If DMAR is configured in but absent, we really do want to make sure that
the dma mask is set appropriately. Otherwise we get mapping failures on
highmem. Spotted by Zhenyu Wang.
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
When graphics dma remapping engine is active, we must fill
gart table with dma address from dmar engine, as now graphics
device access to graphics memory must go through dma remapping
table to get real physical address.
Add this support to all drivers which use intel_i915_insert_entries()
Signed-off-by: Zhenyu Wang <zhenyu.z.wang@intel.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
In commit 07613ba2 ("agp: switch AGP to use page array instead of
unsigned long array") we switched the mask_memory() method to take a
'struct page *' instead of an address. This is painful, because in some
cases it has to be an IOMMU-mapped virtual bus address (in fact,
shouldn't it _always_ be a dma_addr_t returned from pci_map_xxx(), and
we just happen to get lucky most of the time?)
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Otherwise, the high bits to be stuffed in the unused lower bits of the
page address are lost.
Signed-off-by: Pierre Willenbrock <pierre@pirsoft.de>
Signed-off-by: Dave Airlie <airlied@redhat.com>
This switches AGP to use an array of pages for tracking the
pages allocated to the GART. This should enable GEM on PAE to work
a lot better as we can pass highmem pages to the PAT code and it will
do the right thing with them.
Signed-off-by: Dave Airlie <airlied@redhat.com>
Support the Intel 854 Chipset in fbdev.
We test and use the patch on a Thomson IP1101 IPTV-Box. On the VGA-Port
we get a normal signal.
Here is the link to the Mambux-Project: http://www.mambux.de
Cc: Keith Packard <keithp@keithp.com>
Cc: Dave Airlie <airlied@linux.ie>
Cc: Krzysztof Helt <krzysztof.h1@poczta.fm>
Signed-off-by: Stefan Husemann <shusemann@googlemail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
When GTT size is equal to amount of video memory, the amount of GTT
entries is computed lower than zero, which is invalid and leads to
off-by-one error in intel_i915_configure()
Originally posted here:
http://bugzilla.kernel.org/show_bug.cgi?id=12539http://bugzilla.redhat.com/show_bug.cgi?id=445592
Signed-off-by: Lubomir Rintel <lkundrak@v3.sk>
Cc: Lubomir Rintel <lkundrak@v3.sk>
Cc: Dave Airlie <airlied@linux.ie>
Reviewed-by: Eric Anholt <eric@anholt.net>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Update assorted email addresses and related info to point
to a single current, valid address.
additionally
- trivial CREDITS entry updates. (Not that this file means much any more)
- remove arjans dead redhat.com address from powernow driver
Signed-off-by: Dave Jones <davej@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* 'agp-next' of git://git.kernel.org/pub/scm/linux/kernel/git/airlied/agp-2.6:
agp/nvidia: Support agp user-memory on nvidia agp.
agp/amd-k7: Suspend support for AMD K7 GART driver
agp/intel: Reduce extraneous PCI posting reads during init
agp: Fix stolen memory counting on G4X.
Instead of doing a posting read after each GTT entry update, do a single one
at the end of the writes. This should reduce boot time a tiny amount by
avoiding a lot of extra uncached reads.
Signed-off-by: Keith Packard <keithp@keithp.com>
Signed-off-by: Eric Anholt <eric@anholt.net>
Signed-off-by: Dave Airlie <airlied@redhat.com>
On the GM45, the amount of stolen memory mapped to the GTT was underestimated,
even though we had 508KB more available since the GTT doesn't take from
stolen memory. On the non-GM45 G4X, we overestimated how much stolen was
mapped to the GTT by 4KB, resulting in GPU page faults when that page was
accessed.
This update requires a corresponding update to xf86-video-intel to work
correctly.
Signed-off-by: Eric Anholt <eric@anholt.net>
Signed-off-by: Dave Airlie <airlied@redhat.com>
On my Intel chipset (965GM), the GTT is entirely erased across
suspend/resume. This patch simply re-plays the current mapping at resume
time to restore the table.=20
I noticed this once I started relying on persistent GTT mappings across VT
switch in our GEM work -- the old X server and DRM code carefully unbind
all memory from the GTT on VT switch, but GEM does not bother.
I placed the list management and rewrite code in the generic layer on the
assumption that it will be needed on other hardware, but I did not add the
rewrite call to anything other than the Intel resume function.
Keep a list of current GATT mappings. At resume time, rewrite them into
the GATT. This is needed on Intel (at least) as the entire GATT is
cleared across suspend/resume.
[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Keith Packard <keithp@keithp.com>
Cc: Dave Jones <davej@codemonkey.org.uk>
Cc: Andi Kleen <andi@firstfloor.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Convert printks to use dev_printk().
Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Zhenyu Wang <zhenyu.z.wang@intel.com>
Cc: Dave Airlie <airlied@linux.ie>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Dave Airlie <airlied@redhat.com>
This adds missing stolen memory size detect for IGD_GM, be sure to
detect right size as current X intel driver (2.3.2) which has already
worked out.
Signed-off-by: Zhenyu Wang <zhenyu.z.wang@intel.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
besides it apparently being useful only in 2.6.24 (the changes in 2.6.25
really mean that it could be converted back to a single-stage mechanism),
I'm seeing an issue in Xen Dom0 kernels, which is caused by the calling
of gart_to_virt() in the second stage invocations of the destroy function.
I think that besides this being a real issue with Xen (where
unmap_page_from_agp() is not just a page table attribute change), this
also is invalid from a theoretical perspective: One should not assume that
gart_to_virt() is still valid after unmapping a page. So minimally (keeping
the 2-stage mechanism) a patch like the one below would be needed.
Jan
Signed-off-by: Dave Airlie <airlied@redhat.com>
Change origin chipset name i965G_1 to market name G35.
Signed-off-by: Zhenyu Wang <zhenyu.z.wang@intel.com>
Signed-off-by: Dave Airlie <airlied@linux.ie>
The flush code resource handling was having problems where some BIOS
reserve the resource in a pnp block and some don't.
Also there was a bug in that configure was being called at resume
and resetting some of the structs.
Signed-off-by: Dave Airlie <airlied@linux.ie>
This one adds new pci ids for Intel intergrated graphics chipset, with gtt
table access change on it and new gtt table size definition.
Signed-off-by: Zhenyu Wang <zhenyu.z.wang@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Dave Airlie <airlied@linux.ie>
drivers/char/agp/intel-agp.c: In function 'intel_i965_g33_setup_chipset_flush':
drivers/char/agp/intel-agp.c:872: warning: right shift count >= width of type
I wish the agp code wasn't written in a 10,000-column xterm :(
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Dave Airlie <airlied@linux.ie>
This is a bit of a large hammer but it makes sure the chipset is flushed
by writing out 1k of data to an uncached page. We may be able to get better
information in the future on how to this better.
Signed-off-by: Dave Airlie <airlied@redhat.com>
This adds support for flushing the chipsets on the 915, 945, 965 and G33
families of Intel chips.
The BIOS doesn't seem to always allocate the BAR on the 965 chipsets
so I have to use pci resource code to create a resource
It adds an export for pcibios_align_resource.
The set_memory_* and set_pages_* family of API's currently requires the
callers to do a global tlb flush after the function call; forgetting this is
a very nasty deathtrap. This patch moves the global tlb flush into
each of the callers
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
This patch converts various users of change_page_attr() to the new,
more intent driven set_page_*/set_memory_* API set.
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The E7221 chipset is a 915 rebadged for the Intel server line.
Signed-off-by: Dave Airlie <airlied@linux.ie>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
With Andi's clflush fixup, we were getting hangs on server exit, flushing the
mappings after freeing each page helped.
This showed up a race condition where the pages after being freed could be
reused before the agp mappings had been flushed. Flushing after each single
page is a bad thing for future drm work, so make the page destroy a two pass
unmapping all the pages, flushing the mappings, and then destroying the pages.
Signed-off-by: Dave Airlie <airlied@linux.ie>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
This reverts commit f443675aff, which
breaks horribly if you aren't running an unreleased xf86-video-intel
driver out of git.
Signed-off-by: Kyle McMartin <kyle@mcmartin.ca>
Cc: Dave Airlie <airlied@linux.ie>
Cc: Zhenyu Wang <zhenyu.z.wang@intel.com>
Acked-by: Keith Packard <keithp@keithp.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The mask on i830 should be 0x70 always, later chips 0xF0 should be okay.
Signed-off-by: Dave Airlie <airlied@linux.ie>
Acked-by: Zhenyu Wang <zhenyu.z.wang@intel.com>
Cc: Michael Haas <laga@laga.ath.cx>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
G33 has 1MB GTT table range. Fix GTT mapping in case like 512MB aperture
size.
Signed-off-by: Zhenyu Wang <zhenyu.z.wang@intel.com>
Acked-by: Dave Airlie <airlied@linux.ie>
Cc: Dave Jones <davej@codemonkey.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
G33 GTT stolen memory is below graphics data stolen memory and be seperate,
so don't subtract it in stolen mem counting.
Signed-off-by: Zhenyu Wang <zhenyu.z.wang@intel.com>
Acked-by: Dave Airlie <airlied@linux.ie>
Cc: Dave Jones <davej@codemonkey.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
patchset against 2.6.23-rc3.
corrects missing ioremap return checks and balancing on iounmap calls, integrated changes per list
recommendations on the original set of patches..
Signed-off-by: Scott Thompson <postfail <at> hushmail.com>
Signed-off-by: Dave Airlie <airlied@linux.ie>
Fix some missing places to check with device id info, which
should probe the device gart correctly.
Signed-off-by: Wang Zhenyu <zhenyu.z.wang@intel.com>
Signed-off-by: Dave 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 d00806b183.
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Dave Airlie <airlied@redhat.com>
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: Wang Zhenyu <zhenyu.z.wang@intel.com>
Signed-off-by: Dave 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: Wang Zhenyu <zhenyu.z.wang@intel.com>
Signed-off-by: Dave 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: Wang Zhenyu <zhenyu.z.wang@intel.com>
Signed-off-by: Dave 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: Eric Anholt <eric@anholt.net>
Signed-off-by: Wang Zhenyu <zhenyu.z.wang@intel.com>
Signed-off-by: Dave 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: Wang Zhenyu <zhenyu.z.wang@intel.com>
Signed-off-by: Dave 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: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Dave Jones <davej@redhat.com>
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: Wang Zhenyu <zhenyu.z.wang@intel.com>
Signed-off-by: Dave Jones <davej@redhat.com>
Fix whitespace, braces, use kzalloc().
Cc: Dave Airlie <airlied@linux.ie>
Cc: Thomas Hellstrom <thomas@tungstengraphics.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Dave 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: Dave 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: Thomas Hellstrom <thomas@tungstengraphics.com>
Signed-off-by: Dave Jones <davej@redhat.com>
Currently in resuming path graphics device's pci space restore is
behind host bridge, so resume function wrongly accesses graphics
device's space. This makes resuming failure which crashed X.
here's a patch to restore device's pci space early, which makes
resuming ok with X.
Signed-off-by: Wang Zhenyu <zhenyu.z.wang@intel.com>
Signed-off-by: Dave Jones <davej@redhat.com>
Unfortunately there was a typo in one of the patches I sent,
(The one now committed to the agpgart tree).
It may cause a bus error on i810 type hardware.
Signed-off-by: Thomas Hellstrom <thomas@tungstengraphics.com>
Signed-off-by: Dave Jones <davej@redhat.com>
On the G965, the GTT size may be larger than is required to cover the
aperture. (In fact, on all hardware we've seen, the GTT is 512KB to the
aperture's 256MB). A previous commit forced the aperture size to 512MB on
G965 to match GTT, which would likely result in hangs at best if users
tried to rely on agpgart's aperture size information. Instead, we use the
resource length for the aperture size and the system's reported GTT size
when available for the GTT size.
Because the MSAC registers which had been read for aperture size detection
on i9xx chips just cause a change in the resource size, we can use generic
code for aperture detection on all i9xx.
Signed-off-by: Eric Anholt <eric@anholt.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Dave Jones <davej@redhat.com>
This patch is to speed up flipping of pages in and out of the AGP aperture as
needed by the new drm memory manager.
A number of global cache flushes are removed as well as some PCI posting flushes.
The following guidelines have been used:
1) Memory that is only mapped uncached and that has been subject to a global
cache flush after the mapping was changed to uncached does not need any more
cache flushes. Neither before binding to the aperture nor after unbinding.
2) Only do one PCI posting flush after a sequence of writes modifying page
entries in the GATT.
Signed-off-by: Thomas Hellstrom <thomas@tungstengraphics.com>
Signed-off-by: Dave Jones <davej@redhat.com>
Not all graphic page remappers support physical addresses over the 4GB
mark for remapping, so while some do (the AMD64 GART always did, and I
just fixed the i965 to do so properly), we're safest off just forcing
GFP_DMA32 allocations to make sure graphics pages get allocated in the
low 32-bit address space by default.
AGP sub-drivers that really care, and can do better, could just choose
to implement their own allocator (or we could add another "64-bit safe"
default allocator for their use), but quite frankly, you're not likely
to care in practice.
So for now, this trivial change means that we won't be allocating pages
that we can't map correctly by mistake on x86-64.
[ On traditional 32-bit x86, this could never happen, because GFP_KERNEL
would never allocate any highmem memory anyway ]
Acked-by: Andi Kleen <ak@suse.de>
Acked-by: Dave Jones <davej@redhat.com>
Cc: Eric Anholt <eric@anholt.net>
Cc: Keith Packard <keithp@keithp.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This introduces a i965-specific "mask_memory()" function that knows
about the extended physical addresses that the i965 supports. This
allows us to correctly map in physical memory in the >4GB range into the
GTT.
Also simplify/clean-up the i965 case for the aperture sizing by just
returning the fixed 512kB size from "fetch_size()". We don't really
care that not all of the aperture may be visible - the only thing that
cares about the aperture size is the Intel "stolen memory" calculation,
which depends on the fixed size.
Cc: Keith Packard <keithp@keithp.com>
Cc: Eric Anholt <eric@anholt.net>
Cc: Dave Jones <davej@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Here's a very small diff for 945GM support for agpgart.
Patch against 2.6.15.
From: Alan Hourihane <alanh@fairlite.demon.co.uk>
Signed-off-by: Dave Jones <davej@redhat.com>
AGP allocation/deallocation is suffering major performance issues due to
the nature of global_flush_tlb() being called on every change_page_attr()
call.
For small allocations this isn't really seen, but when you start allocating
50000 pages of AGP space, for say, texture memory, then things can take
seconds to complete.
In some cases the situation is doubled or even quadrupled in the time due
to SMP, or a deallocation, then a new reallocation. I've had a case of
upto 20 seconds wait time to deallocate and reallocate AGP space.
This patch fixes the problem by making it the caller's responsibility to
call global_flush_tlb(), and so removes it from every instance of mapping a
page into AGP space until the time that all change_page_attr() changes are
done.
Signed-off-by: Dave Jones <davej@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
From: Laurent Riffard <laurent.riffard@free.fr>
This updates .owner field of struct pci_driver.
This allows SYSFS to create the symlink from the driver to the module which
provides it.
$ tree /sys/bus/pci/drivers/agpgart-via/
/sys/bus/pci/drivers/agpgart-via/
|-- 0000:00:00.0 -> ../../../../devices/pci0000:00/0000:00:00.0
|-- bind
|-- module -> ../../../../module/via_agp
|-- new_id
`-- unbind
Signed-off-by: Laurent Riffard <laurent.riffard@free.fr>
Signed-off-by: Dave Jones <davej@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
When leaving S3 state, the AGP bridge may not have all PCI configuration
registers set in the same way as they were at boot. This should be fixed
by pci_restore_state - however, the APBASE register cannot be set to
conflict with the APSIZE register. If APSIZE is larger than it was before
suspend, pci_restore_state will not restore APBASE correctly. The attached
patch adds an extra item to the agp_bridge_data structure and uses it to
store the value of APBASE. On resume, this is then written after APSIZE
has been set. This patch only touches the path used for Intel chipsets
without integrated graphics, and may need to be extended to work with the
others.
Without this patch, I get the symptoms described in bug 4921 - APBASE ends
up overlapping various PCI devices, and as a result they fail to work after
resume.
Signed-off-by: Matthew Garrett <mjg59@srcf.ucam.org>
Acked-by: Dave Jones <davej@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
When Linux is running on the Xen virtual machine monitor, physical
addresses are virtualised and cannot be directly referenced by the AGP
GART. This patch fixes the GART driver for Xen by adding a layer of
abstraction between physical addresses and 'GART addresses'.
Architecture-specific functions are also defined for allocating and freeing
the GATT. Xen requires this to ensure that table really is contiguous from
the point of view of the GART.
These extra interface functions are defined as 'no-ops' for all existing
architectures that use the GART driver.
Signed-off-by: Keir Fraser <keir@xensource.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Dave Jones <davej@redhat.com>
Attached is a small patch for i945G support against 2.6.11.11.
From: Alan Hourihane <alanh@fairlite.demon.co.uk>
Signed-off-by: Dave Jones <davej@redhat.com>
Initial git repository build. I'm not bothering with the full history,
even though we have it. We can create a separate "historical" git
archive of that later if we want to, and in the meantime it's about
3.2GB when imported into git - space that would just make the early
git days unnecessarily complicated, when we don't have a lot of good
infrastructure for it.
Let it rip!
