diff options
Diffstat (limited to 'drivers/gpu/drm/xe/xe_device.c')
-rw-r--r-- | drivers/gpu/drm/xe/xe_device.c | 106 |
1 files changed, 68 insertions, 38 deletions
diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c index 00191227bc95..e9f3c1a53db2 100644 --- a/drivers/gpu/drm/xe/xe_device.c +++ b/drivers/gpu/drm/xe/xe_device.c @@ -23,8 +23,10 @@ #include "regs/xe_gt_regs.h" #include "regs/xe_regs.h" #include "xe_bo.h" +#include "xe_bo_evict.h" #include "xe_debugfs.h" #include "xe_devcoredump.h" +#include "xe_device_sysfs.h" #include "xe_dma_buf.h" #include "xe_drm_client.h" #include "xe_drv.h" @@ -38,6 +40,7 @@ #include "xe_gt_printk.h" #include "xe_gt_sriov_vf.h" #include "xe_guc.h" +#include "xe_guc_pc.h" #include "xe_hw_engine_group.h" #include "xe_hwmon.h" #include "xe_irq.h" @@ -467,10 +470,9 @@ struct xe_device *xe_device_create(struct pci_dev *pdev, xa_erase(&xe->usm.asid_to_vm, asid); } - spin_lock_init(&xe->pinned.lock); - INIT_LIST_HEAD(&xe->pinned.kernel_bo_present); - INIT_LIST_HEAD(&xe->pinned.external_vram); - INIT_LIST_HEAD(&xe->pinned.evicted); + err = xe_bo_pinned_init(xe); + if (err) + goto err; xe->preempt_fence_wq = alloc_ordered_workqueue("xe-preempt-fence-wq", WQ_MEM_RECLAIM); @@ -505,7 +507,15 @@ ALLOW_ERROR_INJECTION(xe_device_create, ERRNO); /* See xe_pci_probe() */ static bool xe_driver_flr_disabled(struct xe_device *xe) { - return xe_mmio_read32(xe_root_tile_mmio(xe), GU_CNTL_PROTECTED) & DRIVERINT_FLR_DIS; + if (IS_SRIOV_VF(xe)) + return true; + + if (xe_mmio_read32(xe_root_tile_mmio(xe), GU_CNTL_PROTECTED) & DRIVERINT_FLR_DIS) { + drm_info(&xe->drm, "Driver-FLR disabled by BIOS\n"); + return true; + } + + return false; } /* @@ -523,7 +533,7 @@ static bool xe_driver_flr_disabled(struct xe_device *xe) */ static void __xe_driver_flr(struct xe_device *xe) { - const unsigned int flr_timeout = 3 * MICRO; /* specs recommend a 3s wait */ + const unsigned int flr_timeout = 3 * USEC_PER_SEC; /* specs recommend a 3s wait */ struct xe_mmio *mmio = xe_root_tile_mmio(xe); int ret; @@ -569,10 +579,8 @@ static void __xe_driver_flr(struct xe_device *xe) static void xe_driver_flr(struct xe_device *xe) { - if (xe_driver_flr_disabled(xe)) { - drm_info_once(&xe->drm, "BIOS Disabled Driver-FLR\n"); + if (xe_driver_flr_disabled(xe)) return; - } __xe_driver_flr(xe); } @@ -706,7 +714,7 @@ int xe_device_probe_early(struct xe_device *xe) sriov_update_device_info(xe); err = xe_pcode_probe_early(xe); - if (err) { + if (err || xe_survivability_mode_is_requested(xe)) { int save_err = err; /* @@ -729,6 +737,7 @@ int xe_device_probe_early(struct xe_device *xe) return 0; } +ALLOW_ERROR_INJECTION(xe_device_probe_early, ERRNO); /* See xe_pci_probe() */ static int probe_has_flat_ccs(struct xe_device *xe) { @@ -908,6 +917,10 @@ int xe_device_probe(struct xe_device *xe) if (err) goto err_unregister_display; + err = xe_device_sysfs_init(xe); + if (err) + goto err_unregister_display; + xe_debugfs_register(xe); err = xe_hwmon_register(xe); @@ -932,6 +945,8 @@ void xe_device_remove(struct xe_device *xe) xe_display_unregister(xe); drm_dev_unplug(&xe->drm); + + xe_bo_pci_dev_remove_all(xe); } void xe_device_shutdown(struct xe_device *xe) @@ -972,38 +987,15 @@ void xe_device_wmb(struct xe_device *xe) xe_mmio_write32(xe_root_tile_mmio(xe), VF_CAP_REG, 0); } -/** - * xe_device_td_flush() - Flush transient L3 cache entries - * @xe: The device - * - * Display engine has direct access to memory and is never coherent with L3/L4 - * caches (or CPU caches), however KMD is responsible for specifically flushing - * transient L3 GPU cache entries prior to the flip sequence to ensure scanout - * can happen from such a surface without seeing corruption. - * - * Display surfaces can be tagged as transient by mapping it using one of the - * various L3:XD PAT index modes on Xe2. - * - * Note: On non-discrete xe2 platforms, like LNL, the entire L3 cache is flushed - * at the end of each submission via PIPE_CONTROL for compute/render, since SA - * Media is not coherent with L3 and we want to support render-vs-media - * usescases. For other engines like copy/blt the HW internally forces uncached - * behaviour, hence why we can skip the TDF on such platforms. +/* + * Issue a TRANSIENT_FLUSH_REQUEST and wait for completion on each gt. */ -void xe_device_td_flush(struct xe_device *xe) +static void tdf_request_sync(struct xe_device *xe) { - struct xe_gt *gt; unsigned int fw_ref; + struct xe_gt *gt; u8 id; - if (!IS_DGFX(xe) || GRAPHICS_VER(xe) < 20) - return; - - if (XE_WA(xe_root_mmio_gt(xe), 16023588340)) { - xe_device_l2_flush(xe); - return; - } - for_each_gt(gt, xe, id) { if (xe_gt_is_media_type(gt)) continue; @@ -1013,6 +1005,7 @@ void xe_device_td_flush(struct xe_device *xe) return; xe_mmio_write32(>->mmio, XE2_TDF_CTRL, TRANSIENT_FLUSH_REQUEST); + /* * FIXME: We can likely do better here with our choice of * timeout. Currently we just assume the worst case, i.e. 150us, @@ -1043,15 +1036,52 @@ void xe_device_l2_flush(struct xe_device *xe) return; spin_lock(>->global_invl_lock); - xe_mmio_write32(>->mmio, XE2_GLOBAL_INVAL, 0x1); + xe_mmio_write32(>->mmio, XE2_GLOBAL_INVAL, 0x1); if (xe_mmio_wait32(>->mmio, XE2_GLOBAL_INVAL, 0x1, 0x0, 500, NULL, true)) xe_gt_err_once(gt, "Global invalidation timeout\n"); + spin_unlock(>->global_invl_lock); xe_force_wake_put(gt_to_fw(gt), fw_ref); } +/** + * xe_device_td_flush() - Flush transient L3 cache entries + * @xe: The device + * + * Display engine has direct access to memory and is never coherent with L3/L4 + * caches (or CPU caches), however KMD is responsible for specifically flushing + * transient L3 GPU cache entries prior to the flip sequence to ensure scanout + * can happen from such a surface without seeing corruption. + * + * Display surfaces can be tagged as transient by mapping it using one of the + * various L3:XD PAT index modes on Xe2. + * + * Note: On non-discrete xe2 platforms, like LNL, the entire L3 cache is flushed + * at the end of each submission via PIPE_CONTROL for compute/render, since SA + * Media is not coherent with L3 and we want to support render-vs-media + * usescases. For other engines like copy/blt the HW internally forces uncached + * behaviour, hence why we can skip the TDF on such platforms. + */ +void xe_device_td_flush(struct xe_device *xe) +{ + struct xe_gt *root_gt; + + if (!IS_DGFX(xe) || GRAPHICS_VER(xe) < 20) + return; + + root_gt = xe_root_mmio_gt(xe); + if (XE_WA(root_gt, 16023588340)) { + /* A transient flush is not sufficient: flush the L2 */ + xe_device_l2_flush(xe); + } else { + xe_guc_pc_apply_flush_freq_limit(&root_gt->uc.guc.pc); + tdf_request_sync(xe); + xe_guc_pc_remove_flush_freq_limit(&root_gt->uc.guc.pc); + } +} + u32 xe_device_ccs_bytes(struct xe_device *xe, u64 size) { return xe_device_has_flat_ccs(xe) ? |