this uses a new entrypoint to invalidate gart entries instead of using 0.
Changed to rather than pointing to 0 address point empty entry to dummy
page. This might help to avoid hard lockup if for some wrong
reasons GPU try to access unmapped GART entry.
I'm not 100% sure this is going to work, we probably need to allocate
a dummy page and point all the GTT entries at it similiar to what AGP does.
but we can test this first I suppose.
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
r = radeon_gart_table_vram_pin(rdev);
if (r)
return r;
+ radeon_gart_restore(rdev);
/* Setup L2 cache */
WREG32(VM_L2_CNTL, ENABLE_L2_CACHE | ENABLE_L2_FRAGMENT_PROCESSING |
ENABLE_L2_PTE_CACHE_LRU_UPDATE_BY_WRITE |
{
uint32_t tmp;
+ radeon_gart_restore(rdev);
/* discard memory request outside of configured range */
tmp = RREG32(RADEON_AIC_CNTL) | RADEON_DIS_OUT_OF_PCI_GART_ACCESS;
WREG32(RADEON_AIC_CNTL, tmp);
r = radeon_gart_table_vram_pin(rdev);
if (r)
return r;
+ radeon_gart_restore(rdev);
/* discard memory request outside of configured range */
tmp = RADEON_PCIE_TX_GART_UNMAPPED_ACCESS_DISCARD;
WREG32_PCIE(RADEON_PCIE_TX_GART_CNTL, tmp);
r = radeon_gart_table_vram_pin(rdev);
if (r)
return r;
+ radeon_gart_restore(rdev);
/* Setup L2 cache */
WREG32(VM_L2_CNTL, ENABLE_L2_CACHE | ENABLE_L2_FRAGMENT_PROCESSING |
/* AGP */
extern void radeon_agp_disable(struct radeon_device *rdev);
extern int radeon_gart_table_vram_pin(struct radeon_device *rdev);
+extern void radeon_gart_restore(struct radeon_device *rdev);
extern int radeon_modeset_init(struct radeon_device *rdev);
extern void radeon_modeset_fini(struct radeon_device *rdev);
extern bool radeon_card_posted(struct radeon_device *rdev);
extern int r600_cp_resume(struct radeon_device *rdev);
extern void r600_cp_fini(struct radeon_device *rdev);
extern int r600_count_pipe_bits(uint32_t val);
-extern int r600_gart_clear_page(struct radeon_device *rdev, int i);
extern int r600_mc_wait_for_idle(struct radeon_device *rdev);
extern int r600_pcie_gart_init(struct radeon_device *rdev);
extern void r600_pcie_gart_tlb_flush(struct radeon_device *rdev);
int radeon_dummy_page_init(struct radeon_device *rdev)
{
+ if (rdev->dummy_page.page)
+ return 0;
rdev->dummy_page.page = alloc_page(GFP_DMA32 | GFP_KERNEL | __GFP_ZERO);
if (rdev->dummy_page.page == NULL)
return -ENOMEM;
unsigned t;
unsigned p;
int i, j;
+ u64 page_base;
if (!rdev->gart.ready) {
WARN(1, "trying to unbind memory to unitialized GART !\n");
pci_unmap_page(rdev->pdev, rdev->gart.pages_addr[p],
PAGE_SIZE, PCI_DMA_BIDIRECTIONAL);
rdev->gart.pages[p] = NULL;
- rdev->gart.pages_addr[p] = 0;
+ rdev->gart.pages_addr[p] = rdev->dummy_page.addr;
+ page_base = rdev->gart.pages_addr[p];
for (j = 0; j < (PAGE_SIZE / RADEON_GPU_PAGE_SIZE); j++, t++) {
- radeon_gart_set_page(rdev, t, 0);
+ radeon_gart_set_page(rdev, t, page_base);
+ page_base += RADEON_GPU_PAGE_SIZE;
}
}
}
return 0;
}
+void radeon_gart_restore(struct radeon_device *rdev)
+{
+ int i, j, t;
+ u64 page_base;
+
+ for (i = 0, t = 0; i < rdev->gart.num_cpu_pages; i++) {
+ page_base = rdev->gart.pages_addr[i];
+ for (j = 0; j < (PAGE_SIZE / RADEON_GPU_PAGE_SIZE); j++, t++) {
+ radeon_gart_set_page(rdev, t, page_base);
+ page_base += RADEON_GPU_PAGE_SIZE;
+ }
+ }
+ mb();
+ radeon_gart_tlb_flush(rdev);
+}
+
int radeon_gart_init(struct radeon_device *rdev)
{
+ int r, i;
+
if (rdev->gart.pages) {
return 0;
}
DRM_ERROR("Page size is smaller than GPU page size!\n");
return -EINVAL;
}
+ r = radeon_dummy_page_init(rdev);
+ if (r)
+ return r;
/* Compute table size */
rdev->gart.num_cpu_pages = rdev->mc.gtt_size / PAGE_SIZE;
rdev->gart.num_gpu_pages = rdev->mc.gtt_size / RADEON_GPU_PAGE_SIZE;
radeon_gart_fini(rdev);
return -ENOMEM;
}
+ /* set GART entry to point to the dummy page by default */
+ for (i = 0; i < rdev->gart.num_cpu_pages; i++) {
+ rdev->gart.pages_addr[i] = rdev->dummy_page.addr;
+ }
return 0;
}
uint32_t size_reg;
uint32_t tmp;
+ radeon_gart_restore(rdev);
tmp = RREG32_MC(RS690_AIC_CTRL_SCRATCH);
tmp |= RS690_DIS_OUT_OF_PCI_GART_ACCESS;
WREG32_MC(RS690_AIC_CTRL_SCRATCH, tmp);
r = radeon_gart_table_vram_pin(rdev);
if (r)
return r;
+ radeon_gart_restore(rdev);
/* Enable bus master */
tmp = RREG32(R_00004C_BUS_CNTL) & C_00004C_BUS_MASTER_DIS;
WREG32(R_00004C_BUS_CNTL, tmp);
r = radeon_gart_table_vram_pin(rdev);
if (r)
return r;
+ radeon_gart_restore(rdev);
/* Setup L2 cache */
WREG32(VM_L2_CNTL, ENABLE_L2_CACHE | ENABLE_L2_FRAGMENT_PROCESSING |
ENABLE_L2_PTE_CACHE_LRU_UPDATE_BY_WRITE |