Signed-off-by: NDave Airlie <airlied@redhat.com>

Fix some missing places to check with device id info, which
should probe the device gart correctly.
Signed-off-by: NWang Zhenyu <zhenyu.z.wang@intel.com>
Signed-off-by: NDave Airlie <airlied@redhat.com>

AGP should not need to lock pages. They are not protecting any race
because there is no lock_page calls, only SetPageLocked.

This is causing hangs with d00806b1.
Signed-off-by: NNick Piggin <npiggin@suse.de>
Signed-off-by: NDave Airlie <airlied@redhat.com>

Hi,

Coverity spotted a "use after free" bug in
drivers/char/agp/ati-agp.c::ati_create_gatt_pages().

The same one that was in
  drivers/char/agp/amd-k7-agp.c::amd_create_gatt_pages()

The problem is this:
       If "entry = kzalloc(sizeof(struct ati_page_map), GFP_KERNEL);"
fails, then there's a loop in the function to free all entries
allocated so far and break out of the allocation loop. That in itself
is pretty sane, but then the (now freed) 'tables' is assigned to
ati_generic_private.gatt_pages and 'retval' is set to -ENOMEM which
causes ati_free_gatt_pages(); to be called at the end of the function.
The problem with this is that ati_free_gatt_pages() will then loop
'ati_generic_private.num_tables' times and try to free each entry in
tables[] - this is bad since tables has already been freed and
furthermore it will call kfree(tables) at the end - a double free.

This patch removes the freeing loop in ati_create_gatt_pages() and
instead relies entirely on the call to ati_free_gatt_pages() to free
everything we allocated in case of an error. It also sets
ati_generic_private.num_tables to the actual number of entries
allocated instead of just using the value passed in from the caller -
this ensures that ati_free_gatt_pages() will only attempt to free
stuff that was actually allocated.

Note: I'm in no way intimate with this code and I have no way to
actually test this patch (besides compile test it), so while I've
tried to be careful in reading the code and make sure the patch
does the right thing an ACK from someone who actually knows the
code in-depth would be very much appreciated.
Signed-off-by: NJesper Juhl <jesper.juhl@gmail.com>
Signed-off-by: NDave Airlie <airlied@linux.ie>

Instead of all drivers reading pci config space to get the revision
ID, they can now use the pci_device->revision member.

This exposes some issues where drivers where reading a word or a dword
for the revision number, and adding useless error-handling around the
read. Some drivers even just read it for no purpose of all.

In devices where the revision ID is being copied over and used in what
appears to be the equivalent of hotpath, I have left the copy code
and the cached copy as not to influence the driver's performance.

Compile tested with make all{yes,mod}config on x86_64 and i386.
Signed-off-by: NAuke Kok <auke-jan.h.kok@intel.com>
Acked-by: NDave Jones <davej@redhat.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Most of the AGP changes recently have been done in lock-step with
DRM updates, so it's probably easier to have airlied pushing
AGP changes at the same time he does DRM updates.

[Also remove my name from the boot messages.
 Cautionary tale to others: Never do this, when computers
 don't boot, people assume you're responsible even if 15
 other subsystems initialised after yours. :-) ]
Signed-off-by: NDave Jones <davej@redhat.com>
Signed-off-by: NDave Airlie <airlied@linux.ie>

After i915 chip, GMCH has no AGP port. Origin bridge driver in device
table will try to access illegal regs like APBASE, APSIZE, etc. This
may cause problem.

So mark them as NULL in the table, we won't load if no IGD got detect
and bridge has no AGP port.
Signed-off-by: NWang Zhenyu <zhenyu.z.wang@intel.com>
Signed-off-by: NDave Jones <davej@redhat.com>

This patch trys to fix device probe in two cases. First we should
correctly detect device if integrated graphics device is not enabled
or exists, like an add-in card is plugged. Second on some type of intel
GMCH, it might have multiple graphic chip models, like 945GME case, so
we should be sure the detect works through the whole table.
Signed-off-by: NWang Zhenyu <zhenyu.z.wang@intel.com>
Signed-off-by: NDave Jones <davej@redhat.com>

This patch adds pci ids for G33, Q33 and Q35 chips, and update with new
GTT size and stolen mem size detect method on these chips.
Signed-off-by: NWang Zhenyu <zhenyu.z.wang@intel.com>
Signed-off-by: NDave Jones <davej@redhat.com>

Add pci id info for 945GME.
Signed-off-by: NWang Zhenyu <zhenyu.z.wang@intel.com>
Signed-off-by: NDave Jones <davej@redhat.com>

Add pci id info for 965GME/GLE support.
Signed-off-by: NWang Zhenyu <zhenyu.z.wang@intel.com>
Signed-off-by: NDave Jones <davej@redhat.com>

Fixed issues noted by Christoph Hellwig, and I changed device table
scan a bit to allow the case that some models of graphics chips may
have same host bridge type. This type of chip will be added in the future.

This patch cleans up device probe function. Eric Anholt was the original author.
Signed-off-by: NEric Anholt <eric@anholt.net>
Signed-off-by: NWang Zhenyu <zhenyu.z.wang@intel.com>
Signed-off-by: NDave Jones <davej@redhat.com>

Remove volatile type declare for IO mem variables.

A single private gart data is used by all drivers, this
makes it clean. Eric Anholt wrote the original patch.
Signed-off-by: NWang Zhenyu <zhenyu.z.wang@intel.com>
Signed-off-by: NDave Jones <davej@redhat.com>

First thing mm.h does is including sched.h solely for can_do_mlock() inline
function which has "current" dereference inside. By dealing with can_do_mlock()
mm.h can be detached from sched.h which is good. See below, why.

This patch
a) removes unconditional inclusion of sched.h from mm.h
b) makes can_do_mlock() normal function in mm/mlock.c
c) exports can_do_mlock() to not break compilation
d) adds sched.h inclusions back to files that were getting it indirectly.
e) adds less bloated headers to some files (asm/signal.h, jiffies.h) that were
   getting them indirectly

Net result is:
a) mm.h users would get less code to open, read, preprocess, parse, ... if
   they don't need sched.h
b) sched.h stops being dependency for significant number of files:
   on x86_64 allmodconfig touching sched.h results in recompile of 4083 files,
   after patch it's only 3744 (-8.3%).

Cross-compile tested on

	all arm defconfigs, all mips defconfigs, all powerpc defconfigs,
	alpha alpha-up
	arm
	i386 i386-up i386-defconfig i386-allnoconfig
	ia64 ia64-up
	m68k
	mips
	parisc parisc-up
	powerpc powerpc-up
	s390 s390-up
	sparc sparc-up
	sparc64 sparc64-up
	um-x86_64
	x86_64 x86_64-up x86_64-defconfig x86_64-allnoconfig

as well as my two usual configs.
Signed-off-by: NAlexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

there is a wrong id in drivers/char/agp/via-agp.c
#define PCI_DEVICE_ID_VIA_CX700         0x8324
It must be 0x0324

Notice that PCI_DEVICE_ID_VIA_CX700 is also used in
drivers/i2c/busses/i2c-viapro.c and
drivers/ide/pci/via82cxxx.c

So, I think that constant must be renamed to avoid conflicting.
I attached a proposed patch.
Signed-off-by: NDave Jones <davej@redhat.com>

I'm using a custom BIOS to configure the northbridge GART at address
0x80000000, size 2G.  Linux complains:

"Aperture from northbridge cpu 0 beyond 4GB. Ignoring."

I think there's an off-by-two error in arch/x86_64/kernel/aperture.c:

AK: use correct types for i386
Signed-off-by: NAndi Kleen <ak@suse.de>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Convert code that allocs a struct pci_dev to use alloc_pci_dev().
Signed-off-by: NMichael Ellerman <michael@ellerman.id.au>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Under CONFIG_DISCONTIGMEM, assuming that a !pfn_valid() implies all
subsequent pfn-s are also invalid is wrong. Thus replace this by
explicitly checking against the E820 map.

AK: make e820 on x86-64 not initdata
Signed-off-by: NJan Beulich <jbeulich@novell.com>
Signed-off-by: NAndi Kleen <ak@suse.de>
Acked-by: NMark Langsdorf <mark.langsdorf@amd.com>

These are all the remaining instances of get_property.  Simple rename of
get_property to of_get_property.
Signed-off-by: NStephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

Signed-off-by: NAlan Cox <alan@redhat.com>
Signed-off-by: NDave Jones <davej@redhat.com>

Signed-off-by: NAlan Cox <alan@redhat.com>
Signed-off-by: NDave Jones <davej@redhat.com>

drivers/char/agp/sgi-agp.c:51:10: warning: Using plain integer as NULL pointer
Signed-off-by: NDave Jones <davej@redhat.com>

Fix a call to __free_page where __free_pages(, 2) was meant, and do proper
error path handling. Also remove a redundant conditional.
Signed-off-by: NJan Beulich <jbeulich@novell.com>
Signed-off-by: NDave Jones <davej@redhat.com>

Remove an arch-dependent hunk in favor of #define-ing the respective bits in
asm-<arch>/agp.h (allowing easier overriding in para-virtualized environments).
Signed-off-by: NJan Beulich <jbeulich@novell.com>
Signed-off-by: NDave Jones <davej@redhat.com>

add missing calls to global_flush_tlb().
Signed-off-by: NJan Beulich <jbeulich@novell.com>
Signed-off-by: NDave Jones <davej@redhat.com>

For some vendor/id pairs the kernel will autoload both the sis-agp and the
amd64_agp modules as the sis-agp module will load for all sis devices.
This collision causes the bug reported in:
http://bugzilla.novell.com/show_bug.cgi?id=248665

As currently sis_probe does its own matching, requesting the whole range
gains nothing.  The clean fix seems to me to leave the matching to the core
and advertise only the devices actually supported.  This patch does so.
Signed-off-by: NOliver Neukum <oneukum@suse.de>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NDave Jones <davej@redhat.com>

This reverts commit d05c7a80,
which included changes which should go via other subsystem
maintainers.

Signed-off-by: NStephen Rothwell <sfr@canb.auug.org.au>
Acked-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

On G965, I810_PGETBL_CTL is a mmio offset, but we wrongly take it
as pci config space offset in detecting GTT size. This one line patch
fixs this.
Signed-off-by: NWang Zhenyu <zhenyu.z.wang@intel.com>
Signed-off-by: NDave Jones <davej@redhat.com>

Update PCI id info for Intel 965GM chipset.
Signed-off-by: NWang Zhenyu <zhenyu.z.wang@intel.com>
Signed-off-by: NDave Jones <davej@redhat.com>

This fixes the following compile failures of agpgart drivers.
These errors were inserted by the recent AGPGART constification patch.

drivers/char/agp/uninorth-agp.c:492: error: expected '{' before 'const'
drivers/char/agp/uninorth-agp.c:517: error: expected '{' before 'const'
drivers/char/agp/uninorth-agp.c: In function 'agp_uninorth_probe':
drivers/char/agp/uninorth-agp.c:634: error: 'u3_agp_driver' undeclared (first use in this function)
drivers/char/agp/uninorth-agp.c:634: error: (Each undeclared identifier is reported only once
drivers/char/agp/uninorth-agp.c:634: error: for each function it appears in.)
drivers/char/agp/uninorth-agp.c:636: error: 'uninorth_agp_driver' undeclared (first use in this function)
Signed-off-by: NRyusuke Konishi <ryusuke@osrg.net>
Signed-off-by: NDave Jones <davej@redhat.com>

Can't really blame davej for mucking this up... static-ify
it while we're at it, which would have prevented this...
Signed-off-by: NKyle McMartin <kyle@parisc-linux.org>

Make agp_bridge_driver->aperture_sizes and ->masks const.
Also agp_bridge_data->driver
Signed-off-by: NDave Jones <davej@redhat.com>

My previous compat AGP patch broke modular AGPGART.

Test built on;

i386 CONFIG_AGP=y,m
x86_64 CONFIG_AGP=y
ia64 CONFIG_AGP=m
Signed-off-by: NZwane Mwaikambo <zwane@infradead.org>
Cc: Kyle McMartin <kyle@mcmartin.ca>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NDave Jones <davej@redhat.com>

Fix whitespace, braces, use kzalloc().

Cc: Dave Airlie <airlied@linux.ie>
Cc: Thomas Hellstrom <thomas@tungstengraphics.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NDave Jones <davej@redhat.com>

use ARRAY_SIZE macro already defined in kernel.h
Signed-off-by: NAhmed S. Darwish <darwish.07@gmail.com>
Signed-off-by: NDave Jones <davej@redhat.com>

Signed-off-by: NDave Jones <davej@redhat.com>

We don't unmap them on the suspend path, so on resume
trying to remap will fail, and then result in an
oops the next time something tries to access them.
Signed-off-by: NDave Jones <davej@redhat.com>

This patch allows drm to populate an agpgart structure with pages of its own.
It's needed for the new drm memory manager which dynamically flips pages in and out of AGP.

The patch modifies the generic functions as well as the intel agp driver. The intel drm driver is
currently the only one supporting the new memory manager.

Other agp drivers may need some minor fixing up once they have a corresponding memory manager enabled drm driver.

AGP memory types >= AGP_USER_TYPES are not populated by the agpgart driver, but the drm is expected
to do that, as well as taking care of cache- and tlb flushing when needed.

It's not possible to request these types from user space using agpgart ioctls.

The Intel driver also gets a new memory type for pages that can be bound cached to the intel GTT.
Signed-off-by: NThomas Hellstrom <thomas@tungstengraphics.com>
Signed-off-by: NDave Jones <davej@redhat.com>

The following video card requires the agpgart driver ioctl
interface in order to detect video memory.

00:02.0 VGA compatible controller: Intel Corporation Mobile
945GM/GMS/940GML Express Integrated Graphics Controller (rev 03)

Tested on a Thinkpad Z61t, Xorg.0.log from a 32bit debian Xorg is at;

http://montezuma.homeunix.net/Xorg.0.logSigned-off-by: NZwane Mwaikambo <zwane@infradead.org>
Signed-off-by: NDave Jones <davej@redhat.com>