AGPGART DRIVER
- T: git git://people.freedesktop.org/~airlied/linux (part of drm maint)
+ T: git git://anongit.freedesktop.org/drm/drm
S: Maintained
F: drivers/char/agp/
F: include/linux/agp*
F: drivers/crypto/ccp/
F: include/linux/ccp.h
+AMD DISPLAY CORE
+T: git git://people.freedesktop.org/~agd5f/linux
+S: Supported
+F: drivers/gpu/drm/amd/display/
+
AMD FAM15H PROCESSOR POWER MONITORING DRIVER
F: drivers/gpu/drm/amd/include/v9_structs.h
F: include/uapi/linux/kfd_ioctl.h
+AMD POWERPLAY
+S: Supported
+F: drivers/gpu/drm/amd/powerplay/
+T: git git://people.freedesktop.org/~agd5f/linux
+
AMD SEATTLE DEVICE TREE SUPPORT
F: drivers/scsi/esas2r
ATUSB IEEE 802.15.4 RADIO DRIVER
- M: Stefan Schmidt <stefan@osg.samsung.com>
+ M: Stefan Schmidt <stefan@datenfreihafen.org>
S: Maintained
F: drivers/net/ieee802154/atusb.c
DRIVER CORE, KOBJECTS, DEBUGFS AND SYSFS
T: git git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core.git
S: Supported
F: Documentation/kobject.txt
DRM DRIVERS
- T: git git://people.freedesktop.org/~airlied/linux
+ T: git git://anongit.freedesktop.org/drm/drm
B: https://bugs.freedesktop.org/
C: irc://chat.freenode.net/dri-devel
S: Maintained
DRM TTM SUBSYSTEM
T: git git://people.freedesktop.org/~agd5f/linux
S: Maintained
FREESCALE SOC FS_ENET DRIVER
S: Maintained
IEEE 802.15.4 SUBSYSTEM
- M: Stefan Schmidt <stefan@osg.samsung.com>
+ M: Stefan Schmidt <stefan@datenfreihafen.org>
W: http://wpan.cakelab.org/
T: git git://git.kernel.org/pub/scm/linux/kernel/git/sschmidt/wpan.git
F: include/uapi/linux/input-event-codes.h
F: include/linux/input/
F: Documentation/devicetree/bindings/input/
+ F: Documentation/devicetree/bindings/serio/
F: Documentation/input/
INPUT MULTITOUCH (MT) PROTOCOL
- M: Xinming Hu <huxm@marvell.com>
+ M: Xinming Hu <huxinming820@gmail.com>
S: Maintained
F: drivers/net/wireless/marvell/mwifiex/
F: drivers/usb/mtu3/
MEGACHIPS STDPXXXX-GE-B850V3-FW LVDS/DP++ BRIDGES
- M: Peter Senna Tschudin <peter.senna@collabora.com>
+ M: Peter Senna Tschudin <peter.senna@gmail.com>
S: Maintained
NXP TDA998X DRM DRIVER
- S: Supported
+ S: Maintained
T: git git://git.armlinux.org.uk/~rmk/linux-arm.git drm-tda998x-devel
T: git git://git.armlinux.org.uk/~rmk/linux-arm.git drm-tda998x-fixes
F: drivers/gpu/drm/i2c/tda998x_drv.c
F: include/drm/i2c/tda998x.h
+ F: include/dt-bindings/display/tda998x.h
+ K: "nxp,tda998x"
NXP TFA9879 DRIVER
F: arch/hexagon/
QUALCOMM HIDMA DRIVER
- M: Sinan Kaya <okaya@codeaurora.org>
+ M: Sinan Kaya <okaya@kernel.org>
bool switch_start;
bool switch_end;
bool disp_connectors_mapping;
- bool disp_detetion_ports;
+ bool disp_detection_ports;
};
struct amdgpu_atpx {
f->switch_start = mask & ATPX_GRAPHICS_DEVICE_SWITCH_START_NOTIFICATION_SUPPORTED;
f->switch_end = mask & ATPX_GRAPHICS_DEVICE_SWITCH_END_NOTIFICATION_SUPPORTED;
f->disp_connectors_mapping = mask & ATPX_GET_DISPLAY_CONNECTORS_MAPPING_SUPPORTED;
- f->disp_detetion_ports = mask & ATPX_GET_DISPLAY_DETECTION_PORTS_SUPPORTED;
+ f->disp_detection_ports = mask & ATPX_GET_DISPLAY_DETECTION_PORTS_SUPPORTED;
}
/**
{ 0x1002, 0x6900, 0x1002, 0x0124, AMDGPU_PX_QUIRK_FORCE_ATPX },
{ 0x1002, 0x6900, 0x1028, 0x0812, AMDGPU_PX_QUIRK_FORCE_ATPX },
{ 0x1002, 0x6900, 0x1028, 0x0813, AMDGPU_PX_QUIRK_FORCE_ATPX },
+ { 0x1002, 0x6900, 0x1025, 0x125A, AMDGPU_PX_QUIRK_FORCE_ATPX },
{ 0, 0, 0, 0, 0 },
};
* Alex Deucher
* Jerome Glisse
*/
+#include <linux/power_supply.h>
#include <linux/kthread.h>
#include <linux/console.h>
#include <linux/slab.h>
}
/**
- * amdgpu_device_gart_location - try to find GTT location
+ * amdgpu_device_gart_location - try to find GART location
*
* @adev: amdgpu device structure holding all necessary informations
* @mc: memory controller structure holding memory informations
*
- * Function will place try to place GTT before or after VRAM.
+ * Function will place try to place GART before or after VRAM.
*
- * If GTT size is bigger than space left then we ajust GTT size.
+ * If GART size is bigger than space left then we ajust GART size.
* Thus function will never fails.
- *
- * FIXME: when reducing GTT size align new size on power of 2.
*/
void amdgpu_device_gart_location(struct amdgpu_device *adev,
struct amdgpu_gmc *mc)
size_bf = mc->vram_start;
if (size_bf > size_af) {
if (mc->gart_size > size_bf) {
- dev_warn(adev->dev, "limiting GTT\n");
+ dev_warn(adev->dev, "limiting GART\n");
mc->gart_size = size_bf;
}
mc->gart_start = 0;
} else {
if (mc->gart_size > size_af) {
- dev_warn(adev->dev, "limiting GTT\n");
+ dev_warn(adev->dev, "limiting GART\n");
mc->gart_size = size_af;
}
/* VCE doesn't like it when BOs cross a 4GB segment, so align
mc->gart_start = ALIGN(mc->vram_end + 1, 0x100000000ULL);
}
mc->gart_end = mc->gart_start + mc->gart_size - 1;
- dev_info(adev->dev, "GTT: %lluM 0x%016llX - 0x%016llX\n",
+ dev_info(adev->dev, "GART: %lluM 0x%016llX - 0x%016llX\n",
mc->gart_size >> 20, mc->gart_start, mc->gart_end);
}
/**
* amdgpu_device_ip_set_clockgating_state - set the CG state
*
- * @adev: amdgpu_device pointer
+ * @dev: amdgpu_device pointer
* @block_type: Type of hardware IP (SMU, GFX, UVD, etc.)
* @state: clockgating state (gate or ungate)
*
/**
* amdgpu_device_ip_set_powergating_state - set the PG state
*
- * @adev: amdgpu_device pointer
+ * @dev: amdgpu_device pointer
* @block_type: Type of hardware IP (SMU, GFX, UVD, etc.)
* @state: powergating state (gate or ungate)
*
* amdgpu_device_ip_get_ip_block - get a hw IP pointer
*
* @adev: amdgpu_device pointer
- * @block_type: Type of hardware IP (SMU, GFX, UVD, etc.)
+ * @type: Type of hardware IP (SMU, GFX, UVD, etc.)
*
* Returns a pointer to the hardware IP block structure
* if it exists for the asic, otherwise NULL.
if (amdgpu_emu_mode == 1)
return 0;
- r = amdgpu_ib_ring_tests(adev);
- if (r)
- DRM_ERROR("ib ring test failed (%d).\n", r);
-
for (i = 0; i < adev->num_ip_blocks; i++) {
if (!adev->ip_blocks[i].status.valid)
continue;
}
}
- if (adev->powerplay.pp_feature & PP_GFXOFF_MASK) {
- /* enable gfx powergating */
- amdgpu_device_ip_set_powergating_state(adev,
- AMD_IP_BLOCK_TYPE_GFX,
- AMD_PG_STATE_GATE);
- /* enable gfxoff */
- amdgpu_device_ip_set_powergating_state(adev,
- AMD_IP_BLOCK_TYPE_SMC,
- AMD_PG_STATE_GATE);
- }
+ return 0;
+}
+static int amdgpu_device_ip_late_set_pg_state(struct amdgpu_device *adev)
+{
+ int i = 0, r;
+
+ if (amdgpu_emu_mode == 1)
+ return 0;
+
+ for (i = 0; i < adev->num_ip_blocks; i++) {
+ if (!adev->ip_blocks[i].status.valid)
+ continue;
+ /* skip CG for VCE/UVD, it's handled specially */
+ if (adev->ip_blocks[i].version->type != AMD_IP_BLOCK_TYPE_UVD &&
+ adev->ip_blocks[i].version->type != AMD_IP_BLOCK_TYPE_VCE &&
+ adev->ip_blocks[i].version->type != AMD_IP_BLOCK_TYPE_VCN &&
+ adev->ip_blocks[i].version->funcs->set_powergating_state) {
+ /* enable powergating to save power */
+ r = adev->ip_blocks[i].version->funcs->set_powergating_state((void *)adev,
+ AMD_PG_STATE_GATE);
+ if (r) {
+ DRM_ERROR("set_powergating_state(gate) of IP block <%s> failed %d\n",
+ adev->ip_blocks[i].version->funcs->name, r);
+ return r;
+ }
+ }
+ }
return 0;
}
}
}
+ amdgpu_device_ip_late_set_cg_state(adev);
+ amdgpu_device_ip_late_set_pg_state(adev);
+
queue_delayed_work(system_wq, &adev->late_init_work,
msecs_to_jiffies(AMDGPU_RESUME_MS));
adev->ip_blocks[i].version->funcs->name, r);
return r;
}
+ if (adev->powerplay.pp_funcs->set_powergating_by_smu)
+ amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_GFX, false);
r = adev->ip_blocks[i].version->funcs->hw_fini((void *)adev);
/* XXX handle errors */
if (r) {
{
struct amdgpu_device *adev =
container_of(work, struct amdgpu_device, late_init_work.work);
- amdgpu_device_ip_late_set_cg_state(adev);
+ int r;
+
+ r = amdgpu_ib_ring_tests(adev);
+ if (r)
+ DRM_ERROR("ib ring test failed (%d).\n", r);
}
/**
- * amdgpu_device_ip_suspend - run suspend for hardware IPs
+ * amdgpu_device_ip_suspend_phase1 - run suspend for hardware IPs (phase 1)
*
* @adev: amdgpu_device pointer
*
* in each IP into a state suitable for suspend.
* Returns 0 on success, negative error code on failure.
*/
-int amdgpu_device_ip_suspend(struct amdgpu_device *adev)
+static int amdgpu_device_ip_suspend_phase1(struct amdgpu_device *adev)
{
int i, r;
if (amdgpu_sriov_vf(adev))
amdgpu_virt_request_full_gpu(adev, false);
- /* ungate SMC block powergating */
- if (adev->powerplay.pp_feature & PP_GFXOFF_MASK)
- amdgpu_device_ip_set_powergating_state(adev,
- AMD_IP_BLOCK_TYPE_SMC,
- AMD_CG_STATE_UNGATE);
+ for (i = adev->num_ip_blocks - 1; i >= 0; i--) {
+ if (!adev->ip_blocks[i].status.valid)
+ continue;
+ /* displays are handled separately */
+ if (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_DCE) {
+ /* ungate blocks so that suspend can properly shut them down */
+ if (adev->ip_blocks[i].version->funcs->set_clockgating_state) {
+ r = adev->ip_blocks[i].version->funcs->set_clockgating_state((void *)adev,
+ AMD_CG_STATE_UNGATE);
+ if (r) {
+ DRM_ERROR("set_clockgating_state(ungate) of IP block <%s> failed %d\n",
+ adev->ip_blocks[i].version->funcs->name, r);
+ }
+ }
+ /* XXX handle errors */
+ r = adev->ip_blocks[i].version->funcs->suspend(adev);
+ /* XXX handle errors */
+ if (r) {
+ DRM_ERROR("suspend of IP block <%s> failed %d\n",
+ adev->ip_blocks[i].version->funcs->name, r);
+ }
+ }
+ }
+
+ if (amdgpu_sriov_vf(adev))
+ amdgpu_virt_release_full_gpu(adev, false);
+
+ return 0;
+}
+
+/**
+ * amdgpu_device_ip_suspend_phase2 - run suspend for hardware IPs (phase 2)
+ *
+ * @adev: amdgpu_device pointer
+ *
+ * Main suspend function for hardware IPs. The list of all the hardware
+ * IPs that make up the asic is walked, clockgating is disabled and the
+ * suspend callbacks are run. suspend puts the hardware and software state
+ * in each IP into a state suitable for suspend.
+ * Returns 0 on success, negative error code on failure.
+ */
+static int amdgpu_device_ip_suspend_phase2(struct amdgpu_device *adev)
+{
+ int i, r;
+
+ if (amdgpu_sriov_vf(adev))
+ amdgpu_virt_request_full_gpu(adev, false);
/* ungate SMC block first */
r = amdgpu_device_ip_set_clockgating_state(adev, AMD_IP_BLOCK_TYPE_SMC,
DRM_ERROR("set_clockgating_state(ungate) SMC failed %d\n", r);
}
+ /* call smu to disable gfx off feature first when suspend */
+ if (adev->powerplay.pp_funcs->set_powergating_by_smu)
+ amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_GFX, false);
+
for (i = adev->num_ip_blocks - 1; i >= 0; i--) {
if (!adev->ip_blocks[i].status.valid)
continue;
+ /* displays are handled in phase1 */
+ if (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_DCE)
+ continue;
/* ungate blocks so that suspend can properly shut them down */
if (adev->ip_blocks[i].version->type != AMD_IP_BLOCK_TYPE_SMC &&
adev->ip_blocks[i].version->funcs->set_clockgating_state) {
return 0;
}
+/**
+ * amdgpu_device_ip_suspend - run suspend for hardware IPs
+ *
+ * @adev: amdgpu_device pointer
+ *
+ * Main suspend function for hardware IPs. The list of all the hardware
+ * IPs that make up the asic is walked, clockgating is disabled and the
+ * suspend callbacks are run. suspend puts the hardware and software state
+ * in each IP into a state suitable for suspend.
+ * Returns 0 on success, negative error code on failure.
+ */
+int amdgpu_device_ip_suspend(struct amdgpu_device *adev)
+{
+ int r;
+
+ r = amdgpu_device_ip_suspend_phase1(adev);
+ if (r)
+ return r;
+ r = amdgpu_device_ip_suspend_phase2(adev);
+
+ return r;
+}
+
static int amdgpu_device_ip_reinit_early_sriov(struct amdgpu_device *adev)
{
int i, r;
continue;
r = block->version->funcs->hw_init(adev);
- DRM_INFO("RE-INIT: %s %s\n", block->version->funcs->name, r?"failed":"successed");
+ DRM_INFO("RE-INIT: %s %s\n", block->version->funcs->name, r?"failed":"succeeded");
if (r)
return r;
}
continue;
r = block->version->funcs->hw_init(adev);
- DRM_INFO("RE-INIT: %s %s\n", block->version->funcs->name, r?"failed":"successed");
+ DRM_INFO("RE-INIT: %s %s\n", block->version->funcs->name, r?"failed":"succeeded");
if (r)
return r;
}
case CHIP_VEGA10:
case CHIP_VEGA12:
case CHIP_VEGA20:
-#if defined(CONFIG_DRM_AMD_DC_DCN1_0)
+#ifdef CONFIG_X86
case CHIP_RAVEN:
#endif
return amdgpu_dc != 0;
* amdgpu_device_init - initialize the driver
*
* @adev: amdgpu_device pointer
- * @pdev: drm dev pointer
+ * @ddev: drm dev pointer
* @pdev: pci dev pointer
* @flags: driver flags
*
INIT_DELAYED_WORK(&adev->late_init_work,
amdgpu_device_ip_late_init_func_handler);
+ adev->pm.ac_power = power_supply_is_system_supplied() > 0 ? true : false;
+
/* Registers mapping */
/* TODO: block userspace mapping of io register */
if (adev->asic_type >= CHIP_BONAIRE) {
/**
* amdgpu_device_suspend - initiate device suspend
*
- * @pdev: drm dev pointer
- * @state: suspend state
+ * @dev: drm dev pointer
+ * @suspend: suspend state
+ * @fbcon : notify the fbdev of suspend
*
* Puts the hw in the suspend state (all asics).
* Returns 0 for success or an error on failure.
drm_kms_helper_poll_disable(dev);
+ if (fbcon)
+ amdgpu_fbdev_set_suspend(adev, 1);
+
if (!amdgpu_device_has_dc_support(adev)) {
/* turn off display hw */
drm_modeset_lock_all(dev);
drm_helper_connector_dpms(connector, DRM_MODE_DPMS_OFF);
}
drm_modeset_unlock_all(dev);
- }
-
- amdgpu_amdkfd_suspend(adev);
-
- /* unpin the front buffers and cursors */
- list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
- struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
- struct drm_framebuffer *fb = crtc->primary->fb;
- struct amdgpu_bo *robj;
-
- if (amdgpu_crtc->cursor_bo) {
- struct amdgpu_bo *aobj = gem_to_amdgpu_bo(amdgpu_crtc->cursor_bo);
- r = amdgpu_bo_reserve(aobj, true);
- if (r == 0) {
- amdgpu_bo_unpin(aobj);
- amdgpu_bo_unreserve(aobj);
+ /* unpin the front buffers and cursors */
+ list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
+ struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
+ struct drm_framebuffer *fb = crtc->primary->fb;
+ struct amdgpu_bo *robj;
+
+ if (amdgpu_crtc->cursor_bo) {
+ struct amdgpu_bo *aobj = gem_to_amdgpu_bo(amdgpu_crtc->cursor_bo);
+ r = amdgpu_bo_reserve(aobj, true);
+ if (r == 0) {
+ amdgpu_bo_unpin(aobj);
+ amdgpu_bo_unreserve(aobj);
+ }
}
- }
- if (fb == NULL || fb->obj[0] == NULL) {
- continue;
- }
- robj = gem_to_amdgpu_bo(fb->obj[0]);
- /* don't unpin kernel fb objects */
- if (!amdgpu_fbdev_robj_is_fb(adev, robj)) {
- r = amdgpu_bo_reserve(robj, true);
- if (r == 0) {
- amdgpu_bo_unpin(robj);
- amdgpu_bo_unreserve(robj);
+ if (fb == NULL || fb->obj[0] == NULL) {
+ continue;
+ }
+ robj = gem_to_amdgpu_bo(fb->obj[0]);
+ /* don't unpin kernel fb objects */
+ if (!amdgpu_fbdev_robj_is_fb(adev, robj)) {
+ r = amdgpu_bo_reserve(robj, true);
+ if (r == 0) {
+ amdgpu_bo_unpin(robj);
+ amdgpu_bo_unreserve(robj);
+ }
}
}
}
+
+ amdgpu_amdkfd_suspend(adev);
+
+ r = amdgpu_device_ip_suspend_phase1(adev);
+
/* evict vram memory */
amdgpu_bo_evict_vram(adev);
amdgpu_fence_driver_suspend(adev);
- r = amdgpu_device_ip_suspend(adev);
+ r = amdgpu_device_ip_suspend_phase2(adev);
/* evict remaining vram memory
* This second call to evict vram is to evict the gart page table
DRM_ERROR("amdgpu asic reset failed\n");
}
- if (fbcon) {
- console_lock();
- amdgpu_fbdev_set_suspend(adev, 1);
- console_unlock();
- }
return 0;
}
/**
* amdgpu_device_resume - initiate device resume
*
- * @pdev: drm dev pointer
+ * @dev: drm dev pointer
+ * @resume: resume state
+ * @fbcon : notify the fbdev of resume
*
* Bring the hw back to operating state (all asics).
* Returns 0 for success or an error on failure.
if (dev->switch_power_state == DRM_SWITCH_POWER_OFF)
return 0;
- if (fbcon)
- console_lock();
-
if (resume) {
pci_set_power_state(dev->pdev, PCI_D0);
pci_restore_state(dev->pdev);
r = pci_enable_device(dev->pdev);
if (r)
- goto unlock;
+ return r;
}
/* post card */
r = amdgpu_device_ip_resume(adev);
if (r) {
DRM_ERROR("amdgpu_device_ip_resume failed (%d).\n", r);
- goto unlock;
+ return r;
}
amdgpu_fence_driver_resume(adev);
r = amdgpu_device_ip_late_init(adev);
if (r)
- goto unlock;
-
- /* pin cursors */
- list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
- struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
-
- if (amdgpu_crtc->cursor_bo) {
- struct amdgpu_bo *aobj = gem_to_amdgpu_bo(amdgpu_crtc->cursor_bo);
- r = amdgpu_bo_reserve(aobj, true);
- if (r == 0) {
- r = amdgpu_bo_pin(aobj,
- AMDGPU_GEM_DOMAIN_VRAM,
- &amdgpu_crtc->cursor_addr);
- if (r != 0)
- DRM_ERROR("Failed to pin cursor BO (%d)\n", r);
- amdgpu_bo_unreserve(aobj);
+ return r;
+
+ if (!amdgpu_device_has_dc_support(adev)) {
+ /* pin cursors */
+ list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
+ struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
+
+ if (amdgpu_crtc->cursor_bo) {
+ struct amdgpu_bo *aobj = gem_to_amdgpu_bo(amdgpu_crtc->cursor_bo);
+ r = amdgpu_bo_reserve(aobj, true);
+ if (r == 0) {
+ r = amdgpu_bo_pin(aobj, AMDGPU_GEM_DOMAIN_VRAM);
+ if (r != 0)
+ DRM_ERROR("Failed to pin cursor BO (%d)\n", r);
+ amdgpu_crtc->cursor_addr = amdgpu_bo_gpu_offset(aobj);
+ amdgpu_bo_unreserve(aobj);
+ }
}
}
}
if (r)
return r;
+ /* Make sure IB tests flushed */
+ flush_delayed_work(&adev->late_init_work);
+
/* blat the mode back in */
if (fbcon) {
if (!amdgpu_device_has_dc_support(adev)) {
}
drm_modeset_unlock_all(dev);
}
+ amdgpu_fbdev_set_suspend(adev, 0);
}
drm_kms_helper_poll_enable(dev);
#ifdef CONFIG_PM
dev->dev->power.disable_depth--;
#endif
-
- if (fbcon)
- amdgpu_fbdev_set_suspend(adev, 0);
-
-unlock:
- if (fbcon)
- console_unlock();
-
- return r;
+ return 0;
}
/**
* @adev: amdgpu device pointer
*
* attempt to do soft-reset or full-reset and reinitialize Asic
- * return 0 means successed otherwise failed
+ * return 0 means succeeded otherwise failed
*/
static int amdgpu_device_reset(struct amdgpu_device *adev)
{
* amdgpu_device_reset_sriov - reset ASIC for SR-IOV vf
*
* @adev: amdgpu device pointer
+ * @from_hypervisor: request from hypervisor
*
* do VF FLR and reinitialize Asic
- * return 0 means successed otherwise failed
+ * return 0 means succeeded otherwise failed
*/
static int amdgpu_device_reset_sriov(struct amdgpu_device *adev,
bool from_hypervisor)
*
* @adev: amdgpu device pointer
* @job: which job trigger hang
- * @force forces reset regardless of amdgpu_gpu_recovery
+ * @force: forces reset regardless of amdgpu_gpu_recovery
*
* Attempt to reset the GPU if it has hung (all asics).
* Returns 0 for success or an error on failure.
atomic_inc(&adev->gpu_reset_counter);
adev->in_gpu_reset = 1;
+ /* Block kfd */
+ amdgpu_amdkfd_pre_reset(adev);
+
/* block TTM */
resched = ttm_bo_lock_delayed_workqueue(&adev->mman.bdev);
kthread_park(ring->sched.thread);
- if (job && job->ring->idx != i)
+ if (job && job->base.sched == &ring->sched)
continue;
- drm_sched_hw_job_reset(&ring->sched, &job->base);
+ drm_sched_hw_job_reset(&ring->sched, job ? &job->base : NULL);
/* after all hw jobs are reset, hw fence is meaningless, so force_completion */
amdgpu_fence_driver_force_completion(ring);
* or all rings (in the case @job is NULL)
* after above amdgpu_reset accomplished
*/
- if ((!job || job->ring->idx == i) && !r)
+ if ((!job || job->base.sched == &ring->sched) && !r)
drm_sched_job_recovery(&ring->sched);
kthread_unpark(ring->sched.thread);
dev_info(adev->dev, "GPU reset(%d) failed\n", atomic_read(&adev->gpu_reset_counter));
amdgpu_vf_error_put(adev, AMDGIM_ERROR_VF_GPU_RESET_FAIL, 0, r);
} else {
- dev_info(adev->dev, "GPU reset(%d) successed!\n",atomic_read(&adev->gpu_reset_counter));
+ dev_info(adev->dev, "GPU reset(%d) succeeded!\n",atomic_read(&adev->gpu_reset_counter));
}
+ /*unlock kfd */
+ amdgpu_amdkfd_post_reset(adev);
amdgpu_vf_error_trans_all(adev);
adev->in_gpu_reset = 0;
mutex_unlock(&adev->lock_reset);
*/
static void amdgpu_device_get_pcie_info(struct amdgpu_device *adev)
{
- u32 mask;
- int ret;
+ struct pci_dev *pdev;
+ enum pci_bus_speed speed_cap;
+ enum pcie_link_width link_width;
if (amdgpu_pcie_gen_cap)
adev->pm.pcie_gen_mask = amdgpu_pcie_gen_cap;
}
if (adev->pm.pcie_gen_mask == 0) {
- ret = drm_pcie_get_speed_cap_mask(adev->ddev, &mask);
- if (!ret) {
- adev->pm.pcie_gen_mask = (CAIL_ASIC_PCIE_LINK_SPEED_SUPPORT_GEN1 |
+ /* asic caps */
+ pdev = adev->pdev;
+ speed_cap = pcie_get_speed_cap(pdev);
+ if (speed_cap == PCI_SPEED_UNKNOWN) {
+ adev->pm.pcie_gen_mask |= (CAIL_ASIC_PCIE_LINK_SPEED_SUPPORT_GEN1 |
CAIL_ASIC_PCIE_LINK_SPEED_SUPPORT_GEN2 |
CAIL_ASIC_PCIE_LINK_SPEED_SUPPORT_GEN3);
-
- if (mask & DRM_PCIE_SPEED_25)
- adev->pm.pcie_gen_mask |= CAIL_PCIE_LINK_SPEED_SUPPORT_GEN1;
- if (mask & DRM_PCIE_SPEED_50)
- adev->pm.pcie_gen_mask |= CAIL_PCIE_LINK_SPEED_SUPPORT_GEN2;
- if (mask & DRM_PCIE_SPEED_80)
- adev->pm.pcie_gen_mask |= CAIL_PCIE_LINK_SPEED_SUPPORT_GEN3;
} else {
- adev->pm.pcie_gen_mask = AMDGPU_DEFAULT_PCIE_GEN_MASK;
+ if (speed_cap == PCIE_SPEED_16_0GT)
+ adev->pm.pcie_gen_mask |= (CAIL_ASIC_PCIE_LINK_SPEED_SUPPORT_GEN1 |
+ CAIL_ASIC_PCIE_LINK_SPEED_SUPPORT_GEN2 |
+ CAIL_ASIC_PCIE_LINK_SPEED_SUPPORT_GEN3 |
+ CAIL_ASIC_PCIE_LINK_SPEED_SUPPORT_GEN4);
+ else if (speed_cap == PCIE_SPEED_8_0GT)
+ adev->pm.pcie_gen_mask |= (CAIL_ASIC_PCIE_LINK_SPEED_SUPPORT_GEN1 |
+ CAIL_ASIC_PCIE_LINK_SPEED_SUPPORT_GEN2 |
+ CAIL_ASIC_PCIE_LINK_SPEED_SUPPORT_GEN3);
+ else if (speed_cap == PCIE_SPEED_5_0GT)
+ adev->pm.pcie_gen_mask |= (CAIL_ASIC_PCIE_LINK_SPEED_SUPPORT_GEN1 |
+ CAIL_ASIC_PCIE_LINK_SPEED_SUPPORT_GEN2);
+ else
+ adev->pm.pcie_gen_mask |= CAIL_ASIC_PCIE_LINK_SPEED_SUPPORT_GEN1;
+ }
+ /* platform caps */
+ pdev = adev->ddev->pdev->bus->self;
+ speed_cap = pcie_get_speed_cap(pdev);
+ if (speed_cap == PCI_SPEED_UNKNOWN) {
+ adev->pm.pcie_gen_mask |= (CAIL_PCIE_LINK_SPEED_SUPPORT_GEN1 |
+ CAIL_PCIE_LINK_SPEED_SUPPORT_GEN2);
+ } else {
+ if (speed_cap == PCIE_SPEED_16_0GT)
+ adev->pm.pcie_gen_mask |= (CAIL_PCIE_LINK_SPEED_SUPPORT_GEN1 |
+ CAIL_PCIE_LINK_SPEED_SUPPORT_GEN2 |
+ CAIL_PCIE_LINK_SPEED_SUPPORT_GEN3 |
+ CAIL_PCIE_LINK_SPEED_SUPPORT_GEN4);
+ else if (speed_cap == PCIE_SPEED_8_0GT)
+ adev->pm.pcie_gen_mask |= (CAIL_PCIE_LINK_SPEED_SUPPORT_GEN1 |
+ CAIL_PCIE_LINK_SPEED_SUPPORT_GEN2 |
+ CAIL_PCIE_LINK_SPEED_SUPPORT_GEN3);
+ else if (speed_cap == PCIE_SPEED_5_0GT)
+ adev->pm.pcie_gen_mask |= (CAIL_PCIE_LINK_SPEED_SUPPORT_GEN1 |
+ CAIL_PCIE_LINK_SPEED_SUPPORT_GEN2);
+ else
+ adev->pm.pcie_gen_mask |= CAIL_PCIE_LINK_SPEED_SUPPORT_GEN1;
+
}
}
if (adev->pm.pcie_mlw_mask == 0) {
- ret = drm_pcie_get_max_link_width(adev->ddev, &mask);
- if (!ret) {
- switch (mask) {
- case 32:
+ pdev = adev->ddev->pdev->bus->self;
+ link_width = pcie_get_width_cap(pdev);
+ if (link_width == PCIE_LNK_WIDTH_UNKNOWN) {
+ adev->pm.pcie_mlw_mask |= AMDGPU_DEFAULT_PCIE_MLW_MASK;
+ } else {
+ switch (link_width) {
+ case PCIE_LNK_X32:
adev->pm.pcie_mlw_mask = (CAIL_PCIE_LINK_WIDTH_SUPPORT_X32 |
CAIL_PCIE_LINK_WIDTH_SUPPORT_X16 |
CAIL_PCIE_LINK_WIDTH_SUPPORT_X12 |
CAIL_PCIE_LINK_WIDTH_SUPPORT_X2 |
CAIL_PCIE_LINK_WIDTH_SUPPORT_X1);
break;
- case 16:
+ case PCIE_LNK_X16:
adev->pm.pcie_mlw_mask = (CAIL_PCIE_LINK_WIDTH_SUPPORT_X16 |
CAIL_PCIE_LINK_WIDTH_SUPPORT_X12 |
CAIL_PCIE_LINK_WIDTH_SUPPORT_X8 |
CAIL_PCIE_LINK_WIDTH_SUPPORT_X2 |
CAIL_PCIE_LINK_WIDTH_SUPPORT_X1);
break;
- case 12:
+ case PCIE_LNK_X12:
adev->pm.pcie_mlw_mask = (CAIL_PCIE_LINK_WIDTH_SUPPORT_X12 |
CAIL_PCIE_LINK_WIDTH_SUPPORT_X8 |
CAIL_PCIE_LINK_WIDTH_SUPPORT_X4 |
CAIL_PCIE_LINK_WIDTH_SUPPORT_X2 |
CAIL_PCIE_LINK_WIDTH_SUPPORT_X1);
break;
- case 8:
+ case PCIE_LNK_X8:
adev->pm.pcie_mlw_mask = (CAIL_PCIE_LINK_WIDTH_SUPPORT_X8 |
CAIL_PCIE_LINK_WIDTH_SUPPORT_X4 |
CAIL_PCIE_LINK_WIDTH_SUPPORT_X2 |
CAIL_PCIE_LINK_WIDTH_SUPPORT_X1);
break;
- case 4:
+ case PCIE_LNK_X4:
adev->pm.pcie_mlw_mask = (CAIL_PCIE_LINK_WIDTH_SUPPORT_X4 |
CAIL_PCIE_LINK_WIDTH_SUPPORT_X2 |
CAIL_PCIE_LINK_WIDTH_SUPPORT_X1);
break;
- case 2:
+ case PCIE_LNK_X2:
adev->pm.pcie_mlw_mask = (CAIL_PCIE_LINK_WIDTH_SUPPORT_X2 |
CAIL_PCIE_LINK_WIDTH_SUPPORT_X1);
break;
- case 1:
+ case PCIE_LNK_X1:
adev->pm.pcie_mlw_mask = CAIL_PCIE_LINK_WIDTH_SUPPORT_X1;
break;
default:
break;
}
- } else {
- adev->pm.pcie_mlw_mask = AMDGPU_DEFAULT_PCIE_MLW_MASK;
}
}
}
#include "amdgpu_dm_irq.h"
#include "amdgpu_pm.h"
-unsigned long long dm_get_timestamp(struct dc_context *ctx)
-{
- struct timespec64 time;
- getrawmonotonic64(&time);
- return timespec64_to_ns(&time);
-}
unsigned long long dm_get_elapse_time_in_ns(struct dc_context *ctx,
unsigned long long current_time_stamp,
/**** power component interfaces ****/
-bool dm_pp_apply_display_requirements(
- const struct dc_context *ctx,
- const struct dm_pp_display_configuration *pp_display_cfg)
-{
- struct amdgpu_device *adev = ctx->driver_context;
-
- if (adev->pm.dpm_enabled) {
-
- memset(&adev->pm.pm_display_cfg, 0,
- sizeof(adev->pm.pm_display_cfg));
-
- adev->pm.pm_display_cfg.cpu_cc6_disable =
- pp_display_cfg->cpu_cc6_disable;
-
- adev->pm.pm_display_cfg.cpu_pstate_disable =
- pp_display_cfg->cpu_pstate_disable;
-
- adev->pm.pm_display_cfg.cpu_pstate_separation_time =
- pp_display_cfg->cpu_pstate_separation_time;
-
- adev->pm.pm_display_cfg.nb_pstate_switch_disable =
- pp_display_cfg->nb_pstate_switch_disable;
-
- adev->pm.pm_display_cfg.num_display =
- pp_display_cfg->display_count;
- adev->pm.pm_display_cfg.num_path_including_non_display =
- pp_display_cfg->display_count;
-
- adev->pm.pm_display_cfg.min_core_set_clock =
- pp_display_cfg->min_engine_clock_khz/10;
- adev->pm.pm_display_cfg.min_core_set_clock_in_sr =
- pp_display_cfg->min_engine_clock_deep_sleep_khz/10;
- adev->pm.pm_display_cfg.min_mem_set_clock =
- pp_display_cfg->min_memory_clock_khz/10;
-
- adev->pm.pm_display_cfg.multi_monitor_in_sync =
- pp_display_cfg->all_displays_in_sync;
- adev->pm.pm_display_cfg.min_vblank_time =
- pp_display_cfg->avail_mclk_switch_time_us;
-
- adev->pm.pm_display_cfg.display_clk =
- pp_display_cfg->disp_clk_khz/10;
-
- adev->pm.pm_display_cfg.dce_tolerable_mclk_in_active_latency =
- pp_display_cfg->avail_mclk_switch_time_in_disp_active_us;
-
- adev->pm.pm_display_cfg.crtc_index = pp_display_cfg->crtc_index;
- adev->pm.pm_display_cfg.line_time_in_us =
- pp_display_cfg->line_time_in_us;
-
- adev->pm.pm_display_cfg.vrefresh = pp_display_cfg->disp_configs[0].v_refresh;
- adev->pm.pm_display_cfg.crossfire_display_index = -1;
- adev->pm.pm_display_cfg.min_bus_bandwidth = 0;
-
- /* TODO: complete implementation of
- * pp_display_configuration_change().
- * Follow example of:
- * PHM_StoreDALConfigurationData - powerplay\hwmgr\hardwaremanager.c
- * PP_IRI_DisplayConfigurationChange - powerplay\eventmgr\iri.c */
- if (adev->powerplay.pp_funcs->display_configuration_change)
- adev->powerplay.pp_funcs->display_configuration_change(
- adev->powerplay.pp_handle,
- &adev->pm.pm_display_cfg);
-
- /* TODO: replace by a separate call to 'apply display cfg'? */
- amdgpu_pm_compute_clocks(adev);
- }
-
- return true;
-}
-
-static void get_default_clock_levels(
- enum dm_pp_clock_type clk_type,
- struct dm_pp_clock_levels *clks)
-{
- uint32_t disp_clks_in_khz[6] = {
- 300000, 400000, 496560, 626090, 685720, 757900 };
- uint32_t sclks_in_khz[6] = {
- 300000, 360000, 423530, 514290, 626090, 720000 };
- uint32_t mclks_in_khz[2] = { 333000, 800000 };
-
- switch (clk_type) {
- case DM_PP_CLOCK_TYPE_DISPLAY_CLK:
- clks->num_levels = 6;
- memmove(clks->clocks_in_khz, disp_clks_in_khz,
- sizeof(disp_clks_in_khz));
- break;
- case DM_PP_CLOCK_TYPE_ENGINE_CLK:
- clks->num_levels = 6;
- memmove(clks->clocks_in_khz, sclks_in_khz,
- sizeof(sclks_in_khz));
- break;
- case DM_PP_CLOCK_TYPE_MEMORY_CLK:
- clks->num_levels = 2;
- memmove(clks->clocks_in_khz, mclks_in_khz,
- sizeof(mclks_in_khz));
- break;
- default:
- clks->num_levels = 0;
- break;
- }
-}
-
-static enum amd_pp_clock_type dc_to_pp_clock_type(
- enum dm_pp_clock_type dm_pp_clk_type)
-{
- enum amd_pp_clock_type amd_pp_clk_type = 0;
-
- switch (dm_pp_clk_type) {
- case DM_PP_CLOCK_TYPE_DISPLAY_CLK:
- amd_pp_clk_type = amd_pp_disp_clock;
- break;
- case DM_PP_CLOCK_TYPE_ENGINE_CLK:
- amd_pp_clk_type = amd_pp_sys_clock;
- break;
- case DM_PP_CLOCK_TYPE_MEMORY_CLK:
- amd_pp_clk_type = amd_pp_mem_clock;
- break;
- default:
- DRM_ERROR("DM_PPLIB: invalid clock type: %d!\n",
- dm_pp_clk_type);
- break;
- }
-
- return amd_pp_clk_type;
-}
-
-static void pp_to_dc_clock_levels(
- const struct amd_pp_clocks *pp_clks,
- struct dm_pp_clock_levels *dc_clks,
- enum dm_pp_clock_type dc_clk_type)
-{
- uint32_t i;
-
- if (pp_clks->count > DM_PP_MAX_CLOCK_LEVELS) {
- DRM_INFO("DM_PPLIB: Warning: %s clock: number of levels %d exceeds maximum of %d!\n",
- DC_DECODE_PP_CLOCK_TYPE(dc_clk_type),
- pp_clks->count,
- DM_PP_MAX_CLOCK_LEVELS);
-
- dc_clks->num_levels = DM_PP_MAX_CLOCK_LEVELS;
- } else
- dc_clks->num_levels = pp_clks->count;
-
- DRM_INFO("DM_PPLIB: values for %s clock\n",
- DC_DECODE_PP_CLOCK_TYPE(dc_clk_type));
-
- for (i = 0; i < dc_clks->num_levels; i++) {
- DRM_INFO("DM_PPLIB:\t %d\n", pp_clks->clock[i]);
- /* translate 10kHz to kHz */
- dc_clks->clocks_in_khz[i] = pp_clks->clock[i] * 10;
- }
-}
-
-static void pp_to_dc_clock_levels_with_latency(
- const struct pp_clock_levels_with_latency *pp_clks,
- struct dm_pp_clock_levels_with_latency *clk_level_info,
- enum dm_pp_clock_type dc_clk_type)
-{
- uint32_t i;
-
- if (pp_clks->num_levels > DM_PP_MAX_CLOCK_LEVELS) {
- DRM_INFO("DM_PPLIB: Warning: %s clock: number of levels %d exceeds maximum of %d!\n",
- DC_DECODE_PP_CLOCK_TYPE(dc_clk_type),
- pp_clks->num_levels,
- DM_PP_MAX_CLOCK_LEVELS);
-
- clk_level_info->num_levels = DM_PP_MAX_CLOCK_LEVELS;
- } else
- clk_level_info->num_levels = pp_clks->num_levels;
-
- DRM_DEBUG("DM_PPLIB: values for %s clock\n",
- DC_DECODE_PP_CLOCK_TYPE(dc_clk_type));
-
- for (i = 0; i < clk_level_info->num_levels; i++) {
- DRM_DEBUG("DM_PPLIB:\t %d in 10kHz\n", pp_clks->data[i].clocks_in_khz);
- /* translate 10kHz to kHz */
- clk_level_info->data[i].clocks_in_khz = pp_clks->data[i].clocks_in_khz * 10;
- clk_level_info->data[i].latency_in_us = pp_clks->data[i].latency_in_us;
- }
-}
-
-bool dm_pp_get_clock_levels_by_type(
- const struct dc_context *ctx,
- enum dm_pp_clock_type clk_type,
- struct dm_pp_clock_levels *dc_clks)
-{
- struct amdgpu_device *adev = ctx->driver_context;
- void *pp_handle = adev->powerplay.pp_handle;
- struct amd_pp_clocks pp_clks = { 0 };
- struct amd_pp_simple_clock_info validation_clks = { 0 };
- uint32_t i;
-
- if (adev->powerplay.pp_funcs->get_clock_by_type) {
- if (adev->powerplay.pp_funcs->get_clock_by_type(pp_handle,
- dc_to_pp_clock_type(clk_type), &pp_clks)) {
- /* Error in pplib. Provide default values. */
- get_default_clock_levels(clk_type, dc_clks);
- return true;
- }
- }
-
- pp_to_dc_clock_levels(&pp_clks, dc_clks, clk_type);
-
- if (adev->powerplay.pp_funcs->get_display_mode_validation_clocks) {
- if (adev->powerplay.pp_funcs->get_display_mode_validation_clocks(
- pp_handle, &validation_clks)) {
- /* Error in pplib. Provide default values. */
- DRM_INFO("DM_PPLIB: Warning: using default validation clocks!\n");
- validation_clks.engine_max_clock = 72000;
- validation_clks.memory_max_clock = 80000;
- validation_clks.level = 0;
- }
- }
-
- DRM_INFO("DM_PPLIB: Validation clocks:\n");
- DRM_INFO("DM_PPLIB: engine_max_clock: %d\n",
- validation_clks.engine_max_clock);
- DRM_INFO("DM_PPLIB: memory_max_clock: %d\n",
- validation_clks.memory_max_clock);
- DRM_INFO("DM_PPLIB: level : %d\n",
- validation_clks.level);
-
- /* Translate 10 kHz to kHz. */
- validation_clks.engine_max_clock *= 10;
- validation_clks.memory_max_clock *= 10;
-
- /* Determine the highest non-boosted level from the Validation Clocks */
- if (clk_type == DM_PP_CLOCK_TYPE_ENGINE_CLK) {
- for (i = 0; i < dc_clks->num_levels; i++) {
- if (dc_clks->clocks_in_khz[i] > validation_clks.engine_max_clock) {
- /* This clock is higher the validation clock.
- * Than means the previous one is the highest
- * non-boosted one. */
- DRM_INFO("DM_PPLIB: reducing engine clock level from %d to %d\n",
- dc_clks->num_levels, i);
- dc_clks->num_levels = i > 0 ? i : 1;
- break;
- }
- }
- } else if (clk_type == DM_PP_CLOCK_TYPE_MEMORY_CLK) {
- for (i = 0; i < dc_clks->num_levels; i++) {
- if (dc_clks->clocks_in_khz[i] > validation_clks.memory_max_clock) {
- DRM_INFO("DM_PPLIB: reducing memory clock level from %d to %d\n",
- dc_clks->num_levels, i);
- dc_clks->num_levels = i > 0 ? i : 1;
- break;
- }
- }
- }
-
- return true;
-}
-
-bool dm_pp_get_clock_levels_by_type_with_latency(
- const struct dc_context *ctx,
- enum dm_pp_clock_type clk_type,
- struct dm_pp_clock_levels_with_latency *clk_level_info)
-{
- struct amdgpu_device *adev = ctx->driver_context;
- void *pp_handle = adev->powerplay.pp_handle;
- struct pp_clock_levels_with_latency pp_clks = { 0 };
- const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs;
-
- if (!pp_funcs || !pp_funcs->get_clock_by_type_with_latency)
- return false;
-
- if (pp_funcs->get_clock_by_type_with_latency(pp_handle,
- dc_to_pp_clock_type(clk_type),
- &pp_clks))
- return false;
-
- pp_to_dc_clock_levels_with_latency(&pp_clks, clk_level_info, clk_type);
-
- return true;
-}
-
-bool dm_pp_get_clock_levels_by_type_with_voltage(
- const struct dc_context *ctx,
- enum dm_pp_clock_type clk_type,
- struct dm_pp_clock_levels_with_voltage *clk_level_info)
-{
- /* TODO: to be implemented */
- return false;
-}
-
-bool dm_pp_notify_wm_clock_changes(
- const struct dc_context *ctx,
- struct dm_pp_wm_sets_with_clock_ranges *wm_with_clock_ranges)
-{
- /* TODO: to be implemented */
- return false;
-}
-
-bool dm_pp_apply_power_level_change_request(
- const struct dc_context *ctx,
- struct dm_pp_power_level_change_request *level_change_req)
-{
- /* TODO: to be implemented */
- return false;
-}
-
-bool dm_pp_apply_clock_for_voltage_request(
- const struct dc_context *ctx,
- struct dm_pp_clock_for_voltage_req *clock_for_voltage_req)
-{
- /* TODO: to be implemented */
- return false;
-}
-
-bool dm_pp_get_static_clocks(
- const struct dc_context *ctx,
- struct dm_pp_static_clock_info *static_clk_info)
-{
- /* TODO: to be implemented */
- return false;
-}
-
-void dm_pp_get_funcs_rv(
- struct dc_context *ctx,
- struct pp_smu_funcs_rv *funcs)
-{}
--
-/**** end of power component interfaces ****/
#include "dc.h"
#include "dc_link_dp.h"
#include "dm_helpers.h"
+#include "opp.h"
#include "inc/core_types.h"
#include "link_hwss.h"
struct dc_link_settings initial_link_settings,
struct dc_link_settings *current_link_setting,
enum link_training_result training_result);
-static struct dc_link_settings get_common_supported_link_settings (
+static struct dc_link_settings get_common_supported_link_settings(
struct dc_link_settings link_setting_a,
struct dc_link_settings link_setting_b);
uint8_t rate = (uint8_t)
(lt_settings->link_settings.link_rate);
- union down_spread_ctrl downspread = {{0}};
- union lane_count_set lane_count_set = {{0}};
+ union down_spread_ctrl downspread = { {0} };
+ union lane_count_set lane_count_set = { {0} };
uint8_t link_set_buffer[2];
downspread.raw = (uint8_t)
const struct link_training_settings *lt_settings,
enum hw_dp_training_pattern pattern)
{
- union dpcd_training_lane dpcd_lane[LANE_COUNT_DP_MAX] = {{{0}}};
+ union dpcd_training_lane dpcd_lane[LANE_COUNT_DP_MAX] = { { {0} } };
const uint32_t dpcd_base_lt_offset =
DP_TRAINING_PATTERN_SET;
uint8_t dpcd_lt_buffer[5] = {0};
- union dpcd_training_pattern dpcd_pattern = {{0}};
+ union dpcd_training_pattern dpcd_pattern = { {0} };
uint32_t lane;
uint32_t size_in_bytes;
bool edp_workaround = false; /* TODO link_prop.INTERNAL */
link,
DP_TRAINING_PATTERN_SET,
&dpcd_pattern.raw,
- sizeof(dpcd_pattern.raw) );
+ sizeof(dpcd_pattern.raw));
core_link_write_dpcd(
link,
link,
dpcd_base_lt_offset,
dpcd_lt_buffer,
- size_in_bytes + sizeof(dpcd_pattern.raw) );
+ size_in_bytes + sizeof(dpcd_pattern.raw));
link->cur_lane_setting = lt_settings->lane_settings[0];
}
struct link_training_settings *req_settings)
{
uint8_t dpcd_buf[6] = {0};
- union lane_adjust dpcd_lane_adjust[LANE_COUNT_DP_MAX] = {{{0}}};
- struct link_training_settings request_settings = {{0}};
+ union lane_adjust dpcd_lane_adjust[LANE_COUNT_DP_MAX] = { { {0} } };
+ struct link_training_settings request_settings = { {0} };
uint32_t lane;
memset(req_settings, '\0', sizeof(struct link_training_settings));
if (req_drv_setting_changed) {
update_drive_settings(
- lt_settings,req_settings);
+ lt_settings, req_settings);
dc_link_dp_set_drive_settings(link,
lt_settings);
enum hw_dp_training_pattern hw_tr_pattern;
uint32_t retries_ch_eq;
enum dc_lane_count lane_count = lt_settings->link_settings.lane_count;
- union lane_align_status_updated dpcd_lane_status_updated = {{0}};
- union lane_status dpcd_lane_status[LANE_COUNT_DP_MAX] = {{{0}}};
+ union lane_align_status_updated dpcd_lane_status_updated = { {0} };
+ union lane_status dpcd_lane_status[LANE_COUNT_DP_MAX] = { { {0} } };
hw_tr_pattern = get_supported_tp(link);
lt_settings.lane_settings[0].VOLTAGE_SWING,
lt_settings.lane_settings[0].PRE_EMPHASIS);
+ if (status != LINK_TRAINING_SUCCESS)
+ link->ctx->dc->debug_data.ltFailCount++;
+
return status;
}
return max_link_cap;
}
-bool dp_hbr_verify_link_cap(
+bool dp_verify_link_cap(
struct dc_link *link,
struct dc_link_settings *known_limit_link_setting)
{
enum clock_source_id dp_cs_id = CLOCK_SOURCE_ID_EXTERNAL;
enum link_training_result status;
+ if (link->dc->debug.skip_detection_link_training) {
+ link->verified_link_cap = *known_limit_link_setting;
+ return true;
+ }
+
success = false;
skip_link_training = false;
return success;
}
-static struct dc_link_settings get_common_supported_link_settings (
+static struct dc_link_settings get_common_supported_link_settings(
struct dc_link_settings link_setting_a,
struct dc_link_settings link_setting_b)
{
uint32_t lane_count = link_setting->lane_count;
uint32_t kbps = link_rate_in_kbps;
+
kbps *= lane_count;
kbps *= 8; /* 8 bits per byte*/
const struct dc_link_settings *link_setting;
/*always DP fail safe mode*/
- if (timing->pix_clk_khz == (uint32_t)25175 &&
- timing->h_addressable == (uint32_t)640 &&
- timing->v_addressable == (uint32_t)480)
+ if (timing->pix_clk_khz == (uint32_t) 25175 &&
+ timing->h_addressable == (uint32_t) 640 &&
+ timing->v_addressable == (uint32_t) 480)
return true;
/* We always use verified link settings */
irq_data->raw,
sizeof(union hpd_irq_data));
else {
- /* Read 2 bytes at this location,... */
+ /* Read 14 bytes in a single read and then copy only the required fields.
+ * This is more efficient than doing it in two separate AUX reads. */
+
+ uint8_t tmp[DP_SINK_STATUS_ESI - DP_SINK_COUNT_ESI + 1];
+
retval = core_link_read_dpcd(
link,
DP_SINK_COUNT_ESI,
- irq_data->raw,
- 2);
+ tmp,
+ sizeof(tmp));
if (retval != DC_OK)
return retval;
- /* ... then read remaining 4 at the other location */
- retval = core_link_read_dpcd(
- link,
- DP_LANE0_1_STATUS_ESI,
- &irq_data->raw[2],
- 4);
+ irq_data->bytes.sink_cnt.raw = tmp[DP_SINK_COUNT_ESI - DP_SINK_COUNT_ESI];
+ irq_data->bytes.device_service_irq.raw = tmp[DP_DEVICE_SERVICE_IRQ_VECTOR_ESI0 - DP_SINK_COUNT_ESI];
+ irq_data->bytes.lane01_status.raw = tmp[DP_LANE0_1_STATUS_ESI - DP_SINK_COUNT_ESI];
+ irq_data->bytes.lane23_status.raw = tmp[DP_LANE2_3_STATUS_ESI - DP_SINK_COUNT_ESI];
+ irq_data->bytes.lane_status_updated.raw = tmp[DP_LANE_ALIGN_STATUS_UPDATED_ESI - DP_SINK_COUNT_ESI];
+ irq_data->bytes.sink_status.raw = tmp[DP_SINK_STATUS_ESI - DP_SINK_COUNT_ESI];
}
return retval;
dp_retrain_link_dp_test(link, &link_settings, false);
}
- /* TODO hbr2 compliance eye output is unstable
+ /* TODO Raven hbr2 compliance eye output is unstable
* (toggling on and off) with debugger break
* This caueses intermittent PHY automation failure
* Need to look into the root cause */
- static uint8_t force_tps4_for_cp2520 = 1;
-
static void dp_test_send_phy_test_pattern(struct dc_link *link)
{
union phy_test_pattern dpcd_test_pattern;
break;
case PHY_TEST_PATTERN_CP2520_1:
/* CP2520 pattern is unstable, temporarily use TPS4 instead */
- test_pattern = (force_tps4_for_cp2520 == 1) ?
+ test_pattern = (link->dc->caps.force_dp_tps4_for_cp2520 == 1) ?
DP_TEST_PATTERN_TRAINING_PATTERN4 :
DP_TEST_PATTERN_HBR2_COMPLIANCE_EYE;
break;
case PHY_TEST_PATTERN_CP2520_2:
/* CP2520 pattern is unstable, temporarily use TPS4 instead */
- test_pattern = (force_tps4_for_cp2520 == 1) ?
+ test_pattern = (link->dc->caps.force_dp_tps4_for_cp2520 == 1) ?
DP_TEST_PATTERN_TRAINING_PATTERN4 :
DP_TEST_PATTERN_HBR2_COMPLIANCE_EYE;
break;
sizeof(test_response));
}
-bool dc_link_handle_hpd_rx_irq(struct dc_link *link, union hpd_irq_data *out_hpd_irq_dpcd_data)
+bool dc_link_handle_hpd_rx_irq(struct dc_link *link, union hpd_irq_data *out_hpd_irq_dpcd_data, bool *out_link_loss)
{
- union hpd_irq_data hpd_irq_dpcd_data = {{{{0}}}};
+ union hpd_irq_data hpd_irq_dpcd_data = { { { {0} } } };
union device_service_irq device_service_clear = { { 0 } };
- enum dc_status result = DDC_RESULT_UNKNOWN;
+ enum dc_status result;
+
bool status = false;
+
+ if (out_link_loss)
+ *out_link_loss = false;
/* For use cases related to down stream connection status change,
* PSR and device auto test, refer to function handle_sst_hpd_irq
* in DAL2.1*/
true, LINK_TRAINING_ATTEMPTS);
status = false;
+ if (out_link_loss)
+ *out_link_loss = true;
}
if (link->type == dc_connection_active_dongle &&
link->dpcd_caps.branch_hw_revision =
dp_hw_fw_revision.ieee_hw_rev;
+
+ memmove(
+ link->dpcd_caps.branch_fw_revision,
+ dp_hw_fw_revision.ieee_fw_rev,
+ sizeof(dp_hw_fw_revision.ieee_fw_rev));
}
}
{
uint8_t dpcd_data[DP_ADAPTER_CAP - DP_DPCD_REV + 1];
+ struct dp_device_vendor_id sink_id;
union down_stream_port_count down_strm_port_count;
union edp_configuration_cap edp_config_cap;
union dp_downstream_port_present ds_port = { 0 };
enum dc_status status = DC_ERROR_UNEXPECTED;
uint32_t read_dpcd_retry_cnt = 3;
int i;
+ struct dp_sink_hw_fw_revision dp_hw_fw_revision;
memset(dpcd_data, '\0', sizeof(dpcd_data));
memset(&down_strm_port_count,
&link->dpcd_caps.sink_count.raw,
sizeof(link->dpcd_caps.sink_count.raw));
+ /* read sink ieee oui */
+ core_link_read_dpcd(link,
+ DP_SINK_OUI,
+ (uint8_t *)(&sink_id),
+ sizeof(sink_id));
+
+ link->dpcd_caps.sink_dev_id =
+ (sink_id.ieee_oui[0] << 16) +
+ (sink_id.ieee_oui[1] << 8) +
+ (sink_id.ieee_oui[2]);
+
+ memmove(
+ link->dpcd_caps.sink_dev_id_str,
+ sink_id.ieee_device_id,
+ sizeof(sink_id.ieee_device_id));
+
+ core_link_read_dpcd(
+ link,
+ DP_SINK_HW_REVISION_START,
+ (uint8_t *)&dp_hw_fw_revision,
+ sizeof(dp_hw_fw_revision));
+
+ link->dpcd_caps.sink_hw_revision =
+ dp_hw_fw_revision.ieee_hw_rev;
+
+ memmove(
+ link->dpcd_caps.sink_fw_revision,
+ dp_hw_fw_revision.ieee_fw_rev,
+ sizeof(dp_hw_fw_revision.ieee_fw_rev));
+
/* Connectivity log: detection */
CONN_DATA_DETECT(link, dpcd_data, sizeof(dpcd_data), "Rx Caps: ");
pipe_ctx->stream->bit_depth_params = params;
pipe_ctx->stream_res.opp->funcs->
opp_program_bit_depth_reduction(pipe_ctx->stream_res.opp, ¶ms);
-
- pipe_ctx->stream_res.tg->funcs->set_test_pattern(pipe_ctx->stream_res.tg,
+ if (pipe_ctx->stream_res.tg->funcs->set_test_pattern)
+ pipe_ctx->stream_res.tg->funcs->set_test_pattern(pipe_ctx->stream_res.tg,
controller_test_pattern, color_depth);
}
break;
pipe_ctx->stream->bit_depth_params = params;
pipe_ctx->stream_res.opp->funcs->
opp_program_bit_depth_reduction(pipe_ctx->stream_res.opp, ¶ms);
-
- pipe_ctx->stream_res.tg->funcs->set_test_pattern(pipe_ctx->stream_res.tg,
+ if (pipe_ctx->stream_res.tg->funcs->set_test_pattern)
+ pipe_ctx->stream_res.tg->funcs->set_test_pattern(pipe_ctx->stream_res.tg,
CONTROLLER_DP_TEST_PATTERN_VIDEOMODE,
color_depth);
}
#include "inc/compressor.h"
#include "dml/display_mode_lib.h"
-#define DC_VER "3.1.44"
+#define DC_VER "3.1.58"
#define MAX_SURFACES 3
#define MAX_STREAMS 6
uint32_t max_planes;
uint32_t max_downscale_ratio;
uint32_t i2c_speed_in_khz;
+ uint32_t dmdata_alloc_size;
unsigned int max_cursor_size;
unsigned int max_video_width;
int linear_pitch_alignment;
bool is_apu;
bool dual_link_dvi;
bool post_blend_color_processing;
+ bool force_dp_tps4_for_cp2520;
};
struct dc_dcc_surface_param {
bool disable_disp_pll_sharing;
};
+enum visual_confirm {
+ VISUAL_CONFIRM_DISABLE = 0,
+ VISUAL_CONFIRM_SURFACE = 1,
+ VISUAL_CONFIRM_HDR = 2,
+};
+
enum dcc_option {
DCC_ENABLE = 0,
DCC_DISABLE = 1,
WM_REPORT_OVERRIDE = 1,
};
+/*
+ * For any clocks that may differ per pipe
+ * only the max is stored in this structure
+ */
struct dc_clocks {
int dispclk_khz;
int max_supported_dppclk_khz;
int socclk_khz;
int dcfclk_deep_sleep_khz;
int fclk_khz;
+ int phyclk_khz;
};
-struct dc_debug {
- bool surface_visual_confirm;
+struct dc_debug_options {
+ enum visual_confirm visual_confirm;
bool sanity_checks;
bool max_disp_clk;
bool surface_trace;
int urgent_latency_ns;
int percent_of_ideal_drambw;
int dram_clock_change_latency_ns;
+ bool optimized_watermark;
int always_scale;
bool disable_pplib_clock_request;
bool disable_clock_gate;
bool always_use_regamma;
bool p010_mpo_support;
bool recovery_enabled;
+ bool avoid_vbios_exec_table;
+ bool scl_reset_length10;
+ bool hdmi20_disable;
+ bool skip_detection_link_training;
+};
+struct dc_debug_data {
+ uint32_t ltFailCount;
+ uint32_t i2cErrorCount;
+ uint32_t auxErrorCount;
};
+
+
struct dc_state;
struct resource_pool;
struct dce_hwseq;
struct dc_caps caps;
struct dc_cap_funcs cap_funcs;
struct dc_config config;
- struct dc_debug debug;
-
+ struct dc_debug_options debug;
struct dc_context *ctx;
uint8_t link_count;
/* Inputs into BW and WM calculations. */
struct bw_calcs_dceip *bw_dceip;
struct bw_calcs_vbios *bw_vbios;
-#ifdef CONFIG_DRM_AMD_DC_DCN1_0
+#ifdef CONFIG_X86
struct dcn_soc_bounding_box *dcn_soc;
struct dcn_ip_params *dcn_ip;
struct display_mode_lib dml;
bool apply_edp_fast_boot_optimization;
/* FBC compressor */
-#if defined(CONFIG_DRM_AMD_DC_FBC)
struct compressor *fbc_compressor;
-#endif
+
+ struct dc_debug_data debug_data;
};
enum frame_buffer_mode {
TF_TYPE_PREDEFINED,
TF_TYPE_DISTRIBUTED_POINTS,
TF_TYPE_BYPASS,
+ TF_TYPE_HWPWL
};
struct dc_transfer_func_distributed_points {
TRANSFER_FUNCTION_PQ,
TRANSFER_FUNCTION_LINEAR,
TRANSFER_FUNCTION_UNITY,
+ TRANSFER_FUNCTION_HLG,
+ TRANSFER_FUNCTION_HLG12,
+ TRANSFER_FUNCTION_GAMMA22
};
struct dc_transfer_func {
struct kref refcount;
- struct dc_transfer_func_distributed_points tf_pts;
enum dc_transfer_func_type type;
enum dc_transfer_func_predefined tf;
/* FP16 1.0 reference level in nits, default is 80 nits, only for PQ*/
uint32_t sdr_ref_white_level;
struct dc_context *ctx;
+ union {
+ struct pwl_params pwl;
+ struct dc_transfer_func_distributed_points tf_pts;
+ };
};
/*
struct dc_dongle_caps dongle_caps;
uint32_t sink_dev_id;
+ int8_t sink_dev_id_str[6];
+ int8_t sink_hw_revision;
+ int8_t sink_fw_revision[2];
+
uint32_t branch_dev_id;
int8_t branch_dev_name[6];
int8_t branch_hw_revision;
+ int8_t branch_fw_revision[2];
bool allow_invalid_MSA_timing_param;
bool panel_mode_edp;
struct dc_link *link;
struct dc_context *ctx;
+ uint32_t sink_id;
+
/* private to dc_sink.c */
+ // refcount must be the last member in dc_sink, since we want the
+ // sink structure to be logically cloneable up to (but not including)
+ // refcount
struct kref refcount;
-
};
void dc_sink_retain(struct dc_sink *sink);
#include "reg_helper.h"
#include "dce/dce_abm.h"
#include "dce/dce_dmcu.h"
+#include "dce/dce_aux.h"
+
+const struct _vcs_dpi_ip_params_st dcn1_0_ip = {
+ .rob_buffer_size_kbytes = 64,
+ .det_buffer_size_kbytes = 164,
+ .dpte_buffer_size_in_pte_reqs = 42,
+ .dpp_output_buffer_pixels = 2560,
+ .opp_output_buffer_lines = 1,
+ .pixel_chunk_size_kbytes = 8,
+ .pte_enable = 1,
+ .pte_chunk_size_kbytes = 2,
+ .meta_chunk_size_kbytes = 2,
+ .writeback_chunk_size_kbytes = 2,
+ .line_buffer_size_bits = 589824,
+ .max_line_buffer_lines = 12,
+ .IsLineBufferBppFixed = 0,
+ .LineBufferFixedBpp = -1,
+ .writeback_luma_buffer_size_kbytes = 12,
+ .writeback_chroma_buffer_size_kbytes = 8,
+ .max_num_dpp = 4,
+ .max_num_wb = 2,
+ .max_dchub_pscl_bw_pix_per_clk = 4,
+ .max_pscl_lb_bw_pix_per_clk = 2,
+ .max_lb_vscl_bw_pix_per_clk = 4,
+ .max_vscl_hscl_bw_pix_per_clk = 4,
+ .max_hscl_ratio = 4,
+ .max_vscl_ratio = 4,
+ .hscl_mults = 4,
+ .vscl_mults = 4,
+ .max_hscl_taps = 8,
+ .max_vscl_taps = 8,
+ .dispclk_ramp_margin_percent = 1,
+ .underscan_factor = 1.10,
+ .min_vblank_lines = 14,
+ .dppclk_delay_subtotal = 90,
+ .dispclk_delay_subtotal = 42,
+ .dcfclk_cstate_latency = 10,
+ .max_inter_dcn_tile_repeaters = 8,
+ .can_vstartup_lines_exceed_vsync_plus_back_porch_lines_minus_one = 0,
+ .bug_forcing_LC_req_same_size_fixed = 0,
+};
+
+const struct _vcs_dpi_soc_bounding_box_st dcn1_0_soc = {
+ .sr_exit_time_us = 9.0,
+ .sr_enter_plus_exit_time_us = 11.0,
+ .urgent_latency_us = 4.0,
+ .writeback_latency_us = 12.0,
+ .ideal_dram_bw_after_urgent_percent = 80.0,
+ .max_request_size_bytes = 256,
+ .downspread_percent = 0.5,
+ .dram_page_open_time_ns = 50.0,
+ .dram_rw_turnaround_time_ns = 17.5,
+ .dram_return_buffer_per_channel_bytes = 8192,
+ .round_trip_ping_latency_dcfclk_cycles = 128,
+ .urgent_out_of_order_return_per_channel_bytes = 256,
+ .channel_interleave_bytes = 256,
+ .num_banks = 8,
+ .num_chans = 2,
+ .vmm_page_size_bytes = 4096,
+ .dram_clock_change_latency_us = 17.0,
+ .writeback_dram_clock_change_latency_us = 23.0,
+ .return_bus_width_bytes = 64,
+};
#ifndef mmDP0_DP_DPHY_INTERNAL_CTRL
#define mmDP0_DP_DPHY_INTERNAL_CTRL 0x210f
OPP_MASK_SH_LIST_DCN10(_MASK),
};
+#define aux_engine_regs(id)\
+[id] = {\
+ AUX_COMMON_REG_LIST(id), \
+ .AUX_RESET_MASK = 0 \
+}
+
+static const struct dce110_aux_registers aux_engine_regs[] = {
+ aux_engine_regs(0),
+ aux_engine_regs(1),
+ aux_engine_regs(2),
+ aux_engine_regs(3),
+ aux_engine_regs(4),
+ aux_engine_regs(5)
+};
+
#define tf_regs(id)\
[id] = {\
TF_REG_LIST_DCN10(id),\
static const struct resource_caps res_cap = {
.num_timing_generator = 4,
+ .num_opp = 4,
.num_video_plane = 4,
.num_audio = 4,
.num_stream_encoder = 4,
.num_pll = 4,
};
-static const struct dc_debug debug_defaults_drv = {
+static const struct dc_debug_options debug_defaults_drv = {
.sanity_checks = true,
.disable_dmcu = true,
.force_abm_enable = false,
*/
.min_disp_clk_khz = 100000,
- .disable_pplib_clock_request = true,
+ .disable_pplib_clock_request = false,
.disable_pplib_wm_range = false,
.pplib_wm_report_mode = WM_REPORT_DEFAULT,
.pipe_split_policy = MPC_SPLIT_AVOID_MULT_DISP,
.max_downscale_src_width = 3840,
};
-static const struct dc_debug debug_defaults_diags = {
+static const struct dc_debug_options debug_defaults_diags = {
.disable_dmcu = true,
.force_abm_enable = false,
.timing_trace = true,
return &opp->base;
}
+struct engine *dcn10_aux_engine_create(
+ struct dc_context *ctx,
+ uint32_t inst)
+{
+ struct aux_engine_dce110 *aux_engine =
+ kzalloc(sizeof(struct aux_engine_dce110), GFP_KERNEL);
+
+ if (!aux_engine)
+ return NULL;
+
+ dce110_aux_engine_construct(aux_engine, ctx, inst,
+ SW_AUX_TIMEOUT_PERIOD_MULTIPLIER * AUX_TIMEOUT_PERIOD,
+ &aux_engine_regs[inst]);
+
+ return &aux_engine->base.base;
+}
+
static struct mpc *dcn10_mpc_create(struct dc_context *ctx)
{
struct dcn10_mpc *mpc10 = kzalloc(sizeof(struct dcn10_mpc),
hws->masks = &hwseq_mask;
hws->wa.DEGVIDCN10_253 = true;
hws->wa.false_optc_underflow = true;
+ hws->wa.DEGVIDCN10_254 = true;
}
return hws;
}
kfree(DCN10TG_FROM_TG(pool->base.timing_generators[i]));
pool->base.timing_generators[i] = NULL;
}
+
+ if (pool->base.engines[i] != NULL)
+ pool->base.engines[i]->funcs->destroy_engine(&pool->base.engines[i]);
}
for (i = 0; i < pool->base.stream_enc_count; i++)
if (pool->base.dmcu != NULL)
dce_dmcu_destroy(&pool->base.dmcu);
- if (pool->base.display_clock != NULL)
- dce_disp_clk_destroy(&pool->base.display_clock);
+ if (pool->base.dccg != NULL)
+ dce_dccg_destroy(&pool->base.dccg);
kfree(pool->base.pp_smu);
}
return DC_OK;
}
-static struct dc_cap_funcs cap_funcs = {
+static const struct dc_cap_funcs cap_funcs = {
.get_dcc_compression_cap = dcn10_get_dcc_compression_cap
};
-static struct resource_funcs dcn10_res_pool_funcs = {
+static const struct resource_funcs dcn10_res_pool_funcs = {
.destroy = dcn10_destroy_resource_pool,
.link_enc_create = dcn10_link_encoder_create,
.validate_bandwidth = dcn_validate_bandwidth,
dc->caps.max_slave_planes = 1;
dc->caps.is_apu = true;
dc->caps.post_blend_color_processing = false;
+ /* Raven DP PHY HBR2 eye diagram pattern is not stable. Use TP4 */
+ dc->caps.force_dp_tps4_for_cp2520 = true;
if (dc->ctx->dce_environment == DCE_ENV_PRODUCTION_DRV)
dc->debug = debug_defaults_drv;
}
}
- pool->base.display_clock = dce120_disp_clk_create(ctx);
- if (pool->base.display_clock == NULL) {
+ pool->base.dccg = dcn1_dccg_create(ctx);
+ if (pool->base.dccg == NULL) {
dm_error("DC: failed to create display clock!\n");
BREAK_TO_DEBUGGER();
goto fail;
goto fail;
}
+ pool->base.engines[i] = dcn10_aux_engine_create(ctx, i);
+ if (pool->base.engines[i] == NULL) {
+ BREAK_TO_DEBUGGER();
+ dm_error(
+ "DC:failed to create aux engine!!\n");
+ goto fail;
+ }
+
/* check next valid pipe */
j++;
}
ppsmc_pptable->AcgGfxclkSpreadPercent = smc_dpm_table.acggfxclkspreadpercent;
ppsmc_pptable->AcgGfxclkSpreadFreq = smc_dpm_table.acggfxclkspreadfreq;
- /* 0xFFFF will disable the ACG feature */
- if (!(hwmgr->feature_mask & PP_ACG_MASK)) {
- ppsmc_pptable->AcgThresholdFreqHigh = 0xFFFF;
- ppsmc_pptable->AcgThresholdFreqLow = 0xFFFF;
- }
+ ppsmc_pptable->Vr2_I2C_address = smc_dpm_table.Vr2_I2C_address;
+ ppsmc_pptable->Vr2_I2C_address = smc_dpm_table.Vr2_I2C_address;
+
return 0;
}
for (i = 0; i < ETNA_MAX_PIPES; i++) {
struct etnaviv_gpu *gpu = priv->gpu[i];
+ struct drm_sched_rq *rq;
if (gpu) {
- drm_sched_entity_init(&gpu->sched,
- &ctx->sched_entity[i],
- &gpu->sched.sched_rq[DRM_SCHED_PRIORITY_NORMAL],
- NULL);
+ rq = &gpu->sched.sched_rq[DRM_SCHED_PRIORITY_NORMAL];
+ drm_sched_entity_init(&ctx->sched_entity[i],
+ &rq, 1, NULL);
}
}
gpu->lastctx = NULL;
mutex_unlock(&gpu->lock);
- drm_sched_entity_fini(&gpu->sched,
- &ctx->sched_entity[i]);
+ drm_sched_entity_destroy(&ctx->sched_entity[i]);
}
}
},
};
+ static struct platform_device *etnaviv_drm;
+
static int __init etnaviv_init(void)
{
+ struct platform_device *pdev;
int ret;
struct device_node *np;
ret = platform_driver_register(&etnaviv_platform_driver);
if (ret != 0)
- platform_driver_unregister(&etnaviv_gpu_driver);
+ goto unregister_gpu_driver;
/*
* If the DT contains at least one available GPU device, instantiate
for_each_compatible_node(np, NULL, "vivante,gc") {
if (!of_device_is_available(np))
continue;
-
- platform_device_register_simple("etnaviv", -1, NULL, 0);
+ pdev = platform_device_register_simple("etnaviv", -1,
+ NULL, 0);
+ if (IS_ERR(pdev)) {
+ ret = PTR_ERR(pdev);
+ of_node_put(np);
+ goto unregister_platform_driver;
+ }
+ etnaviv_drm = pdev;
of_node_put(np);
break;
}
+ return 0;
+
+ unregister_platform_driver:
+ platform_driver_unregister(&etnaviv_platform_driver);
+ unregister_gpu_driver:
+ platform_driver_unregister(&etnaviv_gpu_driver);
return ret;
}
module_init(etnaviv_init);
static void __exit etnaviv_exit(void)
{
- platform_driver_unregister(&etnaviv_gpu_driver);
+ platform_device_unregister(etnaviv_drm);
platform_driver_unregister(&etnaviv_platform_driver);
+ platform_driver_unregister(&etnaviv_gpu_driver);
}
module_exit(etnaviv_exit);
#include "etnaviv_gem.h"
#include "etnaviv_gpu.h"
#include "etnaviv_sched.h"
+ #include "state.xml.h"
static int etnaviv_job_hang_limit = 0;
module_param_named(job_hang_limit, etnaviv_job_hang_limit, int , 0444);
{
struct etnaviv_gem_submit *submit = to_etnaviv_submit(sched_job);
struct etnaviv_gpu *gpu = submit->gpu;
+ u32 dma_addr;
+ int change;
+
+ /*
+ * If the GPU managed to complete this jobs fence, the timout is
+ * spurious. Bail out.
+ */
+ if (fence_completed(gpu, submit->out_fence->seqno))
+ return;
+
+ /*
+ * If the GPU is still making forward progress on the front-end (which
+ * should never loop) we shift out the timeout to give it a chance to
+ * finish the job.
+ */
+ dma_addr = gpu_read(gpu, VIVS_FE_DMA_ADDRESS);
+ change = dma_addr - gpu->hangcheck_dma_addr;
+ if (change < 0 || change > 16) {
+ gpu->hangcheck_dma_addr = dma_addr;
+ schedule_delayed_work(&sched_job->work_tdr,
+ sched_job->sched->timeout);
+ return;
+ }
/* block scheduler */
kthread_park(gpu->sched.thread);
{
int ret;
- ret = drm_sched_job_init(&submit->sched_job, &submit->gpu->sched,
- sched_entity, submit->cmdbuf.ctx);
+ ret = drm_sched_job_init(&submit->sched_job, sched_entity,
+ submit->cmdbuf.ctx);
if (ret)
return ret;
#define OP_MEDIA_INTERFACE_DESCRIPTOR_LOAD OP_3D_MEDIA(0x2, 0x0, 0x2)
#define OP_MEDIA_GATEWAY_STATE OP_3D_MEDIA(0x2, 0x0, 0x3)
#define OP_MEDIA_STATE_FLUSH OP_3D_MEDIA(0x2, 0x0, 0x4)
+#define OP_MEDIA_POOL_STATE OP_3D_MEDIA(0x2, 0x0, 0x5)
#define OP_MEDIA_OBJECT OP_3D_MEDIA(0x2, 0x1, 0x0)
#define OP_MEDIA_OBJECT_PRT OP_3D_MEDIA(0x2, 0x1, 0x2)
{
struct intel_vgpu *vgpu = s->vgpu;
struct intel_gvt *gvt = vgpu->gvt;
+ u32 ctx_sr_ctl;
if (offset + 4 > gvt->device_info.mmio_size) {
gvt_vgpu_err("%s access to (%x) outside of MMIO range\n",
patch_value(s, cmd_ptr(s, index), VGT_PVINFO_PAGE);
}
+ /* TODO
+ * Right now only scan LRI command on KBL and in inhibit context.
+ * It's good enough to support initializing mmio by lri command in
+ * vgpu inhibit context on KBL.
+ */
+ if (IS_KABYLAKE(s->vgpu->gvt->dev_priv) &&
+ intel_gvt_mmio_is_in_ctx(gvt, offset) &&
+ !strncmp(cmd, "lri", 3)) {
+ intel_gvt_hypervisor_read_gpa(s->vgpu,
+ s->workload->ring_context_gpa + 12, &ctx_sr_ctl, 4);
+ /* check inhibit context */
+ if (ctx_sr_ctl & 1) {
+ u32 data = cmd_val(s, index + 1);
+
+ if (intel_gvt_mmio_has_mode_mask(s->vgpu->gvt, offset))
+ intel_vgpu_mask_mmio_write(vgpu,
+ offset, &data, 4);
+ else
+ vgpu_vreg(vgpu, offset) = data;
+ }
+ }
+
/* TODO: Update the global mask if this MMIO is a masked-MMIO */
intel_gvt_mmio_set_cmd_accessed(gvt, offset);
return 0;
if (!info->async_flip)
return 0;
- if (IS_SKYLAKE(dev_priv) || IS_KABYLAKE(dev_priv)) {
+ if (IS_SKYLAKE(dev_priv)
+ || IS_KABYLAKE(dev_priv)
+ || IS_BROXTON(dev_priv)) {
stride = vgpu_vreg_t(s->vgpu, info->stride_reg) & GENMASK(9, 0);
tile = (vgpu_vreg_t(s->vgpu, info->ctrl_reg) &
GENMASK(12, 10)) >> 10;
set_mask_bits(&vgpu_vreg_t(vgpu, info->surf_reg), GENMASK(31, 12),
info->surf_val << 12);
- if (IS_SKYLAKE(dev_priv) || IS_KABYLAKE(dev_priv)) {
+ if (IS_SKYLAKE(dev_priv)
+ || IS_KABYLAKE(dev_priv)
+ || IS_BROXTON(dev_priv)) {
set_mask_bits(&vgpu_vreg_t(vgpu, info->stride_reg), GENMASK(9, 0),
info->stride_val);
set_mask_bits(&vgpu_vreg_t(vgpu, info->ctrl_reg), GENMASK(12, 10),
if (IS_BROADWELL(dev_priv))
return gen8_decode_mi_display_flip(s, info);
- if (IS_SKYLAKE(dev_priv) || IS_KABYLAKE(dev_priv))
+ if (IS_SKYLAKE(dev_priv)
+ || IS_KABYLAKE(dev_priv)
+ || IS_BROXTON(dev_priv))
return skl_decode_mi_display_flip(s, info);
return -ENODEV;
static int check_mi_display_flip(struct parser_exec_state *s,
struct mi_display_flip_command_info *info)
{
- struct drm_i915_private *dev_priv = s->vgpu->gvt->dev_priv;
-
- if (IS_BROADWELL(dev_priv)
- || IS_SKYLAKE(dev_priv)
- || IS_KABYLAKE(dev_priv))
- return gen8_check_mi_display_flip(s, info);
- return -ENODEV;
+ return gen8_check_mi_display_flip(s, info);
}
static int update_plane_mmio_from_mi_display_flip(
struct parser_exec_state *s,
struct mi_display_flip_command_info *info)
{
- struct drm_i915_private *dev_priv = s->vgpu->gvt->dev_priv;
-
- if (IS_BROADWELL(dev_priv)
- || IS_SKYLAKE(dev_priv)
- || IS_KABYLAKE(dev_priv))
- return gen8_update_plane_mmio_from_mi_display_flip(s, info);
- return -ENODEV;
+ return gen8_update_plane_mmio_from_mi_display_flip(s, info);
}
static int cmd_handler_mi_display_flip(struct parser_exec_state *s)
*/
static int batch_buffer_needs_scan(struct parser_exec_state *s)
{
- struct intel_gvt *gvt = s->vgpu->gvt;
-
- if (IS_BROADWELL(gvt->dev_priv) || IS_SKYLAKE(gvt->dev_priv)
- || IS_KABYLAKE(gvt->dev_priv)) {
- /* BDW decides privilege based on address space */
- if (cmd_val(s, 0) & (1 << 8) &&
+ /* Decide privilege based on address space */
+ if (cmd_val(s, 0) & (1 << 8) &&
!(s->vgpu->scan_nonprivbb & (1 << s->ring_id)))
- return 0;
- }
+ return 0;
return 1;
}
{"MEDIA_STATE_FLUSH", OP_MEDIA_STATE_FLUSH, F_LEN_VAR, R_RCS, D_ALL,
0, 16, NULL},
+ {"MEDIA_POOL_STATE", OP_MEDIA_POOL_STATE, F_LEN_VAR, R_RCS, D_ALL,
+ 0, 16, NULL},
+
{"MEDIA_OBJECT", OP_MEDIA_OBJECT, F_LEN_VAR, R_RCS, D_ALL, 0, 16, NULL},
{"MEDIA_CURBE_LOAD", OP_MEDIA_CURBE_LOAD, F_LEN_VAR, R_RCS, D_ALL,
struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv;
int pipe;
+ if (IS_BROXTON(dev_priv)) {
+ vgpu_vreg_t(vgpu, GEN8_DE_PORT_ISR) &= ~(BXT_DE_PORT_HP_DDIA |
+ BXT_DE_PORT_HP_DDIB |
+ BXT_DE_PORT_HP_DDIC);
+
+ if (intel_vgpu_has_monitor_on_port(vgpu, PORT_A)) {
+ vgpu_vreg_t(vgpu, GEN8_DE_PORT_ISR) |=
+ BXT_DE_PORT_HP_DDIA;
+ }
+
+ if (intel_vgpu_has_monitor_on_port(vgpu, PORT_B)) {
+ vgpu_vreg_t(vgpu, GEN8_DE_PORT_ISR) |=
+ BXT_DE_PORT_HP_DDIB;
+ }
+
+ if (intel_vgpu_has_monitor_on_port(vgpu, PORT_C)) {
+ vgpu_vreg_t(vgpu, GEN8_DE_PORT_ISR) |=
+ BXT_DE_PORT_HP_DDIC;
+ }
+
+ return;
+ }
+
vgpu_vreg_t(vgpu, SDEISR) &= ~(SDE_PORTB_HOTPLUG_CPT |
SDE_PORTC_HOTPLUG_CPT |
SDE_PORTD_HOTPLUG_CPT);
~(TRANS_DDI_BPC_MASK | TRANS_DDI_MODE_SELECT_MASK |
TRANS_DDI_PORT_MASK);
vgpu_vreg_t(vgpu, TRANS_DDI_FUNC_CTL(TRANSCODER_A)) |=
- (TRANS_DDI_BPC_8 | TRANS_DDI_MODE_SELECT_DP_SST |
+ (TRANS_DDI_BPC_8 | TRANS_DDI_MODE_SELECT_DVI |
(PORT_B << TRANS_DDI_PORT_SHIFT) |
TRANS_DDI_FUNC_ENABLE);
if (IS_BROADWELL(dev_priv)) {
~(TRANS_DDI_BPC_MASK | TRANS_DDI_MODE_SELECT_MASK |
TRANS_DDI_PORT_MASK);
vgpu_vreg_t(vgpu, TRANS_DDI_FUNC_CTL(TRANSCODER_A)) |=
- (TRANS_DDI_BPC_8 | TRANS_DDI_MODE_SELECT_DP_SST |
+ (TRANS_DDI_BPC_8 | TRANS_DDI_MODE_SELECT_DVI |
(PORT_C << TRANS_DDI_PORT_SHIFT) |
TRANS_DDI_FUNC_ENABLE);
if (IS_BROADWELL(dev_priv)) {
~(TRANS_DDI_BPC_MASK | TRANS_DDI_MODE_SELECT_MASK |
TRANS_DDI_PORT_MASK);
vgpu_vreg_t(vgpu, TRANS_DDI_FUNC_CTL(TRANSCODER_A)) |=
- (TRANS_DDI_BPC_8 | TRANS_DDI_MODE_SELECT_DP_SST |
+ (TRANS_DDI_BPC_8 | TRANS_DDI_MODE_SELECT_DVI |
(PORT_D << TRANS_DDI_PORT_SHIFT) |
TRANS_DDI_FUNC_ENABLE);
if (IS_BROADWELL(dev_priv)) {
for_each_pipe(dev_priv, pipe) {
vgpu_vreg_t(vgpu, DSPCNTR(pipe)) &= ~DISPLAY_PLANE_ENABLE;
vgpu_vreg_t(vgpu, SPRCTL(pipe)) &= ~SPRITE_ENABLE;
- vgpu_vreg_t(vgpu, CURCNTR(pipe)) &= ~CURSOR_MODE;
- vgpu_vreg_t(vgpu, CURCNTR(pipe)) |= CURSOR_MODE_DISABLE;
+ vgpu_vreg_t(vgpu, CURCNTR(pipe)) &= ~MCURSOR_MODE;
+ vgpu_vreg_t(vgpu, CURCNTR(pipe)) |= MCURSOR_MODE_DISABLE;
}
vgpu_vreg_t(vgpu, PIPECONF(PIPE_A)) |= PIPECONF_ENABLE;
struct intel_gvt_irq *irq = &gvt->irq;
struct intel_vgpu *vgpu;
int pipe, id;
+ int found = false;
- if (WARN_ON(!mutex_is_locked(&gvt->lock)))
- return;
-
+ mutex_lock(&gvt->lock);
for_each_active_vgpu(gvt, vgpu, id) {
for (pipe = 0; pipe < I915_MAX_PIPES; pipe++) {
- if (pipe_is_enabled(vgpu, pipe))
- goto out;
+ if (pipe_is_enabled(vgpu, pipe)) {
+ found = true;
+ break;
+ }
}
+ if (found)
+ break;
}
/* all the pipes are disabled */
- hrtimer_cancel(&irq->vblank_timer.timer);
- return;
-
-out:
- hrtimer_start(&irq->vblank_timer.timer,
- ktime_add_ns(ktime_get(), irq->vblank_timer.period),
- HRTIMER_MODE_ABS);
-
+ if (!found)
+ hrtimer_cancel(&irq->vblank_timer.timer);
+ else
+ hrtimer_start(&irq->vblank_timer.timer,
+ ktime_add_ns(ktime_get(), irq->vblank_timer.period),
+ HRTIMER_MODE_ABS);
+ mutex_unlock(&gvt->lock);
}
static void emulate_vblank_on_pipe(struct intel_vgpu *vgpu, int pipe)
{
int pipe;
+ mutex_lock(&vgpu->vgpu_lock);
for_each_pipe(vgpu->gvt->dev_priv, pipe)
emulate_vblank_on_pipe(vgpu, pipe);
+ mutex_unlock(&vgpu->vgpu_lock);
}
/**
struct intel_vgpu *vgpu;
int id;
- if (WARN_ON(!mutex_is_locked(&gvt->lock)))
- return;
-
+ mutex_lock(&gvt->lock);
for_each_active_vgpu(gvt, vgpu, id)
emulate_vblank(vgpu);
+ mutex_unlock(&gvt->lock);
}
/**
GTT_TYPE_PPGTT_PDE_PT,
GTT_TYPE_PPGTT_PTE_PT,
GTT_TYPE_PPGTT_PTE_2M_ENTRY),
+ /* We take IPS bit as 'PSE' for PTE level. */
GTT_TYPE_TABLE_ENTRY(GTT_TYPE_PPGTT_PTE_PT,
GTT_TYPE_PPGTT_PTE_4K_ENTRY,
GTT_TYPE_PPGTT_PTE_PT,
GTT_TYPE_INVALID,
- GTT_TYPE_INVALID),
+ GTT_TYPE_PPGTT_PTE_64K_ENTRY),
GTT_TYPE_TABLE_ENTRY(GTT_TYPE_PPGTT_PTE_4K_ENTRY,
GTT_TYPE_PPGTT_PTE_4K_ENTRY,
GTT_TYPE_PPGTT_PTE_PT,
GTT_TYPE_INVALID,
- GTT_TYPE_INVALID),
+ GTT_TYPE_PPGTT_PTE_64K_ENTRY),
+ GTT_TYPE_TABLE_ENTRY(GTT_TYPE_PPGTT_PTE_64K_ENTRY,
+ GTT_TYPE_PPGTT_PTE_4K_ENTRY,
+ GTT_TYPE_PPGTT_PTE_PT,
+ GTT_TYPE_INVALID,
+ GTT_TYPE_PPGTT_PTE_64K_ENTRY),
GTT_TYPE_TABLE_ENTRY(GTT_TYPE_PPGTT_PTE_2M_ENTRY,
GTT_TYPE_PPGTT_PDE_ENTRY,
GTT_TYPE_PPGTT_PDE_PT,
#define ADDR_1G_MASK GENMASK_ULL(GTT_HAW - 1, 30)
#define ADDR_2M_MASK GENMASK_ULL(GTT_HAW - 1, 21)
+#define ADDR_64K_MASK GENMASK_ULL(GTT_HAW - 1, 16)
#define ADDR_4K_MASK GENMASK_ULL(GTT_HAW - 1, 12)
+#define GTT_SPTE_FLAG_MASK GENMASK_ULL(62, 52)
+#define GTT_SPTE_FLAG_64K_SPLITED BIT(52) /* splited 64K gtt entry */
+
+#define GTT_64K_PTE_STRIDE 16
+
static unsigned long gen8_gtt_get_pfn(struct intel_gvt_gtt_entry *e)
{
unsigned long pfn;
pfn = (e->val64 & ADDR_1G_MASK) >> PAGE_SHIFT;
else if (e->type == GTT_TYPE_PPGTT_PTE_2M_ENTRY)
pfn = (e->val64 & ADDR_2M_MASK) >> PAGE_SHIFT;
+ else if (e->type == GTT_TYPE_PPGTT_PTE_64K_ENTRY)
+ pfn = (e->val64 & ADDR_64K_MASK) >> PAGE_SHIFT;
else
pfn = (e->val64 & ADDR_4K_MASK) >> PAGE_SHIFT;
return pfn;
} else if (e->type == GTT_TYPE_PPGTT_PTE_2M_ENTRY) {
e->val64 &= ~ADDR_2M_MASK;
pfn &= (ADDR_2M_MASK >> PAGE_SHIFT);
+ } else if (e->type == GTT_TYPE_PPGTT_PTE_64K_ENTRY) {
+ e->val64 &= ~ADDR_64K_MASK;
+ pfn &= (ADDR_64K_MASK >> PAGE_SHIFT);
} else {
e->val64 &= ~ADDR_4K_MASK;
pfn &= (ADDR_4K_MASK >> PAGE_SHIFT);
static bool gen8_gtt_test_pse(struct intel_gvt_gtt_entry *e)
{
- /* Entry doesn't have PSE bit. */
- if (get_pse_type(e->type) == GTT_TYPE_INVALID)
- return false;
+ return !!(e->val64 & _PAGE_PSE);
+}
- e->type = get_entry_type(e->type);
- if (!(e->val64 & _PAGE_PSE))
+static void gen8_gtt_clear_pse(struct intel_gvt_gtt_entry *e)
+{
+ if (gen8_gtt_test_pse(e)) {
+ switch (e->type) {
+ case GTT_TYPE_PPGTT_PTE_2M_ENTRY:
+ e->val64 &= ~_PAGE_PSE;
+ e->type = GTT_TYPE_PPGTT_PDE_ENTRY;
+ break;
+ case GTT_TYPE_PPGTT_PTE_1G_ENTRY:
+ e->type = GTT_TYPE_PPGTT_PDP_ENTRY;
+ e->val64 &= ~_PAGE_PSE;
+ break;
+ default:
+ WARN_ON(1);
+ }
+ }
+}
+
+static bool gen8_gtt_test_ips(struct intel_gvt_gtt_entry *e)
+{
+ if (GEM_WARN_ON(e->type != GTT_TYPE_PPGTT_PDE_ENTRY))
return false;
- e->type = get_pse_type(e->type);
- return true;
+ return !!(e->val64 & GEN8_PDE_IPS_64K);
+}
+
+static void gen8_gtt_clear_ips(struct intel_gvt_gtt_entry *e)
+{
+ if (GEM_WARN_ON(e->type != GTT_TYPE_PPGTT_PDE_ENTRY))
+ return;
+
+ e->val64 &= ~GEN8_PDE_IPS_64K;
}
static bool gen8_gtt_test_present(struct intel_gvt_gtt_entry *e)
e->val64 |= _PAGE_PRESENT;
}
+static bool gen8_gtt_test_64k_splited(struct intel_gvt_gtt_entry *e)
+{
+ return !!(e->val64 & GTT_SPTE_FLAG_64K_SPLITED);
+}
+
+static void gen8_gtt_set_64k_splited(struct intel_gvt_gtt_entry *e)
+{
+ e->val64 |= GTT_SPTE_FLAG_64K_SPLITED;
+}
+
+static void gen8_gtt_clear_64k_splited(struct intel_gvt_gtt_entry *e)
+{
+ e->val64 &= ~GTT_SPTE_FLAG_64K_SPLITED;
+}
+
/*
* Per-platform GMA routines.
*/
.set_present = gtt_entry_set_present,
.test_present = gen8_gtt_test_present,
.test_pse = gen8_gtt_test_pse,
+ .clear_pse = gen8_gtt_clear_pse,
+ .clear_ips = gen8_gtt_clear_ips,
+ .test_ips = gen8_gtt_test_ips,
+ .clear_64k_splited = gen8_gtt_clear_64k_splited,
+ .set_64k_splited = gen8_gtt_set_64k_splited,
+ .test_64k_splited = gen8_gtt_test_64k_splited,
.get_pfn = gen8_gtt_get_pfn,
.set_pfn = gen8_gtt_set_pfn,
};
.gma_to_pml4_index = gen8_gma_to_pml4_index,
};
+/* Update entry type per pse and ips bit. */
+static void update_entry_type_for_real(struct intel_gvt_gtt_pte_ops *pte_ops,
+ struct intel_gvt_gtt_entry *entry, bool ips)
+{
+ switch (entry->type) {
+ case GTT_TYPE_PPGTT_PDE_ENTRY:
+ case GTT_TYPE_PPGTT_PDP_ENTRY:
+ if (pte_ops->test_pse(entry))
+ entry->type = get_pse_type(entry->type);
+ break;
+ case GTT_TYPE_PPGTT_PTE_4K_ENTRY:
+ if (ips)
+ entry->type = get_pse_type(entry->type);
+ break;
+ default:
+ GEM_BUG_ON(!gtt_type_is_entry(entry->type));
+ }
+
+ GEM_BUG_ON(entry->type == GTT_TYPE_INVALID);
+}
+
/*
* MM helpers.
*/
pte_ops->get_entry(guest ? mm->ppgtt_mm.guest_pdps :
mm->ppgtt_mm.shadow_pdps,
entry, index, false, 0, mm->vgpu);
-
- pte_ops->test_pse(entry);
+ update_entry_type_for_real(pte_ops, entry, false);
}
static inline void ppgtt_get_guest_root_entry(struct intel_vgpu_mm *mm,
if (ret)
return ret;
- ops->test_pse(e);
+ update_entry_type_for_real(ops, e, guest ?
+ spt->guest_page.pde_ips : false);
gvt_vdbg_mm("read ppgtt entry, spt type %d, entry type %d, index %lu, value %llx\n",
type, e->type, index, e->val64);
radix_tree_delete(&spt->vgpu->gtt.spt_tree, spt->shadow_page.mfn);
- if (spt->guest_page.oos_page)
- detach_oos_page(spt->vgpu, spt->guest_page.oos_page);
+ if (spt->guest_page.gfn) {
+ if (spt->guest_page.oos_page)
+ detach_oos_page(spt->vgpu, spt->guest_page.oos_page);
- intel_vgpu_unregister_page_track(spt->vgpu, spt->guest_page.gfn);
+ intel_vgpu_unregister_page_track(spt->vgpu, spt->guest_page.gfn);
+ }
list_del_init(&spt->post_shadow_list);
free_spt(spt);
static int reclaim_one_ppgtt_mm(struct intel_gvt *gvt);
+/* Allocate shadow page table without guest page. */
static struct intel_vgpu_ppgtt_spt *ppgtt_alloc_spt(
- struct intel_vgpu *vgpu, int type, unsigned long gfn)
+ struct intel_vgpu *vgpu, intel_gvt_gtt_type_t type)
{
struct device *kdev = &vgpu->gvt->dev_priv->drm.pdev->dev;
struct intel_vgpu_ppgtt_spt *spt = NULL;
spt->shadow_page.vaddr = page_address(spt->shadow_page.page);
spt->shadow_page.mfn = daddr >> I915_GTT_PAGE_SHIFT;
- /*
- * Init guest_page.
- */
- spt->guest_page.type = type;
- spt->guest_page.gfn = gfn;
-
- ret = intel_vgpu_register_page_track(vgpu, spt->guest_page.gfn,
- ppgtt_write_protection_handler, spt);
- if (ret)
- goto err_unmap_dma;
-
ret = radix_tree_insert(&vgpu->gtt.spt_tree, spt->shadow_page.mfn, spt);
if (ret)
- goto err_unreg_page_track;
+ goto err_unmap_dma;
- trace_spt_alloc(vgpu->id, spt, type, spt->shadow_page.mfn, gfn);
return spt;
-err_unreg_page_track:
- intel_vgpu_unregister_page_track(vgpu, spt->guest_page.gfn);
err_unmap_dma:
dma_unmap_page(kdev, daddr, PAGE_SIZE, PCI_DMA_BIDIRECTIONAL);
err_free_spt:
return ERR_PTR(ret);
}
+/* Allocate shadow page table associated with specific gfn. */
+static struct intel_vgpu_ppgtt_spt *ppgtt_alloc_spt_gfn(
+ struct intel_vgpu *vgpu, intel_gvt_gtt_type_t type,
+ unsigned long gfn, bool guest_pde_ips)
+{
+ struct intel_vgpu_ppgtt_spt *spt;
+ int ret;
+
+ spt = ppgtt_alloc_spt(vgpu, type);
+ if (IS_ERR(spt))
+ return spt;
+
+ /*
+ * Init guest_page.
+ */
+ ret = intel_vgpu_register_page_track(vgpu, gfn,
+ ppgtt_write_protection_handler, spt);
+ if (ret) {
+ ppgtt_free_spt(spt);
+ return ERR_PTR(ret);
+ }
+
+ spt->guest_page.type = type;
+ spt->guest_page.gfn = gfn;
+ spt->guest_page.pde_ips = guest_pde_ips;
+
+ trace_spt_alloc(vgpu->id, spt, type, spt->shadow_page.mfn, gfn);
+
+ return spt;
+}
+
#define pt_entry_size_shift(spt) \
((spt)->vgpu->gvt->device_info.gtt_entry_size_shift)
(I915_GTT_PAGE_SIZE >> pt_entry_size_shift(spt))
#define for_each_present_guest_entry(spt, e, i) \
- for (i = 0; i < pt_entries(spt); i++) \
+ for (i = 0; i < pt_entries(spt); \
+ i += spt->guest_page.pde_ips ? GTT_64K_PTE_STRIDE : 1) \
if (!ppgtt_get_guest_entry(spt, e, i) && \
spt->vgpu->gvt->gtt.pte_ops->test_present(e))
#define for_each_present_shadow_entry(spt, e, i) \
- for (i = 0; i < pt_entries(spt); i++) \
+ for (i = 0; i < pt_entries(spt); \
+ i += spt->shadow_page.pde_ips ? GTT_64K_PTE_STRIDE : 1) \
if (!ppgtt_get_shadow_entry(spt, e, i) && \
spt->vgpu->gvt->gtt.pte_ops->test_present(e))
-static void ppgtt_get_spt(struct intel_vgpu_ppgtt_spt *spt)
+#define for_each_shadow_entry(spt, e, i) \
+ for (i = 0; i < pt_entries(spt); \
+ i += (spt->shadow_page.pde_ips ? GTT_64K_PTE_STRIDE : 1)) \
+ if (!ppgtt_get_shadow_entry(spt, e, i))
+
+static inline void ppgtt_get_spt(struct intel_vgpu_ppgtt_spt *spt)
{
int v = atomic_read(&spt->refcount);
trace_spt_refcount(spt->vgpu->id, "inc", spt, v, (v + 1));
-
atomic_inc(&spt->refcount);
}
+static inline int ppgtt_put_spt(struct intel_vgpu_ppgtt_spt *spt)
+{
+ int v = atomic_read(&spt->refcount);
+
+ trace_spt_refcount(spt->vgpu->id, "dec", spt, v, (v - 1));
+ return atomic_dec_return(&spt->refcount);
+}
+
static int ppgtt_invalidate_spt(struct intel_vgpu_ppgtt_spt *spt);
static int ppgtt_invalidate_spt_by_shadow_entry(struct intel_vgpu *vgpu,
pfn = ops->get_pfn(entry);
type = spt->shadow_page.type;
- if (pfn == vgpu->gtt.scratch_pt[type].page_mfn)
+ /* Uninitialized spte or unshadowed spte. */
+ if (!pfn || pfn == vgpu->gtt.scratch_pt[type].page_mfn)
return;
intel_gvt_hypervisor_dma_unmap_guest_page(vgpu, pfn << PAGE_SHIFT);
struct intel_gvt_gtt_entry e;
unsigned long index;
int ret;
- int v = atomic_read(&spt->refcount);
trace_spt_change(spt->vgpu->id, "die", spt,
spt->guest_page.gfn, spt->shadow_page.type);
- trace_spt_refcount(spt->vgpu->id, "dec", spt, v, (v - 1));
-
- if (atomic_dec_return(&spt->refcount) > 0)
+ if (ppgtt_put_spt(spt) > 0)
return 0;
for_each_present_shadow_entry(spt, &e, index) {
gvt_vdbg_mm("invalidate 4K entry\n");
ppgtt_invalidate_pte(spt, &e);
break;
+ case GTT_TYPE_PPGTT_PTE_64K_ENTRY:
+ /* We don't setup 64K shadow entry so far. */
+ WARN(1, "suspicious 64K gtt entry\n");
+ continue;
case GTT_TYPE_PPGTT_PTE_2M_ENTRY:
+ gvt_vdbg_mm("invalidate 2M entry\n");
+ continue;
case GTT_TYPE_PPGTT_PTE_1G_ENTRY:
- WARN(1, "GVT doesn't support 2M/1GB page\n");
+ WARN(1, "GVT doesn't support 1GB page\n");
continue;
case GTT_TYPE_PPGTT_PML4_ENTRY:
case GTT_TYPE_PPGTT_PDP_ENTRY:
return ret;
}
+static bool vgpu_ips_enabled(struct intel_vgpu *vgpu)
+{
+ struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv;
+
+ if (INTEL_GEN(dev_priv) == 9 || INTEL_GEN(dev_priv) == 10) {
+ u32 ips = vgpu_vreg_t(vgpu, GEN8_GAMW_ECO_DEV_RW_IA) &
+ GAMW_ECO_ENABLE_64K_IPS_FIELD;
+
+ return ips == GAMW_ECO_ENABLE_64K_IPS_FIELD;
+ } else if (INTEL_GEN(dev_priv) >= 11) {
+ /* 64K paging only controlled by IPS bit in PTE now. */
+ return true;
+ } else
+ return false;
+}
+
static int ppgtt_populate_spt(struct intel_vgpu_ppgtt_spt *spt);
static struct intel_vgpu_ppgtt_spt *ppgtt_populate_spt_by_guest_entry(
{
struct intel_gvt_gtt_pte_ops *ops = vgpu->gvt->gtt.pte_ops;
struct intel_vgpu_ppgtt_spt *spt = NULL;
+ bool ips = false;
int ret;
GEM_BUG_ON(!gtt_type_is_pt(get_next_pt_type(we->type)));
+ if (we->type == GTT_TYPE_PPGTT_PDE_ENTRY)
+ ips = vgpu_ips_enabled(vgpu) && ops->test_ips(we);
+
spt = intel_vgpu_find_spt_by_gfn(vgpu, ops->get_pfn(we));
- if (spt)
+ if (spt) {
ppgtt_get_spt(spt);
- else {
+
+ if (ips != spt->guest_page.pde_ips) {
+ spt->guest_page.pde_ips = ips;
+
+ gvt_dbg_mm("reshadow PDE since ips changed\n");
+ clear_page(spt->shadow_page.vaddr);
+ ret = ppgtt_populate_spt(spt);
+ if (ret) {
+ ppgtt_put_spt(spt);
+ goto err;
+ }
+ }
+ } else {
int type = get_next_pt_type(we->type);
- spt = ppgtt_alloc_spt(vgpu, type, ops->get_pfn(we));
+ spt = ppgtt_alloc_spt_gfn(vgpu, type, ops->get_pfn(we), ips);
if (IS_ERR(spt)) {
ret = PTR_ERR(spt);
- goto fail;
+ goto err;
}
ret = intel_vgpu_enable_page_track(vgpu, spt->guest_page.gfn);
if (ret)
- goto fail;
+ goto err_free_spt;
ret = ppgtt_populate_spt(spt);
if (ret)
- goto fail;
+ goto err_free_spt;
trace_spt_change(vgpu->id, "new", spt, spt->guest_page.gfn,
spt->shadow_page.type);
}
return spt;
-fail:
+
+err_free_spt:
+ ppgtt_free_spt(spt);
+err:
gvt_vgpu_err("fail: shadow page %p guest entry 0x%llx type %d\n",
spt, we->val64, we->type);
return ERR_PTR(ret);
se->type = ge->type;
se->val64 = ge->val64;
+ /* Because we always split 64KB pages, so clear IPS in shadow PDE. */
+ if (se->type == GTT_TYPE_PPGTT_PDE_ENTRY)
+ ops->clear_ips(se);
+
ops->set_pfn(se, s->shadow_page.mfn);
}
+/**
+ * Return 1 if 2MB huge gtt shadowing is possilbe, 0 if miscondition,
+ * negtive if found err.
+ */
+static int is_2MB_gtt_possible(struct intel_vgpu *vgpu,
+ struct intel_gvt_gtt_entry *entry)
+{
+ struct intel_gvt_gtt_pte_ops *ops = vgpu->gvt->gtt.pte_ops;
+ unsigned long pfn;
+
+ if (!HAS_PAGE_SIZES(vgpu->gvt->dev_priv, I915_GTT_PAGE_SIZE_2M))
+ return 0;
+
+ pfn = intel_gvt_hypervisor_gfn_to_mfn(vgpu, ops->get_pfn(entry));
+ if (pfn == INTEL_GVT_INVALID_ADDR)
+ return -EINVAL;
+
+ return PageTransHuge(pfn_to_page(pfn));
+}
+
+static int split_2MB_gtt_entry(struct intel_vgpu *vgpu,
+ struct intel_vgpu_ppgtt_spt *spt, unsigned long index,
+ struct intel_gvt_gtt_entry *se)
+{
+ struct intel_gvt_gtt_pte_ops *ops = vgpu->gvt->gtt.pte_ops;
+ struct intel_vgpu_ppgtt_spt *sub_spt;
+ struct intel_gvt_gtt_entry sub_se;
+ unsigned long start_gfn;
+ dma_addr_t dma_addr;
+ unsigned long sub_index;
+ int ret;
+
+ gvt_dbg_mm("Split 2M gtt entry, index %lu\n", index);
+
+ start_gfn = ops->get_pfn(se);
+
+ sub_spt = ppgtt_alloc_spt(vgpu, GTT_TYPE_PPGTT_PTE_PT);
+ if (IS_ERR(sub_spt))
+ return PTR_ERR(sub_spt);
+
+ for_each_shadow_entry(sub_spt, &sub_se, sub_index) {
+ ret = intel_gvt_hypervisor_dma_map_guest_page(vgpu,
+ start_gfn + sub_index, PAGE_SIZE, &dma_addr);
+ if (ret) {
+ ppgtt_invalidate_spt(spt);
+ return ret;
+ }
+ sub_se.val64 = se->val64;
+
+ /* Copy the PAT field from PDE. */
+ sub_se.val64 &= ~_PAGE_PAT;
+ sub_se.val64 |= (se->val64 & _PAGE_PAT_LARGE) >> 5;
+
+ ops->set_pfn(&sub_se, dma_addr >> PAGE_SHIFT);
+ ppgtt_set_shadow_entry(sub_spt, &sub_se, sub_index);
+ }
+
+ /* Clear dirty field. */
+ se->val64 &= ~_PAGE_DIRTY;
+
+ ops->clear_pse(se);
+ ops->clear_ips(se);
+ ops->set_pfn(se, sub_spt->shadow_page.mfn);
+ ppgtt_set_shadow_entry(spt, se, index);
+ return 0;
+}
+
+static int split_64KB_gtt_entry(struct intel_vgpu *vgpu,
+ struct intel_vgpu_ppgtt_spt *spt, unsigned long index,
+ struct intel_gvt_gtt_entry *se)
+{
+ struct intel_gvt_gtt_pte_ops *ops = vgpu->gvt->gtt.pte_ops;
+ struct intel_gvt_gtt_entry entry = *se;
+ unsigned long start_gfn;
+ dma_addr_t dma_addr;
+ int i, ret;
+
+ gvt_vdbg_mm("Split 64K gtt entry, index %lu\n", index);
+
+ GEM_BUG_ON(index % GTT_64K_PTE_STRIDE);
+
+ start_gfn = ops->get_pfn(se);
+
+ entry.type = GTT_TYPE_PPGTT_PTE_4K_ENTRY;
+ ops->set_64k_splited(&entry);
+
+ for (i = 0; i < GTT_64K_PTE_STRIDE; i++) {
+ ret = intel_gvt_hypervisor_dma_map_guest_page(vgpu,
+ start_gfn + i, PAGE_SIZE, &dma_addr);
+ if (ret)
+ return ret;
+
+ ops->set_pfn(&entry, dma_addr >> PAGE_SHIFT);
+ ppgtt_set_shadow_entry(spt, &entry, index + i);
+ }
+ return 0;
+}
+
static int ppgtt_populate_shadow_entry(struct intel_vgpu *vgpu,
struct intel_vgpu_ppgtt_spt *spt, unsigned long index,
struct intel_gvt_gtt_entry *ge)
{
struct intel_gvt_gtt_pte_ops *pte_ops = vgpu->gvt->gtt.pte_ops;
struct intel_gvt_gtt_entry se = *ge;
- unsigned long gfn;
+ unsigned long gfn, page_size = PAGE_SIZE;
dma_addr_t dma_addr;
int ret;
case GTT_TYPE_PPGTT_PTE_4K_ENTRY:
gvt_vdbg_mm("shadow 4K gtt entry\n");
break;
+ case GTT_TYPE_PPGTT_PTE_64K_ENTRY:
+ gvt_vdbg_mm("shadow 64K gtt entry\n");
+ /*
+ * The layout of 64K page is special, the page size is
+ * controlled by uper PDE. To be simple, we always split
+ * 64K page to smaller 4K pages in shadow PT.
+ */
+ return split_64KB_gtt_entry(vgpu, spt, index, &se);
case GTT_TYPE_PPGTT_PTE_2M_ENTRY:
+ gvt_vdbg_mm("shadow 2M gtt entry\n");
+ ret = is_2MB_gtt_possible(vgpu, ge);
+ if (ret == 0)
+ return split_2MB_gtt_entry(vgpu, spt, index, &se);
+ else if (ret < 0)
+ return ret;
+ page_size = I915_GTT_PAGE_SIZE_2M;
+ break;
case GTT_TYPE_PPGTT_PTE_1G_ENTRY:
- gvt_vgpu_err("GVT doesn't support 2M/1GB entry\n");
+ gvt_vgpu_err("GVT doesn't support 1GB entry\n");
return -EINVAL;
default:
GEM_BUG_ON(1);
};
/* direct shadow */
- ret = intel_gvt_hypervisor_dma_map_guest_page(vgpu, gfn, &dma_addr);
+ ret = intel_gvt_hypervisor_dma_map_guest_page(vgpu, gfn, page_size,
+ &dma_addr);
if (ret)
return -ENXIO;
ret = ppgtt_invalidate_spt(s);
if (ret)
goto fail;
- } else
+ } else {
+ /* We don't setup 64K shadow entry so far. */
+ WARN(se->type == GTT_TYPE_PPGTT_PTE_64K_ENTRY,
+ "suspicious 64K entry\n");
ppgtt_invalidate_pte(spt, se);
+ }
return 0;
fail:
struct intel_gvt_gtt_pte_ops *ops = vgpu->gvt->gtt.pte_ops;
struct intel_gvt_gtt_entry old_se;
int new_present;
- int ret;
+ int i, ret;
new_present = ops->test_present(we);
goto fail;
if (!new_present) {
- ops->set_pfn(&old_se, vgpu->gtt.scratch_pt[type].page_mfn);
- ppgtt_set_shadow_entry(spt, &old_se, index);
+ /* For 64KB splited entries, we need clear them all. */
+ if (ops->test_64k_splited(&old_se) &&
+ !(index % GTT_64K_PTE_STRIDE)) {
+ gvt_vdbg_mm("remove splited 64K shadow entries\n");
+ for (i = 0; i < GTT_64K_PTE_STRIDE; i++) {
+ ops->clear_64k_splited(&old_se);
+ ops->set_pfn(&old_se,
+ vgpu->gtt.scratch_pt[type].page_mfn);
+ ppgtt_set_shadow_entry(spt, &old_se, index + i);
+ }
+ } else if (old_se.type == GTT_TYPE_PPGTT_PTE_2M_ENTRY ||
+ old_se.type == GTT_TYPE_PPGTT_PTE_1G_ENTRY) {
+ ops->clear_pse(&old_se);
+ ops->set_pfn(&old_se,
+ vgpu->gtt.scratch_pt[type].page_mfn);
+ ppgtt_set_shadow_entry(spt, &old_se, index);
+ } else {
+ ops->set_pfn(&old_se,
+ vgpu->gtt.scratch_pt[type].page_mfn);
+ ppgtt_set_shadow_entry(spt, &old_se, index);
+ }
}
return 0;
ppgtt_get_guest_entry(spt, &we, index);
- ops->test_pse(&we);
+ /*
+ * For page table which has 64K gtt entry, only PTE#0, PTE#16,
+ * PTE#32, ... PTE#496 are used. Unused PTEs update should be
+ * ignored.
+ */
+ if (we.type == GTT_TYPE_PPGTT_PTE_64K_ENTRY &&
+ (index % GTT_64K_PTE_STRIDE)) {
+ gvt_vdbg_mm("Ignore write to unused PTE entry, index %lu\n",
+ index);
+ return 0;
+ }
if (bytes == info->gtt_entry_size) {
ret = ppgtt_handle_guest_write_page_table(spt, &we, index);
vgpu_free_mm(mm);
return ERR_PTR(-ENOMEM);
}
+ mm->ggtt_mm.last_partial_off = -1UL;
return mm;
}
invalidate_ppgtt_mm(mm);
} else {
vfree(mm->ggtt_mm.virtual_ggtt);
+ mm->ggtt_mm.last_partial_off = -1UL;
}
vgpu_free_mm(mm);
memcpy((void *)&e.val64 + (off & (info->gtt_entry_size - 1)), p_data,
bytes);
+ /* If ggtt entry size is 8 bytes, and it's split into two 4 bytes
+ * write, we assume the two 4 bytes writes are consecutive.
+ * Otherwise, we abort and report error
+ */
+ if (bytes < info->gtt_entry_size) {
+ if (ggtt_mm->ggtt_mm.last_partial_off == -1UL) {
+ /* the first partial part*/
+ ggtt_mm->ggtt_mm.last_partial_off = off;
+ ggtt_mm->ggtt_mm.last_partial_data = e.val64;
+ return 0;
+ } else if ((g_gtt_index ==
+ (ggtt_mm->ggtt_mm.last_partial_off >>
+ info->gtt_entry_size_shift)) &&
+ (off != ggtt_mm->ggtt_mm.last_partial_off)) {
+ /* the second partial part */
+
+ int last_off = ggtt_mm->ggtt_mm.last_partial_off &
+ (info->gtt_entry_size - 1);
+
+ memcpy((void *)&e.val64 + last_off,
+ (void *)&ggtt_mm->ggtt_mm.last_partial_data +
+ last_off, bytes);
+
+ ggtt_mm->ggtt_mm.last_partial_off = -1UL;
+ } else {
+ int last_offset;
+
+ gvt_vgpu_err("failed to populate guest ggtt entry: abnormal ggtt entry write sequence, last_partial_off=%lx, offset=%x, bytes=%d, ggtt entry size=%d\n",
+ ggtt_mm->ggtt_mm.last_partial_off, off,
+ bytes, info->gtt_entry_size);
+
+ /* set host ggtt entry to scratch page and clear
+ * virtual ggtt entry as not present for last
+ * partially write offset
+ */
+ last_offset = ggtt_mm->ggtt_mm.last_partial_off &
+ (~(info->gtt_entry_size - 1));
+
+ ggtt_get_host_entry(ggtt_mm, &m, last_offset);
+ ggtt_invalidate_pte(vgpu, &m);
+ ops->set_pfn(&m, gvt->gtt.scratch_mfn);
+ ops->clear_present(&m);
+ ggtt_set_host_entry(ggtt_mm, &m, last_offset);
+ ggtt_invalidate(gvt->dev_priv);
+
+ ggtt_get_guest_entry(ggtt_mm, &e, last_offset);
+ ops->clear_present(&e);
+ ggtt_set_guest_entry(ggtt_mm, &e, last_offset);
+
+ ggtt_mm->ggtt_mm.last_partial_off = off;
+ ggtt_mm->ggtt_mm.last_partial_data = e.val64;
+
+ return 0;
+ }
+ }
+
if (ops->test_present(&e)) {
gfn = ops->get_pfn(&e);
m = e;
}
ret = intel_gvt_hypervisor_dma_map_guest_page(vgpu, gfn,
- &dma_addr);
+ PAGE_SIZE, &dma_addr);
if (ret) {
gvt_vgpu_err("fail to populate guest ggtt entry\n");
/* guest driver may read/write the entry when partial
* GTT_TYPE_PPGTT_PDE_PT level pt, that means this scratch_pt it self
* is GTT_TYPE_PPGTT_PTE_PT, and full filled by scratch page mfn.
*/
- if (type > GTT_TYPE_PPGTT_PTE_PT && type < GTT_TYPE_MAX) {
+ if (type > GTT_TYPE_PPGTT_PTE_PT) {
struct intel_gvt_gtt_entry se;
memset(&se, 0, sizeof(struct intel_gvt_gtt_entry));
gvt_dbg_core("init gtt\n");
- if (IS_BROADWELL(gvt->dev_priv) || IS_SKYLAKE(gvt->dev_priv)
- || IS_KABYLAKE(gvt->dev_priv)) {
- gvt->gtt.pte_ops = &gen8_gtt_pte_ops;
- gvt->gtt.gma_ops = &gen8_gtt_gma_ops;
- } else {
- return -ENODEV;
- }
+ gvt->gtt.pte_ops = &gen8_gtt_pte_ops;
+ gvt->gtt.gma_ops = &gen8_gtt_gma_ops;
page = (void *)get_zeroed_page(GFP_KERNEL);
if (!page) {
void (*clear_present)(struct intel_gvt_gtt_entry *e);
void (*set_present)(struct intel_gvt_gtt_entry *e);
bool (*test_pse)(struct intel_gvt_gtt_entry *e);
+ void (*clear_pse)(struct intel_gvt_gtt_entry *e);
+ bool (*test_ips)(struct intel_gvt_gtt_entry *e);
+ void (*clear_ips)(struct intel_gvt_gtt_entry *e);
+ bool (*test_64k_splited)(struct intel_gvt_gtt_entry *e);
+ void (*clear_64k_splited)(struct intel_gvt_gtt_entry *e);
+ void (*set_64k_splited)(struct intel_gvt_gtt_entry *e);
void (*set_pfn)(struct intel_gvt_gtt_entry *e, unsigned long pfn);
unsigned long (*get_pfn)(struct intel_gvt_gtt_entry *e);
};
GTT_TYPE_GGTT_PTE,
GTT_TYPE_PPGTT_PTE_4K_ENTRY,
+ GTT_TYPE_PPGTT_PTE_64K_ENTRY,
GTT_TYPE_PPGTT_PTE_2M_ENTRY,
GTT_TYPE_PPGTT_PTE_1G_ENTRY,
} ppgtt_mm;
struct {
void *virtual_ggtt;
+ unsigned long last_partial_off;
+ u64 last_partial_data;
} ggtt_mm;
};
};
struct {
intel_gvt_gtt_type_t type;
+ bool pde_ips; /* for 64KB PTEs */
void *vaddr;
struct page *page;
unsigned long mfn;
struct {
intel_gvt_gtt_type_t type;
+ bool pde_ips; /* for 64KB PTEs */
unsigned long gfn;
unsigned long write_cnt;
struct intel_vgpu_oos_page *oos_page;
struct intel_vgpu {
struct intel_gvt *gvt;
+ struct mutex vgpu_lock;
int id;
unsigned long handle; /* vGPU handle used by hypervisor MPT modules */
bool active;
bool pv_notified;
bool failsafe;
unsigned int resetting_eng;
+
+ /* Both sched_data and sched_ctl can be seen a part of the global gvt
+ * scheduler structure. So below 2 vgpu data are protected
+ * by sched_lock, not vgpu_lock.
+ */
void *sched_data;
struct vgpu_sched_ctl sched_ctl;
#define F_CMD_ACCESSED (1 << 5)
/* This reg could be accessed by unaligned address */
#define F_UNALIGN (1 << 6)
+ /* This reg is saved/restored in context */
+ #define F_IN_CTX (1 << 7)
struct gvt_mmio_block *mmio_block;
unsigned int num_mmio_block;
};
struct intel_gvt {
+ /* GVT scope lock, protect GVT itself, and all resource currently
+ * not yet protected by special locks(vgpu and scheduler lock).
+ */
struct mutex lock;
+ /* scheduler scope lock, protect gvt and vgpu schedule related data */
+ struct mutex sched_lock;
+
struct drm_i915_private *dev_priv;
struct idr vgpu_idr; /* vGPU IDR pool */
struct task_struct *service_thread;
wait_queue_head_t service_thread_wq;
+
+ /* service_request is always used in bit operation, we should always
+ * use it with atomic bit ops so that no need to use gvt big lock.
+ */
unsigned long service_request;
struct {
#define gvt_aperture_sz(gvt) (gvt->dev_priv->ggtt.mappable_end)
#define gvt_aperture_pa_base(gvt) (gvt->dev_priv->ggtt.gmadr.start)
-#define gvt_ggtt_gm_sz(gvt) (gvt->dev_priv->ggtt.base.total)
+#define gvt_ggtt_gm_sz(gvt) (gvt->dev_priv->ggtt.vm.total)
#define gvt_ggtt_sz(gvt) \
- ((gvt->dev_priv->ggtt.base.total >> PAGE_SHIFT) << 3)
+ ((gvt->dev_priv->ggtt.vm.total >> PAGE_SHIFT) << 3)
#define gvt_hidden_sz(gvt) (gvt_ggtt_gm_sz(gvt) - gvt_aperture_sz(gvt))
#define gvt_aperture_gmadr_base(gvt) (0)
return gvt->mmio.mmio_attribute[offset >> 2] & F_MODE_MASK;
}
+ /**
+ * intel_gvt_mmio_is_in_ctx - check if a MMIO has in-ctx mask
+ * @gvt: a GVT device
+ * @offset: register offset
+ *
+ * Returns:
+ * True if a MMIO has a in-context mask, false if it isn't.
+ *
+ */
+ static inline bool intel_gvt_mmio_is_in_ctx(
+ struct intel_gvt *gvt, unsigned int offset)
+ {
+ return gvt->mmio.mmio_attribute[offset >> 2] & F_IN_CTX;
+ }
+
+ /**
+ * intel_gvt_mmio_set_in_ctx - mask a MMIO in logical context
+ * @gvt: a GVT device
+ * @offset: register offset
+ *
+ */
+ static inline void intel_gvt_mmio_set_in_ctx(
+ struct intel_gvt *gvt, unsigned int offset)
+ {
+ gvt->mmio.mmio_attribute[offset >> 2] |= F_IN_CTX;
+ }
+
int intel_gvt_debugfs_add_vgpu(struct intel_vgpu *vgpu);
void intel_gvt_debugfs_remove_vgpu(struct intel_vgpu *vgpu);
int intel_gvt_debugfs_init(struct intel_gvt *gvt);
return D_SKL;
else if (IS_KABYLAKE(gvt->dev_priv))
return D_KBL;
+ else if (IS_BROXTON(gvt->dev_priv))
+ return D_BXT;
return 0;
}
return 0;
}
+static int gamw_echo_dev_rw_ia_write(struct intel_vgpu *vgpu,
+ unsigned int offset, void *p_data, unsigned int bytes)
+{
+ u32 ips = (*(u32 *)p_data) & GAMW_ECO_ENABLE_64K_IPS_FIELD;
+
+ if (INTEL_GEN(vgpu->gvt->dev_priv) <= 10) {
+ if (ips == GAMW_ECO_ENABLE_64K_IPS_FIELD)
+ gvt_dbg_core("vgpu%d: ips enabled\n", vgpu->id);
+ else if (!ips)
+ gvt_dbg_core("vgpu%d: ips disabled\n", vgpu->id);
+ else {
+ /* All engines must be enabled together for vGPU,
+ * since we don't know which engine the ppgtt will
+ * bind to when shadowing.
+ */
+ gvt_vgpu_err("Unsupported IPS setting %x, cannot enable 64K gtt.\n",
+ ips);
+ return -EINVAL;
+ }
+ }
+
+ write_vreg(vgpu, offset, p_data, bytes);
+ return 0;
+}
+
static int fence_mmio_read(struct intel_vgpu *vgpu, unsigned int off,
void *p_data, unsigned int bytes)
{
new = CALC_MODE_MASK_REG(old, *(u32 *)p_data);
if (IS_SKYLAKE(vgpu->gvt->dev_priv)
- || IS_KABYLAKE(vgpu->gvt->dev_priv)) {
+ || IS_KABYLAKE(vgpu->gvt->dev_priv)
+ || IS_BROXTON(vgpu->gvt->dev_priv)) {
switch (offset) {
case FORCEWAKE_RENDER_GEN9_REG:
ack_reg_offset = FORCEWAKE_ACK_RENDER_GEN9_REG;
}
}
+ /* vgpu_lock already hold by emulate mmio r/w */
intel_gvt_reset_vgpu_locked(vgpu, false, engine_mask);
/* sw will wait for the device to ack the reset request */
vgpu_vreg(vgpu, offset) |= I965_PIPECONF_ACTIVE;
else
vgpu_vreg(vgpu, offset) &= ~I965_PIPECONF_ACTIVE;
+ /* vgpu_lock already hold by emulate mmio r/w */
+ mutex_unlock(&vgpu->vgpu_lock);
intel_gvt_check_vblank_emulation(vgpu->gvt);
+ mutex_lock(&vgpu->vgpu_lock);
return 0;
}
data = vgpu_vreg(vgpu, offset);
if ((IS_SKYLAKE(vgpu->gvt->dev_priv)
- || IS_KABYLAKE(vgpu->gvt->dev_priv))
+ || IS_KABYLAKE(vgpu->gvt->dev_priv)
+ || IS_BROXTON(vgpu->gvt->dev_priv))
&& offset != _REG_SKL_DP_AUX_CH_CTL(port_index)) {
/* SKL DPB/C/D aux ctl register changed */
return 0;
ret = handle_g2v_notification(vgpu, data);
break;
/* add xhot and yhot to handled list to avoid error log */
- case 0x78830:
- case 0x78834:
+ case _vgtif_reg(cursor_x_hot):
+ case _vgtif_reg(cursor_y_hot):
case _vgtif_reg(pdp[0].lo):
case _vgtif_reg(pdp[0].hi):
case _vgtif_reg(pdp[1].lo):
*data0 = 0x1e1a1100;
else
*data0 = 0x61514b3d;
+ } else if (IS_BROXTON(vgpu->gvt->dev_priv)) {
+ /**
+ * "Read memory latency" command on gen9.
+ * Below memory latency values are read
+ * from Broxton MRB.
+ */
+ if (!*data0)
+ *data0 = 0x16080707;
+ else
+ *data0 = 0x16161616;
}
break;
case SKL_PCODE_CDCLK_CONTROL:
{
u32 v = *(u32 *)p_data;
- v &= (1 << 31) | (1 << 29) | (1 << 9) |
- (1 << 7) | (1 << 5) | (1 << 3) | (1 << 1);
+ if (IS_BROXTON(vgpu->gvt->dev_priv))
+ v &= (1 << 31) | (1 << 29);
+ else
+ v &= (1 << 31) | (1 << 29) | (1 << 9) |
+ (1 << 7) | (1 << 5) | (1 << 3) | (1 << 1);
v |= (v >> 1);
return intel_vgpu_default_mmio_write(vgpu, offset, &v, bytes);
return 0;
}
+static int bxt_de_pll_enable_write(struct intel_vgpu *vgpu,
+ unsigned int offset, void *p_data, unsigned int bytes)
+{
+ u32 v = *(u32 *)p_data;
+
+ if (v & BXT_DE_PLL_PLL_ENABLE)
+ v |= BXT_DE_PLL_LOCK;
+
+ vgpu_vreg(vgpu, offset) = v;
+
+ return 0;
+}
+
+static int bxt_port_pll_enable_write(struct intel_vgpu *vgpu,
+ unsigned int offset, void *p_data, unsigned int bytes)
+{
+ u32 v = *(u32 *)p_data;
+
+ if (v & PORT_PLL_ENABLE)
+ v |= PORT_PLL_LOCK;
+
+ vgpu_vreg(vgpu, offset) = v;
+
+ return 0;
+}
+
+static int bxt_phy_ctl_family_write(struct intel_vgpu *vgpu,
+ unsigned int offset, void *p_data, unsigned int bytes)
+{
+ u32 v = *(u32 *)p_data;
+ u32 data = v & COMMON_RESET_DIS ? BXT_PHY_LANE_ENABLED : 0;
+
+ vgpu_vreg(vgpu, _BXT_PHY_CTL_DDI_A) = data;
+ vgpu_vreg(vgpu, _BXT_PHY_CTL_DDI_B) = data;
+ vgpu_vreg(vgpu, _BXT_PHY_CTL_DDI_C) = data;
+
+ vgpu_vreg(vgpu, offset) = v;
+
+ return 0;
+}
+
+static int bxt_port_tx_dw3_read(struct intel_vgpu *vgpu,
+ unsigned int offset, void *p_data, unsigned int bytes)
+{
+ u32 v = vgpu_vreg(vgpu, offset);
+
+ v &= ~UNIQUE_TRANGE_EN_METHOD;
+
+ vgpu_vreg(vgpu, offset) = v;
+
+ return intel_vgpu_default_mmio_read(vgpu, offset, p_data, bytes);
+}
+
+static int bxt_pcs_dw12_grp_write(struct intel_vgpu *vgpu,
+ unsigned int offset, void *p_data, unsigned int bytes)
+{
+ u32 v = *(u32 *)p_data;
+
+ if (offset == _PORT_PCS_DW12_GRP_A || offset == _PORT_PCS_DW12_GRP_B) {
+ vgpu_vreg(vgpu, offset - 0x600) = v;
+ vgpu_vreg(vgpu, offset - 0x800) = v;
+ } else {
+ vgpu_vreg(vgpu, offset - 0x400) = v;
+ vgpu_vreg(vgpu, offset - 0x600) = v;
+ }
+
+ vgpu_vreg(vgpu, offset) = v;
+
+ return 0;
+}
+
+static int bxt_gt_disp_pwron_write(struct intel_vgpu *vgpu,
+ unsigned int offset, void *p_data, unsigned int bytes)
+{
+ u32 v = *(u32 *)p_data;
+
+ if (v & BIT(0)) {
+ vgpu_vreg_t(vgpu, BXT_PORT_CL1CM_DW0(DPIO_PHY0)) &=
+ ~PHY_RESERVED;
+ vgpu_vreg_t(vgpu, BXT_PORT_CL1CM_DW0(DPIO_PHY0)) |=
+ PHY_POWER_GOOD;
+ }
+
+ if (v & BIT(1)) {
+ vgpu_vreg_t(vgpu, BXT_PORT_CL1CM_DW0(DPIO_PHY1)) &=
+ ~PHY_RESERVED;
+ vgpu_vreg_t(vgpu, BXT_PORT_CL1CM_DW0(DPIO_PHY1)) |=
+ PHY_POWER_GOOD;
+ }
+
+
+ vgpu_vreg(vgpu, offset) = v;
+
+ return 0;
+}
+
+static int bxt_edp_psr_imr_iir_write(struct intel_vgpu *vgpu,
+ unsigned int offset, void *p_data, unsigned int bytes)
+{
+ vgpu_vreg(vgpu, offset) = 0;
+ return 0;
+}
+
static int mmio_read_from_hw(struct intel_vgpu *vgpu,
unsigned int offset, void *p_data, unsigned int bytes)
{
MMIO_RING_DFH(RING_HWSTAM, D_ALL, F_CMD_ACCESS, NULL, NULL);
- MMIO_GM_RDR(RENDER_HWS_PGA_GEN7, D_ALL, NULL, NULL);
+ MMIO_DH(GEN8_GAMW_ECO_DEV_RW_IA, D_BDW_PLUS, NULL,
+ gamw_echo_dev_rw_ia_write);
+
MMIO_GM_RDR(BSD_HWS_PGA_GEN7, D_ALL, NULL, NULL);
MMIO_GM_RDR(BLT_HWS_PGA_GEN7, D_ALL, NULL, NULL);
MMIO_GM_RDR(VEBOX_HWS_PGA_GEN7, D_ALL, NULL, NULL);
MMIO_D(_MMIO(0x45504), D_SKL_PLUS);
MMIO_D(_MMIO(0x45520), D_SKL_PLUS);
MMIO_D(_MMIO(0x46000), D_SKL_PLUS);
- MMIO_DH(_MMIO(0x46010), D_SKL | D_KBL, NULL, skl_lcpll_write);
- MMIO_DH(_MMIO(0x46014), D_SKL | D_KBL, NULL, skl_lcpll_write);
- MMIO_D(_MMIO(0x6C040), D_SKL | D_KBL);
- MMIO_D(_MMIO(0x6C048), D_SKL | D_KBL);
- MMIO_D(_MMIO(0x6C050), D_SKL | D_KBL);
- MMIO_D(_MMIO(0x6C044), D_SKL | D_KBL);
- MMIO_D(_MMIO(0x6C04C), D_SKL | D_KBL);
- MMIO_D(_MMIO(0x6C054), D_SKL | D_KBL);
- MMIO_D(_MMIO(0x6c058), D_SKL | D_KBL);
- MMIO_D(_MMIO(0x6c05c), D_SKL | D_KBL);
- MMIO_DH(_MMIO(0x6c060), D_SKL | D_KBL, dpll_status_read, NULL);
+ MMIO_DH(_MMIO(0x46010), D_SKL_PLUS, NULL, skl_lcpll_write);
+ MMIO_DH(_MMIO(0x46014), D_SKL_PLUS, NULL, skl_lcpll_write);
+ MMIO_D(_MMIO(0x6C040), D_SKL_PLUS);
+ MMIO_D(_MMIO(0x6C048), D_SKL_PLUS);
+ MMIO_D(_MMIO(0x6C050), D_SKL_PLUS);
+ MMIO_D(_MMIO(0x6C044), D_SKL_PLUS);
+ MMIO_D(_MMIO(0x6C04C), D_SKL_PLUS);
+ MMIO_D(_MMIO(0x6C054), D_SKL_PLUS);
+ MMIO_D(_MMIO(0x6c058), D_SKL_PLUS);
+ MMIO_D(_MMIO(0x6c05c), D_SKL_PLUS);
+ MMIO_DH(_MMIO(0x6c060), D_SKL_PLUS, dpll_status_read, NULL);
MMIO_DH(SKL_PS_WIN_POS(PIPE_A, 0), D_SKL_PLUS, NULL, pf_write);
MMIO_DH(SKL_PS_WIN_POS(PIPE_A, 1), D_SKL_PLUS, NULL, pf_write);
MMIO_D(_MMIO(0x7239c), D_SKL_PLUS);
MMIO_D(_MMIO(0x7039c), D_SKL_PLUS);
- MMIO_D(_MMIO(0x8f074), D_SKL | D_KBL);
- MMIO_D(_MMIO(0x8f004), D_SKL | D_KBL);
- MMIO_D(_MMIO(0x8f034), D_SKL | D_KBL);
+ MMIO_D(_MMIO(0x8f074), D_SKL_PLUS);
+ MMIO_D(_MMIO(0x8f004), D_SKL_PLUS);
+ MMIO_D(_MMIO(0x8f034), D_SKL_PLUS);
- MMIO_D(_MMIO(0xb11c), D_SKL | D_KBL);
+ MMIO_D(_MMIO(0xb11c), D_SKL_PLUS);
- MMIO_D(_MMIO(0x51000), D_SKL | D_KBL);
+ MMIO_D(_MMIO(0x51000), D_SKL_PLUS);
MMIO_D(_MMIO(0x6c00c), D_SKL_PLUS);
- MMIO_F(_MMIO(0xc800), 0x7f8, F_CMD_ACCESS, 0, 0, D_SKL | D_KBL, NULL, NULL);
- MMIO_F(_MMIO(0xb020), 0x80, F_CMD_ACCESS, 0, 0, D_SKL | D_KBL, NULL, NULL);
+ MMIO_F(_MMIO(0xc800), 0x7f8, F_CMD_ACCESS, 0, 0, D_SKL_PLUS,
+ NULL, NULL);
+ MMIO_F(_MMIO(0xb020), 0x80, F_CMD_ACCESS, 0, 0, D_SKL_PLUS,
+ NULL, NULL);
MMIO_D(RPM_CONFIG0, D_SKL_PLUS);
MMIO_D(_MMIO(0xd08), D_SKL_PLUS);
MMIO_D(RC6_LOCATION, D_SKL_PLUS);
MMIO_DFH(_MMIO(0x20e0), D_SKL_PLUS, F_MODE_MASK, NULL, NULL);
- MMIO_DFH(_MMIO(0x20ec), D_SKL_PLUS, F_MODE_MASK | F_CMD_ACCESS, NULL, NULL);
+ MMIO_DFH(_MMIO(0x20ec), D_SKL_PLUS, F_MODE_MASK | F_CMD_ACCESS,
+ NULL, NULL);
/* TRTT */
- MMIO_DFH(_MMIO(0x4de0), D_SKL | D_KBL, F_CMD_ACCESS, NULL, NULL);
- MMIO_DFH(_MMIO(0x4de4), D_SKL | D_KBL, F_CMD_ACCESS, NULL, NULL);
- MMIO_DFH(_MMIO(0x4de8), D_SKL | D_KBL, F_CMD_ACCESS, NULL, NULL);
- MMIO_DFH(_MMIO(0x4dec), D_SKL | D_KBL, F_CMD_ACCESS, NULL, NULL);
- MMIO_DFH(_MMIO(0x4df0), D_SKL | D_KBL, F_CMD_ACCESS, NULL, NULL);
- MMIO_DFH(_MMIO(0x4df4), D_SKL | D_KBL, F_CMD_ACCESS, NULL, gen9_trtte_write);
- MMIO_DH(_MMIO(0x4dfc), D_SKL | D_KBL, NULL, gen9_trtt_chicken_write);
+ MMIO_DFH(_MMIO(0x4de0), D_SKL_PLUS, F_CMD_ACCESS, NULL, NULL);
+ MMIO_DFH(_MMIO(0x4de4), D_SKL_PLUS, F_CMD_ACCESS, NULL, NULL);
+ MMIO_DFH(_MMIO(0x4de8), D_SKL_PLUS, F_CMD_ACCESS, NULL, NULL);
+ MMIO_DFH(_MMIO(0x4dec), D_SKL_PLUS, F_CMD_ACCESS, NULL, NULL);
+ MMIO_DFH(_MMIO(0x4df0), D_SKL_PLUS, F_CMD_ACCESS, NULL, NULL);
+ MMIO_DFH(_MMIO(0x4df4), D_SKL_PLUS, F_CMD_ACCESS,
+ NULL, gen9_trtte_write);
+ MMIO_DH(_MMIO(0x4dfc), D_SKL_PLUS, NULL, gen9_trtt_chicken_write);
- MMIO_D(_MMIO(0x45008), D_SKL | D_KBL);
+ MMIO_D(_MMIO(0x45008), D_SKL_PLUS);
- MMIO_D(_MMIO(0x46430), D_SKL | D_KBL);
+ MMIO_D(_MMIO(0x46430), D_SKL_PLUS);
- MMIO_D(_MMIO(0x46520), D_SKL | D_KBL);
+ MMIO_D(_MMIO(0x46520), D_SKL_PLUS);
- MMIO_D(_MMIO(0xc403c), D_SKL | D_KBL);
+ MMIO_D(_MMIO(0xc403c), D_SKL_PLUS);
MMIO_D(_MMIO(0xb004), D_SKL_PLUS);
MMIO_DH(DMA_CTRL, D_SKL_PLUS, NULL, dma_ctrl_write);
MMIO_D(_MMIO(0x65900), D_SKL_PLUS);
- MMIO_D(_MMIO(0x1082c0), D_SKL | D_KBL);
- MMIO_D(_MMIO(0x4068), D_SKL | D_KBL);
- MMIO_D(_MMIO(0x67054), D_SKL | D_KBL);
- MMIO_D(_MMIO(0x6e560), D_SKL | D_KBL);
- MMIO_D(_MMIO(0x6e554), D_SKL | D_KBL);
- MMIO_D(_MMIO(0x2b20), D_SKL | D_KBL);
- MMIO_D(_MMIO(0x65f00), D_SKL | D_KBL);
- MMIO_D(_MMIO(0x65f08), D_SKL | D_KBL);
- MMIO_D(_MMIO(0x320f0), D_SKL | D_KBL);
+ MMIO_D(_MMIO(0x1082c0), D_SKL_PLUS);
+ MMIO_D(_MMIO(0x4068), D_SKL_PLUS);
+ MMIO_D(_MMIO(0x67054), D_SKL_PLUS);
+ MMIO_D(_MMIO(0x6e560), D_SKL_PLUS);
+ MMIO_D(_MMIO(0x6e554), D_SKL_PLUS);
+ MMIO_D(_MMIO(0x2b20), D_SKL_PLUS);
+ MMIO_D(_MMIO(0x65f00), D_SKL_PLUS);
+ MMIO_D(_MMIO(0x65f08), D_SKL_PLUS);
+ MMIO_D(_MMIO(0x320f0), D_SKL_PLUS);
MMIO_D(_MMIO(0x70034), D_SKL_PLUS);
MMIO_D(_MMIO(0x71034), D_SKL_PLUS);
MMIO_D(_MMIO(0x44500), D_SKL_PLUS);
MMIO_DFH(GEN9_CSFE_CHICKEN1_RCS, D_SKL_PLUS, F_CMD_ACCESS, NULL, NULL);
- MMIO_DFH(GEN8_HDC_CHICKEN1, D_SKL | D_KBL, F_MODE_MASK | F_CMD_ACCESS,
+ MMIO_DFH(GEN8_HDC_CHICKEN1, D_SKL_PLUS, F_MODE_MASK | F_CMD_ACCESS,
NULL, NULL);
MMIO_D(_MMIO(0x4ab8), D_KBL);
- MMIO_D(_MMIO(0x2248), D_SKL_PLUS | D_KBL);
+ MMIO_D(_MMIO(0x2248), D_KBL | D_SKL);
+
+ return 0;
+}
+
+static int init_bxt_mmio_info(struct intel_gvt *gvt)
+{
+ struct drm_i915_private *dev_priv = gvt->dev_priv;
+ int ret;
+
+ MMIO_F(_MMIO(0x80000), 0x3000, 0, 0, 0, D_BXT, NULL, NULL);
+
+ MMIO_D(GEN7_SAMPLER_INSTDONE, D_BXT);
+ MMIO_D(GEN7_ROW_INSTDONE, D_BXT);
+ MMIO_D(GEN8_FAULT_TLB_DATA0, D_BXT);
+ MMIO_D(GEN8_FAULT_TLB_DATA1, D_BXT);
+ MMIO_D(ERROR_GEN6, D_BXT);
+ MMIO_D(DONE_REG, D_BXT);
+ MMIO_D(EIR, D_BXT);
+ MMIO_D(PGTBL_ER, D_BXT);
+ MMIO_D(_MMIO(0x4194), D_BXT);
+ MMIO_D(_MMIO(0x4294), D_BXT);
+ MMIO_D(_MMIO(0x4494), D_BXT);
+
+ MMIO_RING_D(RING_PSMI_CTL, D_BXT);
+ MMIO_RING_D(RING_DMA_FADD, D_BXT);
+ MMIO_RING_D(RING_DMA_FADD_UDW, D_BXT);
+ MMIO_RING_D(RING_IPEHR, D_BXT);
+ MMIO_RING_D(RING_INSTPS, D_BXT);
+ MMIO_RING_D(RING_BBADDR_UDW, D_BXT);
+ MMIO_RING_D(RING_BBSTATE, D_BXT);
+ MMIO_RING_D(RING_IPEIR, D_BXT);
+
+ MMIO_F(SOFT_SCRATCH(0), 16 * 4, 0, 0, 0, D_BXT, NULL, NULL);
+
+ MMIO_DH(BXT_P_CR_GT_DISP_PWRON, D_BXT, NULL, bxt_gt_disp_pwron_write);
+ MMIO_D(BXT_RP_STATE_CAP, D_BXT);
+ MMIO_DH(BXT_PHY_CTL_FAMILY(DPIO_PHY0), D_BXT,
+ NULL, bxt_phy_ctl_family_write);
+ MMIO_DH(BXT_PHY_CTL_FAMILY(DPIO_PHY1), D_BXT,
+ NULL, bxt_phy_ctl_family_write);
+ MMIO_D(BXT_PHY_CTL(PORT_A), D_BXT);
+ MMIO_D(BXT_PHY_CTL(PORT_B), D_BXT);
+ MMIO_D(BXT_PHY_CTL(PORT_C), D_BXT);
+ MMIO_DH(BXT_PORT_PLL_ENABLE(PORT_A), D_BXT,
+ NULL, bxt_port_pll_enable_write);
+ MMIO_DH(BXT_PORT_PLL_ENABLE(PORT_B), D_BXT,
+ NULL, bxt_port_pll_enable_write);
+ MMIO_DH(BXT_PORT_PLL_ENABLE(PORT_C), D_BXT, NULL,
+ bxt_port_pll_enable_write);
+
+ MMIO_D(BXT_PORT_CL1CM_DW0(DPIO_PHY0), D_BXT);
+ MMIO_D(BXT_PORT_CL1CM_DW9(DPIO_PHY0), D_BXT);
+ MMIO_D(BXT_PORT_CL1CM_DW10(DPIO_PHY0), D_BXT);
+ MMIO_D(BXT_PORT_CL1CM_DW28(DPIO_PHY0), D_BXT);
+ MMIO_D(BXT_PORT_CL1CM_DW30(DPIO_PHY0), D_BXT);
+ MMIO_D(BXT_PORT_CL2CM_DW6(DPIO_PHY0), D_BXT);
+ MMIO_D(BXT_PORT_REF_DW3(DPIO_PHY0), D_BXT);
+ MMIO_D(BXT_PORT_REF_DW6(DPIO_PHY0), D_BXT);
+ MMIO_D(BXT_PORT_REF_DW8(DPIO_PHY0), D_BXT);
+
+ MMIO_D(BXT_PORT_CL1CM_DW0(DPIO_PHY1), D_BXT);
+ MMIO_D(BXT_PORT_CL1CM_DW9(DPIO_PHY1), D_BXT);
+ MMIO_D(BXT_PORT_CL1CM_DW10(DPIO_PHY1), D_BXT);
+ MMIO_D(BXT_PORT_CL1CM_DW28(DPIO_PHY1), D_BXT);
+ MMIO_D(BXT_PORT_CL1CM_DW30(DPIO_PHY1), D_BXT);
+ MMIO_D(BXT_PORT_CL2CM_DW6(DPIO_PHY1), D_BXT);
+ MMIO_D(BXT_PORT_REF_DW3(DPIO_PHY1), D_BXT);
+ MMIO_D(BXT_PORT_REF_DW6(DPIO_PHY1), D_BXT);
+ MMIO_D(BXT_PORT_REF_DW8(DPIO_PHY1), D_BXT);
+
+ MMIO_D(BXT_PORT_PLL_EBB_0(DPIO_PHY0, DPIO_CH0), D_BXT);
+ MMIO_D(BXT_PORT_PLL_EBB_4(DPIO_PHY0, DPIO_CH0), D_BXT);
+ MMIO_D(BXT_PORT_PCS_DW10_LN01(DPIO_PHY0, DPIO_CH0), D_BXT);
+ MMIO_D(BXT_PORT_PCS_DW10_GRP(DPIO_PHY0, DPIO_CH0), D_BXT);
+ MMIO_D(BXT_PORT_PCS_DW12_LN01(DPIO_PHY0, DPIO_CH0), D_BXT);
+ MMIO_D(BXT_PORT_PCS_DW12_LN23(DPIO_PHY0, DPIO_CH0), D_BXT);
+ MMIO_DH(BXT_PORT_PCS_DW12_GRP(DPIO_PHY0, DPIO_CH0), D_BXT,
+ NULL, bxt_pcs_dw12_grp_write);
+ MMIO_D(BXT_PORT_TX_DW2_LN0(DPIO_PHY0, DPIO_CH0), D_BXT);
+ MMIO_D(BXT_PORT_TX_DW2_GRP(DPIO_PHY0, DPIO_CH0), D_BXT);
+ MMIO_DH(BXT_PORT_TX_DW3_LN0(DPIO_PHY0, DPIO_CH0), D_BXT,
+ bxt_port_tx_dw3_read, NULL);
+ MMIO_D(BXT_PORT_TX_DW3_GRP(DPIO_PHY0, DPIO_CH0), D_BXT);
+ MMIO_D(BXT_PORT_TX_DW4_LN0(DPIO_PHY0, DPIO_CH0), D_BXT);
+ MMIO_D(BXT_PORT_TX_DW4_GRP(DPIO_PHY0, DPIO_CH0), D_BXT);
+ MMIO_D(BXT_PORT_TX_DW14_LN(DPIO_PHY0, DPIO_CH0, 0), D_BXT);
+ MMIO_D(BXT_PORT_TX_DW14_LN(DPIO_PHY0, DPIO_CH0, 1), D_BXT);
+ MMIO_D(BXT_PORT_TX_DW14_LN(DPIO_PHY0, DPIO_CH0, 2), D_BXT);
+ MMIO_D(BXT_PORT_TX_DW14_LN(DPIO_PHY0, DPIO_CH0, 3), D_BXT);
+ MMIO_D(BXT_PORT_PLL(DPIO_PHY0, DPIO_CH0, 0), D_BXT);
+ MMIO_D(BXT_PORT_PLL(DPIO_PHY0, DPIO_CH0, 1), D_BXT);
+ MMIO_D(BXT_PORT_PLL(DPIO_PHY0, DPIO_CH0, 2), D_BXT);
+ MMIO_D(BXT_PORT_PLL(DPIO_PHY0, DPIO_CH0, 3), D_BXT);
+ MMIO_D(BXT_PORT_PLL(DPIO_PHY0, DPIO_CH0, 6), D_BXT);
+ MMIO_D(BXT_PORT_PLL(DPIO_PHY0, DPIO_CH0, 8), D_BXT);
+ MMIO_D(BXT_PORT_PLL(DPIO_PHY0, DPIO_CH0, 9), D_BXT);
+ MMIO_D(BXT_PORT_PLL(DPIO_PHY0, DPIO_CH0, 10), D_BXT);
+
+ MMIO_D(BXT_PORT_PLL_EBB_0(DPIO_PHY0, DPIO_CH1), D_BXT);
+ MMIO_D(BXT_PORT_PLL_EBB_4(DPIO_PHY0, DPIO_CH1), D_BXT);
+ MMIO_D(BXT_PORT_PCS_DW10_LN01(DPIO_PHY0, DPIO_CH1), D_BXT);
+ MMIO_D(BXT_PORT_PCS_DW10_GRP(DPIO_PHY0, DPIO_CH1), D_BXT);
+ MMIO_D(BXT_PORT_PCS_DW12_LN01(DPIO_PHY0, DPIO_CH1), D_BXT);
+ MMIO_D(BXT_PORT_PCS_DW12_LN23(DPIO_PHY0, DPIO_CH1), D_BXT);
+ MMIO_DH(BXT_PORT_PCS_DW12_GRP(DPIO_PHY0, DPIO_CH1), D_BXT,
+ NULL, bxt_pcs_dw12_grp_write);
+ MMIO_D(BXT_PORT_TX_DW2_LN0(DPIO_PHY0, DPIO_CH1), D_BXT);
+ MMIO_D(BXT_PORT_TX_DW2_GRP(DPIO_PHY0, DPIO_CH1), D_BXT);
+ MMIO_DH(BXT_PORT_TX_DW3_LN0(DPIO_PHY0, DPIO_CH1), D_BXT,
+ bxt_port_tx_dw3_read, NULL);
+ MMIO_D(BXT_PORT_TX_DW3_GRP(DPIO_PHY0, DPIO_CH1), D_BXT);
+ MMIO_D(BXT_PORT_TX_DW4_LN0(DPIO_PHY0, DPIO_CH1), D_BXT);
+ MMIO_D(BXT_PORT_TX_DW4_GRP(DPIO_PHY0, DPIO_CH1), D_BXT);
+ MMIO_D(BXT_PORT_TX_DW14_LN(DPIO_PHY0, DPIO_CH1, 0), D_BXT);
+ MMIO_D(BXT_PORT_TX_DW14_LN(DPIO_PHY0, DPIO_CH1, 1), D_BXT);
+ MMIO_D(BXT_PORT_TX_DW14_LN(DPIO_PHY0, DPIO_CH1, 2), D_BXT);
+ MMIO_D(BXT_PORT_TX_DW14_LN(DPIO_PHY0, DPIO_CH1, 3), D_BXT);
+ MMIO_D(BXT_PORT_PLL(DPIO_PHY0, DPIO_CH1, 0), D_BXT);
+ MMIO_D(BXT_PORT_PLL(DPIO_PHY0, DPIO_CH1, 1), D_BXT);
+ MMIO_D(BXT_PORT_PLL(DPIO_PHY0, DPIO_CH1, 2), D_BXT);
+ MMIO_D(BXT_PORT_PLL(DPIO_PHY0, DPIO_CH1, 3), D_BXT);
+ MMIO_D(BXT_PORT_PLL(DPIO_PHY0, DPIO_CH1, 6), D_BXT);
+ MMIO_D(BXT_PORT_PLL(DPIO_PHY0, DPIO_CH1, 8), D_BXT);
+ MMIO_D(BXT_PORT_PLL(DPIO_PHY0, DPIO_CH1, 9), D_BXT);
+ MMIO_D(BXT_PORT_PLL(DPIO_PHY0, DPIO_CH1, 10), D_BXT);
+
+ MMIO_D(BXT_PORT_PLL_EBB_0(DPIO_PHY1, DPIO_CH0), D_BXT);
+ MMIO_D(BXT_PORT_PLL_EBB_4(DPIO_PHY1, DPIO_CH0), D_BXT);
+ MMIO_D(BXT_PORT_PCS_DW10_LN01(DPIO_PHY1, DPIO_CH0), D_BXT);
+ MMIO_D(BXT_PORT_PCS_DW10_GRP(DPIO_PHY1, DPIO_CH0), D_BXT);
+ MMIO_D(BXT_PORT_PCS_DW12_LN01(DPIO_PHY1, DPIO_CH0), D_BXT);
+ MMIO_D(BXT_PORT_PCS_DW12_LN23(DPIO_PHY1, DPIO_CH0), D_BXT);
+ MMIO_DH(BXT_PORT_PCS_DW12_GRP(DPIO_PHY1, DPIO_CH0), D_BXT,
+ NULL, bxt_pcs_dw12_grp_write);
+ MMIO_D(BXT_PORT_TX_DW2_LN0(DPIO_PHY1, DPIO_CH0), D_BXT);
+ MMIO_D(BXT_PORT_TX_DW2_GRP(DPIO_PHY1, DPIO_CH0), D_BXT);
+ MMIO_DH(BXT_PORT_TX_DW3_LN0(DPIO_PHY1, DPIO_CH0), D_BXT,
+ bxt_port_tx_dw3_read, NULL);
+ MMIO_D(BXT_PORT_TX_DW3_GRP(DPIO_PHY1, DPIO_CH0), D_BXT);
+ MMIO_D(BXT_PORT_TX_DW4_LN0(DPIO_PHY1, DPIO_CH0), D_BXT);
+ MMIO_D(BXT_PORT_TX_DW4_GRP(DPIO_PHY1, DPIO_CH0), D_BXT);
+ MMIO_D(BXT_PORT_TX_DW14_LN(DPIO_PHY1, DPIO_CH0, 0), D_BXT);
+ MMIO_D(BXT_PORT_TX_DW14_LN(DPIO_PHY1, DPIO_CH0, 1), D_BXT);
+ MMIO_D(BXT_PORT_TX_DW14_LN(DPIO_PHY1, DPIO_CH0, 2), D_BXT);
+ MMIO_D(BXT_PORT_TX_DW14_LN(DPIO_PHY1, DPIO_CH0, 3), D_BXT);
+ MMIO_D(BXT_PORT_PLL(DPIO_PHY1, DPIO_CH0, 0), D_BXT);
+ MMIO_D(BXT_PORT_PLL(DPIO_PHY1, DPIO_CH0, 1), D_BXT);
+ MMIO_D(BXT_PORT_PLL(DPIO_PHY1, DPIO_CH0, 2), D_BXT);
+ MMIO_D(BXT_PORT_PLL(DPIO_PHY1, DPIO_CH0, 3), D_BXT);
+ MMIO_D(BXT_PORT_PLL(DPIO_PHY1, DPIO_CH0, 6), D_BXT);
+ MMIO_D(BXT_PORT_PLL(DPIO_PHY1, DPIO_CH0, 8), D_BXT);
+ MMIO_D(BXT_PORT_PLL(DPIO_PHY1, DPIO_CH0, 9), D_BXT);
+ MMIO_D(BXT_PORT_PLL(DPIO_PHY1, DPIO_CH0, 10), D_BXT);
+
+ MMIO_D(BXT_DE_PLL_CTL, D_BXT);
+ MMIO_DH(BXT_DE_PLL_ENABLE, D_BXT, NULL, bxt_de_pll_enable_write);
+ MMIO_D(BXT_DSI_PLL_CTL, D_BXT);
+ MMIO_D(BXT_DSI_PLL_ENABLE, D_BXT);
+
+ MMIO_D(GEN9_CLKGATE_DIS_0, D_BXT);
+
+ MMIO_D(HSW_TVIDEO_DIP_GCP(TRANSCODER_A), D_BXT);
+ MMIO_D(HSW_TVIDEO_DIP_GCP(TRANSCODER_B), D_BXT);
+ MMIO_D(HSW_TVIDEO_DIP_GCP(TRANSCODER_C), D_BXT);
+
+ MMIO_DH(EDP_PSR_IMR, D_BXT, NULL, bxt_edp_psr_imr_iir_write);
+ MMIO_DH(EDP_PSR_IIR, D_BXT, NULL, bxt_edp_psr_imr_iir_write);
+
+ MMIO_D(RC6_CTX_BASE, D_BXT);
+
+ MMIO_D(GEN8_PUSHBUS_CONTROL, D_BXT);
+ MMIO_D(GEN8_PUSHBUS_ENABLE, D_BXT);
+ MMIO_D(GEN8_PUSHBUS_SHIFT, D_BXT);
+ MMIO_D(GEN6_GFXPAUSE, D_BXT);
+ MMIO_D(GEN8_L3SQCREG1, D_BXT);
+
+ MMIO_DFH(GEN9_CTX_PREEMPT_REG, D_BXT, F_CMD_ACCESS, NULL, NULL);
return 0;
}
ret = init_skl_mmio_info(gvt);
if (ret)
goto err;
+ } else if (IS_BROXTON(dev_priv)) {
+ ret = init_broadwell_mmio_info(gvt);
+ if (ret)
+ goto err;
+ ret = init_skl_mmio_info(gvt);
+ if (ret)
+ goto err;
+ ret = init_bxt_mmio_info(gvt);
+ if (ret)
+ goto err;
}
gvt->mmio.mmio_block = mmio_blocks;
return 0;
}
+ /**
+ * intel_vgpu_mask_mmio_write - write mask register
+ * @vgpu: a vGPU
+ * @offset: access offset
+ * @p_data: write data buffer
+ * @bytes: access data length
+ *
+ * Returns:
+ * Zero on success, negative error code if failed.
+ */
+ int intel_vgpu_mask_mmio_write(struct intel_vgpu *vgpu, unsigned int offset,
+ void *p_data, unsigned int bytes)
+ {
+ u32 mask, old_vreg;
+
+ old_vreg = vgpu_vreg(vgpu, offset);
+ write_vreg(vgpu, offset, p_data, bytes);
+ mask = vgpu_vreg(vgpu, offset) >> 16;
+ vgpu_vreg(vgpu, offset) = (old_vreg & ~mask) |
+ (vgpu_vreg(vgpu, offset) & mask);
+
+ return 0;
+ }
+
/**
* intel_gvt_in_force_nonpriv_whitelist - if a mmio is in whitelist to be
* force-nopriv register
#define D_BDW (1 << 0)
#define D_SKL (1 << 1)
#define D_KBL (1 << 2)
+#define D_BXT (1 << 3)
-#define D_GEN9PLUS (D_SKL | D_KBL)
-#define D_GEN8PLUS (D_BDW | D_SKL | D_KBL)
+#define D_GEN9PLUS (D_SKL | D_KBL | D_BXT)
+#define D_GEN8PLUS (D_BDW | D_SKL | D_KBL | D_BXT)
-#define D_SKL_PLUS (D_SKL | D_KBL)
-#define D_BDW_PLUS (D_BDW | D_SKL | D_KBL)
+#define D_SKL_PLUS (D_SKL | D_KBL | D_BXT)
+#define D_BDW_PLUS (D_BDW | D_SKL | D_KBL | D_BXT)
#define D_PRE_SKL (D_BDW)
-#define D_ALL (D_BDW | D_SKL | D_KBL)
+#define D_ALL (D_BDW | D_SKL | D_KBL | D_BXT)
typedef int (*gvt_mmio_func)(struct intel_vgpu *, unsigned int, void *,
unsigned int);
int intel_vgpu_mmio_reg_rw(struct intel_vgpu *vgpu, unsigned int offset,
void *pdata, unsigned int bytes, bool is_read);
+ int intel_vgpu_mask_mmio_write(struct intel_vgpu *vgpu, unsigned int offset,
+ void *p_data, unsigned int bytes);
#endif
*/
fw = intel_uncore_forcewake_for_reg(dev_priv, reg,
FW_REG_READ | FW_REG_WRITE);
- if (ring_id == RCS && (IS_SKYLAKE(dev_priv) || IS_KABYLAKE(dev_priv)))
+ if (ring_id == RCS && (IS_SKYLAKE(dev_priv) ||
+ IS_KABYLAKE(dev_priv) || IS_BROXTON(dev_priv)))
fw |= FORCEWAKE_RENDER;
intel_uncore_forcewake_get(dev_priv, fw);
if (WARN_ON(ring_id >= ARRAY_SIZE(regs)))
return;
- if (IS_KABYLAKE(dev_priv) && ring_id == RCS)
+ if ((IS_KABYLAKE(dev_priv) || IS_BROXTON(dev_priv)) && ring_id == RCS)
return;
if (!pre && !gen9_render_mocs.initialized)
#define CTX_CONTEXT_CONTROL_VAL 0x03
-bool is_inhibit_context(struct i915_gem_context *ctx, int ring_id)
+bool is_inhibit_context(struct intel_context *ce)
{
- u32 *reg_state = ctx->__engine[ring_id].lrc_reg_state;
+ const u32 *reg_state = ce->lrc_reg_state;
u32 inhibit_mask =
_MASKED_BIT_ENABLE(CTX_CTRL_ENGINE_CTX_RESTORE_INHIBIT);
u32 old_v, new_v;
dev_priv = pre ? pre->gvt->dev_priv : next->gvt->dev_priv;
- if (IS_SKYLAKE(dev_priv) || IS_KABYLAKE(dev_priv))
+ if (IS_SKYLAKE(dev_priv)
+ || IS_KABYLAKE(dev_priv)
+ || IS_BROXTON(dev_priv))
switch_mocs(pre, next, ring_id);
for (mmio = dev_priv->gvt->engine_mmio_list.mmio;
* state image on kabylake, it's initialized by lri command and
* save or restore with context together.
*/
- if (IS_KABYLAKE(dev_priv) && mmio->in_context)
+ if ((IS_KABYLAKE(dev_priv) || IS_BROXTON(dev_priv))
+ && mmio->in_context)
continue;
// save
* itself.
*/
if (mmio->in_context &&
- !is_inhibit_context(s->shadow_ctx, ring_id))
+ !is_inhibit_context(&s->shadow_ctx->__engine[ring_id]))
continue;
if (mmio->mask)
{
struct engine_mmio *mmio;
- if (IS_SKYLAKE(gvt->dev_priv) || IS_KABYLAKE(gvt->dev_priv))
+ if (IS_SKYLAKE(gvt->dev_priv) ||
+ IS_KABYLAKE(gvt->dev_priv) ||
+ IS_BROXTON(gvt->dev_priv))
gvt->engine_mmio_list.mmio = gen9_engine_mmio_list;
else
gvt->engine_mmio_list.mmio = gen8_engine_mmio_list;
for (mmio = gvt->engine_mmio_list.mmio;
i915_mmio_reg_valid(mmio->reg); mmio++) {
- if (mmio->in_context)
+ if (mmio->in_context) {
gvt->engine_mmio_list.ctx_mmio_count[mmio->ring_id]++;
+ intel_gvt_mmio_set_in_ctx(gvt, mmio->reg.reg);
+ }
}
}
{
struct nouveau_cli *cli = (void *)device->object.client;
struct nv50_disp_core_channel_dma_v0 *args = data;
+ u8 type = NVIF_MEM_COHERENT;
int ret;
mutex_init(&dmac->lock);
- ret = nvif_mem_init_map(&cli->mmu, NVIF_MEM_COHERENT, 0x1000,
- &dmac->push);
+ /* Pascal added support for 47-bit physical addresses, but some
+ * parts of EVO still only accept 40-bit PAs.
+ *
+ * To avoid issues on systems with large amounts of RAM, and on
+ * systems where an IOMMU maps pages at a high address, we need
+ * to allocate push buffers in VRAM instead.
+ *
+ * This appears to match NVIDIA's behaviour on Pascal.
+ */
+ if (device->info.family == NV_DEVICE_INFO_V0_PASCAL)
+ type |= NVIF_MEM_VRAM;
+
+ ret = nvif_mem_init_map(&cli->mmu, type, 0x1000, &dmac->push);
if (ret)
return ret;
evo_kick(u32 *push, struct nv50_dmac *evoc)
{
struct nv50_dmac *dmac = evoc;
+
+ /* Push buffer fetches are not coherent with BAR1, we need to ensure
+ * writes have been flushed right through to VRAM before writing PUT.
+ */
+ if (dmac->push.type & NVIF_MEM_VRAM) {
+ struct nvif_device *device = dmac->base.device;
+ nvif_wr32(&device->object, 0x070000, 0x00000001);
+ nvif_msec(device, 2000,
+ if (!(nvif_rd32(&device->object, 0x070000) & 0x00000002))
+ break;
+ );
+ }
+
nvif_wr32(&dmac->base.user, 0x0000, (push - dmac->ptr) << 2);
mutex_unlock(&dmac->lock);
}
"dac-%04x-%04x", dcbe->hasht, dcbe->hashm);
drm_encoder_helper_add(encoder, &nv50_dac_help);
- drm_mode_connector_attach_encoder(connector, encoder);
+ drm_connector_attach_encoder(connector, encoder);
return 0;
}
int ret = 0;
mstc->edid = drm_dp_mst_get_edid(&mstc->connector, mstc->port->mgr, mstc->port);
- drm_mode_connector_update_edid_property(&mstc->connector, mstc->edid);
+ drm_connector_update_edid_property(&mstc->connector, mstc->edid);
if (mstc->edid)
ret = drm_add_edid_modes(&mstc->connector, mstc->edid);
nouveau_conn_attach_properties(&mstc->connector);
for (i = 0; i < ARRAY_SIZE(mstm->msto) && mstm->msto[i]; i++)
- drm_mode_connector_attach_encoder(&mstc->connector, &mstm->msto[i]->encoder);
+ drm_connector_attach_encoder(&mstc->connector, &mstm->msto[i]->encoder);
drm_object_attach_property(&mstc->connector.base, dev->mode_config.path_property, 0);
drm_object_attach_property(&mstc->connector.base, dev->mode_config.tile_property, 0);
- drm_mode_connector_set_path_property(&mstc->connector, path);
+ drm_connector_set_path_property(&mstc->connector, path);
return 0;
}
mstc->port = NULL;
drm_modeset_unlock(&drm->dev->mode_config.connection_mutex);
- drm_connector_unreference(&mstc->connector);
+ drm_connector_put(&mstc->connector);
}
static void
"sor-%04x-%04x", dcbe->hasht, dcbe->hashm);
drm_encoder_helper_add(encoder, &nv50_sor_help);
- drm_mode_connector_attach_encoder(connector, encoder);
+ drm_connector_attach_encoder(connector, encoder);
if (dcbe->type == DCB_OUTPUT_DP) {
struct nv50_disp *disp = nv50_disp(encoder->dev);
"pior-%04x-%04x", dcbe->hasht, dcbe->hashm);
drm_encoder_helper_add(encoder, &nv50_pior_help);
- drm_mode_connector_attach_encoder(connector, encoder);
+ drm_connector_attach_encoder(connector, encoder);
return 0;
}
*****************************************************************************/
static void
- nv50_disp_atomic_commit_core(struct nouveau_drm *drm, u32 *interlock)
+ nv50_disp_atomic_commit_core(struct drm_atomic_state *state, u32 *interlock)
{
+ struct nouveau_drm *drm = nouveau_drm(state->dev);
struct nv50_disp *disp = nv50_disp(drm->dev);
struct nv50_core *core = disp->core;
struct nv50_mstm *mstm;
}
}
+ static void
+ nv50_disp_atomic_commit_wndw(struct drm_atomic_state *state, u32 *interlock)
+ {
+ struct drm_plane_state *new_plane_state;
+ struct drm_plane *plane;
+ int i;
+
+ for_each_new_plane_in_state(state, plane, new_plane_state, i) {
+ struct nv50_wndw *wndw = nv50_wndw(plane);
+ if (interlock[wndw->interlock.type] & wndw->interlock.data) {
+ if (wndw->func->update)
+ wndw->func->update(wndw, interlock);
+ }
+ }
+ }
+
static void
nv50_disp_atomic_commit_tail(struct drm_atomic_state *state)
{
help->disable(encoder);
interlock[NV50_DISP_INTERLOCK_CORE] |= 1;
if (outp->flush_disable) {
- nv50_disp_atomic_commit_core(drm, interlock);
+ nv50_disp_atomic_commit_wndw(state, interlock);
+ nv50_disp_atomic_commit_core(state, interlock);
memset(interlock, 0x00, sizeof(interlock));
}
}
/* Flush disable. */
if (interlock[NV50_DISP_INTERLOCK_CORE]) {
if (atom->flush_disable) {
- for_each_new_plane_in_state(state, plane, new_plane_state, i) {
- struct nv50_wndw *wndw = nv50_wndw(plane);
- if (interlock[wndw->interlock.type] & wndw->interlock.data) {
- if (wndw->func->update)
- wndw->func->update(wndw, interlock);
- }
- }
-
- nv50_disp_atomic_commit_core(drm, interlock);
+ nv50_disp_atomic_commit_wndw(state, interlock);
+ nv50_disp_atomic_commit_core(state, interlock);
memset(interlock, 0x00, sizeof(interlock));
}
}
}
/* Flush update. */
- for_each_new_plane_in_state(state, plane, new_plane_state, i) {
- struct nv50_wndw *wndw = nv50_wndw(plane);
- if (interlock[wndw->interlock.type] & wndw->interlock.data) {
- if (wndw->func->update)
- wndw->func->update(wndw, interlock);
- }
- }
+ nv50_disp_atomic_commit_wndw(state, interlock);
if (interlock[NV50_DISP_INTERLOCK_CORE]) {
if (interlock[NV50_DISP_INTERLOCK_BASE] ||
+ interlock[NV50_DISP_INTERLOCK_OVLY] ||
+ interlock[NV50_DISP_INTERLOCK_WNDW] ||
!atom->state.legacy_cursor_update)
- nv50_disp_atomic_commit_core(drm, interlock);
+ nv50_disp_atomic_commit_core(state, interlock);
else
disp->core->func->update(disp->core, interlock, false);
}
nv50_disp_atomic_commit_tail(state);
drm_for_each_crtc(crtc, dev) {
- if (crtc->state->enable) {
+ if (crtc->state->active) {
if (!drm->have_disp_power_ref) {
drm->have_disp_power_ref = true;
return 0;
kfree(disp);
}
- MODULE_PARM_DESC(atomic, "Expose atomic ioctl (default: disabled)");
- static int nouveau_atomic = 0;
- module_param_named(atomic, nouveau_atomic, int, 0400);
-
int
nv50_display_create(struct drm_device *dev)
{
disp->disp = &nouveau_display(dev)->disp;
dev->mode_config.funcs = &nv50_disp_func;
dev->driver->driver_features |= DRIVER_PREFER_XBGR_30BPP;
- if (nouveau_atomic)
- dev->driver->driver_features |= DRIVER_ATOMIC;
/* small shared memory area we use for notifiers and semaphores */
ret = nouveau_bo_new(&drm->client, 4096, 0x1000, TTM_PL_FLAG_VRAM,
connector->funcs->destroy(connector);
}
+ /* Disable vblank irqs aggressively for power-saving, safe on nv50+ */
+ dev->vblank_disable_immediate = true;
+
out:
if (ret)
nv50_display_destroy(dev);
struct nouveau_encoder *
find_encoder(struct drm_connector *connector, int type)
{
- struct drm_device *dev = connector->dev;
struct nouveau_encoder *nv_encoder;
struct drm_encoder *enc;
- int i, id;
-
- for (i = 0; i < DRM_CONNECTOR_MAX_ENCODER; i++) {
- id = connector->encoder_ids[i];
- if (!id)
- break;
+ int i;
- enc = drm_encoder_find(dev, NULL, id);
- if (!enc)
- continue;
+ drm_connector_for_each_possible_encoder(connector, enc, i) {
nv_encoder = nouveau_encoder(enc);
if (type == DCB_OUTPUT_ANY ||
struct nouveau_connector *nv_connector = nouveau_connector(connector);
struct nouveau_drm *drm = nouveau_drm(dev);
struct nvkm_gpio *gpio = nvxx_gpio(&drm->client.device);
- struct nouveau_encoder *nv_encoder;
+ struct nouveau_encoder *nv_encoder = NULL;
struct drm_encoder *encoder;
int i, panel = -ENODEV;
}
}
- for (i = 0; nv_encoder = NULL, i < DRM_CONNECTOR_MAX_ENCODER; i++) {
- int id = connector->encoder_ids[i];
- if (id == 0)
- break;
-
- encoder = drm_encoder_find(dev, NULL, id);
- if (!encoder)
- continue;
+ drm_connector_for_each_possible_encoder(connector, encoder, i) {
nv_encoder = nouveau_encoder(encoder);
if (nv_encoder->dcb->type == DCB_OUTPUT_DP) {
/* Cleanup the previous EDID block. */
if (nv_connector->edid) {
- drm_mode_connector_update_edid_property(connector, NULL);
+ drm_connector_update_edid_property(connector, NULL);
kfree(nv_connector->edid);
nv_connector->edid = NULL;
}
else
nv_connector->edid = drm_get_edid(connector, i2c);
- drm_mode_connector_update_edid_property(connector,
+ drm_connector_update_edid_property(connector,
nv_connector->edid);
if (!nv_connector->edid) {
NV_ERROR(drm, "DDC responded, but no EDID for %s\n",
/* Cleanup the previous EDID block. */
if (nv_connector->edid) {
- drm_mode_connector_update_edid_property(connector, NULL);
+ drm_connector_update_edid_property(connector, NULL);
kfree(nv_connector->edid);
nv_connector->edid = NULL;
}
status = connector_status_unknown;
#endif
- drm_mode_connector_update_edid_property(connector, nv_connector->edid);
+ drm_connector_update_edid_property(connector, nv_connector->edid);
nouveau_connector_set_encoder(connector, nv_encoder);
return status;
}
struct nouveau_display *disp = nouveau_display(dev);
struct nouveau_connector *nv_connector = NULL;
struct drm_connector *connector;
+ struct drm_connector_list_iter conn_iter;
int type, ret = 0;
bool dummy;
- list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
+ drm_connector_list_iter_begin(dev, &conn_iter);
+ nouveau_for_each_non_mst_connector_iter(connector, &conn_iter) {
nv_connector = nouveau_connector(connector);
- if (nv_connector->index == index)
+ if (nv_connector->index == index) {
+ drm_connector_list_iter_end(&conn_iter);
return connector;
+ }
}
+ drm_connector_list_iter_end(&conn_iter);
nv_connector = kzalloc(sizeof(*nv_connector), GFP_KERNEL);
if (!nv_connector)
struct nouveau_framebuffer *fb = nouveau_framebuffer(drm_fb);
if (fb->nvbo)
- drm_gem_object_unreference_unlocked(&fb->nvbo->gem);
+ drm_gem_object_put_unlocked(&fb->nvbo->gem);
drm_framebuffer_cleanup(drm_fb);
kfree(fb);
if (ret == 0)
return &fb->base;
- drm_gem_object_unreference_unlocked(gem);
+ drm_gem_object_put_unlocked(gem);
return ERR_PTR(ret);
}
struct nouveau_display *disp = nouveau_display(dev);
struct nouveau_drm *drm = nouveau_drm(dev);
struct drm_connector *connector;
+ struct drm_connector_list_iter conn_iter;
int ret;
ret = disp->init(dev);
return ret;
/* enable hotplug interrupts */
- list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
+ drm_connector_list_iter_begin(dev, &conn_iter);
+ nouveau_for_each_non_mst_connector_iter(connector, &conn_iter) {
struct nouveau_connector *conn = nouveau_connector(connector);
nvif_notify_get(&conn->hpd);
}
+ drm_connector_list_iter_end(&conn_iter);
/* enable flip completion events */
nvif_notify_get(&drm->flip);
struct nouveau_display *disp = nouveau_display(dev);
struct nouveau_drm *drm = nouveau_drm(dev);
struct drm_connector *connector;
+ struct drm_connector_list_iter conn_iter;
if (!suspend) {
if (drm_drv_uses_atomic_modeset(dev))
nvif_notify_put(&drm->flip);
/* disable hotplug interrupts */
- list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
+ drm_connector_list_iter_begin(dev, &conn_iter);
+ nouveau_for_each_non_mst_connector_iter(connector, &conn_iter) {
struct nouveau_connector *conn = nouveau_connector(connector);
nvif_notify_put(&conn->hpd);
}
+ drm_connector_list_iter_end(&conn_iter);
drm_kms_helper_poll_disable(dev);
disp->fini(dev);
return ret;
ret = drm_gem_handle_create(file_priv, &bo->gem, &args->handle);
- drm_gem_object_unreference_unlocked(&bo->gem);
+ drm_gem_object_put_unlocked(&bo->gem);
return ret;
}
if (gem) {
struct nouveau_bo *bo = nouveau_gem_object(gem);
*poffset = drm_vma_node_offset_addr(&bo->bo.vma_node);
- drm_gem_object_unreference_unlocked(gem);
+ drm_gem_object_put_unlocked(gem);
return 0;
}
int nouveau_modeset = -1;
module_param_named(modeset, nouveau_modeset, int, 0400);
+ MODULE_PARM_DESC(atomic, "Expose atomic ioctl (default: disabled)");
+ static int nouveau_atomic = 0;
+ module_param_named(atomic, nouveau_atomic, int, 0400);
+
MODULE_PARM_DESC(runpm, "disable (0), force enable (1), optimus only default (-1)");
static int nouveau_runtime_pm = -1;
module_param_named(runpm, nouveau_runtime_pm, int, 0400);
pci_set_master(pdev);
+ if (nouveau_atomic)
+ driver_pci.driver_features |= DRIVER_ATOMIC;
+
ret = drm_get_pci_dev(pdev, pent, &driver_pci);
if (ret) {
nvkm_device_del(&device);
static int
nouveau_pmops_runtime_idle(struct device *dev)
{
- struct pci_dev *pdev = to_pci_dev(dev);
- struct drm_device *drm_dev = pci_get_drvdata(pdev);
- struct nouveau_drm *drm = nouveau_drm(drm_dev);
- struct drm_crtc *crtc;
-
if (!nouveau_pmops_runtime()) {
pm_runtime_forbid(dev);
return -EBUSY;
}
- list_for_each_entry(crtc, &drm->dev->mode_config.crtc_list, head) {
- if (crtc->enabled) {
- DRM_DEBUG_DRIVER("failing to power off - crtc active\n");
- return -EBUSY;
- }
- }
pm_runtime_mark_last_busy(dev);
pm_runtime_autosuspend(dev);
/* we don't want the main rpm_idle to call suspend - we want to autosuspend */
get_task_comm(tmpname, current);
snprintf(name, sizeof(name), "%s[%d]", tmpname, pid_nr(fpriv->pid));
- if (!(cli = kzalloc(sizeof(*cli), GFP_KERNEL)))
- return ret;
+ if (!(cli = kzalloc(sizeof(*cli), GFP_KERNEL))) {
+ ret = -ENOMEM;
+ goto done;
+ }
ret = nouveau_cli_init(drm, name, cli);
if (ret)
}
/* drop reference from allocate - handle holds it now */
- drm_gem_object_unreference_unlocked(&nvbo->gem);
+ drm_gem_object_put_unlocked(&nvbo->gem);
return ret;
}
list_del(&nvbo->entry);
nvbo->reserved_by = NULL;
ttm_bo_unreserve(&nvbo->bo);
- drm_gem_object_unreference_unlocked(&nvbo->gem);
+ drm_gem_object_put_unlocked(&nvbo->gem);
}
}
nvbo = nouveau_gem_object(gem);
if (nvbo == res_bo) {
res_bo = NULL;
- drm_gem_object_unreference_unlocked(gem);
+ drm_gem_object_put_unlocked(gem);
continue;
}
if (nvbo->reserved_by && nvbo->reserved_by == file_priv) {
NV_PRINTK(err, cli, "multiple instances of buffer %d on "
"validation list\n", b->handle);
- drm_gem_object_unreference_unlocked(gem);
+ drm_gem_object_put_unlocked(gem);
ret = -EINVAL;
break;
}
struct nouveau_bo *nvbo;
uint32_t data;
- if (unlikely(r->bo_index > req->nr_buffers)) {
+ if (unlikely(r->bo_index >= req->nr_buffers)) {
NV_PRINTK(err, cli, "reloc bo index invalid\n");
ret = -EINVAL;
break;
if (b->presumed.valid)
continue;
- if (unlikely(r->reloc_bo_index > req->nr_buffers)) {
+ if (unlikely(r->reloc_bo_index >= req->nr_buffers)) {
NV_PRINTK(err, cli, "reloc container bo index invalid\n");
ret = -EINVAL;
break;
ret = lret;
nouveau_bo_sync_for_cpu(nvbo);
- drm_gem_object_unreference_unlocked(gem);
+ drm_gem_object_put_unlocked(gem);
return ret;
}
nvbo = nouveau_gem_object(gem);
nouveau_bo_sync_for_device(nvbo);
- drm_gem_object_unreference_unlocked(gem);
+ drm_gem_object_put_unlocked(gem);
return 0;
}
return -ENOENT;
ret = nouveau_gem_info(file_priv, gem, req);
- drm_gem_object_unreference_unlocked(gem);
+ drm_gem_object_put_unlocked(gem);
return ret;
}
obj-$(CONFIG_DRM_SUN4I) += sun4i_tv.o
obj-$(CONFIG_DRM_SUN4I) += sun6i_drc.o
- obj-$(CONFIG_DRM_SUN4I_BACKEND) += sun4i-backend.o sun4i-frontend.o
+ obj-$(CONFIG_DRM_SUN4I_BACKEND) += sun4i-backend.o
+ ifdef CONFIG_DRM_SUN4I_BACKEND
+ obj-$(CONFIG_DRM_SUN4I) += sun4i-frontend.o
+ endif
obj-$(CONFIG_DRM_SUN4I_HDMI) += sun4i-drm-hdmi.o
obj-$(CONFIG_DRM_SUN6I_DSI) += sun6i-dsi.o
obj-$(CONFIG_DRM_SUN8I_DW_HDMI) += sun8i-drm-hdmi.o
obj-$(CONFIG_DRM_SUN8I_MIXER) += sun8i-mixer.o
+obj-$(CONFIG_DRM_SUN8I_TCON_TOP) += sun8i_tcon_top.o
}
EXPORT_SYMBOL(pci_unmap_iospace);
+ static void devm_pci_unmap_iospace(struct device *dev, void *ptr)
+ {
+ struct resource **res = ptr;
+
+ pci_unmap_iospace(*res);
+ }
+
+ /**
+ * devm_pci_remap_iospace - Managed pci_remap_iospace()
+ * @dev: Generic device to remap IO address for
+ * @res: Resource describing the I/O space
+ * @phys_addr: physical address of range to be mapped
+ *
+ * Managed pci_remap_iospace(). Map is automatically unmapped on driver
+ * detach.
+ */
+ int devm_pci_remap_iospace(struct device *dev, const struct resource *res,
+ phys_addr_t phys_addr)
+ {
+ const struct resource **ptr;
+ int error;
+
+ ptr = devres_alloc(devm_pci_unmap_iospace, sizeof(*ptr), GFP_KERNEL);
+ if (!ptr)
+ return -ENOMEM;
+
+ error = pci_remap_iospace(res, phys_addr);
+ if (error) {
+ devres_free(ptr);
+ } else {
+ *ptr = res;
+ devres_add(dev, ptr);
+ }
+
+ return error;
+ }
+ EXPORT_SYMBOL(devm_pci_remap_iospace);
+
/**
* devm_pci_remap_cfgspace - Managed pci_remap_cfgspace()
* @dev: Generic device to remap IO address for
return PCI_SPEED_UNKNOWN;
}
+EXPORT_SYMBOL(pcie_get_speed_cap);
/**
* pcie_get_width_cap - query for the PCI device's link width capability
return PCIE_LNK_WIDTH_UNKNOWN;
}
+EXPORT_SYMBOL(pcie_get_width_cap);
/**
* pcie_bandwidth_capable - calculate a PCI device's link bandwidth capability
PCI_SPEED_UNKNOWN = 0xff,
};
+enum pci_bus_speed pcie_get_speed_cap(struct pci_dev *dev);
+enum pcie_link_width pcie_get_width_cap(struct pci_dev *dev);
+
struct pci_cap_saved_data {
u16 cap_nr;
bool cap_extended;
unsigned long pci_address_to_pio(phys_addr_t addr);
phys_addr_t pci_pio_to_address(unsigned long pio);
int pci_remap_iospace(const struct resource *res, phys_addr_t phys_addr);
+ int devm_pci_remap_iospace(struct device *dev, const struct resource *res,
+ phys_addr_t phys_addr);
void pci_unmap_iospace(struct resource *res);
void __iomem *devm_pci_remap_cfgspace(struct device *dev,
resource_size_t offset,
projects / linux.git / commitdiff