This series adds support for the gpu found in the Snapdragon 7c Gen 3 compute platform. This gpu is similar to the exisiting a660 gpu with minor delta in the programing sequence. As the Adreno GPUs are moving away from a numeric chipid based naming scheme to a string, it was decided to use 0x06030500 as the chip id of this gpu to communicate to the userspace driver.
Changes in v4: - Move adreno_cmp_rev() to patch-2/3 to fix compilation - Minor updates to commit msg (Rob)
Changes in v3: - Add a cover letter.
Changes in v2: - Use rev to identify SKU - Introduce adreno_is_a660_family() (Rob) - Remove revn for 7c3 (Rob) - Remove CPR register programing since they are not required for 7c3
Akhil P Oommen (3): drm/msm/a6xx: Fix llcc configuration for a660 gpu drm/msm/a6xx: Use rev to identify SKU drm/msm/a6xx: Add support for Adreno 7c Gen 3 gpu
drivers/gpu/drm/msm/adreno/a6xx_gmu.c | 8 ++- drivers/gpu/drm/msm/adreno/a6xx_gmu.h | 1 + drivers/gpu/drm/msm/adreno/a6xx_gpu.c | 87 ++++++++++++++++++------------ drivers/gpu/drm/msm/adreno/a6xx_hfi.c | 32 +++++++++++ drivers/gpu/drm/msm/adreno/adreno_device.c | 27 +++++++--- drivers/gpu/drm/msm/adreno/adreno_gpu.h | 18 ++++++- 6 files changed, 129 insertions(+), 44 deletions(-)
Add the missing scache_cntl0 register programing which is required for a660 gpu.
Signed-off-by: Akhil P Oommen akhilpo@codeaurora.org ---
(no changes since v1)
drivers/gpu/drm/msm/adreno/a6xx_gpu.c | 46 ++++++++++++++++++++--------------- 1 file changed, 27 insertions(+), 19 deletions(-)
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c index 9c5e461..183b9f9 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c @@ -1383,13 +1383,13 @@ static void a6xx_llc_activate(struct a6xx_gpu *a6xx_gpu) { struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; struct msm_gpu *gpu = &adreno_gpu->base; - u32 cntl1_regval = 0; + u32 gpu_scid, cntl1_regval = 0;
if (IS_ERR(a6xx_gpu->llc_mmio)) return;
if (!llcc_slice_activate(a6xx_gpu->llc_slice)) { - u32 gpu_scid = llcc_get_slice_id(a6xx_gpu->llc_slice); + gpu_scid = llcc_get_slice_id(a6xx_gpu->llc_slice);
gpu_scid &= 0x1f; cntl1_regval = (gpu_scid << 0) | (gpu_scid << 5) | (gpu_scid << 10) | @@ -1409,26 +1409,34 @@ static void a6xx_llc_activate(struct a6xx_gpu *a6xx_gpu) } }
- if (cntl1_regval) { + if (!cntl1_regval) + return; + + /* + * Program the slice IDs for the various GPU blocks and GPU MMU + * pagetables + */ + if (!a6xx_gpu->have_mmu500) { + a6xx_llc_write(a6xx_gpu, + REG_A6XX_CX_MISC_SYSTEM_CACHE_CNTL_1, cntl1_regval); + /* - * Program the slice IDs for the various GPU blocks and GPU MMU - * pagetables + * Program cacheability overrides to not allocate cache + * lines on a write miss */ - if (a6xx_gpu->have_mmu500) - gpu_rmw(gpu, REG_A6XX_GBIF_SCACHE_CNTL1, GENMASK(24, 0), - cntl1_regval); - else { - a6xx_llc_write(a6xx_gpu, - REG_A6XX_CX_MISC_SYSTEM_CACHE_CNTL_1, cntl1_regval); - - /* - * Program cacheability overrides to not allocate cache - * lines on a write miss - */ - a6xx_llc_rmw(a6xx_gpu, - REG_A6XX_CX_MISC_SYSTEM_CACHE_CNTL_0, 0xF, 0x03); - } + a6xx_llc_rmw(a6xx_gpu, + REG_A6XX_CX_MISC_SYSTEM_CACHE_CNTL_0, 0xF, 0x03); + return; } + + gpu_rmw(gpu, REG_A6XX_GBIF_SCACHE_CNTL1, GENMASK(24, 0), cntl1_regval); + + /* On A660, the SCID programming for UCHE traffic is done in + * A6XX_GBIF_SCACHE_CNTL0[14:10] + */ + if (adreno_is_a660(adreno_gpu)) + gpu_rmw(gpu, REG_A6XX_GBIF_SCACHE_CNTL0, (0x1f << 10) | + (1 << 8), (gpu_scid << 10) | (1 << 8)); }
static void a6xx_llc_slices_destroy(struct a6xx_gpu *a6xx_gpu)
Use rev instead of revn to identify the SKU. This is in preparation to the introduction of 7c3 gpu which won't have a revn.
Signed-off-by: Akhil P Oommen akhilpo@codeaurora.org ---
Changes in v4: - Move adreno_cmp_rev() here to fix compilation
drivers/gpu/drm/msm/adreno/a6xx_gpu.c | 11 +++++------ drivers/gpu/drm/msm/adreno/adreno_device.c | 16 ++++++++++------ drivers/gpu/drm/msm/adreno/adreno_gpu.h | 4 ++++ 3 files changed, 19 insertions(+), 12 deletions(-)
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c index 183b9f9..0da1a66 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c @@ -1675,11 +1675,11 @@ static u32 a618_get_speed_bin(u32 fuse) return UINT_MAX; }
-static u32 fuse_to_supp_hw(struct device *dev, u32 revn, u32 fuse) +static u32 fuse_to_supp_hw(struct device *dev, struct adreno_rev rev, u32 fuse) { u32 val = UINT_MAX;
- if (revn == 618) + if (adreno_cmp_rev(ADRENO_REV(6, 1, 8, ANY_ID), rev)) val = a618_get_speed_bin(fuse);
if (val == UINT_MAX) { @@ -1692,8 +1692,7 @@ static u32 fuse_to_supp_hw(struct device *dev, u32 revn, u32 fuse) return (1 << val); }
-static int a6xx_set_supported_hw(struct device *dev, struct a6xx_gpu *a6xx_gpu, - u32 revn) +static int a6xx_set_supported_hw(struct device *dev, struct adreno_rev rev) { u32 supp_hw = UINT_MAX; u16 speedbin; @@ -1714,7 +1713,7 @@ static int a6xx_set_supported_hw(struct device *dev, struct a6xx_gpu *a6xx_gpu, } speedbin = le16_to_cpu(speedbin);
- supp_hw = fuse_to_supp_hw(dev, revn, speedbin); + supp_hw = fuse_to_supp_hw(dev, rev, speedbin);
done: ret = devm_pm_opp_set_supported_hw(dev, &supp_hw, 1); @@ -1785,7 +1784,7 @@ struct msm_gpu *a6xx_gpu_init(struct drm_device *dev)
a6xx_llc_slices_init(pdev, a6xx_gpu);
- ret = a6xx_set_supported_hw(&pdev->dev, a6xx_gpu, info->revn); + ret = a6xx_set_supported_hw(&pdev->dev, config->rev); if (ret) { a6xx_destroy(&(a6xx_gpu->base.base)); return ERR_PTR(ret); diff --git a/drivers/gpu/drm/msm/adreno/adreno_device.c b/drivers/gpu/drm/msm/adreno/adreno_device.c index 6dad801..7e6fafe 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_device.c +++ b/drivers/gpu/drm/msm/adreno/adreno_device.c @@ -8,8 +8,6 @@
#include "adreno_gpu.h"
-#define ANY_ID 0xff - bool hang_debug = false; MODULE_PARM_DESC(hang_debug, "Dump registers when hang is detected (can be slow!)"); module_param_named(hang_debug, hang_debug, bool, 0600); @@ -325,6 +323,15 @@ static inline bool _rev_match(uint8_t entry, uint8_t id) return (entry == ANY_ID) || (entry == id); }
+bool adreno_cmp_rev(struct adreno_rev rev1, struct adreno_rev rev2) +{ + + return _rev_match(rev1.core, rev2.core) && + _rev_match(rev1.major, rev2.major) && + _rev_match(rev1.minor, rev2.minor) && + _rev_match(rev1.patchid, rev2.patchid); +} + const struct adreno_info *adreno_info(struct adreno_rev rev) { int i; @@ -332,10 +339,7 @@ const struct adreno_info *adreno_info(struct adreno_rev rev) /* identify gpu: */ for (i = 0; i < ARRAY_SIZE(gpulist); i++) { const struct adreno_info *info = &gpulist[i]; - if (_rev_match(info->rev.core, rev.core) && - _rev_match(info->rev.major, rev.major) && - _rev_match(info->rev.minor, rev.minor) && - _rev_match(info->rev.patchid, rev.patchid)) + if (adreno_cmp_rev(info->rev, rev)) return info; }
diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.h b/drivers/gpu/drm/msm/adreno/adreno_gpu.h index 8dbe0d1..a368a16 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_gpu.h +++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.h @@ -42,6 +42,8 @@ struct adreno_rev { uint8_t patchid; };
+#define ANY_ID 0xff + #define ADRENO_REV(core, major, minor, patchid) \ ((struct adreno_rev){ core, major, minor, patchid })
@@ -141,6 +143,8 @@ struct adreno_platform_config { __ret; \ })
+bool adreno_cmp_rev(struct adreno_rev rev1, struct adreno_rev rev2); + static inline bool adreno_is_a2xx(struct adreno_gpu *gpu) { return (gpu->revn < 300);
This patch adds support for the gpu found in the Snapdragon 7c Gen 3 compute platform. This gpu is similar to the exisiting a660 gpu with minor delta in the programing sequence. As the Adreno GPUs are moving away from a numeric chipid based naming scheme to a string, it was decided to use 0x06030500 as the chip id of this gpu to communicate to the userspace driver.
Signed-off-by: Akhil P Oommen akhilpo@codeaurora.org ---
Changes in v4: - Move out adreno_cmp_rev() to patch-2/3 to fix compilation
Changes in v2: - Introduce adreno_is_a660_family() (Rob) - Remove revn for 7c3 (Rob) - Remove CPR register programing since they are not required for 7c3
drivers/gpu/drm/msm/adreno/a6xx_gmu.c | 8 ++++++-- drivers/gpu/drm/msm/adreno/a6xx_gmu.h | 1 + drivers/gpu/drm/msm/adreno/a6xx_gpu.c | 32 ++++++++++++++++++++---------- drivers/gpu/drm/msm/adreno/a6xx_hfi.c | 32 ++++++++++++++++++++++++++++++ drivers/gpu/drm/msm/adreno/adreno_device.c | 11 ++++++++++ drivers/gpu/drm/msm/adreno/adreno_gpu.h | 14 ++++++++++++- 6 files changed, 84 insertions(+), 14 deletions(-)
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c index b349692..70ba3bf 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c +++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c @@ -519,7 +519,7 @@ static void a6xx_gmu_rpmh_init(struct a6xx_gmu *gmu) if (!pdcptr) goto err;
- if (adreno_is_a650(adreno_gpu) || adreno_is_a660(adreno_gpu)) + if (adreno_is_a650(adreno_gpu) || adreno_is_a660_family(adreno_gpu)) pdc_in_aop = true; else if (adreno_is_a618(adreno_gpu) || adreno_is_a640(adreno_gpu)) pdc_address_offset = 0x30090; @@ -933,6 +933,7 @@ int a6xx_gmu_resume(struct a6xx_gpu *a6xx_gpu)
/* Use a known rate to bring up the GMU */ clk_set_rate(gmu->core_clk, 200000000); + clk_set_rate(gmu->hub_clk, 150000000); ret = clk_bulk_prepare_enable(gmu->nr_clocks, gmu->clocks); if (ret) { pm_runtime_put(gmu->gxpd); @@ -1393,6 +1394,9 @@ static int a6xx_gmu_clocks_probe(struct a6xx_gmu *gmu) gmu->core_clk = msm_clk_bulk_get_clock(gmu->clocks, gmu->nr_clocks, "gmu");
+ gmu->hub_clk = msm_clk_bulk_get_clock(gmu->clocks, + gmu->nr_clocks, "hub"); + return 0; }
@@ -1504,7 +1508,7 @@ int a6xx_gmu_init(struct a6xx_gpu *a6xx_gpu, struct device_node *node) * are otherwise unused by a660. */ gmu->dummy.size = SZ_4K; - if (adreno_is_a660(adreno_gpu)) { + if (adreno_is_a660_family(adreno_gpu)) { ret = a6xx_gmu_memory_alloc(gmu, &gmu->debug, SZ_4K * 7, 0x60400000); if (ret) goto err_memory; diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.h b/drivers/gpu/drm/msm/adreno/a6xx_gmu.h index 71dfa600..3c74f64 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.h +++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.h @@ -66,6 +66,7 @@ struct a6xx_gmu { int nr_clocks; struct clk_bulk_data *clocks; struct clk *core_clk; + struct clk *hub_clk;
/* current performance index set externally */ int current_perf_index; diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c index 0da1a66..1881e09 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c @@ -652,7 +652,7 @@ static void a6xx_set_cp_protect(struct msm_gpu *gpu) regs = a650_protect; count = ARRAY_SIZE(a650_protect); count_max = 48; - } else if (adreno_is_a660(adreno_gpu)) { + } else if (adreno_is_a660_family(adreno_gpu)) { regs = a660_protect; count = ARRAY_SIZE(a660_protect); count_max = 48; @@ -694,6 +694,13 @@ static void a6xx_set_ubwc_config(struct msm_gpu *gpu) uavflagprd_inv = 2; }
+ if (adreno_is_7c3(adreno_gpu)) { + lower_bit = 1; + amsbc = 1; + rgb565_predicator = 1; + uavflagprd_inv = 2; + } + gpu_write(gpu, REG_A6XX_RB_NC_MODE_CNTL, rgb565_predicator << 11 | amsbc << 4 | lower_bit << 1); gpu_write(gpu, REG_A6XX_TPL1_NC_MODE_CNTL, lower_bit << 1); @@ -787,7 +794,7 @@ static bool a6xx_ucode_check_version(struct a6xx_gpu *a6xx_gpu, DRM_DEV_ERROR(&gpu->pdev->dev, "a650 SQE ucode is too old. Have version %x need at least %x\n", buf[0] & 0xfff, 0x095); - } else if (adreno_is_a660(adreno_gpu)) { + } else if (adreno_is_a660_family(adreno_gpu)) { ret = true; } else { DRM_DEV_ERROR(&gpu->pdev->dev, @@ -941,7 +948,7 @@ static int a6xx_hw_init(struct msm_gpu *gpu) gpu_write(gpu, REG_A6XX_CP_ROQ_THRESHOLDS_2, 0x010000c0); gpu_write(gpu, REG_A6XX_CP_ROQ_THRESHOLDS_1, 0x8040362c);
- if (adreno_is_a660(adreno_gpu)) + if (adreno_is_a660_family(adreno_gpu)) gpu_write(gpu, REG_A6XX_CP_LPAC_PROG_FIFO_SIZE, 0x00000020);
/* Setting the mem pool size */ @@ -950,10 +957,10 @@ static int a6xx_hw_init(struct msm_gpu *gpu) /* Setting the primFifo thresholds default values, * and vccCacheSkipDis=1 bit (0x200) for A640 and newer */ - if (adreno_is_a650(adreno_gpu) || adreno_is_a660(adreno_gpu)) - gpu_write(gpu, REG_A6XX_PC_DBG_ECO_CNTL, 0x00300200); - else if (adreno_is_a640(adreno_gpu)) + if (adreno_is_a640(adreno_gpu) || adreno_is_7c3(adreno_gpu)) gpu_write(gpu, REG_A6XX_PC_DBG_ECO_CNTL, 0x00200200); + else if (adreno_is_a650(adreno_gpu) || adreno_is_a660(adreno_gpu)) + gpu_write(gpu, REG_A6XX_PC_DBG_ECO_CNTL, 0x00300200); else gpu_write(gpu, REG_A6XX_PC_DBG_ECO_CNTL, 0x00180000);
@@ -990,13 +997,15 @@ static int a6xx_hw_init(struct msm_gpu *gpu) /* Protect registers from the CP */ a6xx_set_cp_protect(gpu);
- if (adreno_is_a660(adreno_gpu)) { + if (adreno_is_a660_family(adreno_gpu)) { gpu_write(gpu, REG_A6XX_CP_CHICKEN_DBG, 0x1); gpu_write(gpu, REG_A6XX_RBBM_GBIF_CLIENT_QOS_CNTL, 0x0); - /* Set dualQ + disable afull for A660 GPU but not for A635 */ - gpu_write(gpu, REG_A6XX_UCHE_CMDQ_CONFIG, 0x66906); }
+ /* Set dualQ + disable afull for A660 GPU */ + if (adreno_is_a660(adreno_gpu)) + gpu_write(gpu, REG_A6XX_UCHE_CMDQ_CONFIG, 0x66906); + /* Enable expanded apriv for targets that support it */ if (gpu->hw_apriv) { gpu_write(gpu, REG_A6XX_CP_APRIV_CNTL, @@ -1434,7 +1443,7 @@ static void a6xx_llc_activate(struct a6xx_gpu *a6xx_gpu) /* On A660, the SCID programming for UCHE traffic is done in * A6XX_GBIF_SCACHE_CNTL0[14:10] */ - if (adreno_is_a660(adreno_gpu)) + if (adreno_is_a660_family(adreno_gpu)) gpu_rmw(gpu, REG_A6XX_GBIF_SCACHE_CNTL0, (0x1f << 10) | (1 << 8), (gpu_scid << 10) | (1 << 8)); } @@ -1779,7 +1788,8 @@ struct msm_gpu *a6xx_gpu_init(struct drm_device *dev) */ info = adreno_info(config->rev);
- if (info && (info->revn == 650 || info->revn == 660)) + if (info && (info->revn == 650 || info->revn == 660 || + adreno_cmp_rev(ADRENO_REV(6, 3, 5, ANY_ID), info->rev))) adreno_gpu->base.hw_apriv = true;
a6xx_llc_slices_init(pdev, a6xx_gpu); diff --git a/drivers/gpu/drm/msm/adreno/a6xx_hfi.c b/drivers/gpu/drm/msm/adreno/a6xx_hfi.c index 9194337..1451c2b 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_hfi.c +++ b/drivers/gpu/drm/msm/adreno/a6xx_hfi.c @@ -382,6 +382,36 @@ static void a660_build_bw_table(struct a6xx_hfi_msg_bw_table *msg) msg->cnoc_cmds_data[1][0] = 0x60000001; }
+static void adreno_7c3_build_bw_table(struct a6xx_hfi_msg_bw_table *msg) +{ + /* + * Send a single "off" entry just to get things running + * TODO: bus scaling + */ + msg->bw_level_num = 1; + + msg->ddr_cmds_num = 3; + msg->ddr_wait_bitmask = 0x07; + + msg->ddr_cmds_addrs[0] = 0x50004; + msg->ddr_cmds_addrs[1] = 0x50000; + msg->ddr_cmds_addrs[2] = 0x50088; + + msg->ddr_cmds_data[0][0] = 0x40000000; + msg->ddr_cmds_data[0][1] = 0x40000000; + msg->ddr_cmds_data[0][2] = 0x40000000; + + /* + * These are the CX (CNOC) votes - these are used by the GMU but the + * votes are known and fixed for the target + */ + msg->cnoc_cmds_num = 1; + msg->cnoc_wait_bitmask = 0x01; + + msg->cnoc_cmds_addrs[0] = 0x5006c; + msg->cnoc_cmds_data[0][0] = 0x40000000; + msg->cnoc_cmds_data[1][0] = 0x60000001; +} static void a6xx_build_bw_table(struct a6xx_hfi_msg_bw_table *msg) { /* Send a single "off" entry since the 630 GMU doesn't do bus scaling */ @@ -432,6 +462,8 @@ static int a6xx_hfi_send_bw_table(struct a6xx_gmu *gmu) a640_build_bw_table(&msg); else if (adreno_is_a650(adreno_gpu)) a650_build_bw_table(&msg); + else if (adreno_is_7c3(adreno_gpu)) + adreno_7c3_build_bw_table(&msg); else if (adreno_is_a660(adreno_gpu)) a660_build_bw_table(&msg); else diff --git a/drivers/gpu/drm/msm/adreno/adreno_device.c b/drivers/gpu/drm/msm/adreno/adreno_device.c index 7e6fafe..7b9d605 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_device.c +++ b/drivers/gpu/drm/msm/adreno/adreno_device.c @@ -298,6 +298,17 @@ static const struct adreno_info gpulist[] = { .init = a6xx_gpu_init, .zapfw = "a660_zap.mdt", .hwcg = a660_hwcg, + }, { + .rev = ADRENO_REV(6, 3, 5, ANY_ID), + .name = "Adreno 7c Gen 3", + .fw = { + [ADRENO_FW_SQE] = "a660_sqe.fw", + [ADRENO_FW_GMU] = "a660_gmu.bin", + }, + .gmem = SZ_512K, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .init = a6xx_gpu_init, + .hwcg = a660_hwcg, }, };
diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.h b/drivers/gpu/drm/msm/adreno/adreno_gpu.h index a368a16..1393f21 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_gpu.h +++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.h @@ -251,15 +251,27 @@ static inline int adreno_is_a650(struct adreno_gpu *gpu) return gpu->revn == 650; }
+static inline int adreno_is_7c3(struct adreno_gpu *gpu) +{ + /* The order of args is important here to handle ANY_ID correctly */ + return adreno_cmp_rev(ADRENO_REV(6, 3, 5, ANY_ID), gpu->rev); +} + static inline int adreno_is_a660(struct adreno_gpu *gpu) { return gpu->revn == 660; }
+static inline int adreno_is_a660_family(struct adreno_gpu *gpu) +{ + return adreno_is_a660(gpu) || adreno_is_7c3(gpu); +} + /* check for a650, a660, or any derivatives */ static inline int adreno_is_a650_family(struct adreno_gpu *gpu) { - return gpu->revn == 650 || gpu->revn == 620 || gpu->revn == 660; + return gpu->revn == 650 || gpu->revn == 620 || + adreno_is_a660_family(gpu); }
int adreno_get_param(struct msm_gpu *gpu, uint32_t param, uint64_t *value);
dri-devel@lists.freedesktop.org
-
Akhil P Oommen