diff options
Diffstat (limited to 'drivers/gpu/drm/radeon')
30 files changed, 414 insertions, 273 deletions
diff --git a/drivers/gpu/drm/radeon/atombios_crtc.c b/drivers/gpu/drm/radeon/atombios_crtc.c index 30d242b25078..ed644a4f6f57 100644 --- a/drivers/gpu/drm/radeon/atombios_crtc.c +++ b/drivers/gpu/drm/radeon/atombios_crtc.c @@ -1851,10 +1851,9 @@ static int radeon_atom_pick_pll(struct drm_crtc *crtc) return pll; } /* otherwise, pick one of the plls */ - if ((rdev->family == CHIP_KAVERI) || - (rdev->family == CHIP_KABINI) || + if ((rdev->family == CHIP_KABINI) || (rdev->family == CHIP_MULLINS)) { - /* KB/KV/ML has PPLL1 and PPLL2 */ + /* KB/ML has PPLL1 and PPLL2 */ pll_in_use = radeon_get_pll_use_mask(crtc); if (!(pll_in_use & (1 << ATOM_PPLL2))) return ATOM_PPLL2; @@ -1863,7 +1862,7 @@ static int radeon_atom_pick_pll(struct drm_crtc *crtc) DRM_ERROR("unable to allocate a PPLL\n"); return ATOM_PPLL_INVALID; } else { - /* CI has PPLL0, PPLL1, and PPLL2 */ + /* CI/KV has PPLL0, PPLL1, and PPLL2 */ pll_in_use = radeon_get_pll_use_mask(crtc); if (!(pll_in_use & (1 << ATOM_PPLL2))) return ATOM_PPLL2; @@ -2039,6 +2038,7 @@ int atombios_crtc_mode_set(struct drm_crtc *crtc, atombios_crtc_set_base(crtc, x, y, old_fb); atombios_overscan_setup(crtc, mode, adjusted_mode); atombios_scaler_setup(crtc); + radeon_cursor_reset(crtc); /* update the hw version fpr dpm */ radeon_crtc->hw_mode = *adjusted_mode; @@ -2154,6 +2154,7 @@ static void atombios_crtc_disable(struct drm_crtc *crtc) case ATOM_PPLL0: /* disable the ppll */ if ((rdev->family == CHIP_ARUBA) || + (rdev->family == CHIP_KAVERI) || (rdev->family == CHIP_BONAIRE) || (rdev->family == CHIP_HAWAII)) atombios_crtc_program_pll(crtc, radeon_crtc->crtc_id, radeon_crtc->pll_id, diff --git a/drivers/gpu/drm/radeon/atombios_dp.c b/drivers/gpu/drm/radeon/atombios_dp.c index 11ba9d21b89b..db42a670f995 100644 --- a/drivers/gpu/drm/radeon/atombios_dp.c +++ b/drivers/gpu/drm/radeon/atombios_dp.c @@ -492,6 +492,10 @@ int radeon_dp_mode_valid_helper(struct drm_connector *connector, struct radeon_connector_atom_dig *dig_connector; int dp_clock; + if ((mode->clock > 340000) && + (!radeon_connector_is_dp12_capable(connector))) + return MODE_CLOCK_HIGH; + if (!radeon_connector->con_priv) return MODE_CLOCK_HIGH; dig_connector = radeon_connector->con_priv; diff --git a/drivers/gpu/drm/radeon/ci_dpm.c b/drivers/gpu/drm/radeon/ci_dpm.c index 3f898d020ae6..f373a81ba3d5 100644 --- a/drivers/gpu/drm/radeon/ci_dpm.c +++ b/drivers/gpu/drm/radeon/ci_dpm.c @@ -937,7 +937,7 @@ static void ci_fan_ctrl_set_static_mode(struct radeon_device *rdev, u32 mode) tmp |= TMIN(0); WREG32_SMC(CG_FDO_CTRL2, tmp); - tmp = RREG32_SMC(CG_FDO_CTRL2) & FDO_PWM_MODE_MASK; + tmp = RREG32_SMC(CG_FDO_CTRL2) & ~FDO_PWM_MODE_MASK; tmp |= FDO_PWM_MODE(mode); WREG32_SMC(CG_FDO_CTRL2, tmp); } @@ -1162,7 +1162,7 @@ static int ci_fan_ctrl_set_fan_speed_rpm(struct radeon_device *rdev, tmp |= TARGET_PERIOD(tach_period); WREG32_SMC(CG_TACH_CTRL, tmp); - ci_fan_ctrl_set_static_mode(rdev, FDO_PWM_MODE_STATIC); + ci_fan_ctrl_set_static_mode(rdev, FDO_PWM_MODE_STATIC_RPM); return 0; } @@ -1178,7 +1178,7 @@ static void ci_fan_ctrl_set_default_mode(struct radeon_device *rdev) tmp |= FDO_PWM_MODE(pi->fan_ctrl_default_mode); WREG32_SMC(CG_FDO_CTRL2, tmp); - tmp = RREG32_SMC(CG_FDO_CTRL2) & TMIN_MASK; + tmp = RREG32_SMC(CG_FDO_CTRL2) & ~TMIN_MASK; tmp |= TMIN(pi->t_min); WREG32_SMC(CG_FDO_CTRL2, tmp); pi->fan_ctrl_is_in_default_mode = true; @@ -5849,7 +5849,6 @@ int ci_dpm_init(struct radeon_device *rdev) rdev->pm.dpm.dyn_state.max_clock_voltage_on_ac; pi->fan_ctrl_is_in_default_mode = true; - rdev->pm.dpm.fan.ucode_fan_control = false; return 0; } diff --git a/drivers/gpu/drm/radeon/cikd.h b/drivers/gpu/drm/radeon/cikd.h index e4e88ca8b82e..03003f8a6de6 100644 --- a/drivers/gpu/drm/radeon/cikd.h +++ b/drivers/gpu/drm/radeon/cikd.h @@ -213,18 +213,18 @@ #define CG_FDO_CTRL0 0xC0300064 #define FDO_STATIC_DUTY(x) ((x) << 0) -#define FDO_STATIC_DUTY_MASK 0x0000000F +#define FDO_STATIC_DUTY_MASK 0x000000FF #define FDO_STATIC_DUTY_SHIFT 0 #define CG_FDO_CTRL1 0xC0300068 #define FMAX_DUTY100(x) ((x) << 0) -#define FMAX_DUTY100_MASK 0x0000000F +#define FMAX_DUTY100_MASK 0x000000FF #define FMAX_DUTY100_SHIFT 0 #define CG_FDO_CTRL2 0xC030006C #define TMIN(x) ((x) << 0) -#define TMIN_MASK 0x0000000F +#define TMIN_MASK 0x000000FF #define TMIN_SHIFT 0 #define FDO_PWM_MODE(x) ((x) << 11) -#define FDO_PWM_MODE_MASK (3 << 11) +#define FDO_PWM_MODE_MASK (7 << 11) #define FDO_PWM_MODE_SHIFT 11 #define TACH_PWM_RESP_RATE(x) ((x) << 25) #define TACH_PWM_RESP_RATE_MASK (0x7f << 25) @@ -2156,4 +2156,6 @@ #define ATC_VM_APERTURE1_HIGH_ADDR 0x330Cu #define ATC_VM_APERTURE1_LOW_ADDR 0x3304u +#define IH_VMID_0_LUT 0x3D40u + #endif diff --git a/drivers/gpu/drm/radeon/dce3_1_afmt.c b/drivers/gpu/drm/radeon/dce3_1_afmt.c index 2fe8cfc966d9..bafdf92a5732 100644 --- a/drivers/gpu/drm/radeon/dce3_1_afmt.c +++ b/drivers/gpu/drm/radeon/dce3_1_afmt.c @@ -103,7 +103,7 @@ static void dce3_2_afmt_write_sad_regs(struct drm_encoder *encoder) } sad_count = drm_edid_to_sad(radeon_connector->edid, &sads); - if (sad_count < 0) { + if (sad_count <= 0) { DRM_ERROR("Couldn't read SADs: %d\n", sad_count); return; } diff --git a/drivers/gpu/drm/radeon/evergreen_cs.c b/drivers/gpu/drm/radeon/evergreen_cs.c index 5c8b358f9fba..924b1b7ab455 100644 --- a/drivers/gpu/drm/radeon/evergreen_cs.c +++ b/drivers/gpu/drm/radeon/evergreen_cs.c @@ -35,7 +35,7 @@ #define MIN(a,b) (((a)<(b))?(a):(b)) int r600_dma_cs_next_reloc(struct radeon_cs_parser *p, - struct radeon_cs_reloc **cs_reloc); + struct radeon_bo_list **cs_reloc); struct evergreen_cs_track { u32 group_size; u32 nbanks; @@ -1094,7 +1094,7 @@ static int evergreen_cs_parse_packet0(struct radeon_cs_parser *p, static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx) { struct evergreen_cs_track *track = (struct evergreen_cs_track *)p->track; - struct radeon_cs_reloc *reloc; + struct radeon_bo_list *reloc; u32 last_reg; u32 m, i, tmp, *ib; int r; @@ -1792,7 +1792,7 @@ static bool evergreen_is_safe_reg(struct radeon_cs_parser *p, u32 reg, u32 idx) static int evergreen_packet3_check(struct radeon_cs_parser *p, struct radeon_cs_packet *pkt) { - struct radeon_cs_reloc *reloc; + struct radeon_bo_list *reloc; struct evergreen_cs_track *track; volatile u32 *ib; unsigned idx; @@ -2661,7 +2661,7 @@ int evergreen_cs_parse(struct radeon_cs_parser *p) p->track = NULL; return r; } - } while (p->idx < p->chunks[p->chunk_ib_idx].length_dw); + } while (p->idx < p->chunk_ib->length_dw); #if 0 for (r = 0; r < p->ib.length_dw; r++) { printk(KERN_INFO "%05d 0x%08X\n", r, p->ib.ptr[r]); @@ -2684,8 +2684,8 @@ int evergreen_cs_parse(struct radeon_cs_parser *p) **/ int evergreen_dma_cs_parse(struct radeon_cs_parser *p) { - struct radeon_cs_chunk *ib_chunk = &p->chunks[p->chunk_ib_idx]; - struct radeon_cs_reloc *src_reloc, *dst_reloc, *dst2_reloc; + struct radeon_cs_chunk *ib_chunk = p->chunk_ib; + struct radeon_bo_list *src_reloc, *dst_reloc, *dst2_reloc; u32 header, cmd, count, sub_cmd; volatile u32 *ib = p->ib.ptr; u32 idx; @@ -3100,7 +3100,7 @@ int evergreen_dma_cs_parse(struct radeon_cs_parser *p) DRM_ERROR("Unknown packet type %d at %d !\n", cmd, idx); return -EINVAL; } - } while (p->idx < p->chunks[p->chunk_ib_idx].length_dw); + } while (p->idx < p->chunk_ib->length_dw); #if 0 for (r = 0; r < p->ib->length_dw; r++) { printk(KERN_INFO "%05d 0x%08X\n", r, p->ib.ptr[r]); diff --git a/drivers/gpu/drm/radeon/kv_dpm.c b/drivers/gpu/drm/radeon/kv_dpm.c index 9b42001295ba..e3e9c10cfba9 100644 --- a/drivers/gpu/drm/radeon/kv_dpm.c +++ b/drivers/gpu/drm/radeon/kv_dpm.c @@ -2745,13 +2745,11 @@ int kv_dpm_init(struct radeon_device *rdev) pi->enable_auto_thermal_throttling = true; pi->disable_nb_ps3_in_battery = false; if (radeon_bapm == -1) { - /* There are stability issues reported on with - * bapm enabled on an asrock system. - */ - if (rdev->pdev->subsystem_vendor == 0x1849) - pi->bapm_enable = false; - else + /* only enable bapm on KB, ML by default */ + if (rdev->family == CHIP_KABINI || rdev->family == CHIP_MULLINS) pi->bapm_enable = true; + else + pi->bapm_enable = false; } else if (radeon_bapm == 0) { pi->bapm_enable = false; } else { diff --git a/drivers/gpu/drm/radeon/mkregtable.c b/drivers/gpu/drm/radeon/mkregtable.c index 4a85bb644e24..b928c17bdeed 100644 --- a/drivers/gpu/drm/radeon/mkregtable.c +++ b/drivers/gpu/drm/radeon/mkregtable.c @@ -347,7 +347,7 @@ static inline void list_splice_tail_init(struct list_head *list, * list_entry - get the struct for this entry * @ptr: the &struct list_head pointer. * @type: the type of the struct this is embedded in. - * @member: the name of the list_struct within the struct. + * @member: the name of the list_head within the struct. */ #define list_entry(ptr, type, member) \ container_of(ptr, type, member) @@ -356,7 +356,7 @@ static inline void list_splice_tail_init(struct list_head *list, * list_first_entry - get the first element from a list * @ptr: the list head to take the element from. * @type: the type of the struct this is embedded in. - * @member: the name of the list_struct within the struct. + * @member: the name of the list_head within the struct. * * Note, that list is expected to be not empty. */ @@ -406,7 +406,7 @@ static inline void list_splice_tail_init(struct list_head *list, * list_for_each_entry - iterate over list of given type * @pos: the type * to use as a loop cursor. * @head: the head for your list. - * @member: the name of the list_struct within the struct. + * @member: the name of the list_head within the struct. */ #define list_for_each_entry(pos, head, member) \ for (pos = list_entry((head)->next, typeof(*pos), member); \ @@ -417,7 +417,7 @@ static inline void list_splice_tail_init(struct list_head *list, * list_for_each_entry_reverse - iterate backwards over list of given type. * @pos: the type * to use as a loop cursor. * @head: the head for your list. - * @member: the name of the list_struct within the struct. + * @member: the name of the list_head within the struct. */ #define list_for_each_entry_reverse(pos, head, member) \ for (pos = list_entry((head)->prev, typeof(*pos), member); \ @@ -428,7 +428,7 @@ static inline void list_splice_tail_init(struct list_head *list, * list_prepare_entry - prepare a pos entry for use in list_for_each_entry_continue() * @pos: the type * to use as a start point * @head: the head of the list - * @member: the name of the list_struct within the struct. + * @member: the name of the list_head within the struct. * * Prepares a pos entry for use as a start point in list_for_each_entry_continue(). */ @@ -439,7 +439,7 @@ static inline void list_splice_tail_init(struct list_head *list, * list_for_each_entry_continue - continue iteration over list of given type * @pos: the type * to use as a loop cursor. * @head: the head for your list. - * @member: the name of the list_struct within the struct. + * @member: the name of the list_head within the struct. * * Continue to iterate over list of given type, continuing after * the current position. @@ -453,7 +453,7 @@ static inline void list_splice_tail_init(struct list_head *list, * list_for_each_entry_continue_reverse - iterate backwards from the given point * @pos: the type * to use as a loop cursor. * @head: the head for your list. - * @member: the name of the list_struct within the struct. + * @member: the name of the list_head within the struct. * * Start to iterate over list of given type backwards, continuing after * the current position. @@ -467,7 +467,7 @@ static inline void list_splice_tail_init(struct list_head *list, * list_for_each_entry_from - iterate over list of given type from the current point * @pos: the type * to use as a loop cursor. * @head: the head for your list. - * @member: the name of the list_struct within the struct. + * @member: the name of the list_head within the struct. * * Iterate over list of given type, continuing from current position. */ @@ -480,7 +480,7 @@ static inline void list_splice_tail_init(struct list_head *list, * @pos: the type * to use as a loop cursor. * @n: another type * to use as temporary storage * @head: the head for your list. - * @member: the name of the list_struct within the struct. + * @member: the name of the list_head within the struct. */ #define list_for_each_entry_safe(pos, n, head, member) \ for (pos = list_entry((head)->next, typeof(*pos), member), \ @@ -493,7 +493,7 @@ static inline void list_splice_tail_init(struct list_head *list, * @pos: the type * to use as a loop cursor. * @n: another type * to use as temporary storage * @head: the head for your list. - * @member: the name of the list_struct within the struct. + * @member: the name of the list_head within the struct. * * Iterate over list of given type, continuing after current point, * safe against removal of list entry. @@ -509,7 +509,7 @@ static inline void list_splice_tail_init(struct list_head *list, * @pos: the type * to use as a loop cursor. * @n: another type * to use as temporary storage * @head: the head for your list. - * @member: the name of the list_struct within the struct. + * @member: the name of the list_head within the struct. * * Iterate over list of given type from current point, safe against * removal of list entry. @@ -524,7 +524,7 @@ static inline void list_splice_tail_init(struct list_head *list, * @pos: the type * to use as a loop cursor. * @n: another type * to use as temporary storage * @head: the head for your list. - * @member: the name of the list_struct within the struct. + * @member: the name of the list_head within the struct. * * Iterate backwards over list of given type, safe against removal * of list entry. diff --git a/drivers/gpu/drm/radeon/r100.c b/drivers/gpu/drm/radeon/r100.c index b53b31a7b76f..74f06d540591 100644 --- a/drivers/gpu/drm/radeon/r100.c +++ b/drivers/gpu/drm/radeon/r100.c @@ -1254,7 +1254,7 @@ int r100_reloc_pitch_offset(struct radeon_cs_parser *p, int r; u32 tile_flags = 0; u32 tmp; - struct radeon_cs_reloc *reloc; + struct radeon_bo_list *reloc; u32 value; r = radeon_cs_packet_next_reloc(p, &reloc, 0); @@ -1293,7 +1293,7 @@ int r100_packet3_load_vbpntr(struct radeon_cs_parser *p, int idx) { unsigned c, i; - struct radeon_cs_reloc *reloc; + struct radeon_bo_list *reloc; struct r100_cs_track *track; int r = 0; volatile uint32_t *ib; @@ -1542,7 +1542,7 @@ static int r100_packet0_check(struct radeon_cs_parser *p, struct radeon_cs_packet *pkt, unsigned idx, unsigned reg) { - struct radeon_cs_reloc *reloc; + struct radeon_bo_list *reloc; struct r100_cs_track *track; volatile uint32_t *ib; uint32_t tmp; @@ -1901,7 +1901,7 @@ int r100_cs_track_check_pkt3_indx_buffer(struct radeon_cs_parser *p, static int r100_packet3_check(struct radeon_cs_parser *p, struct radeon_cs_packet *pkt) { - struct radeon_cs_reloc *reloc; + struct radeon_bo_list *reloc; struct r100_cs_track *track; unsigned idx; volatile uint32_t *ib; @@ -2061,7 +2061,7 @@ int r100_cs_parse(struct radeon_cs_parser *p) } if (r) return r; - } while (p->idx < p->chunks[p->chunk_ib_idx].length_dw); + } while (p->idx < p->chunk_ib->length_dw); return 0; } diff --git a/drivers/gpu/drm/radeon/r200.c b/drivers/gpu/drm/radeon/r200.c index 732d4938aab7..c70e6d5bcd19 100644 --- a/drivers/gpu/drm/radeon/r200.c +++ b/drivers/gpu/drm/radeon/r200.c @@ -146,7 +146,7 @@ int r200_packet0_check(struct radeon_cs_parser *p, struct radeon_cs_packet *pkt, unsigned idx, unsigned reg) { - struct radeon_cs_reloc *reloc; + struct radeon_bo_list *reloc; struct r100_cs_track *track; volatile uint32_t *ib; uint32_t tmp; diff --git a/drivers/gpu/drm/radeon/r300.c b/drivers/gpu/drm/radeon/r300.c index 1bc4704034ce..064ad5569cca 100644 --- a/drivers/gpu/drm/radeon/r300.c +++ b/drivers/gpu/drm/radeon/r300.c @@ -598,7 +598,7 @@ static int r300_packet0_check(struct radeon_cs_parser *p, struct radeon_cs_packet *pkt, unsigned idx, unsigned reg) { - struct radeon_cs_reloc *reloc; + struct radeon_bo_list *reloc; struct r100_cs_track *track; volatile uint32_t *ib; uint32_t tmp, tile_flags = 0; @@ -1142,7 +1142,7 @@ fail: static int r300_packet3_check(struct radeon_cs_parser *p, struct radeon_cs_packet *pkt) { - struct radeon_cs_reloc *reloc; + struct radeon_bo_list *reloc; struct r100_cs_track *track; volatile uint32_t *ib; unsigned idx; @@ -1283,7 +1283,7 @@ int r300_cs_parse(struct radeon_cs_parser *p) if (r) { return r; } - } while (p->idx < p->chunks[p->chunk_ib_idx].length_dw); + } while (p->idx < p->chunk_ib->length_dw); return 0; } diff --git a/drivers/gpu/drm/radeon/r600_cs.c b/drivers/gpu/drm/radeon/r600_cs.c index c47537a1ddba..acc1f99c84d9 100644 --- a/drivers/gpu/drm/radeon/r600_cs.c +++ b/drivers/gpu/drm/radeon/r600_cs.c @@ -969,7 +969,7 @@ static int r600_cs_parse_packet0(struct radeon_cs_parser *p, static int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx) { struct r600_cs_track *track = (struct r600_cs_track *)p->track; - struct radeon_cs_reloc *reloc; + struct radeon_bo_list *reloc; u32 m, i, tmp, *ib; int r; @@ -1626,7 +1626,7 @@ static bool r600_is_safe_reg(struct radeon_cs_parser *p, u32 reg, u32 idx) static int r600_packet3_check(struct radeon_cs_parser *p, struct radeon_cs_packet *pkt) { - struct radeon_cs_reloc *reloc; + struct radeon_bo_list *reloc; struct r600_cs_track *track; volatile u32 *ib; unsigned idx; @@ -2316,7 +2316,7 @@ int r600_cs_parse(struct radeon_cs_parser *p) p->track = NULL; return r; } - } while (p->idx < p->chunks[p->chunk_ib_idx].length_dw); + } while (p->idx < p->chunk_ib->length_dw); #if 0 for (r = 0; r < p->ib.length_dw; r++) { printk(KERN_INFO "%05d 0x%08X\n", r, p->ib.ptr[r]); @@ -2351,10 +2351,10 @@ static void r600_cs_parser_fini(struct radeon_cs_parser *parser, int error) static int r600_cs_parser_relocs_legacy(struct radeon_cs_parser *p) { - if (p->chunk_relocs_idx == -1) { + if (p->chunk_relocs == NULL) { return 0; } - p->relocs = kzalloc(sizeof(struct radeon_cs_reloc), GFP_KERNEL); + p->relocs = kzalloc(sizeof(struct radeon_bo_list), GFP_KERNEL); if (p->relocs == NULL) { return -ENOMEM; } @@ -2398,7 +2398,7 @@ int r600_cs_legacy(struct drm_device *dev, void *data, struct drm_file *filp, /* Copy the packet into the IB, the parser will read from the * input memory (cached) and write to the IB (which can be * uncached). */ - ib_chunk = &parser.chunks[parser.chunk_ib_idx]; + ib_chunk = parser.chunk_ib; parser.ib.length_dw = ib_chunk->length_dw; *l = parser.ib.length_dw; if (copy_from_user(ib, ib_chunk->user_ptr, ib_chunk->length_dw * 4)) { @@ -2435,24 +2435,24 @@ void r600_cs_legacy_init(void) * GPU offset using the provided start. **/ int r600_dma_cs_next_reloc(struct radeon_cs_parser *p, - struct radeon_cs_reloc **cs_reloc) + struct radeon_bo_list **cs_reloc) { struct radeon_cs_chunk *relocs_chunk; unsigned idx; *cs_reloc = NULL; - if (p->chunk_relocs_idx == -1) { + if (p->chunk_relocs == NULL) { DRM_ERROR("No relocation chunk !\n"); return -EINVAL; } - relocs_chunk = &p->chunks[p->chunk_relocs_idx]; + relocs_chunk = p->chunk_relocs; idx = p->dma_reloc_idx; if (idx >= p->nrelocs) { DRM_ERROR("Relocs at %d after relocations chunk end %d !\n", idx, p->nrelocs); return -EINVAL; } - *cs_reloc = p->relocs_ptr[idx]; + *cs_reloc = &p->relocs[idx]; p->dma_reloc_idx++; return 0; } @@ -2472,8 +2472,8 @@ int r600_dma_cs_next_reloc(struct radeon_cs_parser *p, **/ int r600_dma_cs_parse(struct radeon_cs_parser *p) { - struct radeon_cs_chunk *ib_chunk = &p->chunks[p->chunk_ib_idx]; - struct radeon_cs_reloc *src_reloc, *dst_reloc; + struct radeon_cs_chunk *ib_chunk = p->chunk_ib; + struct radeon_bo_list *src_reloc, *dst_reloc; u32 header, cmd, count, tiled; volatile u32 *ib = p->ib.ptr; u32 idx, idx_value; @@ -2619,7 +2619,7 @@ int r600_dma_cs_parse(struct radeon_cs_parser *p) DRM_ERROR("Unknown packet type %d at %d !\n", cmd, idx); return -EINVAL; } - } while (p->idx < p->chunks[p->chunk_ib_idx].length_dw); + } while (p->idx < p->chunk_ib->length_dw); #if 0 for (r = 0; r < p->ib->length_dw; r++) { printk(KERN_INFO "%05d 0x%08X\n", r, p->ib.ptr[r]); diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h index 3207bb60715e..54529b837afa 100644 --- a/drivers/gpu/drm/radeon/radeon.h +++ b/drivers/gpu/drm/radeon/radeon.h @@ -450,6 +450,15 @@ struct radeon_mman { #endif }; +struct radeon_bo_list { + struct radeon_bo *robj; + struct ttm_validate_buffer tv; + uint64_t gpu_offset; + unsigned prefered_domains; + unsigned allowed_domains; + uint32_t tiling_flags; +}; + /* bo virtual address in a specific vm */ struct radeon_bo_va { /* protected by bo being reserved */ @@ -920,6 +929,9 @@ struct radeon_vm { struct rb_root va; + /* protecting invalidated and freed */ + spinlock_t status_lock; + /* BOs moved, but not yet updated in the PT */ struct list_head invalidated; @@ -1044,19 +1056,7 @@ void cayman_dma_fini(struct radeon_device *rdev); /* * CS. */ -struct radeon_cs_reloc { - struct drm_gem_object *gobj; - struct radeon_bo *robj; - struct ttm_validate_buffer tv; - uint64_t gpu_offset; - unsigned prefered_domains; - unsigned allowed_domains; - uint32_t tiling_flags; - uint32_t handle; -}; - struct radeon_cs_chunk { - uint32_t chunk_id; uint32_t length_dw; uint32_t *kdata; void __user *user_ptr; @@ -1074,16 +1074,15 @@ struct radeon_cs_parser { unsigned idx; /* relocations */ unsigned nrelocs; - struct radeon_cs_reloc *relocs; - struct radeon_cs_reloc **relocs_ptr; - struct radeon_cs_reloc *vm_bos; + struct radeon_bo_list *relocs; + struct radeon_bo_list *vm_bos; struct list_head validated; unsigned dma_reloc_idx; /* indices of various chunks */ - int chunk_ib_idx; - int chunk_relocs_idx; - int chunk_flags_idx; - int chunk_const_ib_idx; + struct radeon_cs_chunk *chunk_ib; + struct radeon_cs_chunk *chunk_relocs; + struct radeon_cs_chunk *chunk_flags; + struct radeon_cs_chunk *chunk_const_ib; struct radeon_ib ib; struct radeon_ib const_ib; void *track; @@ -1097,7 +1096,7 @@ struct radeon_cs_parser { static inline u32 radeon_get_ib_value(struct radeon_cs_parser *p, int idx) { - struct radeon_cs_chunk *ibc = &p->chunks[p->chunk_ib_idx]; + struct radeon_cs_chunk *ibc = p->chunk_ib; if (ibc->kdata) return ibc->kdata[idx]; @@ -2975,7 +2974,7 @@ int radeon_vm_manager_init(struct radeon_device *rdev); void radeon_vm_manager_fini(struct radeon_device *rdev); int radeon_vm_init(struct radeon_device *rdev, struct radeon_vm *vm); void radeon_vm_fini(struct radeon_device *rdev, struct radeon_vm *vm); -struct radeon_cs_reloc *radeon_vm_get_bos(struct radeon_device *rdev, +struct radeon_bo_list *radeon_vm_get_bos(struct radeon_device *rdev, struct radeon_vm *vm, struct list_head *head); struct radeon_fence *radeon_vm_grab_id(struct radeon_device *rdev, @@ -3089,7 +3088,7 @@ bool radeon_cs_packet_next_is_pkt3_nop(struct radeon_cs_parser *p); void radeon_cs_dump_packet(struct radeon_cs_parser *p, struct radeon_cs_packet *pkt); int radeon_cs_packet_next_reloc(struct radeon_cs_parser *p, - struct radeon_cs_reloc **cs_reloc, + struct radeon_bo_list **cs_reloc, int nomm); int r600_cs_common_vline_parse(struct radeon_cs_parser *p, uint32_t *vline_start_end, diff --git a/drivers/gpu/drm/radeon/radeon_cs.c b/drivers/gpu/drm/radeon/radeon_cs.c index 75f22e5e999f..c830863bc98a 100644 --- a/drivers/gpu/drm/radeon/radeon_cs.c +++ b/drivers/gpu/drm/radeon/radeon_cs.c @@ -77,22 +77,18 @@ static int radeon_cs_parser_relocs(struct radeon_cs_parser *p) struct drm_device *ddev = p->rdev->ddev; struct radeon_cs_chunk *chunk; struct radeon_cs_buckets buckets; - unsigned i, j; - bool duplicate, need_mmap_lock = false; + unsigned i; + bool need_mmap_lock = false; int r; - if (p->chunk_relocs_idx == -1) { + if (p->chunk_relocs == NULL) { return 0; } - chunk = &p->chunks[p->chunk_relocs_idx]; + chunk = p->chunk_relocs; p->dma_reloc_idx = 0; /* FIXME: we assume that each relocs use 4 dwords */ p->nrelocs = chunk->length_dw / 4; - p->relocs_ptr = kcalloc(p->nrelocs, sizeof(void *), GFP_KERNEL); - if (p->relocs_ptr == NULL) { - return -ENOMEM; - } - p->relocs = kcalloc(p->nrelocs, sizeof(struct radeon_cs_reloc), GFP_KERNEL); + p->relocs = kcalloc(p->nrelocs, sizeof(struct radeon_bo_list), GFP_KERNEL); if (p->relocs == NULL) { return -ENOMEM; } @@ -101,31 +97,17 @@ static int radeon_cs_parser_relocs(struct radeon_cs_parser *p) for (i = 0; i < p->nrelocs; i++) { struct drm_radeon_cs_reloc *r; + struct drm_gem_object *gobj; unsigned priority; - duplicate = false; r = (struct drm_radeon_cs_reloc *)&chunk->kdata[i*4]; - for (j = 0; j < i; j++) { - if (r->handle == p->relocs[j].handle) { - p->relocs_ptr[i] = &p->relocs[j]; - duplicate = true; - break; - } - } - if (duplicate) { - p->relocs[i].handle = 0; - continue; - } - - p->relocs[i].gobj = drm_gem_object_lookup(ddev, p->filp, - r->handle); - if (p->relocs[i].gobj == NULL) { + gobj = drm_gem_object_lookup(ddev, p->filp, r->handle); + if (gobj == NULL) { DRM_ERROR("gem object lookup failed 0x%x\n", r->handle); return -ENOENT; } - p->relocs_ptr[i] = &p->relocs[i]; - p->relocs[i].robj = gem_to_radeon_bo(p->relocs[i].gobj); + p->relocs[i].robj = gem_to_radeon_bo(gobj); /* The userspace buffer priorities are from 0 to 15. A higher * number means the buffer is more important. @@ -184,7 +166,6 @@ static int radeon_cs_parser_relocs(struct radeon_cs_parser *p) p->relocs[i].tv.bo = &p->relocs[i].robj->tbo; p->relocs[i].tv.shared = !r->write_domain; - p->relocs[i].handle = r->handle; radeon_cs_buckets_add(&buckets, &p->relocs[i].tv.head, priority); @@ -251,22 +232,19 @@ static int radeon_cs_get_ring(struct radeon_cs_parser *p, u32 ring, s32 priority static int radeon_cs_sync_rings(struct radeon_cs_parser *p) { - int i, r = 0; + struct radeon_bo_list *reloc; + int r; - for (i = 0; i < p->nrelocs; i++) { + list_for_each_entry(reloc, &p->validated, tv.head) { struct reservation_object *resv; - if (!p->relocs[i].robj) - continue; - - resv = p->relocs[i].robj->tbo.resv; + resv = reloc->robj->tbo.resv; r = radeon_sync_resv(p->rdev, &p->ib.sync, resv, - p->relocs[i].tv.shared); - + reloc->tv.shared); if (r) - break; + return r; } - return r; + return 0; } /* XXX: note that this is called from the legacy UMS CS ioctl as well */ @@ -286,10 +264,10 @@ int radeon_cs_parser_init(struct radeon_cs_parser *p, void *data) p->idx = 0; p->ib.sa_bo = NULL; p->const_ib.sa_bo = NULL; - p->chunk_ib_idx = -1; - p->chunk_relocs_idx = -1; - p->chunk_flags_idx = -1; - p->chunk_const_ib_idx = -1; + p->chunk_ib = NULL; + p->chunk_relocs = NULL; + p->chunk_flags = NULL; + p->chunk_const_ib = NULL; p->chunks_array = kcalloc(cs->num_chunks, sizeof(uint64_t), GFP_KERNEL); if (p->chunks_array == NULL) { return -ENOMEM; @@ -316,24 +294,23 @@ int radeon_cs_parser_init(struct radeon_cs_parser *p, void *data) return -EFAULT; } p->chunks[i].length_dw = user_chunk.length_dw; - p->chunks[i].chunk_id = user_chunk.chunk_id; - if (p->chunks[i].chunk_id == RADEON_CHUNK_ID_RELOCS) { - p->chunk_relocs_idx = i; + if (user_chunk.chunk_id == RADEON_CHUNK_ID_RELOCS) { + p->chunk_relocs = &p->chunks[i]; } - if (p->chunks[i].chunk_id == RADEON_CHUNK_ID_IB) { - p->chunk_ib_idx = i; + if (user_chunk.chunk_id == RADEON_CHUNK_ID_IB) { + p->chunk_ib = &p->chunks[i]; /* zero length IB isn't useful */ if (p->chunks[i].length_dw == 0) return -EINVAL; } - if (p->chunks[i].chunk_id == RADEON_CHUNK_ID_CONST_IB) { - p->chunk_const_ib_idx = i; + if (user_chunk.chunk_id == RADEON_CHUNK_ID_CONST_IB) { + p->chunk_const_ib = &p->chunks[i]; /* zero length CONST IB isn't useful */ if (p->chunks[i].length_dw == 0) return -EINVAL; } - if (p->chunks[i].chunk_id == RADEON_CHUNK_ID_FLAGS) { - p->chunk_flags_idx = i; + if (user_chunk.chunk_id == RADEON_CHUNK_ID_FLAGS) { + p->chunk_flags = &p->chunks[i]; /* zero length flags aren't useful */ if (p->chunks[i].length_dw == 0) return -EINVAL; @@ -342,10 +319,10 @@ int radeon_cs_parser_init(struct radeon_cs_parser *p, void *data) size = p->chunks[i].length_dw; cdata = (void __user *)(unsigned long)user_chunk.chunk_data; p->chunks[i].user_ptr = cdata; - if (p->chunks[i].chunk_id == RADEON_CHUNK_ID_CONST_IB) + if (user_chunk.chunk_id == RADEON_CHUNK_ID_CONST_IB) continue; - if (p->chunks[i].chunk_id == RADEON_CHUNK_ID_IB) { + if (user_chunk.chunk_id == RADEON_CHUNK_ID_IB) { if (!p->rdev || !(p->rdev->flags & RADEON_IS_AGP)) continue; } @@ -358,7 +335,7 @@ int radeon_cs_parser_init(struct radeon_cs_parser *p, void *data) if (copy_from_user(p->chunks[i].kdata, cdata, size)) { return -EFAULT; } - if (p->chunks[i].chunk_id == RADEON_CHUNK_ID_FLAGS) { + if (user_chunk.chunk_id == RADEON_CHUNK_ID_FLAGS) { p->cs_flags = p->chunks[i].kdata[0]; if (p->chunks[i].length_dw > 1) ring = p->chunks[i].kdata[1]; @@ -399,8 +376,8 @@ int radeon_cs_parser_init(struct radeon_cs_parser *p, void *data) static int cmp_size_smaller_first(void *priv, struct list_head *a, struct list_head *b) { - struct radeon_cs_reloc *la = list_entry(a, struct radeon_cs_reloc, tv.head); - struct radeon_cs_reloc *lb = list_entry(b, struct radeon_cs_reloc, tv.head); + struct radeon_bo_list *la = list_entry(a, struct radeon_bo_list, tv.head); + struct radeon_bo_list *lb = list_entry(b, struct radeon_bo_list, tv.head); /* Sort A before B if A is smaller. */ return (int)la->robj->tbo.num_pages - (int)lb->robj->tbo.num_pages; @@ -441,13 +418,15 @@ static void radeon_cs_parser_fini(struct radeon_cs_parser *parser, int error, bo if (parser->relocs != NULL) { for (i = 0; i < parser->nrelocs; i++) { - if (parser->relocs[i].gobj) - drm_gem_object_unreference_unlocked(parser->relocs[i].gobj); + struct radeon_bo *bo = parser->relocs[i].robj; + if (bo == NULL) + continue; + + drm_gem_object_unreference_unlocked(&bo->gem_base); } } kfree(parser->track); kfree(parser->relocs); - kfree(parser->relocs_ptr); drm_free_large(parser->vm_bos); for (i = 0; i < parser->nchunks; i++) drm_free_large(parser->chunks[i].kdata); @@ -462,7 +441,7 @@ static int radeon_cs_ib_chunk(struct radeon_device *rdev, { int r; - if (parser->chunk_ib_idx == -1) + if (parser->chunk_ib == NULL) return 0; if (parser->cs_flags & RADEON_CS_USE_VM) @@ -505,9 +484,6 @@ static int radeon_bo_vm_update_pte(struct radeon_cs_parser *p, if (r) return r; - radeon_sync_resv(p->rdev, &p->ib.sync, vm->page_directory->tbo.resv, - true); - r = radeon_vm_clear_freed(rdev, vm); if (r) return r; @@ -525,10 +501,6 @@ static int radeon_bo_vm_update_pte(struct radeon_cs_parser *p, for (i = 0; i < p->nrelocs; i++) { struct radeon_bo *bo; - /* ignore duplicates */ - if (p->relocs_ptr[i] != &p->relocs[i]) - continue; - bo = p->relocs[i].robj; bo_va = radeon_vm_bo_find(vm, bo); if (bo_va == NULL) { @@ -553,7 +525,7 @@ static int radeon_cs_ib_vm_chunk(struct radeon_device *rdev, struct radeon_vm *vm = &fpriv->vm; int r; - if (parser->chunk_ib_idx == -1) + if (parser->chunk_ib == NULL) return 0; if ((parser->cs_flags & RADEON_CS_USE_VM) == 0) return 0; @@ -587,7 +559,7 @@ static int radeon_cs_ib_vm_chunk(struct radeon_device *rdev, } if ((rdev->family >= CHIP_TAHITI) && - (parser->chunk_const_ib_idx != -1)) { + (parser->chunk_const_ib != NULL)) { r = radeon_ib_schedule(rdev, &parser->ib, &parser->const_ib, true); } else { r = radeon_ib_schedule(rdev, &parser->ib, NULL, true); @@ -614,7 +586,7 @@ static int radeon_cs_ib_fill(struct radeon_device *rdev, struct radeon_cs_parser struct radeon_vm *vm = NULL; int r; - if (parser->chunk_ib_idx == -1) + if (parser->chunk_ib == NULL) return 0; if (parser->cs_flags & RADEON_CS_USE_VM) { @@ -622,8 +594,8 @@ static int radeon_cs_ib_fill(struct radeon_device *rdev, struct radeon_cs_parser vm = &fpriv->vm; if ((rdev->family >= CHIP_TAHITI) && - (parser->chunk_const_ib_idx != -1)) { - ib_chunk = &parser->chunks[parser->chunk_const_ib_idx]; + (parser->chunk_const_ib != NULL)) { + ib_chunk = parser->chunk_const_ib; if (ib_chunk->length_dw > RADEON_IB_VM_MAX_SIZE) { DRM_ERROR("cs IB CONST too big: %d\n", ib_chunk->length_dw); return -EINVAL; @@ -642,13 +614,13 @@ static int radeon_cs_ib_fill(struct radeon_device *rdev, struct radeon_cs_parser return -EFAULT; } - ib_chunk = &parser->chunks[parser->chunk_ib_idx]; + ib_chunk = parser->chunk_ib; if (ib_chunk->length_dw > RADEON_IB_VM_MAX_SIZE) { DRM_ERROR("cs IB too big: %d\n", ib_chunk->length_dw); return -EINVAL; } } - ib_chunk = &parser->chunks[parser->chunk_ib_idx]; + ib_chunk = parser->chunk_ib; r = radeon_ib_get(rdev, parser->ring, &parser->ib, vm, ib_chunk->length_dw * 4); @@ -740,7 +712,7 @@ int radeon_cs_packet_parse(struct radeon_cs_parser *p, struct radeon_cs_packet *pkt, unsigned idx) { - struct radeon_cs_chunk *ib_chunk = &p->chunks[p->chunk_ib_idx]; + struct radeon_cs_chunk *ib_chunk = p->chunk_ib; struct radeon_device *rdev = p->rdev; uint32_t header; @@ -834,7 +806,7 @@ void radeon_cs_dump_packet(struct radeon_cs_parser *p, * GPU offset using the provided start. **/ int radeon_cs_packet_next_reloc(struct radeon_cs_parser *p, - struct radeon_cs_reloc **cs_reloc, + struct radeon_bo_list **cs_reloc, int nomm) { struct radeon_cs_chunk *relocs_chunk; @@ -842,12 +814,12 @@ int radeon_cs_packet_next_reloc(struct radeon_cs_parser *p, unsigned idx; int r; - if (p->chunk_relocs_idx == -1) { + if (p->chunk_relocs == NULL) { DRM_ERROR("No relocation chunk !\n"); return -EINVAL; } *cs_reloc = NULL; - relocs_chunk = &p->chunks[p->chunk_relocs_idx]; + relocs_chunk = p->chunk_relocs; r = radeon_cs_packet_parse(p, &p3reloc, p->idx); if (r) return r; @@ -873,6 +845,6 @@ int radeon_cs_packet_next_reloc(struct radeon_cs_parser *p, (u64)relocs_chunk->kdata[idx + 3] << 32; (*cs_reloc)->gpu_offset |= relocs_chunk->kdata[idx + 0]; } else - *cs_reloc = p->relocs_ptr[(idx / 4)]; + *cs_reloc = &p->relocs[(idx / 4)]; return 0; } diff --git a/drivers/gpu/drm/radeon/radeon_cursor.c b/drivers/gpu/drm/radeon/radeon_cursor.c index 85f38ee11888..45e54060ee97 100644 --- a/drivers/gpu/drm/radeon/radeon_cursor.c +++ b/drivers/gpu/drm/radeon/radeon_cursor.c @@ -227,11 +227,24 @@ int radeon_crtc_cursor_move(struct drm_crtc *crtc, return ret; } -static void radeon_set_cursor(struct drm_crtc *crtc, struct drm_gem_object *obj, - uint64_t gpu_addr, int hot_x, int hot_y) +static int radeon_set_cursor(struct drm_crtc *crtc, struct drm_gem_object *obj) { struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc); struct radeon_device *rdev = crtc->dev->dev_private; + struct radeon_bo *robj = gem_to_radeon_bo(obj); + uint64_t gpu_addr; + int ret; + + ret = radeon_bo_reserve(robj, false); + if (unlikely(ret != 0)) + goto fail; + /* Only 27 bit offset for legacy cursor */ + ret = radeon_bo_pin_restricted(robj, RADEON_GEM_DOMAIN_VRAM, + ASIC_IS_AVIVO(rdev) ? 0 : 1 << 27, + &gpu_addr); + radeon_bo_unreserve(robj); + if (ret) + goto fail; if (ASIC_IS_DCE4(rdev)) { WREG32(EVERGREEN_CUR_SURFACE_ADDRESS_HIGH + radeon_crtc->crtc_offset, @@ -253,18 +266,12 @@ static void radeon_set_cursor(struct drm_crtc *crtc, struct drm_gem_object *obj, WREG32(RADEON_CUR_OFFSET + radeon_crtc->crtc_offset, radeon_crtc->legacy_cursor_offset); } - if (hot_x != radeon_crtc->cursor_hot_x || - hot_y != radeon_crtc->cursor_hot_y) { - int x, y; - - x = radeon_crtc->cursor_x + radeon_crtc->cursor_hot_x - hot_x; - y = radeon_crtc->cursor_y + radeon_crtc->cursor_hot_y - hot_y; + return 0; - radeon_cursor_move_locked(crtc, x, y); +fail: + drm_gem_object_unreference_unlocked(obj); - radeon_crtc->cursor_hot_x = hot_x; - radeon_crtc->cursor_hot_y = hot_y; - } + return ret; } int radeon_crtc_cursor_set2(struct drm_crtc *crtc, @@ -276,10 +283,7 @@ int radeon_crtc_cursor_set2(struct drm_crtc *crtc, int32_t hot_y) { struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc); - struct radeon_device *rdev = crtc->dev->dev_private; struct drm_gem_object *obj; - struct radeon_bo *robj; - uint64_t gpu_addr; int ret; if (!handle) { @@ -301,41 +305,76 @@ int radeon_crtc_cursor_set2(struct drm_crtc *crtc, return -ENOENT; } - robj = gem_to_radeon_bo(obj); - ret = radeon_bo_reserve(robj, false); - if (unlikely(ret != 0)) - goto fail; - /* Only 27 bit offset for legacy cursor */ - ret = radeon_bo_pin_restricted(robj, RADEON_GEM_DOMAIN_VRAM, - ASIC_IS_AVIVO(rdev) ? 0 : 1 << 27, - &gpu_addr); - radeon_bo_unreserve(robj); - if (ret) - goto fail; - radeon_crtc->cursor_width = width; radeon_crtc->cursor_height = height; radeon_lock_cursor(crtc, true); - radeon_set_cursor(crtc, obj, gpu_addr, hot_x, hot_y); - radeon_show_cursor(crtc); + + if (hot_x != radeon_crtc->cursor_hot_x || + hot_y != radeon_crtc->cursor_hot_y) { + int x, y; + + x = radeon_crtc->cursor_x + radeon_crtc->cursor_hot_x - hot_x; + y = radeon_crtc->cursor_y + radeon_crtc->cursor_hot_y - hot_y; + + radeon_cursor_move_locked(crtc, x, y); + + radeon_crtc->cursor_hot_x = hot_x; + radeon_crtc->cursor_hot_y = hot_y; + } + + ret = radeon_set_cursor(crtc, obj); + + if (ret) + DRM_ERROR("radeon_set_cursor returned %d, not changing cursor\n", + ret); + else + radeon_show_cursor(crtc); + radeon_lock_cursor(crtc, false); unpin: if (radeon_crtc->cursor_bo) { - robj = gem_to_radeon_bo(radeon_crtc->cursor_bo); + struct radeon_bo *robj = gem_to_radeon_bo(radeon_crtc->cursor_bo); ret = radeon_bo_reserve(robj, false); if (likely(ret == 0)) { radeon_bo_unpin(robj); radeon_bo_unreserve(robj); } - drm_gem_object_unreference_unlocked(radeon_crtc->cursor_bo); + if (radeon_crtc->cursor_bo != obj) + drm_gem_object_unreference_unlocked(radeon_crtc->cursor_bo); } radeon_crtc->cursor_bo = obj; return 0; -fail: - drm_gem_object_unreference_unlocked(obj); +} - return ret; +/** + * radeon_cursor_reset - Re-set the current cursor, if any. + * + * @crtc: drm crtc + * + * If the CRTC passed in currently has a cursor assigned, this function + * makes sure it's visible. + */ +void radeon_cursor_reset(struct drm_crtc *crtc) +{ + struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc); + int ret; + + if (radeon_crtc->cursor_bo) { + radeon_lock_cursor(crtc, true); + + radeon_cursor_move_locked(crtc, radeon_crtc->cursor_x, + radeon_crtc->cursor_y); + + ret = radeon_set_cursor(crtc, radeon_crtc->cursor_bo); + if (ret) + DRM_ERROR("radeon_set_cursor returned %d, not showing " + "cursor\n", ret); + else + radeon_show_cursor(crtc); + + radeon_lock_cursor(crtc, false); + } } diff --git a/drivers/gpu/drm/radeon/radeon_fb.c b/drivers/gpu/drm/radeon/radeon_fb.c index 0ea1db83d573..29b9220ec399 100644 --- a/drivers/gpu/drm/radeon/radeon_fb.c +++ b/drivers/gpu/drm/radeon/radeon_fb.c @@ -48,10 +48,40 @@ struct radeon_fbdev { struct radeon_device *rdev; }; +/** + * radeon_fb_helper_set_par - Hide cursor on CRTCs used by fbdev. + * + * @info: fbdev info + * + * This function hides the cursor on all CRTCs used by fbdev. + */ +static int radeon_fb_helper_set_par(struct fb_info *info) +{ + int ret; + + ret = drm_fb_helper_set_par(info); + + /* XXX: with universal plane support fbdev will automatically disable + * all non-primary planes (including the cursor) + */ + if (ret == 0) { + struct drm_fb_helper *fb_helper = info->par; + int i; + + for (i = 0; i < fb_helper->crtc_count; i++) { + struct drm_crtc *crtc = fb_helper->crtc_info[i].mode_set.crtc; + + radeon_crtc_cursor_set2(crtc, NULL, 0, 0, 0, 0, 0); + } + } + + return ret; +} + static struct fb_ops radeonfb_ops = { .owner = THIS_MODULE, .fb_check_var = drm_fb_helper_check_var, - .fb_set_par = drm_fb_helper_set_par, + .fb_set_par = radeon_fb_helper_set_par, .fb_fillrect = cfb_fillrect, .fb_copyarea = cfb_copyarea, .fb_imageblit = cfb_imageblit, diff --git a/drivers/gpu/drm/radeon/radeon_gem.c b/drivers/gpu/drm/radeon/radeon_gem.c index 12cfaeac1205..a46f73737994 100644 --- a/drivers/gpu/drm/radeon/radeon_gem.c +++ b/drivers/gpu/drm/radeon/radeon_gem.c @@ -394,10 +394,9 @@ int radeon_gem_set_domain_ioctl(struct drm_device *dev, void *data, return r; } -static int radeon_mode_mmap(struct drm_file *filp, - struct drm_device *dev, - uint32_t handle, bool dumb, - uint64_t *offset_p) +int radeon_mode_dumb_mmap(struct drm_file *filp, + struct drm_device *dev, + uint32_t handle, uint64_t *offset_p) { struct drm_gem_object *gobj; struct radeon_bo *robj; @@ -406,14 +405,6 @@ static int radeon_mode_mmap(struct drm_file *filp, if (gobj == NULL) { return -ENOENT; } - - /* - * We don't allow dumb mmaps on objects created using another - * interface. - */ - WARN_ONCE(dumb && !(gobj->dumb || gobj->import_attach), - "Illegal dumb map of GPU buffer.\n"); - robj = gem_to_radeon_bo(gobj); if (radeon_ttm_tt_has_userptr(robj->tbo.ttm)) { drm_gem_object_unreference_unlocked(gobj); @@ -424,20 +415,12 @@ static int radeon_mode_mmap(struct drm_file *filp, return 0; } -int radeon_mode_dumb_mmap(struct drm_file *filp, - struct drm_device *dev, - uint32_t handle, uint64_t *offset_p) -{ - return radeon_mode_mmap(filp, dev, handle, true, offset_p); -} - int radeon_gem_mmap_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) { struct drm_radeon_gem_mmap *args = data; - return radeon_mode_mmap(filp, dev, args->handle, false, - &args->addr_ptr); + return radeon_mode_dumb_mmap(filp, dev, args->handle, &args->addr_ptr); } int radeon_gem_busy_ioctl(struct drm_device *dev, void *data, @@ -548,7 +531,7 @@ static void radeon_gem_va_update_vm(struct radeon_device *rdev, struct radeon_bo_va *bo_va) { struct ttm_validate_buffer tv, *entry; - struct radeon_cs_reloc *vm_bos; + struct radeon_bo_list *vm_bos; struct ww_acquire_ctx ticket; struct list_head list; unsigned domain; @@ -564,7 +547,7 @@ static void radeon_gem_va_update_vm(struct radeon_device *rdev, if (!vm_bos) return; - r = ttm_eu_reserve_buffers(&ticket, &list, true); + r = ttm_eu_reserve_buffers(&ticket, &list, true, NULL); if (r) goto error_free; @@ -763,7 +746,6 @@ int radeon_mode_dumb_create(struct drm_file *file_priv, return -ENOMEM; r = drm_gem_handle_create(file_priv, gobj, &handle); - gobj->dumb = true; /* drop reference from allocate - handle holds it now */ drm_gem_object_unreference_unlocked(gobj); if (r) { diff --git a/drivers/gpu/drm/radeon/radeon_kfd.c b/drivers/gpu/drm/radeon/radeon_kfd.c index 065d02068ec3..8bf87f1203cc 100644 --- a/drivers/gpu/drm/radeon/radeon_kfd.c +++ b/drivers/gpu/drm/radeon/radeon_kfd.c @@ -28,6 +28,8 @@ #include "cikd.h" #include "cik_reg.h" #include "radeon_kfd.h" +#include "radeon_ucode.h" +#include <linux/firmware.h> #define CIK_PIPE_PER_MEC (4) @@ -49,6 +51,7 @@ static uint64_t get_vmem_size(struct kgd_dev *kgd); static uint64_t get_gpu_clock_counter(struct kgd_dev *kgd); static uint32_t get_max_engine_clock_in_mhz(struct kgd_dev *kgd); +static uint16_t get_fw_version(struct kgd_dev *kgd, enum kgd_engine_type type); /* * Register access functions @@ -69,7 +72,7 @@ static int kgd_init_pipeline(struct kgd_dev *kgd, uint32_t pipe_id, static int kgd_hqd_load(struct kgd_dev *kgd, void *mqd, uint32_t pipe_id, uint32_t queue_id, uint32_t __user *wptr); -static bool kgd_hqd_is_occupies(struct kgd_dev *kgd, uint64_t queue_address, +static bool kgd_hqd_is_occupied(struct kgd_dev *kgd, uint64_t queue_address, uint32_t pipe_id, uint32_t queue_id); static int kgd_hqd_destroy(struct kgd_dev *kgd, uint32_t reset_type, @@ -89,14 +92,16 @@ static const struct kfd2kgd_calls kfd2kgd = { .init_memory = kgd_init_memory, .init_pipeline = kgd_init_pipeline, .hqd_load = kgd_hqd_load, - .hqd_is_occupies = kgd_hqd_is_occupies, + .hqd_is_occupied = kgd_hqd_is_occupied, .hqd_destroy = kgd_hqd_destroy, + .get_fw_version = get_fw_version }; static const struct kgd2kfd_calls *kgd2kfd; bool radeon_kfd_init(void) { +#if defined(CONFIG_HSA_AMD_MODULE) bool (*kgd2kfd_init_p)(unsigned, const struct kfd2kgd_calls*, const struct kgd2kfd_calls**); @@ -113,6 +118,17 @@ bool radeon_kfd_init(void) } return true; +#elif defined(CONFIG_HSA_AMD) + if (!kgd2kfd_init(KFD_INTERFACE_VERSION, &kfd2kgd, &kgd2kfd)) { + kgd2kfd = NULL; + + return false; + } + + return true; +#else + return false; +#endif } void radeon_kfd_fini(void) @@ -374,6 +390,10 @@ static int kgd_set_pasid_vmid_mapping(struct kgd_dev *kgd, unsigned int pasid, cpu_relax(); write_register(kgd, ATC_VMID_PASID_MAPPING_UPDATE_STATUS, 1U << vmid); + /* Mapping vmid to pasid also for IH block */ + write_register(kgd, IH_VMID_0_LUT + vmid * sizeof(uint32_t), + pasid_mapping); + return 0; } @@ -513,7 +533,7 @@ static int kgd_hqd_load(struct kgd_dev *kgd, void *mqd, uint32_t pipe_id, return 0; } -static bool kgd_hqd_is_occupies(struct kgd_dev *kgd, uint64_t queue_address, +static bool kgd_hqd_is_occupied(struct kgd_dev *kgd, uint64_t queue_address, uint32_t pipe_id, uint32_t queue_id) { uint32_t act; @@ -552,6 +572,7 @@ static int kgd_hqd_destroy(struct kgd_dev *kgd, uint32_t reset_type, if (timeout == 0) { pr_err("kfd: cp queue preemption time out (%dms)\n", temp); + release_queue(kgd); return -ETIME; } msleep(20); @@ -561,3 +582,52 @@ static int kgd_hqd_destroy(struct kgd_dev *kgd, uint32_t reset_type, release_queue(kgd); return 0; } + +static uint16_t get_fw_version(struct kgd_dev *kgd, enum kgd_engine_type type) +{ + struct radeon_device *rdev = (struct radeon_device *) kgd; + const union radeon_firmware_header *hdr; + + BUG_ON(kgd == NULL || rdev->mec_fw == NULL); + + switch (type) { + case KGD_ENGINE_PFP: + hdr = (const union radeon_firmware_header *) rdev->pfp_fw->data; + break; + + case KGD_ENGINE_ME: + hdr = (const union radeon_firmware_header *) rdev->me_fw->data; + break; + + case KGD_ENGINE_CE: + hdr = (const union radeon_firmware_header *) rdev->ce_fw->data; + break; + + case KGD_ENGINE_MEC1: + hdr = (const union radeon_firmware_header *) rdev->mec_fw->data; + break; + + case KGD_ENGINE_MEC2: + hdr = (const union radeon_firmware_header *) + rdev->mec2_fw->data; + break; + + case KGD_ENGINE_RLC: + hdr = (const union radeon_firmware_header *) rdev->rlc_fw->data; + break; + + case KGD_ENGINE_SDMA: + hdr = (const union radeon_firmware_header *) + rdev->sdma_fw->data; + break; + + default: + return 0; + } + + if (hdr == NULL) + return 0; + + /* Only 12 bit in use*/ + return hdr->common.ucode_version; +} diff --git a/drivers/gpu/drm/radeon/radeon_kms.c b/drivers/gpu/drm/radeon/radeon_kms.c index f4dd26ae33e5..3cf9c1fa6475 100644 --- a/drivers/gpu/drm/radeon/radeon_kms.c +++ b/drivers/gpu/drm/radeon/radeon_kms.c @@ -800,6 +800,8 @@ int radeon_get_vblank_timestamp_kms(struct drm_device *dev, int crtc, /* Get associated drm_crtc: */ drmcrtc = &rdev->mode_info.crtcs[crtc]->base; + if (!drmcrtc) + return -EINVAL; /* Helper routine in DRM core does all the work: */ return drm_calc_vbltimestamp_from_scanoutpos(dev, crtc, max_error, diff --git a/drivers/gpu/drm/radeon/radeon_legacy_crtc.c b/drivers/gpu/drm/radeon/radeon_legacy_crtc.c index cafb1ccf2ec3..678b4386540d 100644 --- a/drivers/gpu/drm/radeon/radeon_legacy_crtc.c +++ b/drivers/gpu/drm/radeon/radeon_legacy_crtc.c @@ -1054,6 +1054,7 @@ static int radeon_crtc_mode_set(struct drm_crtc *crtc, DRM_ERROR("Mode need scaling but only first crtc can do that.\n"); } } + radeon_cursor_reset(crtc); return 0; } diff --git a/drivers/gpu/drm/radeon/radeon_mode.h b/drivers/gpu/drm/radeon/radeon_mode.h index f3d87cdd5c9d..390db897f322 100644 --- a/drivers/gpu/drm/radeon/radeon_mode.h +++ b/drivers/gpu/drm/radeon/radeon_mode.h @@ -818,6 +818,7 @@ extern int radeon_crtc_cursor_set2(struct drm_crtc *crtc, int32_t hot_y); extern int radeon_crtc_cursor_move(struct drm_crtc *crtc, int x, int y); +extern void radeon_cursor_reset(struct drm_crtc *crtc); extern int radeon_get_crtc_scanoutpos(struct drm_device *dev, int crtc, unsigned int flags, diff --git a/drivers/gpu/drm/radeon/radeon_object.c b/drivers/gpu/drm/radeon/radeon_object.c index 87b00d902bf7..86fc56434b28 100644 --- a/drivers/gpu/drm/radeon/radeon_object.c +++ b/drivers/gpu/drm/radeon/radeon_object.c @@ -233,6 +233,13 @@ int radeon_bo_create(struct radeon_device *rdev, if (!(rdev->flags & RADEON_IS_PCIE)) bo->flags &= ~(RADEON_GEM_GTT_WC | RADEON_GEM_GTT_UC); +#ifdef CONFIG_X86_32 + /* XXX: Write-combined CPU mappings of GTT seem broken on 32-bit + * See https://bugs.freedesktop.org/show_bug.cgi?id=84627 + */ + bo->flags &= ~RADEON_GEM_GTT_WC; +#endif + radeon_ttm_placement_from_domain(bo, domain); /* Kernel allocation are uninterruptible */ down_read(&rdev->pm.mclk_lock); @@ -502,28 +509,26 @@ int radeon_bo_list_validate(struct radeon_device *rdev, struct ww_acquire_ctx *ticket, struct list_head *head, int ring) { - struct radeon_cs_reloc *lobj; - struct radeon_bo *bo; + struct radeon_bo_list *lobj; + struct list_head duplicates; int r; u64 bytes_moved = 0, initial_bytes_moved; u64 bytes_moved_threshold = radeon_bo_get_threshold_for_moves(rdev); - r = ttm_eu_reserve_buffers(ticket, head, true); + INIT_LIST_HEAD(&duplicates); + r = ttm_eu_reserve_buffers(ticket, head, true, &duplicates); if (unlikely(r != 0)) { return r; } list_for_each_entry(lobj, head, tv.head) { - bo = lobj->robj; + struct radeon_bo *bo = lobj->robj; if (!bo->pin_count) { u32 domain = lobj->prefered_domains; u32 allowed = lobj->allowed_domains; u32 current_domain = radeon_mem_type_to_domain(bo->tbo.mem.mem_type); - WARN_ONCE(bo->gem_base.dumb, - "GPU use of dumb buffer is illegal.\n"); - /* Check if this buffer will be moved and don't move it * if we have moved too many buffers for this IB already. * @@ -562,6 +567,12 @@ int radeon_bo_list_validate(struct radeon_device *rdev, lobj->gpu_offset = radeon_bo_gpu_offset(bo); lobj->tiling_flags = bo->tiling_flags; } + + list_for_each_entry(lobj, &duplicates, tv.head) { + lobj->gpu_offset = radeon_bo_gpu_offset(lobj->robj); + lobj->tiling_flags = lobj->robj->tiling_flags; + } + return 0; } diff --git a/drivers/gpu/drm/radeon/radeon_state.c b/drivers/gpu/drm/radeon/radeon_state.c index 535403e0c8a2..15aee723db77 100644 --- a/drivers/gpu/drm/radeon/radeon_state.c +++ b/drivers/gpu/drm/radeon/radeon_state.c @@ -1703,7 +1703,7 @@ static int radeon_cp_dispatch_texture(struct drm_device * dev, u32 format; u32 *buffer; const u8 __user *data; - int size, dwords, tex_width, blit_width, spitch; + unsigned int size, dwords, tex_width, blit_width, spitch; u32 height; int i; u32 texpitch, microtile; diff --git a/drivers/gpu/drm/radeon/radeon_trace.h b/drivers/gpu/drm/radeon/radeon_trace.h index 9db74a96ef61..ce075cb08cb2 100644 --- a/drivers/gpu/drm/radeon/radeon_trace.h +++ b/drivers/gpu/drm/radeon/radeon_trace.h @@ -38,7 +38,7 @@ TRACE_EVENT(radeon_cs, TP_fast_assign( __entry->ring = p->ring; - __entry->dw = p->chunks[p->chunk_ib_idx].length_dw; + __entry->dw = p->chunk_ib->length_dw; __entry->fences = radeon_fence_count_emitted( p->rdev, p->ring); ), diff --git a/drivers/gpu/drm/radeon/radeon_ttm.c b/drivers/gpu/drm/radeon/radeon_ttm.c index cbe7b32d181c..d02aa1d0f588 100644 --- a/drivers/gpu/drm/radeon/radeon_ttm.c +++ b/drivers/gpu/drm/radeon/radeon_ttm.c @@ -196,7 +196,7 @@ static void radeon_evict_flags(struct ttm_buffer_object *bo, rbo = container_of(bo, struct radeon_bo, tbo); switch (bo->mem.mem_type) { case TTM_PL_VRAM: - if (rbo->rdev->ring[RADEON_RING_TYPE_GFX_INDEX].ready == false) + if (rbo->rdev->ring[radeon_copy_ring_index(rbo->rdev)].ready == false) radeon_ttm_placement_from_domain(rbo, RADEON_GEM_DOMAIN_CPU); else if (rbo->rdev->mc.visible_vram_size < rbo->rdev->mc.real_vram_size && bo->mem.start < (rbo->rdev->mc.visible_vram_size >> PAGE_SHIFT)) { diff --git a/drivers/gpu/drm/radeon/radeon_uvd.c b/drivers/gpu/drm/radeon/radeon_uvd.c index 11b662469253..c10b2aec6450 100644 --- a/drivers/gpu/drm/radeon/radeon_uvd.c +++ b/drivers/gpu/drm/radeon/radeon_uvd.c @@ -488,12 +488,12 @@ static int radeon_uvd_cs_reloc(struct radeon_cs_parser *p, unsigned buf_sizes[], bool *has_msg_cmd) { struct radeon_cs_chunk *relocs_chunk; - struct radeon_cs_reloc *reloc; + struct radeon_bo_list *reloc; unsigned idx, cmd, offset; uint64_t start, end; int r; - relocs_chunk = &p->chunks[p->chunk_relocs_idx]; + relocs_chunk = p->chunk_relocs; offset = radeon_get_ib_value(p, data0); idx = radeon_get_ib_value(p, data1); if (idx >= relocs_chunk->length_dw) { @@ -502,7 +502,7 @@ static int radeon_uvd_cs_reloc(struct radeon_cs_parser *p, return -EINVAL; } - reloc = p->relocs_ptr[(idx / 4)]; + reloc = &p->relocs[(idx / 4)]; start = reloc->gpu_offset; end = start + radeon_bo_size(reloc->robj); start += offset; @@ -610,13 +610,13 @@ int radeon_uvd_cs_parse(struct radeon_cs_parser *p) [0x00000003] = 2048, }; - if (p->chunks[p->chunk_ib_idx].length_dw % 16) { + if (p->chunk_ib->length_dw % 16) { DRM_ERROR("UVD IB length (%d) not 16 dwords aligned!\n", - p->chunks[p->chunk_ib_idx].length_dw); + p->chunk_ib->length_dw); return -EINVAL; } - if (p->chunk_relocs_idx == -1) { + if (p->chunk_relocs == NULL) { DRM_ERROR("No relocation chunk !\n"); return -EINVAL; } @@ -640,7 +640,7 @@ int radeon_uvd_cs_parse(struct radeon_cs_parser *p) DRM_ERROR("Unknown packet type %d !\n", pkt.type); return -EINVAL; } - } while (p->idx < p->chunks[p->chunk_ib_idx].length_dw); + } while (p->idx < p->chunk_ib->length_dw); if (!has_msg_cmd) { DRM_ERROR("UVD-IBs need a msg command!\n"); diff --git a/drivers/gpu/drm/radeon/radeon_vce.c b/drivers/gpu/drm/radeon/radeon_vce.c index 9e85757d5599..976fe432f4e2 100644 --- a/drivers/gpu/drm/radeon/radeon_vce.c +++ b/drivers/gpu/drm/radeon/radeon_vce.c @@ -453,11 +453,11 @@ int radeon_vce_cs_reloc(struct radeon_cs_parser *p, int lo, int hi, unsigned size) { struct radeon_cs_chunk *relocs_chunk; - struct radeon_cs_reloc *reloc; + struct radeon_bo_list *reloc; uint64_t start, end, offset; unsigned idx; - relocs_chunk = &p->chunks[p->chunk_relocs_idx]; + relocs_chunk = p->chunk_relocs; offset = radeon_get_ib_value(p, lo); idx = radeon_get_ib_value(p, hi); @@ -467,7 +467,7 @@ int radeon_vce_cs_reloc(struct radeon_cs_parser *p, int lo, int hi, return -EINVAL; } - reloc = p->relocs_ptr[(idx / 4)]; + reloc = &p->relocs[(idx / 4)]; start = reloc->gpu_offset; end = start + radeon_bo_size(reloc->robj); start += offset; @@ -534,7 +534,7 @@ int radeon_vce_cs_parse(struct radeon_cs_parser *p) uint32_t *size = &tmp; int i, r; - while (p->idx < p->chunks[p->chunk_ib_idx].length_dw) { + while (p->idx < p->chunk_ib->length_dw) { uint32_t len = radeon_get_ib_value(p, p->idx); uint32_t cmd = radeon_get_ib_value(p, p->idx + 1); diff --git a/drivers/gpu/drm/radeon/radeon_vm.c b/drivers/gpu/drm/radeon/radeon_vm.c index 0b10f3a03ce2..cde48c42b30a 100644 --- a/drivers/gpu/drm/radeon/radeon_vm.c +++ b/drivers/gpu/drm/radeon/radeon_vm.c @@ -125,41 +125,37 @@ void radeon_vm_manager_fini(struct radeon_device *rdev) * Add the page directory to the list of BOs to * validate for command submission (cayman+). */ -struct radeon_cs_reloc *radeon_vm_get_bos(struct radeon_device *rdev, +struct radeon_bo_list *radeon_vm_get_bos(struct radeon_device *rdev, struct radeon_vm *vm, struct list_head *head) { - struct radeon_cs_reloc *list; + struct radeon_bo_list *list; unsigned i, idx; list = drm_malloc_ab(vm->max_pde_used + 2, - sizeof(struct radeon_cs_reloc)); + sizeof(struct radeon_bo_list)); if (!list) return NULL; /* add the vm page table to the list */ - list[0].gobj = NULL; list[0].robj = vm->page_directory; list[0].prefered_domains = RADEON_GEM_DOMAIN_VRAM; list[0].allowed_domains = RADEON_GEM_DOMAIN_VRAM; list[0].tv.bo = &vm->page_directory->tbo; list[0].tv.shared = true; list[0].tiling_flags = 0; - list[0].handle = 0; list_add(&list[0].tv.head, head); for (i = 0, idx = 1; i <= vm->max_pde_used; i++) { if (!vm->page_tables[i].bo) continue; - list[idx].gobj = NULL; list[idx].robj = vm->page_tables[i].bo; list[idx].prefered_domains = RADEON_GEM_DOMAIN_VRAM; list[idx].allowed_domains = RADEON_GEM_DOMAIN_VRAM; list[idx].tv.bo = &list[idx].robj->tbo; list[idx].tv.shared = true; list[idx].tiling_flags = 0; - list[idx].handle = 0; list_add(&list[idx++].tv.head, head); } @@ -491,7 +487,9 @@ int radeon_vm_bo_set_addr(struct radeon_device *rdev, tmp->vm = vm; tmp->addr = bo_va->addr; tmp->bo = radeon_bo_ref(bo_va->bo); + spin_lock(&vm->status_lock); list_add(&tmp->vm_status, &vm->freed); + spin_unlock(&vm->status_lock); } interval_tree_remove(&bo_va->it, &vm->va); @@ -802,11 +800,11 @@ static void radeon_vm_frag_ptes(struct radeon_device *rdev, * * Global and local mutex must be locked! */ -static void radeon_vm_update_ptes(struct radeon_device *rdev, - struct radeon_vm *vm, - struct radeon_ib *ib, - uint64_t start, uint64_t end, - uint64_t dst, uint32_t flags) +static int radeon_vm_update_ptes(struct radeon_device *rdev, + struct radeon_vm *vm, + struct radeon_ib *ib, + uint64_t start, uint64_t end, + uint64_t dst, uint32_t flags) { uint64_t mask = RADEON_VM_PTE_COUNT - 1; uint64_t last_pte = ~0, last_dst = ~0; @@ -819,8 +817,12 @@ static void radeon_vm_update_ptes(struct radeon_device *rdev, struct radeon_bo *pt = vm->page_tables[pt_idx].bo; unsigned nptes; uint64_t pte; + int r; radeon_sync_resv(rdev, &ib->sync, pt->tbo.resv, true); + r = reservation_object_reserve_shared(pt->tbo.resv); + if (r) + return r; if ((addr & ~mask) == (end & ~mask)) nptes = end - addr; @@ -854,6 +856,8 @@ static void radeon_vm_update_ptes(struct radeon_device *rdev, last_pte + 8 * count, last_dst, flags); } + + return 0; } /** @@ -878,7 +882,7 @@ static void radeon_vm_fence_pts(struct radeon_vm *vm, end >>= radeon_vm_block_size; for (i = start; i <= end; ++i) - radeon_bo_fence(vm->page_tables[i].bo, fence, false); + radeon_bo_fence(vm->page_tables[i].bo, fence, true); } /** @@ -911,7 +915,9 @@ int radeon_vm_bo_update(struct radeon_device *rdev, return -EINVAL; } + spin_lock(&vm->status_lock); list_del_init(&bo_va->vm_status); + spin_unlock(&vm->status_lock); bo_va->flags &= ~RADEON_VM_PAGE_VALID; bo_va->flags &= ~RADEON_VM_PAGE_SYSTEM; @@ -987,9 +993,13 @@ int radeon_vm_bo_update(struct radeon_device *rdev, radeon_sync_fence(&ib.sync, vm->ids[i].last_id_use); } - radeon_vm_update_ptes(rdev, vm, &ib, bo_va->it.start, - bo_va->it.last + 1, addr, - radeon_vm_page_flags(bo_va->flags)); + r = radeon_vm_update_ptes(rdev, vm, &ib, bo_va->it.start, + bo_va->it.last + 1, addr, + radeon_vm_page_flags(bo_va->flags)); + if (r) { + radeon_ib_free(rdev, &ib); + return r; + } radeon_asic_vm_pad_ib(rdev, &ib); WARN_ON(ib.length_dw > ndw); @@ -1022,17 +1032,25 @@ int radeon_vm_bo_update(struct radeon_device *rdev, int radeon_vm_clear_freed(struct radeon_device *rdev, struct radeon_vm *vm) { - struct radeon_bo_va *bo_va, *tmp; + struct radeon_bo_va *bo_va; int r; - list_for_each_entry_safe(bo_va, tmp, &vm->freed, vm_status) { + spin_lock(&vm->status_lock); + while (!list_empty(&vm->freed)) { + bo_va = list_first_entry(&vm->freed, + struct radeon_bo_va, vm_status); + spin_unlock(&vm->status_lock); + r = radeon_vm_bo_update(rdev, bo_va, NULL); radeon_bo_unref(&bo_va->bo); radeon_fence_unref(&bo_va->last_pt_update); kfree(bo_va); if (r) return r; + + spin_lock(&vm->status_lock); } + spin_unlock(&vm->status_lock); return 0; } @@ -1051,14 +1069,23 @@ int radeon_vm_clear_freed(struct radeon_device *rdev, int radeon_vm_clear_invalids(struct radeon_device *rdev, struct radeon_vm *vm) { - struct radeon_bo_va *bo_va, *tmp; + struct radeon_bo_va *bo_va; int r; - list_for_each_entry_safe(bo_va, tmp, &vm->invalidated, vm_status) { + spin_lock(&vm->status_lock); + while (!list_empty(&vm->invalidated)) { + bo_va = list_first_entry(&vm->invalidated, + struct radeon_bo_va, vm_status); + spin_unlock(&vm->status_lock); + r = radeon_vm_bo_update(rdev, bo_va, NULL); if (r) return r; + + spin_lock(&vm->status_lock); } + spin_unlock(&vm->status_lock); + return 0; } @@ -1081,6 +1108,7 @@ void radeon_vm_bo_rmv(struct radeon_device *rdev, mutex_lock(&vm->mutex); interval_tree_remove(&bo_va->it, &vm->va); + spin_lock(&vm->status_lock); list_del(&bo_va->vm_status); if (bo_va->addr) { @@ -1090,6 +1118,7 @@ void radeon_vm_bo_rmv(struct radeon_device *rdev, radeon_fence_unref(&bo_va->last_pt_update); kfree(bo_va); } + spin_unlock(&vm->status_lock); mutex_unlock(&vm->mutex); } @@ -1110,10 +1139,10 @@ void radeon_vm_bo_invalidate(struct radeon_device *rdev, list_for_each_entry(bo_va, &bo->va, bo_list) { if (bo_va->addr) { - mutex_lock(&bo_va->vm->mutex); + spin_lock(&bo_va->vm->status_lock); list_del(&bo_va->vm_status); list_add(&bo_va->vm_status, &bo_va->vm->invalidated); - mutex_unlock(&bo_va->vm->mutex); + spin_unlock(&bo_va->vm->status_lock); } } } @@ -1141,6 +1170,7 @@ int radeon_vm_init(struct radeon_device *rdev, struct radeon_vm *vm) } mutex_init(&vm->mutex); vm->va = RB_ROOT; + spin_lock_init(&vm->status_lock); INIT_LIST_HEAD(&vm->invalidated); INIT_LIST_HEAD(&vm->freed); diff --git a/drivers/gpu/drm/radeon/si_dpm.c b/drivers/gpu/drm/radeon/si_dpm.c index cf4c420b5572..32e354b8b0ab 100644 --- a/drivers/gpu/drm/radeon/si_dpm.c +++ b/drivers/gpu/drm/radeon/si_dpm.c @@ -5893,7 +5893,7 @@ static void si_fan_ctrl_set_static_mode(struct radeon_device *rdev, u32 mode) tmp |= TMIN(0); WREG32(CG_FDO_CTRL2, tmp); - tmp = RREG32(CG_FDO_CTRL2) & FDO_PWM_MODE_MASK; + tmp = RREG32(CG_FDO_CTRL2) & ~FDO_PWM_MODE_MASK; tmp |= FDO_PWM_MODE(mode); WREG32(CG_FDO_CTRL2, tmp); } @@ -6098,7 +6098,7 @@ static int si_fan_ctrl_set_fan_speed_rpm(struct radeon_device *rdev, tmp |= TARGET_PERIOD(tach_period); WREG32(CG_TACH_CTRL, tmp); - si_fan_ctrl_set_static_mode(rdev, FDO_PWM_MODE_STATIC); + si_fan_ctrl_set_static_mode(rdev, FDO_PWM_MODE_STATIC_RPM); return 0; } @@ -6114,7 +6114,7 @@ static void si_fan_ctrl_set_default_mode(struct radeon_device *rdev) tmp |= FDO_PWM_MODE(si_pi->fan_ctrl_default_mode); WREG32(CG_FDO_CTRL2, tmp); - tmp = RREG32(CG_FDO_CTRL2) & TMIN_MASK; + tmp = RREG32(CG_FDO_CTRL2) & ~TMIN_MASK; tmp |= TMIN(si_pi->t_min); WREG32(CG_FDO_CTRL2, tmp); si_pi->fan_ctrl_is_in_default_mode = true; diff --git a/drivers/gpu/drm/radeon/sid.h b/drivers/gpu/drm/radeon/sid.h index c549c16a4fe4..4069be89e585 100644 --- a/drivers/gpu/drm/radeon/sid.h +++ b/drivers/gpu/drm/radeon/sid.h @@ -208,18 +208,18 @@ #define CG_FDO_CTRL0 0x754 #define FDO_STATIC_DUTY(x) ((x) << 0) -#define FDO_STATIC_DUTY_MASK 0x0000000F +#define FDO_STATIC_DUTY_MASK 0x000000FF #define FDO_STATIC_DUTY_SHIFT 0 #define CG_FDO_CTRL1 0x758 #define FMAX_DUTY100(x) ((x) << 0) -#define FMAX_DUTY100_MASK 0x0000000F +#define FMAX_DUTY100_MASK 0x000000FF #define FMAX_DUTY100_SHIFT 0 #define CG_FDO_CTRL2 0x75C #define TMIN(x) ((x) << 0) -#define TMIN_MASK 0x0000000F +#define TMIN_MASK 0x000000FF #define TMIN_SHIFT 0 #define FDO_PWM_MODE(x) ((x) << 11) -#define FDO_PWM_MODE_MASK (3 << 11) +#define FDO_PWM_MODE_MASK (7 << 11) #define FDO_PWM_MODE_SHIFT 11 #define TACH_PWM_RESP_RATE(x) ((x) << 25) #define TACH_PWM_RESP_RATE_MASK (0x7f << 25) |