Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

These currently just call the existing ones in nouveau_drv.c, but will be
extended in upcoming commits.  This needed to be separated from the current
code as there will be some header clashes until things are ported.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Won't necessarily be a drm_mm_node in the future, and I can't think of any
good reason to not use the offset from the bo struct.  There may have been
some reason once apon a time, but, separate commit just in case.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Reported-by: NMaarten Lankhorst <maarten.lankhorst@canonical.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
Signed-off-by: NMaarten Lankhorst <maarten.lankhorst@canonical.com>

Setting dev_mapping (pointer to the address_space structure
used for memory mappings) to the address_space of the first
opener's inode and then failing if other openers come in
through a different inode has a few restrictions that are
eliminated by this patch.

If we already have valid dev_mapping and we spot an opener
with different i_node, we force its i_mapping pointer to the
already established address_space structure (first opener's
inode). This will make all mappings from drm device hang off
the same address_space object.

Some benefits (things that now work and didn't work
before) of this patch are:

 * user space can mknod and use any number of device
   nodes and they will all work fine as long as the major
   device number is that of the drm module.
 * user space can even remove the first opener's device
   nodes and mknod the new one and the applications and
   windowing system will still work.
 * GPU drivers can safely assume that dev->dev_mapping is
   correct address_space and just blindly copy it
   into their (private) bdev.dev_mapping

For reference, some discussion that lead to this patch can
be found here:

http://lists.freedesktop.org/archives/dri-devel/2012-April/022283.htmlSigned-off-by: NIlija Hadzic <ihadzic@research.bell-labs.com>
Signed-off-by: NDave Airlie <airlied@redhat.com>

Now have a somewhat simpler semaphore sync implementation for nv17:nv84,
and a switched to using semaphores as fences on nv84+ and making use of
the hardware's >= acquire operation.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

This adds prime->fd and fd->prime support to nouveau,
it passes the SG object to TTM, and then populates the
GART entries using it.

v2: add stubbed kmap + use new function to fill out pages array
for faulting + add reimport test.
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Signed-off-by: NDave Airlie <airlied@redhat.com>

There was once good reasons for wanting the drm to be able to use M2MF etc
on user channels, but they're not relevant anymore.  For the general
buffer move case, we've already lost by transferring between vram/sysmem
already so the context switching overhead is minimal in comparison.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Due to a race it was possible for a fence to be destroyed while another
thread was trying to synchronise with it.  If this happened in the fallback
non-semaphore path, it lead to the following oops due to fence->channel
being NULL.

BUG: unable to handle kernel NULL pointer dereference at   (null)
IP: [<fa9632ce>] nouveau_fence_update+0xe/0xe0 [nouveau]
*pde = a649c067
SMP
Modules linked in: fuse nouveau(O) ttm(O) drm_kms_helper(O) drm(O) mxm_wmi video wmi netconsole configfs lockd bnep bluetooth rfkill ip6t_REJECT nf_conntrack_ipv6 nf_defrag_ipv6 nf_conntrack_ipv4 nf_defrag_ipv4 xt_state nf_conntrack ip6table_filter ip6_tables snd_hda_codec_realtek snd_hda_intel snd_hda_cobinfmt_misc uinput ata_generic pata_acpi pata_aet2c_algo_bit i2c_core [last unloaded: wmi]

Pid: 2255, comm: gnome-shell Tainted: G           O 3.2.0-0.rc5.git0.1.fc17.i686 #1 System manufacturer System Product Name/M2A-VM
EIP: 0060:[<fa9632ce>] EFLAGS: 00010296 CPU: 1
EIP is at nouveau_fence_update+0xe/0xe0 [nouveau]
EAX: 00000000 EBX: ddfc6dd0 ECX: dd111580 EDX: 00000000
ESI: 00003e80 EDI: dd111580 EBP: dd121d00 ESP: dd121ce8
 DS: 007b ES: 007b FS: 00d8 GS: 00e0 SS: 0068
Process gnome-shell (pid: 2255, ti=dd120000 task=dd111580 task.ti=dd120000)
Stack:
 7dc86c76 00000000 00003e80 ddfc6dd0 00003e80 dd111580 dd121d0c fa96371f
 00000000 dd121d3c fa963773 dd111580 01000246 000ec53d 00000000 ddfc6dd0
 00001f40 00000000 ddfc6dd0 00000010 dc7df840 dd121d6c fa9639a0 00000000
Call Trace:
 [<fa96371f>] __nouveau_fence_signalled+0x1f/0x30 [nouveau]
 [<fa963773>] __nouveau_fence_wait+0x43/0xd0 [nouveau]
 [<fa9639a0>] nouveau_fence_sync+0x1a0/0x1c0 [nouveau]
 [<fa964046>] validate_list+0x176/0x300 [nouveau]
 [<f7d9c9c0>] ? ttm_bo_mem_put+0x30/0x30 [ttm]
 [<fa964b8a>] nouveau_gem_ioctl_pushbuf+0x48a/0xfd0 [nouveau]
 [<c0406481>] ? die+0x31/0x80
 [<f7c93d98>] drm_ioctl+0x388/0x490 [drm]
 [<c0406481>] ? die+0x31/0x80
 [<fa964700>] ? nouveau_gem_ioctl_new+0x150/0x150 [nouveau]
 [<c0635c7b>] ? file_has_perm+0xcb/0xe0
 [<f7c93a10>] ? drm_copy_field+0x80/0x80 [drm]
 [<c0564f56>] do_vfs_ioctl+0x86/0x5b0
 [<c0406481>] ? die+0x31/0x80
 [<c0635f22>] ? selinux_file_ioctl+0x62/0x130
 [<c0554f30>] ? fget_light+0x30/0x340
 [<c05654ef>] sys_ioctl+0x6f/0x80
 [<c099e3a4>] syscall_call+0x7/0xb
 [<c0406481>] ? die+0x31/0x80
 [<c0406481>] ? die+0x31/0x80
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
Cc: stable@vger.kernel.org

This reverts commit dfadbbdb.

Further upstream discussion between Marek and Thomas decided this wasn't
fully baked and needed further work, so revert it before it hits mainline.
Signed-off-by: NDave Airlie <airlied@redhat.com>

Sometimes we want to know whether a buffer is busy and wait for it (bo_wait).
However, sometimes it would be more useful to be able to query whether
a buffer is busy and being either read or written, and wait until it's stopped
being either read or written. The point of this is to be able to avoid
unnecessary waiting, e.g. if a GPU has written something to a buffer and is now
reading that buffer, and a CPU wants to map that buffer for read, it needs to
only wait for the last write. If there were no write, there wouldn't be any
waiting needed.

This, or course, requires user space drivers to send read/write flags
with each relocation (like we have read/write domains in radeon, so we can
actually use those for something useful now).

Now how this patch works:

The read/write flags should passed to ttm_validate_buffer. TTM maintains
separate sync objects of the last read and write for each buffer, in addition
to the sync object of the last use of a buffer. ttm_bo_wait then operates
with one the sync objects.
Signed-off-by: NMarek Olšák <maraeo@gmail.com>
Reviewed-by: NJerome Glisse <jglisse@redhat.com>
Signed-off-by: NDave Airlie <airlied@redhat.com>

Signed-off-by: NMarcin Slusarz <marcin.slusarz@gmail.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

GPU virtual addresses are constant now so this should never be getting hit
anyway and userspace shouldn't break from them being ignored.

This is being done in preference to teaching the code how to deal with BOs
that exist at different virtual addresses within separate VMs.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Userspace hasn't passed us a channel_hint for a long long time now, and
there isn't actually a need to do so anymore anyway.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Removes the need to disable IRQs to lookup channel struct on every pushbuf
ioctl, among others.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NJan Engelhardt <jengelh@medozas.de>
Signed-off-by: NDave Airlie <airlied@redhat.com>

When "buffer in list" check does not pass, don't free validation lists - they were
not initialized yet.

Fixes this oops:

[drm] nouveau 0000:02:00.0: push 105 buffer not in list
BUG: unable to handle kernel NULL pointer dereference at 000000000000057c
IP: [<ffffffff81236aa4>] do_raw_spin_lock+0x14/0x13c
PGD 1ac6cb067 PUD 1aaa52067 PMD 0
CPU 0
Modules linked in: nouveau ttm drm_kms_helper snd_hda_codec_realtek snd_hda_intel snd_hda_codec

Pid: 6265, comm: OilRush_x86 Not tainted 2.6.38-rc6-nv+ #632 System manufacturer System Product Name/P6T SE
RIP: 0010:[<ffffffff81236aa4>]  [<ffffffff81236aa4>] do_raw_spin_lock+0x14/0x13c
(...)
Process OilRush_x86 (pid: 6265, threadinfo ffff8801a6aee000, task ffff8801a26c0000)
 0000000000000000 ffff8801ac74c618 0000000000000000 0000000000000578
 0000000000000000 ffff8801ac74c618 0000000000000000 ffff8801bd9d0000
 [<ffffffff81417f78>] _raw_spin_lock+0x1e/0x22
 [<ffffffffa00a2746>] nouveau_bo_fence+0x2e/0x60 [nouveau]
 [<ffffffffa00a540b>] validate_fini_list+0x35/0xeb [nouveau]
 [<ffffffffa00a54d3>] validate_fini+0x12/0x31 [nouveau]
 [<ffffffffa00a6386>] nouveau_gem_ioctl_pushbuf+0xe94/0xf6b [nouveau]
 [<ffffffff8141ac56>] ? sub_preempt_count+0x9e/0xb2
 [<ffffffff81417e94>] ? _raw_spin_unlock_irqrestore+0x30/0x4d
 [<ffffffff8105dea2>] ? __wake_up+0x3f/0x48
 [<ffffffff812aebb4>] drm_ioctl+0x289/0x361
 [<ffffffff8141ac56>] ? sub_preempt_count+0x9e/0xb2
 [<ffffffffa00a54f2>] ? nouveau_gem_ioctl_pushbuf+0x0/0xf6b [nouveau]
 [<ffffffff8141ac56>] ? sub_preempt_count+0x9e/0xb2
 [<ffffffffa010caa2>] nouveau_compat_ioctl+0x16/0x1c [nouveau]
 [<ffffffff81142c0d>] compat_sys_ioctl+0x1c8/0x12d7
 [<ffffffff814179ca>] ? trace_hardirqs_off_thunk+0x3a/0x6c
 [<ffffffff81058099>] sysenter_dispatch+0x7/0x30
 [<ffffffff8141798e>] ? trace_hardirqs_on_thunk+0x3a/0x3c
RIP  [<ffffffff81236aa4>] do_raw_spin_lock+0x14/0x13c
 RSP <ffff8801a6aefb88>
---[ end trace 0014d5d93e6147e1 ]---

Additionally, don't call validate_fini twice in case of validation failure.
Signed-off-by: NMarcin Slusarz <marcin.slusarz@gmail.com>
Signed-off-by: NMaarten Maathuis <madman2003@gmail.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Upcoming patches are going to enable full support for buffers that keep
a constant GPU virtual address whenever they're validated for use by
the GPU.

In order for this to work properly while keeping support for large pages,
we need to know if it's ever going to be possible for a buffer to end
up in GART, and if so, disable large pages for the buffer's VMA.

This is a new restriction that's not present in earlier kernel's, but
should not break userspace as the current code never attempts to validate
buffers into a memtype other than it was created with.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

'mappable' isn't really used at all, nor is it necessary anymore as the
bo code is capable of moving buffers to mappable vram as required.

'no_vm' isn't necessary anymore either, any places that don't want to be
mapped into a GPU address space should allocate the VRAM directly instead.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

NVC0 will be able to share some of nv50's paths this way.  This also makes
it the card-specific vram code responsible for deciding if a given set
of tile_flags is valid, rather than duplicating the allowed types in
nv50_vram.c and nouveau_gem.c
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

This will be used later to fixup bo.offset with a buffer's fixed GPU
virtual address.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

nv0x-nv4x should be mostly fine, nv50 doesn't work yet.
Signed-off-by: NFrancisco Jerez <currojerez@riseup.net>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

nouveau_fence_* functions are not type safe, which could lead to bugs.
Additionally every use of nouveau_fence_unref had to cast struct
nouveau_fence to void **.
Fix it by renaming old functions and creating static inline functions with
new prototypes. We still need old functions, because we pass function
pointers to ttm.
As we are wrapping functions, drop unused "void *arg" parameter where possible.
Signed-off-by: NMarcin Slusarz <marcin.slusarz@gmail.com>
Signed-off-by: NFrancisco Jerez <currojerez@riseup.net>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

No other driver uses this, and userspace should be responsible for handling
locking between them if they share BOs.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

The bo lock used only to protect the bo sync object members, and since it
is a per bo lock, fencing a buffer list will see a lot of locks and unlocks.
Replace it with a per-device lock that protects the sync object members on
*all* bos. Reading and setting these members will always be very quick, so
the risc of heavy lock contention is microscopic. Note that waiting for
sync objects will always take place outside of this lock.

The bo device fence lock will eventually be replaced with a seqlock /
rcu mechanism so we can determine that a bo is idle under a
rcu / read seqlock.

However this change will allow us to batch fencing and unreserving of
buffers with a minimal amount of locking.
Signed-off-by: NThomas Hellstrom <thellstrom@vmware.com>
Reviewed-by: NJerome Glisse <j.glisse@gmail.com>
Signed-off-by: NDave Airlie <airlied@redhat.com>

This will be needed for Z compression and to take smarter placement
decisions.
Signed-off-by: NFrancisco Jerez <currojerez@riseup.net>
Acked-by: NBen Skeggs <bskeggs@redhat.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Mainly to make room for inter-channel sync.
Signed-off-by: NFrancisco Jerez <currojerez@riseup.net>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

There were lots of places being inconsistent since handle count
looked like a kref but it really wasn't.

Fix this my just making handle count an atomic on the object,
and have it increase the normal object kref.

Now i915/radeon/nouveau drivers can drop the normal reference on
userspace object creation, and have the handle hold it.

This patch fixes a memory leak or corruption on unload, because
the driver had no way of knowing if a handle had been actually
added for this object, and the fbcon object needed to know this
to clean itself up properly.
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NDave Airlie <airlied@redhat.com>

We don't need it and it can lead to lock order inversions with respect
to drm_global_mutex, potentially causing dead locks.
Signed-off-by: NFrancisco Jerez <currojerez@riseup.net>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

If we keep hold of the mutex here, the process which currently holds the
buffer object will never be able to release it, causing a deadlock.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

The "return" command is buggy on the original nv20, it jumps back to
the caller address as expected, but it doesn't clear the subroutine
active bit making the subsequent pushbuf calls fail with a "stack"
overflow.
Signed-off-by: NFrancisco Jerez <currojerez@riseup.net>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>