| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
The context needs to finish before everything else.
Signed-off-by: Christian König <[email protected]>
Acked-by: Alex Deucher <[email protected]>
Reviewed-by: Jammy Zhou <[email protected]>
Reviewed-by: Chunming Zhou <[email protected]>
|
|
This aids handling buffers moves with the scheduler.
Signed-off-by: Chunming Zhou <[email protected]>
Reviewed-by: Christian K?nig <[email protected]>
|
|
The fence in the array may be skipped if wait_all is false,
thus the related callback is not initialized with list head.
So removing this kind callback will cause NULL pointer reference.
Signed-off-by: Junwei Zhang <[email protected]>
Reviewed-by: Jammy Zhou <[email protected]>
|
|
when eviction is happening, if don't handle
dependency, then the fence could be dead off.
Signed-off-by: Chunming Zhou <[email protected]>
Reviewed-by: Jammy Zhou <[email protected]>
Reviewed-by: Christian K?nig <[email protected]>
|
|
Signed-off-by: Chunming Zhou <[email protected]>
Reviewed-by: Christian K?nig <[email protected]>
|
|
Signed-off-by: Chunming Zhou <[email protected]>
Reviewed-by: Christian K?nig <[email protected]>
|
|
Signed-off-by: Chunming Zhou <[email protected]>
Reviewed-by: Christian K?nig <[email protected]>
|
|
Entity don't live as long as scheduler fences.
Signed-off-by: Christian König <[email protected]>
Reviewed-by: Chunming Zhou <[email protected]>
|
|
Calling schedule() is probably the worse things we can do.
Signed-off-by: Christian König <[email protected]>
Reviewed-by: Chunming Zhou <[email protected]>
|
|
Not used any more.
Signed-off-by: Christian König <[email protected]>
Reviewed-by: Chunming Zhou <[email protected]>
|
|
Signed-off-by: Christian König <[email protected]>
Reviewed-by: Chunming Zhou <[email protected]>
|
|
Signed-off-by: Christian König <[email protected]>
Reviewed-by: Chunming Zhou <[email protected]>
|
|
Signed-off-by: Christian König <[email protected]>
Reviewed-by: Chunming Zhou <[email protected]>
|
|
Signed-off-by: Christian König <[email protected]>
Signed-off-by: Chunming Zhou <[email protected]>
Reviewed-by: Christian K?nig <[email protected]>
|
|
Simply not used any more. Only keep 32bit atomic for fence sequence numbering.
v2: trivial rebase
Signed-off-by: Christian König <[email protected]>
Reviewed-by: Alex Deucher <[email protected]> (v1)
Reviewed-by: Jammy Zhou <[email protected]> (v1)
Reviewed-by: Chunming Zhou <[email protected]> (v1)
|
|
Signed-off-by: Chunming Zhou <[email protected]>
Reviewed-by: Christian K?nig <[email protected]>
|
|
Rename the function and update the related code with this modified function.
Add the new parameter of bool wait_all.
If wait_all is true, it will return when all fences are signaled or timeout.
If wait_all is false, it will return when any fence is signaled or timeout.
Signed-off-by: Junwei Zhang <[email protected]>
Reviewed-by: Monk Liu <[email protected]>
Reviewed-by: Jammy Zhou <[email protected]>
Reviewed-by: Christian König <[email protected]>
|
|
Use pci_alloc_consistent rather than kzalloc since we
need 256 byte aligned memory for the ring buffer.
v2: fix copy paste typo in free function noticed
by Jammy.
bug:
https://bugs.freedesktop.org/show_bug.cgi?id=91749
Reviewed-by: Christian König <[email protected]>
Reviewed-by: Jammy Zhou <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
Cc: [email protected]
|
|
This is a port of:
DRM - radeon: Don't link train DisplayPort on HPD until we get the dpcd
to amdgpu.
Signed-off-by: Alex Deucher <[email protected]>
Cc: [email protected]
|
|
Stop double freeing the the BO list by pulling the content
of amdgpu_cs_parser_prepare_job() into the IOCTL function again.
v2: better commit message
Signed-off-by: Christian König <[email protected]>
Reviewed-by: Chunming Zhou <[email protected]> (v1)
Reviewed-by: Alex Deucher <[email protected]>
|
|
The problem now is that we don't necessarily call amdgpu_ib_get()
in some error paths and so work with uninitialized data.
Better require that the memory is already zeroed.
v2: better commit message
Signed-off-by: Christian König <[email protected]>
Reviewed-by: Chunming Zhou <[email protected]> (v1)
Reviewed-by: Alex Deucher <[email protected]>
|
|
Fixes a whole bunch of lockdep warnings.
Signed-off-by: Christian König <[email protected]>
Reviewed-by: Chunming Zhou <[email protected]>
|
|
Signed-off-by: Chunming Zhou <[email protected]>
Reviewed-by: Christian K?nig <[email protected]>
|
|
Signed-off-by: Chunming Zhou <[email protected]>
Reviewed-by: Christian K?nig <[email protected]>
|
|
Signed-off-by: Chunming Zhou <[email protected]>
Reviewed-by: Christian K?nig <[email protected]>
|
|
Signed-off-by: Chunming Zhou <[email protected]>
Reviewed-by: Christian K?nig <[email protected]>
|
|
Used by mesa, etc. for profiling.
Reviewed-by: Michel Dänzer <[email protected]>
Signed-off-by: Marek Olšák <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
Pagetables can be moved and therefore the page directory update can be necessary
for the current cs even if none of the the bo's are moved. In that scenario
there is no fence between the sdma0 and gfx ring, so we add one.
v2 (chk): rebased
Signed-off-by: Bas Nieuwenhuizen <[email protected]>
Signed-off-by: Christian König <[email protected]>
Reviewed-by: Alex Deucher <[email protected]>
|
|
Rework run queue implementation, especially remove the odd list handling.
v2: cleanup the code only, no algorithem change.
Signed-off-by: Christian König <[email protected]>
Reviewed-by: Chunming Zhou <[email protected]>
|
|
Signed-off-by: Chunming Zhou <[email protected]>
Reviewed-by: Christian K?nig <[email protected]>
|
|
fix the bug that there is duplicated bo_update_mapping issued
Signed-off-by: monk.liu <[email protected]>
Reviewed-by: Chunming Zhou <[email protected]>
Reviewed-by: Christian König <[email protected]>
|
|
bo_list_clone() will take a lot of time when bo_list hold too much
elements, like above 7000
Signed-off-by: Monk.Liu <[email protected]>
Reviewed-by: Chunming Zhou <[email protected]>
Reviewed-by: Jammy Zhou <[email protected]>
|
|
It's not validated yet and causes more harm than good.
Avoids spurious resets.
Reviewed-by: Leo Liu <[email protected]>
Reviewed-by: Christian König <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
remaining timeout returned by amdgpu_fence_wait_any can be larger than
max int value, thus the truncated 32 bit value in r ends up being
negative while its original long value is positive.
Signed-off-by: monk.liu <[email protected]>
Reviewed-by: Michel Dänzer <[email protected]>
Reviewed-by: Christian König <[email protected]>
Reviewed-by: Jammy Zhou <[email protected]>
|
|
fix fence is released when pass to **fence sometimes.
add reference for it.
Signed-off-by: Chunming Zhou <[email protected]>
Reviewed-by: Christian K?nig <[email protected]>
|
|
Otherwise we might see corruption.
Signed-off-by: Christian König <[email protected]>
Reviewed-by: Alex Deucher <[email protected]>
|
|
Signed-off-by: Christian König <[email protected]>
Reviewed-by: Alex Deucher <[email protected]>
Tested-and-Reviewed-by: Leo Liu <[email protected]>
|
|
Not used any more.
v2: remove amd_sched_emit as well.
Signed-off-by: Christian König <[email protected]>
Reviewed-by: Chunming Zhou <[email protected]>
|
|
Similar to radeon, except that amdgpu doesn't even use struct_mutex to
protect anything like the shared z buffer (sane gpu architecture,
yay!). And the code already grabs the globa adev->ring_lock, so this
code can't race with itself. Which makes struct_mutex completely
redundnant. Remove it.
Cc: Alex Deucher <[email protected]>
Cc: "Christian König" <[email protected]>
Reviewed-by: Thierry Reding <[email protected]>
Signed-off-by: Daniel Vetter <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
It really doesn't protect anything which doesn't have other locks
already. Also this is run from driver unload code so not much need for
locks anyway.
Same changes as for radeon really.
Cc: Alex Deucher <[email protected]>
Cc: "Christian König" <[email protected]>
Reviewed-by: Thierry Reding <[email protected]>
Signed-off-by: Daniel Vetter <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
The workaround simply doesn't work because VM mappings
are controlled by userspace not the kernel.
Additional to that this is just a performance problem
which happens if you have holes in your VM mapping.
v2: adjust virtual addr alignment as well.
v3: fix trivial warning
Signed-off-by: Christian König <[email protected]>
Reviewed-by: Monk Liu <[email protected]> (v1)
Reviewed-by: Jammy Zhou <[email protected]> (v2)
|
|
Signed-off-by: Christian König <[email protected]>
Reviewed-by: Alex Deucher <[email protected]>
|
|
Looks like that somehow got missed while during porting the radeon changes.
Signed-off-by: Christian König <[email protected]>
Reviewed-by: Alex Deucher <[email protected]>
|
|
It was just a wrapper for fence_wait anyway.
Signed-off-by: Christian König <[email protected]>
Reviewed-by: Alex Deucher <[email protected]>
|
|
Signed-off-by: Christian König <[email protected]>
Reviewed-by: Alex Deucher <[email protected]>
|
|
The common kernel function does the same thing.
Signed-off-by: Christian König <[email protected]>
Reviewed-by: Alex Deucher <[email protected]>
|
|
scheduler fence is based on kernel fence framework.
v2: squash in Christian's build fix
Signed-off-by: Chunming Zhou <[email protected]>
Reviewed-by: Christian K?nig <[email protected]>
|
|
Signed-off-by: Chunming Zhou <[email protected]>
Reviewed-by: Christian K?nig <[email protected]>
|
|
Signed-off-by: Chunming Zhou <[email protected]>
Reviewed-by: Christian K?nig <[email protected]>
|
|
v2: rebased
Signed-off-by: Christian König <[email protected]>
Reviewed-by: Alex Deucher <[email protected]> (v1)
Reviewed-by: Chunming Zhou <[email protected]>
|