GitLab

Ditto, don't forget ggtt entries during reset.
Signed-off-by: Changbin Du <changbin.du@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>

We have canceled dma map for ppgtt entries. Also we need to do it for
ggtt entries when them are invalidated.

This can fix task hung issue as:
[13517.791767] INFO: task gvt_service_thr:1081 blocked for more than 120 seconds.
[13517.792584] Not tainted 4.14.15+ #3
[13517.793417] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[13517.794267] gvt_service_thr D 0 1081 2 0x80000000
[13517.795132] Call Trace:
[13517.795996] ? __schedule+0x493/0x77b
[13517.796859] schedule+0x79/0x82
[13517.797740] schedule_preempt_disabled+0x5/0x6
[13517.798614] __mutex_lock.isra.0+0x2b5/0x445
[13517.799504] ? __switch_to_asm+0x24/0x60
[13517.800381] ? intel_gvt_cleanup+0x10/0x10
[13517.801261] ? intel_gvt_schedule+0x19/0x2b9
[13517.802107] intel_gvt_schedule+0x19/0x2b9
[13517.802954] ? intel_gvt_cleanup+0x10/0x10
[13517.803824] gvt_service_thread+0xe3/0x10d
[13517.804704] ? wait_woken+0x68/0x68
[13517.805588] kthread+0x118/0x120
[13517.806478] ? kthread_create_on_node+0x3a/0x3a
[13517.807381] ? call_usermodehelper_exec_async+0x113/0x11a
[13517.808307] ret_from_fork+0x35/0x40

v3: split out ggtt reset case.
v2: also unmap ggtt during reset.
Signed-off-by: Changbin Du <changbin.du@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>

As different OSes might handling GVT PPGTT creation/destroy notification
differently during a vGPU reset. A better approach is invalidating all
vGPU PPGTT mm objects during vGPU reset.
Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>

Trivial fix to spelling mistake in gvt_err error message text.
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>

The implementation of current kvmgt implicitly setup dma mapping at MPT
API gfn_to_mfn. First this design against the API's original purpose.
Second, there is no unmap hit in this design. The result is that the
dma mapping keep growing larger and larger. For mutl-vm case, they will
consume IOMMU IOVA low 4GB address space quickly and so tons of rbtree
entries crated in the IOMMU IOVA allocator. Finally, single IOVA
allocation can take as long as ~70ms. Such latency is intolerable.

To address both above issues, this patch introduced two new MPT API:
  o dma_map_guest_page - setup dma map for guest page
  o dma_unmap_guest_page - cancel dma map for guest page

The kvmgt implements these 2 API. And to reduce dma setup overhead for
duplicated pages (eg. scratch pages), two caches are used: one is for
mapping gfn to struct gvt_dma, another is for mapping dma addr to
struct gvt_dma.

With these 2 new API, the gtt now is able to cancel dma mapping when page
table is invalidated. The dma mapping is not in a gradual increase now.

v2: follow the old logic for VFIO_IOMMU_NOTIFY_DMA_UNMAP at this point.

Cc: Hang Yuan <hang.yuan@intel.com>
Cc: Xiong Zhang <xiong.y.zhang@intel.com>
Signed-off-by: Changbin Du <changbin.du@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>

Define the masks better.
Signed-off-by: Changbin Du <changbin.du@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>

We don't know how many page tables will be shadowed. It varies
considerably corresponding to guest load. Radix tree is a better
choice for us. Since Page Frame Number is used as key so most of
the bits are common.

Here is some performance data (duration in us) of looking up a
element:
Before: (aka. ppgtt_find_shadow_page)
 0.308 0.292 0.246 0.432 0.143 ... 0.311 0.225 0.382 0.199 0.325
After: (aka. intel_vgpu_find_spt_by_mfn)
 0.106 0.106 0.107 0.106 0.105 0.107 ... 0.107 0.109 0.105 0.108

This time I didn't get the early data of hash table. The data is
measured when desktop is shown.

As last change, the overall benchmark almost is not changed, but
we get better scalability.
Signed-off-by: Changbin Du <changbin.du@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>

This patch provide generic page_track infrastructure for write-protected
guest page. The old page_track logic gets rewrote and now stays in a new
standalone page_track.c. This page track infrastructure can be both used
by vGUC and GTT shadowing.

The important change is that it uses radix tree instead of hash table.
We don't have a predictable number of pages that will be tracked.

Here is some performance data (duration in us) of looking up a element:
Before: (aka. intel_vgpu_find_tracked_page)
 0.091 0.089 0.090 ... 0.093 0.091 0.087 ... 0.292 0.285 0.292 0.291
After: (aka. intel_vgpu_find_page_track)
 0.104 0.105 0.100 0.102 0.102 0.100 ... 0.101 0.101 0.105 0.105

The hash table has good performance at beginning, but turns bad with
more pages being tracked even no 3D applications are running. As
expected, radix tree has stable duration and very quick.

The overall benchmark (tested with Heaven Benchmark) marginally improved
since this is not the bottleneck. What we benefit more from this change
is scalability.
Signed-off-by: Changbin Du <changbin.du@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>

Don't extend page_track to mpt layer. Keep MPT simple and clean.
Meanwhile remove gtt.n_tracked_guest_page which doesn't make much
sense.

v2: clean up gtt.n_tracked_guest_page.
Signed-off-by: Changbin Du <changbin.du@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>

The target structure of some functions is struct intel_vgpu_ppgtt_spt and
their names are xxx_shadow_page. It should be xxx_shadow_page_table. Let's
use short name 'spt' instead to reduce the length. As well as the hash
table name.
Signed-off-by: Changbin Du <changbin.du@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>

This is a another big one and the GVT shadow page management code is
heavily refined.

The new code only use struct intel_vgpu_ppgtt_spt to represent a vgpu
shadow page table - w/ or wo/ a guest page associated with. A pure shadow
page (no guest page associated) will be used to shadow splited 2M huge
gtt. In this case, the spt.guest_page.gfn should be a zero.

To search a existed shadow page table, we have two new interfaces:
 - intel_vgpu_find_spt_by_gfn(), find a spt by guest gfn. It must not
   be a pure spt.
 - intel_vgpu_find_spt_by_mfn, Find the spt using shadow page mfn in
   shadowed PTE.

The oos_page management is remained as what is was.

v2: Split some changes into small standalone patches.
Signed-off-by: Changbin Du <changbin.du@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>

Make the shadow PTE population code clear. Later we will add huge gtt
support based on this.

v2:
  - rebase to latest code.
Signed-off-by: Changbin Du <changbin.du@intel.com>
Reviewed-by: Zhi Wang <zhi.wang@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>

GTT entry has similar format with the CPU PTE. We'd prefer named macro
instead of hardcode.
Signed-off-by: Changbin Du <changbin.du@intel.com>
Reviewed-by: Zhi Wang <zhi.a.wang@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>

Factor out these two interfaces so we can kill some duplicated code in
scheduler.c.

v2:
  - rename to intel_vgpu_{get,put}_ppgtt_mm
  - refine handle_g2v_notification
Signed-off-by: Changbin Du <changbin.du@intel.com>
Reviewed-by: Zhi Wang <zhi.a.wang@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>

Accurate names help to avoid confusing so improve readability.
Signed-off-by: Changbin Du <changbin.du@intel.com>
Reviewed-by: Zhi Wang <zhi.a.wang@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>

This add a new macro gvt_vdbg_mm() to print more verbose logs for
gtt shadowing. The added verbose logs are very useful for debugging.
gvt_vdbg_mm() only comes into effect if VERBOSE_DEBUG is defined by
the developer.
Signed-off-by: Changbin Du <changbin.du@intel.com>
Reviewed-by: Zhi Wang <zhi.a.wang@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>

Less code and use existed helper ggtt_set_host_entry.
Signed-off-by: Changbin Du <changbin.du@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>

Separate ggtt and ppgtt since they are different. A little more code but
straightforward.

And move these helpers to gtt.c since that is the only client.
Signed-off-by: Changbin Du <changbin.du@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>

If we manage an object with a reference count, then its life cycle
must flow the reference count operations. Meanwhile, change the
operation functions to generic name *put* and *get*.
Signed-off-by: Changbin Du <changbin.du@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>

This is a big one and the GVT shadow graphic memory management code is
heavily refined. The new code is more straightforward with less code.

The struct intel_vgpu_mm is restructured to be clearly defined, use
accurate names and some of the original fields are removed which are
really redundant.

Now we only manage ppgtt mm object with mm->ppgtt_mm.lru_list. No need
to mix ppgtt and ggtt together, since one vGPU only has one ggtt object.

v4: Don't invoke ppgtt_free_all_shadow_page before intel_vgpu_destroy_all_ppgtt_mm.
v3: Add GVT_RING_CTX_NR_PDPS to avoid confusing about the PDPs.
v2: Split some changes into small standalone patches.
Signed-off-by: Changbin Du <changbin.du@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>

GVT may receive partial write on one guest PTE update. Validate gfn
not to translate incomplete gfn. This avoids some unnecessary error
messages incurred by the incomplete gfn translating. Also fix the
bug that the whole PPGTT shadow page update is aborted on any invalid
gfn entry.

gfn validation relys on hypervisor's help. Add one MPT module function
to provide the function.
Signed-off-by: Hang Yuan <hang.yuan@intel.com>
Reviewed-by: Zhi Wang <zhi.a.wang@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>

A shadow page table entry needs to be cleared after being set as
post-sync. This patch fixes the recent error reported in Win7-32 test.

Fixes: 2707e444 ("drm/i915/gvt: vGPU graphics memory virtualization")
Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>
CC: Stable <stable@vger.kernel.org>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>

It's a bit confusing that page write protect handler is live in
mmio emulation handler. This moves it to stand alone gvt ops.

Also remove unnecessary check of write protected page access
in mmio read handler and cleanup handling of failsafe case.

v2: rebase
Reviewed-by: Xiong Zhang <xiong.y.zhang@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>

We had previous hack that tried to accept either i915_reg_t or offset
value to access vGPU virtual/shadow regs which broke that purpose to
be type safe in context. This one trys to explicitly separate the usage
of typed mmio reg with real offset.

Old vgpu_vreg(offset) helper is used only for offset now with new
vgpu_vreg_t(reg) is used for i915_reg_t only. Convert left usage
of that to new helper.

Also fixed left KASAN warning issues caused by previous hack.

v2: rebase, fixup against recent mmio switch change
Reviewed-by: Zhi Wang <zhi.a.wang@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>

For ADDR_4K_MASK, bit[45..12] should be 1, all other bits
should be 0. The current definition wrongly set bit[46] as 1
also. This path fixes this.

v2: Add commit message, fixes and cc stable.(Zhenyu)

Fixes: 2707e444("drm/i915/gvt: vGPU graphics memory virtualization")
Signed-off-by: Xiong Zhang <xiong.y.zhang@intel.com>
Cc: stable@vger.kernel.org
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>

This reverts commit b20d09886fd1b74cd2255d846029a049e524db14.

This caused windows driver boot errors for invalid page address.
Revert for now.
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
Cc: Zhi Wang <zhi.a.wang@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>

Refine previously broken PPGTT scratch. Scratch PTE was no correctly
handled and also the handling of scratch entries in page table walk was
not well organized, which brings gaps of introducing lazy shadow.
Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>

We need ops->set_present() during generating a new scratch page table
entry.
Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>

Need to figure out page table type of current level by GTT entry type
during getting a scratch page table entry.
Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>

During a vGPU reset, the scratch page table shouldn't be cleared, what
needs to be cleared should be the scratch page.
Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>

As we want to re-use intel_vgpu_shadow_page in buidling scrach page table
and we don't want to put scrach page table page into hash table, a new
param is introduced to give the caller a choice to decide if a shadow page
should be put into hash table.
Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>

As there is already an I915_GTT_PAGE_SIZE marco in i915, let GVT-g use it
as well. Also this patch re-names some GTT marcos with additional prefix.
Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>

As the data structure of "intel_vgpu_guest_page" will become much heavier
in future, it's better to factor out the guest memory page track mechnisim
as early as possible.
Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>

Replace the plain bit usage with BIT() to make klockwork happy.

Cc: Deng Hongyi <hongyi.deng@intel.com>
Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>

Remove the "INDEX" suffix from PPAT marcos as they are bits actually, not
indexes.
Suggested-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>
Cc: Ben Widawsky <benjamin.widawsky@intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/1505392783-4084-2-git-send-email-zhi.a.wang@intel.com

When it is failed in shadow_mm, the pin_count should rollback
to the original states before return.

v2:
- split the mixed several error paths for better review. (Zhenyu)

v3:
  increase the pincount after shadow success. (Zhenyu)
Signed-off-by: fred gao <fred.gao@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>

Guest i915 full ppgtt functionality was blocking by an issue, which would
lead to gpu hardware hang. Guest i915 driver may update the ppgtt table
just before this workload is going to be submitted to the hardware by
device model. This case wasn't handled well by device model before, due
to the small time window between removing old ppgtt entry and adding the
new one. Errors occur when the workload is executed by hardware during
that small time window. This patch is to remove this time window by adding
the new ppgtt entry first and then remove the old one.

Changes in v2:
- Move VGT_CAPS_FULL_PPGTT introduction to patch 2/4. (Joonas)

Changes since v2:
- Divide the whole patch set into two separate patch series, with one
  patch in i915 side to check guest i915 full ppgtt capability and enable
  it when this capability is supported by the device model, and the other
  one in gvt side which fixs the blocking issue and enables the device
  model to provide the capability to guest. And this patch focuses on gvt
  side. (Joonas)
- Change the title from "reorder the shadow ppgtt update process by adding
  entry first" to "Fix guest i915 full ppgtt blocking issue". (Tina)

Changes since v3:
- Rebase to the latest branch.

Changes since v4:
- Tested by Tina Zhang.

Changes since v5:
- Rebase to the latest branch.

v6:
- Update full 48bit ppgtt definition

Cc: Tina Zhang <tina.zhang@intel.com>
Signed-off-by: Tina Zhang <tina.zhang@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>

When doing the VGPU reset, we don't need to do the gtt/ppgtt reset.
This will make the GVT to do the ppgtt shadow every time for
a workload and caused really bad performance after a VGPU reset.
This patch will make sure ppgtt clean only happen at device module
level reset to fix this.
Signed-off-by: Chuanxiao Dong <chuanxiao.dong@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>

When debugging the gtt code, found the intel_vgpu_gma_to_gpa() can
translate any given GMA though the GMA is not valid. This because
the GTT ops suppress the possible errors, which may result in an
invalid PT entry is retrieved by upper caller.

This patch changed the prototype of pte ops to propagate status to
callers. Then we make sure the GTT walker stop as early as when
a error is detected to prevent undefined behavior.
Signed-off-by: Changbin Du <changbin.du@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>

It will causes memory leak, if the function setup_spt_oos() fail,
in the function intel_gvt_init_gtt(),
which allocated by get_zeroed_page() and mapped by dma_map_page().

Unmap and free the page,  after STP oos initialize fail,
it will fix this issue.
Signed-off-by: Zhou, Wenjia <zhiyuan_zhu@htc.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>