drm: Ifdef out alpha-specific logic in radeon.

Guessing this is handled transparently by bus_space(9) when we ask
for prefetchable linear mappings. Need an alpha wizard to review!

drm: Nix driver-specific DRM_FILE_PAGE_OFFSET defs and tests.

This was consolidated upstream into the vma manager.

PR kern/56812

Cast addr to uint64_t before printing.
Ok: riastradh@

radeon: Fix dma_addressing_limited call.

Stub dma_addressing_limited, fixup references for caller

This might turn out to be a mistake if we actually need to distinguish
if a device is only capable of 32-bit DMA.


Author: Maya Rashish <maya@NetBSD.org>

'r' is just for us, so let's declare it


Author: Maya Rashish <maya@NetBSD.org>

match the 'else' case here to get vma_node


Author: Maya Rashish <maya@NetBSD.org>

Fix up radeon.

Enough to make it compile. (Next up: make it link.)

Merge linux-drm-v5-6-rc3

Fix possible bus_dmamap_load(9) leak. PR/55127

"Looks good to me" from riastradh@.
Note it was also commented "that code path is likely to be reached"
so maybe pullups are not necessary.

Implement drm userptr with uvm_vslock & bus_dmamap_load_uio.

branches: 1.15.6;
Draft support for drm prime.

Sprinkle AGP conditionals where needed.

Kill legacy mmap in radeon like was done upstream.

Ifdef out some more radeon userptr code.

Disable radeon userptr for now.

merge linux-drm-v4-4-143

branches: 1.9.16; 1.9.18;
Don't assign value to rdev unless we're going to use the value.

Therefore, don't declare the variable unless we're going to assign it a
value.

Rework ttm tt swapin/swapout logic.

Rather than handling `swapping in/out' here, per se, we let uvm do
that, we interpret `swap out' as `deactivate pages', and we add
generic ttm operations to wire and unwire pages, for the ttm_tt
driver to use.

This fixes certain graphics buffer eviction logic, which enables
nouveau to suspend/resume on one of my machines. (The machine
doesn't resume overall for other reasons, but the nouveau device
suspends and resumes in isolation.)

XXX pullup to netbsd-7 after a couple weeks

KASSERT about page-alignment on initialization too.

apply a patch from rjs@, which is necessary for radeondrmkms on i386:

set DRM_FILE_PAGE_OFFSET to 0xa0000000UL for 32 bit platforms.

branches: 1.5.2; 1.5.4; 1.5.6; 1.5.8;
Make sure the low bits, not the high bits, of the offset are zero!

Let the drm driver decide what part of what object gets mmapped.

Pass the file around too so radeon/ttm can get at it to verify access.

Add drm_gem_or_legacy_mmap_object for drivers to choose the previous
behaviour, like i915.

Actually hook up the radeon_ttm uvm pager ops!

First whack at radeon driver.

No hardware to test yet, but it builds.

branches: 1.1.1;
Initial revision

drm: Nix driver-specific DRM_FILE_PAGE_OFFSET defs and tests.

This was consolidated upstream into the vma manager.

PR kern/56812

Cast addr to uint64_t before printing.
Ok: riastradh@

radeon: Fix dma_addressing_limited call.

Stub dma_addressing_limited, fixup references for caller

This might turn out to be a mistake if we actually need to distinguish
if a device is only capable of 32-bit DMA.


Author: Maya Rashish <maya@NetBSD.org>

'r' is just for us, so let's declare it


Author: Maya Rashish <maya@NetBSD.org>

match the 'else' case here to get vma_node


Author: Maya Rashish <maya@NetBSD.org>

Fix up radeon.

Enough to make it compile. (Next up: make it link.)

Merge linux-drm-v5-6-rc3

Fix possible bus_dmamap_load(9) leak. PR/55127

"Looks good to me" from riastradh@.
Note it was also commented "that code path is likely to be reached"
so maybe pullups are not necessary.

Implement drm userptr with uvm_vslock & bus_dmamap_load_uio.

branches: 1.15.6;
Draft support for drm prime.

Sprinkle AGP conditionals where needed.

Kill legacy mmap in radeon like was done upstream.

Ifdef out some more radeon userptr code.

Disable radeon userptr for now.

merge linux-drm-v4-4-143

branches: 1.9.16; 1.9.18;
Don't assign value to rdev unless we're going to use the value.

Therefore, don't declare the variable unless we're going to assign it a
value.

Rework ttm tt swapin/swapout logic.

Rather than handling `swapping in/out' here, per se, we let uvm do
that, we interpret `swap out' as `deactivate pages', and we add
generic ttm operations to wire and unwire pages, for the ttm_tt
driver to use.

This fixes certain graphics buffer eviction logic, which enables
nouveau to suspend/resume on one of my machines. (The machine
doesn't resume overall for other reasons, but the nouveau device
suspends and resumes in isolation.)

XXX pullup to netbsd-7 after a couple weeks

KASSERT about page-alignment on initialization too.

apply a patch from rjs@, which is necessary for radeondrmkms on i386:

set DRM_FILE_PAGE_OFFSET to 0xa0000000UL for 32 bit platforms.

branches: 1.5.2; 1.5.4; 1.5.6; 1.5.8;
Make sure the low bits, not the high bits, of the offset are zero!

Let the drm driver decide what part of what object gets mmapped.

Pass the file around too so radeon/ttm can get at it to verify access.

Add drm_gem_or_legacy_mmap_object for drivers to choose the previous
behaviour, like i915.

Actually hook up the radeon_ttm uvm pager ops!

First whack at radeon driver.

No hardware to test yet, but it builds.

branches: 1.1.1;
Initial revision

Cast addr to uint64_t before printing.
Ok: riastradh@

radeon: Fix dma_addressing_limited call.

Stub dma_addressing_limited, fixup references for caller

This might turn out to be a mistake if we actually need to distinguish
if a device is only capable of 32-bit DMA.


Author: Maya Rashish <maya@NetBSD.org>

'r' is just for us, so let's declare it


Author: Maya Rashish <maya@NetBSD.org>

match the 'else' case here to get vma_node


Author: Maya Rashish <maya@NetBSD.org>

Fix up radeon.

Enough to make it compile. (Next up: make it link.)

Merge linux-drm-v5-6-rc3

Fix possible bus_dmamap_load(9) leak. PR/55127

"Looks good to me" from riastradh@.
Note it was also commented "that code path is likely to be reached"
so maybe pullups are not necessary.

Implement drm userptr with uvm_vslock & bus_dmamap_load_uio.

branches: 1.15.6;
Draft support for drm prime.

Sprinkle AGP conditionals where needed.

Kill legacy mmap in radeon like was done upstream.

Ifdef out some more radeon userptr code.

Disable radeon userptr for now.

merge linux-drm-v4-4-143

branches: 1.9.16; 1.9.18;
Don't assign value to rdev unless we're going to use the value.

Therefore, don't declare the variable unless we're going to assign it a
value.

Rework ttm tt swapin/swapout logic.

Rather than handling `swapping in/out' here, per se, we let uvm do
that, we interpret `swap out' as `deactivate pages', and we add
generic ttm operations to wire and unwire pages, for the ttm_tt
driver to use.

This fixes certain graphics buffer eviction logic, which enables
nouveau to suspend/resume on one of my machines. (The machine
doesn't resume overall for other reasons, but the nouveau device
suspends and resumes in isolation.)

XXX pullup to netbsd-7 after a couple weeks

KASSERT about page-alignment on initialization too.

apply a patch from rjs@, which is necessary for radeondrmkms on i386:

set DRM_FILE_PAGE_OFFSET to 0xa0000000UL for 32 bit platforms.

branches: 1.5.2; 1.5.4; 1.5.6; 1.5.8;
Make sure the low bits, not the high bits, of the offset are zero!

Let the drm driver decide what part of what object gets mmapped.

Pass the file around too so radeon/ttm can get at it to verify access.

Add drm_gem_or_legacy_mmap_object for drivers to choose the previous
behaviour, like i915.

Actually hook up the radeon_ttm uvm pager ops!

First whack at radeon driver.

No hardware to test yet, but it builds.

branches: 1.1.1;
Initial revision

radeon: Fix dma_addressing_limited call.

Stub dma_addressing_limited, fixup references for caller

This might turn out to be a mistake if we actually need to distinguish
if a device is only capable of 32-bit DMA.


Author: Maya Rashish <maya@NetBSD.org>

'r' is just for us, so let's declare it


Author: Maya Rashish <maya@NetBSD.org>

match the 'else' case here to get vma_node


Author: Maya Rashish <maya@NetBSD.org>

Fix up radeon.

Enough to make it compile. (Next up: make it link.)

Merge linux-drm-v5-6-rc3

Fix possible bus_dmamap_load(9) leak. PR/55127

"Looks good to me" from riastradh@.
Note it was also commented "that code path is likely to be reached"
so maybe pullups are not necessary.

Implement drm userptr with uvm_vslock & bus_dmamap_load_uio.

branches: 1.15.6;
Draft support for drm prime.

Sprinkle AGP conditionals where needed.

Kill legacy mmap in radeon like was done upstream.

Ifdef out some more radeon userptr code.

Disable radeon userptr for now.

merge linux-drm-v4-4-143

branches: 1.9.16; 1.9.18;
Don't assign value to rdev unless we're going to use the value.

Therefore, don't declare the variable unless we're going to assign it a
value.

Rework ttm tt swapin/swapout logic.

Rather than handling `swapping in/out' here, per se, we let uvm do
that, we interpret `swap out' as `deactivate pages', and we add
generic ttm operations to wire and unwire pages, for the ttm_tt
driver to use.

This fixes certain graphics buffer eviction logic, which enables
nouveau to suspend/resume on one of my machines. (The machine
doesn't resume overall for other reasons, but the nouveau device
suspends and resumes in isolation.)

XXX pullup to netbsd-7 after a couple weeks

KASSERT about page-alignment on initialization too.

apply a patch from rjs@, which is necessary for radeondrmkms on i386:

set DRM_FILE_PAGE_OFFSET to 0xa0000000UL for 32 bit platforms.

branches: 1.5.2; 1.5.4; 1.5.6; 1.5.8;
Make sure the low bits, not the high bits, of the offset are zero!

Let the drm driver decide what part of what object gets mmapped.

Pass the file around too so radeon/ttm can get at it to verify access.

Add drm_gem_or_legacy_mmap_object for drivers to choose the previous
behaviour, like i915.

Actually hook up the radeon_ttm uvm pager ops!

First whack at radeon driver.

No hardware to test yet, but it builds.

branches: 1.1.1;
Initial revision

Fix possible bus_dmamap_load(9) leak. PR/55127

"Looks good to me" from riastradh@.
Note it was also commented "that code path is likely to be reached"
so maybe pullups are not necessary.

Implement drm userptr with uvm_vslock & bus_dmamap_load_uio.

branches: 1.15.6;
Draft support for drm prime.

Sprinkle AGP conditionals where needed.

Kill legacy mmap in radeon like was done upstream.

Ifdef out some more radeon userptr code.

Disable radeon userptr for now.

merge linux-drm-v4-4-143

branches: 1.9.16; 1.9.18;
Don't assign value to rdev unless we're going to use the value.

Therefore, don't declare the variable unless we're going to assign it a
value.

Rework ttm tt swapin/swapout logic.

Rather than handling `swapping in/out' here, per se, we let uvm do
that, we interpret `swap out' as `deactivate pages', and we add
generic ttm operations to wire and unwire pages, for the ttm_tt
driver to use.

This fixes certain graphics buffer eviction logic, which enables
nouveau to suspend/resume on one of my machines. (The machine
doesn't resume overall for other reasons, but the nouveau device
suspends and resumes in isolation.)

XXX pullup to netbsd-7 after a couple weeks

KASSERT about page-alignment on initialization too.

apply a patch from rjs@, which is necessary for radeondrmkms on i386:

set DRM_FILE_PAGE_OFFSET to 0xa0000000UL for 32 bit platforms.

branches: 1.5.2; 1.5.4; 1.5.6; 1.5.8;
Make sure the low bits, not the high bits, of the offset are zero!

Let the drm driver decide what part of what object gets mmapped.

Pass the file around too so radeon/ttm can get at it to verify access.

Add drm_gem_or_legacy_mmap_object for drivers to choose the previous
behaviour, like i915.

Actually hook up the radeon_ttm uvm pager ops!

First whack at radeon driver.

No hardware to test yet, but it builds.

branches: 1.1.1;
Initial revision

Implement drm userptr with uvm_vslock & bus_dmamap_load_uio.

Draft support for drm prime.

Sprinkle AGP conditionals where needed.

Kill legacy mmap in radeon like was done upstream.

Ifdef out some more radeon userptr code.

Disable radeon userptr for now.

merge linux-drm-v4-4-143

branches: 1.9.16; 1.9.18;
Don't assign value to rdev unless we're going to use the value.

Therefore, don't declare the variable unless we're going to assign it a
value.

Rework ttm tt swapin/swapout logic.

Rather than handling `swapping in/out' here, per se, we let uvm do
that, we interpret `swap out' as `deactivate pages', and we add
generic ttm operations to wire and unwire pages, for the ttm_tt
driver to use.

This fixes certain graphics buffer eviction logic, which enables
nouveau to suspend/resume on one of my machines. (The machine
doesn't resume overall for other reasons, but the nouveau device
suspends and resumes in isolation.)

XXX pullup to netbsd-7 after a couple weeks

KASSERT about page-alignment on initialization too.

apply a patch from rjs@, which is necessary for radeondrmkms on i386:

set DRM_FILE_PAGE_OFFSET to 0xa0000000UL for 32 bit platforms.

branches: 1.5.2; 1.5.4; 1.5.6; 1.5.8;
Make sure the low bits, not the high bits, of the offset are zero!

Let the drm driver decide what part of what object gets mmapped.

Pass the file around too so radeon/ttm can get at it to verify access.

Add drm_gem_or_legacy_mmap_object for drivers to choose the previous
behaviour, like i915.

Actually hook up the radeon_ttm uvm pager ops!

First whack at radeon driver.

No hardware to test yet, but it builds.

branches: 1.1.1;
Initial revision

Draft support for drm prime.

Sprinkle AGP conditionals where needed.

Kill legacy mmap in radeon like was done upstream.

Ifdef out some more radeon userptr code.

Disable radeon userptr for now.

merge linux-drm-v4-4-143

branches: 1.9.16;
Don't assign value to rdev unless we're going to use the value.

Therefore, don't declare the variable unless we're going to assign it a
value.

Rework ttm tt swapin/swapout logic.

Rather than handling `swapping in/out' here, per se, we let uvm do
that, we interpret `swap out' as `deactivate pages', and we add
generic ttm operations to wire and unwire pages, for the ttm_tt
driver to use.

This fixes certain graphics buffer eviction logic, which enables
nouveau to suspend/resume on one of my machines. (The machine
doesn't resume overall for other reasons, but the nouveau device
suspends and resumes in isolation.)

XXX pullup to netbsd-7 after a couple weeks

KASSERT about page-alignment on initialization too.

apply a patch from rjs@, which is necessary for radeondrmkms on i386:

set DRM_FILE_PAGE_OFFSET to 0xa0000000UL for 32 bit platforms.

branches: 1.5.2; 1.5.4; 1.5.6; 1.5.8;
Make sure the low bits, not the high bits, of the offset are zero!

Let the drm driver decide what part of what object gets mmapped.

Pass the file around too so radeon/ttm can get at it to verify access.

Add drm_gem_or_legacy_mmap_object for drivers to choose the previous
behaviour, like i915.

Actually hook up the radeon_ttm uvm pager ops!

First whack at radeon driver.

No hardware to test yet, but it builds.

branches: 1.1.1;
Initial revision

Don't assign value to rdev unless we're going to use the value.

Therefore, don't declare the variable unless we're going to assign it a
value.

Rework ttm tt swapin/swapout logic.

Rather than handling `swapping in/out' here, per se, we let uvm do
that, we interpret `swap out' as `deactivate pages', and we add
generic ttm operations to wire and unwire pages, for the ttm_tt
driver to use.

This fixes certain graphics buffer eviction logic, which enables
nouveau to suspend/resume on one of my machines. (The machine
doesn't resume overall for other reasons, but the nouveau device
suspends and resumes in isolation.)

XXX pullup to netbsd-7 after a couple weeks

KASSERT about page-alignment on initialization too.

apply a patch from rjs@, which is necessary for radeondrmkms on i386:

set DRM_FILE_PAGE_OFFSET to 0xa0000000UL for 32 bit platforms.

branches: 1.5.2; 1.5.4; 1.5.6; 1.5.8;
Make sure the low bits, not the high bits, of the offset are zero!

Let the drm driver decide what part of what object gets mmapped.

Pass the file around too so radeon/ttm can get at it to verify access.

Add drm_gem_or_legacy_mmap_object for drivers to choose the previous
behaviour, like i915.

Actually hook up the radeon_ttm uvm pager ops!

First whack at radeon driver.

No hardware to test yet, but it builds.

branches: 1.1.1;
Initial revision