Cc: Evgeniy Polyakov <johnpol@2ka.mipt.ru>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Andi Kleen <andi@firstfloor.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Currently, netlink_broadcast() reports errors to the caller if no
messages at all were delivered:
1) If, at least, one message has been delivered correctly, returns 0.
2) Otherwise, if no messages at all were delivered due to skb_clone()
failure, return -ENOBUFS.
3) Otherwise, if there are no listeners, return -ESRCH.
With this patch, the caller knows if the delivery of any of the
messages to the listeners have failed:
1) If it fails to deliver any message (for whatever reason), return
-ENOBUFS.
2) Otherwise, if all messages were delivered OK, returns 0.
3) Otherwise, if no listeners, return -ESRCH.
In the current ctnetlink code and in Netfilter in general, we can add
reliable logging and connection tracking event delivery by dropping the
packets whose events were not successfully delivered over Netlink. Of
course, this option would be settable via /proc as this approach reduces
performance (in terms of filtered connections per seconds by a stateful
firewall) but providing reliable logging and event delivery (for
conntrackd) in return.
This patch also changes some clients of netlink_broadcast() that
may report ENOBUFS errors via printk. This error handling is not
of any help. Instead, the userspace daemons that are listening to
those netlink messages should resync themselves with the kernel-side
if they hit ENOBUFS.
BTW, netlink_broadcast() clients include those that call
cn_netlink_send(), nlmsg_multicast() and genlmsg_multicast() since they
internally call netlink_broadcast() and return its error value.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Some BIOSes return error codes when queried for information about
modes from their own modelist. uvesafb treats this as an error
case and bails out.
Change this behavior so that broken modes do not prevent the driver
from working. Only the failure to retrieve information about any
usable video mode is considered to be an error case.
Signed-off-by: Michal Januszewski <spock@gentoo.org>
Cc: Krzysztof Helt <krzysztof.h1@poczta.fm>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Make more drivers use the "mode_option" parameter. This one is quite new
so drop the old "mode" parameter before someone starts using it seriously.
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Fix bug identified by Daniel Marjamki: `m' is leaked on the error path.
Addresses http://bugzilla.kernel.org/show_bug.cgi?id=10452
Cc: Daniel Marjamki <danielm77@spray.se>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Cc: "Randy.Dunlap" <rdunlap@xenotime.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Don't treat valid modes returned by fb_find_mode() (best-fit modes, default
modes or the first valid mode) as errors.
Currently, when fb_find_mode() finds a valid mode belonging to one of the
above-mentioned classes, uvesafb will ignore it and will try to set a 640x480
video mode. The expected behaviour (introduced by this patch) would be to use
the valid mode returned by fb_find_mode() instead.
Signed-off-by: Michal Januszewski <spock@gentoo.org>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Fix following warnings:
WARNING: drivers/video/built-in.o(.text+0x7c64a): Section mismatch in reference from the function param_set_scroll() to the variable .devinit.data:ypan
WARNING: drivers/video/built-in.o(.text+0x7c65d): Section mismatch in reference from the function param_set_scroll() to the variable .devinit.data:ypan
WARNING: drivers/video/built-in.o(.text+0x7c679): Section mismatch in reference from the function param_set_scroll() to the variable .devinit.data:ypan
WARNING: drivers/video/built-in.o(.text+0x7c699): Section mismatch in reference from the function param_set_scroll() to the variable .devinit.data:ypan
WARNING: drivers/video/built-in.o(.text+0x7c69f): Section mismatch in reference from the function param_set_scroll() to the variable .devinit.data:ypan
WARNING: drivers/built-in.o(.text+0xa3676): Section mismatch in reference from the function param_set_scroll() to the variable .devinit.data:ypan
WARNING: drivers/built-in.o(.text+0xa3689): Section mismatch in reference from the function param_set_scroll() to the variable .devinit.data:ypan
WARNING: drivers/built-in.o(.text+0xa36a5): Section mismatch in reference from the function param_set_scroll() to the variable .devinit.data:ypan
WARNING: drivers/built-in.o(.text+0xa36c5): Section mismatch in reference from the function param_set_scroll() to the variable .devinit.data:ypan
WARNING: drivers/built-in.o(.text+0xa36cb): Section mismatch in reference from the function param_set_scroll() to the variable .devinit.data:ypan
WARNING: vmlinux.o(.text+0x4a079a): Section mismatch in reference from the function param_set_scroll() to the variable .devinit.data:ypan
WARNING: vmlinux.o(.text+0x4a07ad): Section mismatch in reference from the function param_set_scroll() to the variable .devinit.data:ypan
WARNING: vmlinux.o(.text+0x4a07c9): Section mismatch in reference from the function param_set_scroll() to the variable .devinit.data:ypan
WARNING: vmlinux.o(.text+0x4a07e9): Section mismatch in reference from the function param_set_scroll() to the variable .devinit.data:ypan
WARNING: vmlinux.o(.text+0x4a07ef): Section mismatch in reference from the function param_set_scroll() to the variable .devinit.data:ypan
Remove __devinitdata annotation from the variable ypan.
Signed-off-by: Sergio Luis <sergio@larces.uece.br>
Cc: Michal Januszewski <spock@gentoo.org>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Some cleanups in uvesafb:
- The custom module_param() get/set functions don't need to be inlined
since it is referred to via a pointer in a struct.
- don't end a #define with a ';'
- remove one of the single quote marks in "''ypan'"
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Cc: Michal Januszewski <spock@gentoo.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Mark uvesafb_init_mtrr() as __devinit since its caller is __devinit
and since it accesses __devinitdata.
WARNING: vmlinux.o(.text+0x4df80e): Section mismatch: reference to .init.data: (between 'uvesafb_init_mtrr' and 'uvesafb_show_vbe_ver')
Variable 'blank' cannot be __devinitdata since it is referenced in an
fb_ops method that could be called at any time.
WARNING: vmlinux.o(.text+0x4dfc1e): Section mismatch: reference to .init.data:blank (between 'param_set_scroll' and 'vesa_setpalette')
WARNING: vmlinux.o(.text+0x4dfc24): Section mismatch: reference to .init.data:blank (between 'param_set_scroll' and 'vesa_setpalette')
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Variables that are only used in #ifdef CONFIG_X86 should also only be
declared there.
Signed-off-by: Frank Lichtenheld <frank@lichtenheld.de>
Cc: Michal Januszewski <spock@gentoo.org>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
uvesafb is an enhanced version of vesafb. It uses a userspace helper (v86d)
to execute calls to the x86 Video BIOS functions. The driver is not limited
to any specific arch and whether it works on a given arch or not depends on
that arch being supported by the userspace daemon. It has been tested on
x86_32 and x86_64.
A single BIOS call is represented by an instance of the uvesafb_ktask
structure. This structure contains a buffer, a completion struct and a
uvesafb_task substructure, containing the values of the x86 registers, a flags
field and a field indicating the length of the buffer. Whenever a BIOS call
is made in the driver, uvesafb_exec() builds a message using the uvesafb_task
substructure and the contents of the buffer. This message is then assigned a
random ack number and sent to the userspace daemon using the connector
interface.
The message's sequence number is used as an index for the uvfb_tasks array,
which provides a mapping from the messages coming from userspace to the
in-kernel uvesafb_ktask structs.
The userspace daemon performs the requested operation and sends a reply in the
form of a uvesafb_task struct and, optionally, a buffer. The seq and ack
numbers in the reply should be exactly the same as those in the request.
Each message from userspace is processed by uvesafb_cn_callback() and after
passing a few sanity checks leads to the completion of a BIOS call request.
Signed-off-by: Michal Januszewski <spock@gentoo.org>
Signed-off-by: Antonino Daplas <adaplas@gmail.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Paulo Marques <pmarques@grupopie.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
