static int radeon_cs_ib_fill(struct radeon_device *rdev, struct radeon_cs_parser *parser)
{
struct radeon_cs_chunk *ib_chunk;
struct radeon_vm *vm = NULL;
int r;
if (parser->chunk_ib_idx == -1)
return 0;
if (parser->cs_flags & RADEON_CS_USE_VM) {
struct radeon_fpriv *fpriv = parser->filp->driver_priv;
vm = &fpriv->vm;
if ((rdev->family >= CHIP_TAHITI) &&
(parser->chunk_const_ib_idx != -1)) {
ib_chunk = &parser->chunks[parser->chunk_const_ib_idx];
if (ib_chunk->length_dw > RADEON_IB_VM_MAX_SIZE) {
DRM_ERROR("cs IB CONST too big: %d\n", ib_chunk->length_dw);
return -EINVAL;
}
r = radeon_ib_get(rdev, parser->ring, &parser->const_ib,
vm, ib_chunk->length_dw * 4);
if (r) {
DRM_ERROR("Failed to get const ib !\n");
return r;
}
parser->const_ib.is_const_ib = true;
parser->const_ib.length_dw = ib_chunk->length_dw;
if (copy_from_user(parser->const_ib.ptr,
ib_chunk->user_ptr,
ib_chunk->length_dw * 4))
return -EFAULT;
}
ib_chunk = &parser->chunks[parser->chunk_ib_idx];
if (ib_chunk->length_dw > RADEON_IB_VM_MAX_SIZE) {
DRM_ERROR("cs IB too big: %d\n", ib_chunk->length_dw);
return -EINVAL;
}
}
ib_chunk = &parser->chunks[parser->chunk_ib_idx];
r = radeon_ib_get(rdev, parser->ring, &parser->ib,
vm, ib_chunk->length_dw * 4);
if (r) {
DRM_ERROR("Failed to get ib !\n");
return r;
}
parser->ib.length_dw = ib_chunk->length_dw;
if (ib_chunk->kdata)
memcpy(parser->ib.ptr, ib_chunk->kdata, ib_chunk->length_dw * 4);
else if (copy_from_user(parser->ib.ptr, ib_chunk->user_ptr, ib_chunk->length_dw * 4))
return -EFAULT;
return 0;
}
/**
* cik_sdma_ib_test - test an IB on the DMA engine
*
* @rdev: radeon_device pointer
* @ring: radeon_ring structure holding ring information
*
* Test a simple IB in the DMA ring (CIK).
* Returns 0 on success, error on failure.
*/
int cik_sdma_ib_test(struct radeon_device *rdev, struct radeon_ring *ring)
{
struct radeon_ib ib;
unsigned i;
unsigned index;
int r;
u32 tmp = 0;
u64 gpu_addr;
if (ring->idx == R600_RING_TYPE_DMA_INDEX)
index = R600_WB_DMA_RING_TEST_OFFSET;
else
index = CAYMAN_WB_DMA1_RING_TEST_OFFSET;
gpu_addr = rdev->wb.gpu_addr + index;
tmp = 0xCAFEDEAD;
rdev->wb.wb[index/4] = cpu_to_le32(tmp);
r = radeon_ib_get(rdev, ring->idx, &ib, NULL, 256);
if (r) {
DRM_ERROR("radeon: failed to get ib (%d).\n", r);
return r;
}
ib.ptr[0] = SDMA_PACKET(SDMA_OPCODE_WRITE, SDMA_WRITE_SUB_OPCODE_LINEAR, 0);
ib.ptr[1] = lower_32_bits(gpu_addr);
ib.ptr[2] = upper_32_bits(gpu_addr);
ib.ptr[3] = 1;
ib.ptr[4] = 0xDEADBEEF;
ib.length_dw = 5;
r = radeon_ib_schedule(rdev, &ib, NULL, false);
if (r) {
radeon_ib_free(rdev, &ib);
DRM_ERROR("radeon: failed to schedule ib (%d).\n", r);
return r;
}
r = radeon_fence_wait(ib.fence, false);
if (r) {
DRM_ERROR("radeon: fence wait failed (%d).\n", r);
return r;
}
for (i = 0; i < rdev->usec_timeout; i++) {
tmp = le32_to_cpu(rdev->wb.wb[index/4]);
if (tmp == 0xDEADBEEF)
break;
DRM_UDELAY(1);
}
if (i < rdev->usec_timeout) {
DRM_INFO("ib test on ring %d succeeded in %u usecs\n", ib.fence->ring, i);
} else {
DRM_ERROR("radeon: ib test failed (0x%08X)\n", tmp);
r = -EINVAL;
}
radeon_ib_free(rdev, &ib);
return r;
}
/**
* radeon_vce_get_create_msg - generate a VCE create msg
*
* @rdev: radeon_device pointer
* @ring: ring we should submit the msg to
* @handle: VCE session handle to use
* @fence: optional fence to return
*
* Open up a stream for HW test
*/
int radeon_vce_get_create_msg(struct radeon_device *rdev, int ring,
uint32_t handle, struct radeon_fence **fence)
{
const unsigned ib_size_dw = 1024;
struct radeon_ib ib;
uint64_t dummy;
int i, r;
r = radeon_ib_get(rdev, ring, &ib, NULL, ib_size_dw * 4);
if (r) {
DRM_ERROR("radeon: failed to get ib (%d).\n", r);
return r;
}
dummy = ib.gpu_addr + 1024;
/* stitch together an VCE create msg */
ib.length_dw = 0;
ib.ptr[ib.length_dw++] = 0x0000000c; /* len */
ib.ptr[ib.length_dw++] = 0x00000001; /* session cmd */
ib.ptr[ib.length_dw++] = handle;
ib.ptr[ib.length_dw++] = 0x00000030; /* len */
ib.ptr[ib.length_dw++] = 0x01000001; /* create cmd */
ib.ptr[ib.length_dw++] = 0x00000000;
ib.ptr[ib.length_dw++] = 0x00000042;
ib.ptr[ib.length_dw++] = 0x0000000a;
ib.ptr[ib.length_dw++] = 0x00000001;
ib.ptr[ib.length_dw++] = 0x00000080;
ib.ptr[ib.length_dw++] = 0x00000060;
ib.ptr[ib.length_dw++] = 0x00000100;
ib.ptr[ib.length_dw++] = 0x00000100;
ib.ptr[ib.length_dw++] = 0x0000000c;
ib.ptr[ib.length_dw++] = 0x00000000;
ib.ptr[ib.length_dw++] = 0x00000014; /* len */
ib.ptr[ib.length_dw++] = 0x05000005; /* feedback buffer */
ib.ptr[ib.length_dw++] = upper_32_bits(dummy);
ib.ptr[ib.length_dw++] = dummy;
ib.ptr[ib.length_dw++] = 0x00000001;
for (i = ib.length_dw; i < ib_size_dw; ++i)
ib.ptr[i] = 0x0;
r = radeon_ib_schedule(rdev, &ib, NULL);
if (r) {
DRM_ERROR("radeon: failed to schedule ib (%d).\n", r);
}
if (fence)
*fence = radeon_fence_ref(ib.fence);
radeon_ib_free(rdev, &ib);
return r;
}
/**
* cik_sdma_ib_test - test an IB on the DMA engine
*
* @rdev: radeon_device pointer
* @ring: radeon_ring structure holding ring information
*
* Test a simple IB in the DMA ring (CIK).
* Returns 0 on success, error on failure.
*/
int cik_sdma_ib_test(struct radeon_device *rdev, struct radeon_ring *ring)
{
struct radeon_ib ib;
unsigned i;
int r;
void __iomem *ptr = (void *)rdev->vram_scratch.ptr;
u32 tmp = 0;
if (!ptr) {
DRM_ERROR("invalid vram scratch pointer\n");
return -EINVAL;
}
tmp = 0xCAFEDEAD;
writel(tmp, ptr);
r = radeon_ib_get(rdev, ring->idx, &ib, NULL, 256);
if (r) {
DRM_ERROR("radeon: failed to get ib (%d).\n", r);
return r;
}
ib.ptr[0] = SDMA_PACKET(SDMA_OPCODE_WRITE, SDMA_WRITE_SUB_OPCODE_LINEAR, 0);
ib.ptr[1] = rdev->vram_scratch.gpu_addr & 0xfffffffc;
ib.ptr[2] = upper_32_bits(rdev->vram_scratch.gpu_addr);
ib.ptr[3] = 1;
ib.ptr[4] = 0xDEADBEEF;
ib.length_dw = 5;
r = radeon_ib_schedule(rdev, &ib, NULL);
if (r) {
radeon_ib_free(rdev, &ib);
DRM_ERROR("radeon: failed to schedule ib (%d).\n", r);
return r;
}
r = radeon_fence_wait(ib.fence, false);
if (r) {
DRM_ERROR("radeon: fence wait failed (%d).\n", r);
return r;
}
for (i = 0; i < rdev->usec_timeout; i++) {
tmp = readl(ptr);
if (tmp == 0xDEADBEEF)
break;
DRM_UDELAY(1);
}
if (i < rdev->usec_timeout) {
DRM_INFO("ib test on ring %d succeeded in %u usecs\n", ib.fence->ring, i);
} else {
DRM_ERROR("radeon: ib test failed (0x%08X)\n", tmp);
r = -EINVAL;
}
radeon_ib_free(rdev, &ib);
return r;
}
/**
* radeon_vce_get_destroy_msg - generate a VCE destroy msg
*
* @rdev: radeon_device pointer
* @ring: ring we should submit the msg to
* @handle: VCE session handle to use
* @fence: optional fence to return
*
* Close up a stream for HW test or if userspace failed to do so
*/
int radeon_vce_get_destroy_msg(struct radeon_device *rdev, int ring,
uint32_t handle, struct radeon_fence **fence)
{
const unsigned ib_size_dw = 1024;
struct radeon_ib ib;
uint64_t dummy;
int i, r;
r = radeon_ib_get(rdev, ring, &ib, NULL, ib_size_dw * 4);
if (r) {
DRM_ERROR("radeon: failed to get ib (%d).\n", r);
return r;
}
dummy = ib.gpu_addr + 1024;
/* stitch together an VCE destroy msg */
ib.length_dw = 0;
ib.ptr[ib.length_dw++] = cpu_to_le32(0x0000000c); /* len */
ib.ptr[ib.length_dw++] = cpu_to_le32(0x00000001); /* session cmd */
ib.ptr[ib.length_dw++] = cpu_to_le32(handle);
ib.ptr[ib.length_dw++] = cpu_to_le32(0x00000014); /* len */
ib.ptr[ib.length_dw++] = cpu_to_le32(0x05000005); /* feedback buffer */
ib.ptr[ib.length_dw++] = cpu_to_le32(upper_32_bits(dummy));
ib.ptr[ib.length_dw++] = cpu_to_le32(dummy);
ib.ptr[ib.length_dw++] = cpu_to_le32(0x00000001);
ib.ptr[ib.length_dw++] = cpu_to_le32(0x00000008); /* len */
ib.ptr[ib.length_dw++] = cpu_to_le32(0x02000001); /* destroy cmd */
for (i = ib.length_dw; i < ib_size_dw; ++i)
ib.ptr[i] = cpu_to_le32(0x0);
r = radeon_ib_schedule(rdev, &ib, NULL, false);
if (r) {
DRM_ERROR("radeon: failed to schedule ib (%d).\n", r);
}
if (fence)
*fence = radeon_fence_ref(ib.fence);
radeon_ib_free(rdev, &ib);
return r;
}
int radeon_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
{
struct radeon_device *rdev = dev->dev_private;
struct radeon_cs_parser parser;
struct radeon_cs_chunk *ib_chunk;
int r;
mutex_lock(&rdev->cs_mutex);
/* initialize parser */
memset(&parser, 0, sizeof(struct radeon_cs_parser));
parser.filp = filp;
parser.rdev = rdev;
parser.dev = rdev->dev;
parser.family = rdev->family;
r = radeon_cs_parser_init(&parser, data);
if (r) {
DRM_ERROR("Failed to initialize parser !\n");
radeon_cs_parser_fini(&parser, r);
mutex_unlock(&rdev->cs_mutex);
return r;
}
r = radeon_ib_get(rdev, &parser.ib);
if (r) {
DRM_ERROR("Failed to get ib !\n");
radeon_cs_parser_fini(&parser, r);
mutex_unlock(&rdev->cs_mutex);
return r;
}
r = radeon_cs_parser_relocs(&parser);
if (r) {
if (r != -ERESTARTSYS)
DRM_ERROR("Failed to parse relocation %d!\n", r);
radeon_cs_parser_fini(&parser, r);
mutex_unlock(&rdev->cs_mutex);
return r;
}
/* Copy the packet into the IB, the parser will read from the
* input memory (cached) and write to the IB (which can be
* uncached). */
ib_chunk = &parser.chunks[parser.chunk_ib_idx];
parser.ib->length_dw = ib_chunk->length_dw;
r = radeon_cs_parse(&parser);
if (r || parser.parser_error) {
DRM_ERROR("Invalid command stream !\n");
radeon_cs_parser_fini(&parser, r);
mutex_unlock(&rdev->cs_mutex);
return r;
}
r = radeon_cs_finish_pages(&parser);
if (r) {
DRM_ERROR("Invalid command stream !\n");
radeon_cs_parser_fini(&parser, r);
mutex_unlock(&rdev->cs_mutex);
return r;
}
r = radeon_ib_schedule(rdev, parser.ib);
if (r) {
DRM_ERROR("Failed to schedule IB !\n");
}
radeon_cs_parser_fini(&parser, r);
mutex_unlock(&rdev->cs_mutex);
return r;
}
int radeon_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
{
struct radeon_device *rdev = dev->dev_private;
struct radeon_cs_parser parser;
struct radeon_cs_chunk *ib_chunk;
int r;
mutex_lock(&rdev->cs_mutex);
if (rdev->gpu_lockup) {
mutex_unlock(&rdev->cs_mutex);
return -EINVAL;
}
memset(&parser, 0, sizeof(struct radeon_cs_parser));
parser.filp = filp;
parser.rdev = rdev;
r = radeon_cs_parser_init(&parser, data);
if (r) {
DRM_ERROR("Failed to initialize parser !\n");
radeon_cs_parser_fini(&parser, r);
mutex_unlock(&rdev->cs_mutex);
return r;
}
r = radeon_ib_get(rdev, &parser.ib);
if (r) {
DRM_ERROR("Failed to get ib !\n");
radeon_cs_parser_fini(&parser, r);
mutex_unlock(&rdev->cs_mutex);
return r;
}
r = radeon_cs_parser_relocs(&parser);
if (r) {
DRM_ERROR("Failed to parse relocation !\n");
radeon_cs_parser_fini(&parser, r);
mutex_unlock(&rdev->cs_mutex);
return r;
}
ib_chunk = &parser.chunks[parser.chunk_ib_idx];
parser.ib->length_dw = ib_chunk->length_dw;
r = radeon_cs_parse(&parser);
if (r || parser.parser_error) {
DRM_ERROR("Invalid command stream !\n");
radeon_cs_parser_fini(&parser, r);
mutex_unlock(&rdev->cs_mutex);
return r;
}
r = radeon_cs_finish_pages(&parser);
if (r) {
DRM_ERROR("Invalid command stream !\n");
radeon_cs_parser_fini(&parser, r);
mutex_unlock(&rdev->cs_mutex);
return r;
}
r = radeon_ib_schedule(rdev, parser.ib);
if (r) {
DRM_ERROR("Faild to schedule IB !\n");
}
radeon_cs_parser_fini(&parser, r);
mutex_unlock(&rdev->cs_mutex);
return r;
}
