--- /dev/null
+/*
+ * Copyright 2018 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+
+#include "amdgpu.h"
+#include "amdgpu_discovery.h"
+#include "soc15_common.h"
+#include "soc15_hw_ip.h"
+#include "nbio/nbio_2_3_offset.h"
+#include "discovery.h"
+
+#define mmMM_INDEX 0x0
+#define mmMM_INDEX_HI 0x6
+#define mmMM_DATA 0x1
+#define HW_ID_MAX 300
+
+const char *hw_id_names[HW_ID_MAX] = {
+ [MP1_HWID] = "MP1",
+ [MP2_HWID] = "MP2",
+ [THM_HWID] = "THM",
+ [SMUIO_HWID] = "SMUIO",
+ [FUSE_HWID] = "FUSE",
+ [CLKA_HWID] = "CLKA",
+ [PWR_HWID] = "PWR",
+ [GC_HWID] = "GC",
+ [UVD_HWID] = "UVD",
+ [AUDIO_AZ_HWID] = "AUDIO_AZ",
+ [ACP_HWID] = "ACP",
+ [DCI_HWID] = "DCI",
+ [DMU_HWID] = "DMU",
+ [DCO_HWID] = "DCO",
+ [DIO_HWID] = "DIO",
+ [XDMA_HWID] = "XDMA",
+ [DCEAZ_HWID] = "DCEAZ",
+ [DAZ_HWID] = "DAZ",
+ [SDPMUX_HWID] = "SDPMUX",
+ [NTB_HWID] = "NTB",
+ [IOHC_HWID] = "IOHC",
+ [L2IMU_HWID] = "L2IMU",
+ [VCE_HWID] = "VCE",
+ [MMHUB_HWID] = "MMHUB",
+ [ATHUB_HWID] = "ATHUB",
+ [DBGU_NBIO_HWID] = "DBGU_NBIO",
+ [DFX_HWID] = "DFX",
+ [DBGU0_HWID] = "DBGU0",
+ [DBGU1_HWID] = "DBGU1",
+ [OSSSYS_HWID] = "OSSSYS",
+ [HDP_HWID] = "HDP",
+ [SDMA0_HWID] = "SDMA0",
+ [SDMA1_HWID] = "SDMA1",
+ [ISP_HWID] = "ISP",
+ [DBGU_IO_HWID] = "DBGU_IO",
+ [DF_HWID] = "DF",
+ [CLKB_HWID] = "CLKB",
+ [FCH_HWID] = "FCH",
+ [DFX_DAP_HWID] = "DFX_DAP",
+ [L1IMU_PCIE_HWID] = "L1IMU_PCIE",
+ [L1IMU_NBIF_HWID] = "L1IMU_NBIF",
+ [L1IMU_IOAGR_HWID] = "L1IMU_IOAGR",
+ [L1IMU3_HWID] = "L1IMU3",
+ [L1IMU4_HWID] = "L1IMU4",
+ [L1IMU5_HWID] = "L1IMU5",
+ [L1IMU6_HWID] = "L1IMU6",
+ [L1IMU7_HWID] = "L1IMU7",
+ [L1IMU8_HWID] = "L1IMU8",
+ [L1IMU9_HWID] = "L1IMU9",
+ [L1IMU10_HWID] = "L1IMU10",
+ [L1IMU11_HWID] = "L1IMU11",
+ [L1IMU12_HWID] = "L1IMU12",
+ [L1IMU13_HWID] = "L1IMU13",
+ [L1IMU14_HWID] = "L1IMU14",
+ [L1IMU15_HWID] = "L1IMU15",
+ [WAFLC_HWID] = "WAFLC",
+ [FCH_USB_PD_HWID] = "FCH_USB_PD",
+ [PCIE_HWID] = "PCIE",
+ [PCS_HWID] = "PCS",
+ [DDCL_HWID] = "DDCL",
+ [SST_HWID] = "SST",
+ [IOAGR_HWID] = "IOAGR",
+ [NBIF_HWID] = "NBIF",
+ [IOAPIC_HWID] = "IOAPIC",
+ [SYSTEMHUB_HWID] = "SYSTEMHUB",
+ [NTBCCP_HWID] = "NTBCCP",
+ [UMC_HWID] = "UMC",
+ [SATA_HWID] = "SATA",
+ [USB_HWID] = "USB",
+ [CCXSEC_HWID] = "CCXSEC",
+ [XGMI_HWID] = "XGMI",
+ [XGBE_HWID] = "XGBE",
+ [MP0_HWID] = "MP0",
+};
+
+static int hw_id_map[MAX_HWIP] = {
+ [GC_HWIP] = GC_HWID,
+ [HDP_HWIP] = HDP_HWID,
+ [SDMA0_HWIP] = SDMA0_HWID,
+ [SDMA1_HWIP] = SDMA1_HWID,
+ [MMHUB_HWIP] = MMHUB_HWID,
+ [ATHUB_HWIP] = ATHUB_HWID,
+ [NBIO_HWIP] = DBGU_NBIO_HWID,
+ [MP0_HWIP] = MP0_HWID,
+ [MP1_HWIP] = MP1_HWID,
+ [UVD_HWIP] = UVD_HWID,
+ [VCE_HWIP] = VCE_HWID,
+ [DF_HWIP] = DF_HWID,
+ [DCE_HWIP] = DCEAZ_HWID,
+ [OSSSYS_HWIP] = OSSSYS_HWID,
+ [SMUIO_HWIP] = SMUIO_HWID,
+ [PWR_HWIP] = PWR_HWID,
+ [NBIF_HWIP] = NBIF_HWID,
+ [THM_HWIP] = THM_HWID,
+ [CLK_HWIP] = CLKA_HWID,
+};
+
+static int amdgpu_discovery_read_binary(struct amdgpu_device *adev, uint8_t *binary)
+{
+ uint32_t *p = (uint32_t *)binary;
+ uint64_t vram_size = RREG32_SOC15(NBIO, 0,
+ mmRCC_DEV0_EPF0_RCC_CONFIG_MEMSIZE) * 1024 * 1024;
+ uint64_t pos = vram_size - BINARY_MAX_SIZE;
+ unsigned long flags;
+
+ while (pos < vram_size) {
+ spin_lock_irqsave(&adev->mmio_idx_lock, flags);
+ WREG32_NO_KIQ(mmMM_INDEX, ((uint32_t)pos) | 0x80000000);
+ WREG32_NO_KIQ(mmMM_INDEX_HI, pos >> 31);
+ *p++ = RREG32_NO_KIQ(mmMM_DATA);
+ spin_unlock_irqrestore(&adev->mmio_idx_lock, flags);
+ pos += 4;
+ }
+
+ return 0;
+}
+
+static uint16_t amdgpu_discovery_calculate_checksum(uint8_t *data, uint32_t size)
+{
+ uint16_t checksum = 0;
+ int i;
+
+ for (i = 0; i < size; i++)
+ checksum += data[i];
+
+ return checksum;
+}
+
+static inline bool amdgpu_discovery_verify_checksum(uint8_t *data, uint32_t size,
+ uint16_t expected)
+{
+ return !!(amdgpu_discovery_calculate_checksum(data, size) == expected);
+}
+
+int amdgpu_discovery_init(struct amdgpu_device *adev)
+{
+ struct table_info *info;
+ struct binary_header *bhdr;
+ struct ip_discovery_header *ihdr;
+ struct gpu_info_header *ghdr;
+ uint16_t offset;
+ uint16_t size;
+ uint16_t checksum;
+ int r;
+
+ adev->discovery = kzalloc(BINARY_MAX_SIZE, GFP_KERNEL);
+ if (!adev->discovery)
+ return -ENOMEM;
+
+ r = amdgpu_discovery_read_binary(adev, adev->discovery);
+ if (r) {
+ DRM_ERROR("failed to read ip discovery binary\n");
+ goto out;
+ }
+
+ bhdr = (struct binary_header *)(adev->discovery + PSP_HEADER_SIZE);
+
+ if (le32_to_cpu(bhdr->binary_signature) != BINARY_SIGNATURE) {
+ DRM_ERROR("invalid ip discovery binary signature\n");
+ r = -EINVAL;
+ goto out;
+ }
+
+ offset = PSP_HEADER_SIZE +
+ offsetof(struct binary_header, binary_checksum) +
+ sizeof(bhdr->binary_checksum);
+ size = bhdr->binary_size - offset;
+ checksum = bhdr->binary_checksum;
+
+ if (!amdgpu_discovery_verify_checksum(adev->discovery + offset,
+ size, checksum)) {
+ DRM_ERROR("invalid ip discovery binary checksum\n");
+ r = -EINVAL;
+ goto out;
+ }
+
+ info = &bhdr->table_list[IP_DISCOVERY];
+ offset = le16_to_cpu(info->offset);
+ checksum = le16_to_cpu(info->checksum);
+ ihdr = (struct ip_discovery_header *)(adev->discovery + offset);
+
+ if (le32_to_cpu(ihdr->signature) != DISCOVERY_TABLE_SIGNATURE) {
+ DRM_ERROR("invalid ip discovery data table signature\n");
+ r = -EINVAL;
+ goto out;
+ }
+
+ if (!amdgpu_discovery_verify_checksum(adev->discovery + offset,
+ ihdr->size, checksum)) {
+ DRM_ERROR("invalid ip discovery data table checksum\n");
+ r = -EINVAL;
+ goto out;
+ }
+
+ info = &bhdr->table_list[GC];
+ offset = le16_to_cpu(info->offset);
+ checksum = le16_to_cpu(info->checksum);
+ ghdr = (struct gpu_info_header *)(adev->discovery + offset);
+
+ if (!amdgpu_discovery_verify_checksum(adev->discovery + offset,
+ ghdr->size, checksum)) {
+ DRM_ERROR("invalid gc data table checksum\n");
+ r = -EINVAL;
+ goto out;
+ }
+
+ return 0;
+
+out:
+ kfree(adev->discovery);
+ adev->discovery = NULL;
+
+ return r;
+}
+
+void amdgpu_discovery_fini(struct amdgpu_device *adev)
+{
+ kfree(adev->discovery);
+ adev->discovery = NULL;
+}
+
+int amdgpu_discovery_reg_base_init(struct amdgpu_device *adev)
+{
+ struct binary_header *bhdr;
+ struct ip_discovery_header *ihdr;
+ struct die_header *dhdr;
+ struct ip *ip;
+ uint16_t die_offset;
+ uint16_t ip_offset;
+ uint16_t num_dies;
+ uint16_t num_ips;
+ uint8_t num_base_address;
+ int hw_ip;
+ int hw_id;
+ int i, j, k;
+
+ if (!adev->discovery) {
+ DRM_ERROR("ip discovery uninitialized\n");
+ return -EINVAL;
+ }
+
+ bhdr = (struct binary_header *)(adev->discovery + PSP_HEADER_SIZE);
+ ihdr = (struct ip_discovery_header *)(adev->discovery +
+ le16_to_cpu(bhdr->table_list[IP_DISCOVERY].offset));
+ num_dies = le16_to_cpu(ihdr->num_dies);
+
+ for (hw_ip = 0; hw_ip < MAX_HWIP; hw_ip++) {
+ hw_id = hw_id_map[hw_ip];
+
+ for (i = 0; i < num_dies; i++) {
+ die_offset = le16_to_cpu(ihdr->die_info[i].die_offset);
+ dhdr = (struct die_header *)(adev->discovery + die_offset);
+ num_ips = le16_to_cpu(dhdr->num_ips);
+ ip_offset = die_offset + sizeof(*dhdr);
+
+ for (j = 0; j < num_ips; j++) {
+ ip = (struct ip *)(adev->discovery + ip_offset);
+ num_base_address = ip->num_base_address;
+
+ if (le16_to_cpu(ip->hw_id) == hw_id) {
+ DRM_DEBUG("%s(%d) v%d.%d.%d:\n",
+ hw_id_names[hw_id], hw_id,
+ ip->major, ip->minor,
+ ip->revision);
+
+ for (k = 0; k < num_base_address; k++) {
+ /*
+ * convert the endianness and unit (in dword) of base addresses in place,
+ * so that we don't need to convert them when accessing adev->reg_offset.
+ */
+ ip->base_address[k] = le32_to_cpu(ip->base_address[k]) >> 2;
+ DRM_DEBUG("\t0x%08x\n", ip->base_address[k] << 2);
+ }
+
+ adev->reg_offset[hw_ip][ip->number_instance] =
+ ip->base_address;
+ }
+
+ ip_offset += sizeof(*ip) + 4 * (ip->num_base_address - 1);
+ }
+ }
+ }
+
+ return 0;
+}
+
+int amdgpu_discovery_get_ip_version(struct amdgpu_device *adev, int hw_id,
+ int *major, int *minor)
+{
+ struct binary_header *bhdr;
+ struct ip_discovery_header *ihdr;
+ struct die_header *dhdr;
+ struct ip *ip;
+ uint16_t die_offset;
+ uint16_t ip_offset;
+ uint16_t num_dies;
+ uint16_t num_ips;
+ int i, j;
+
+ if (!adev->discovery) {
+ DRM_ERROR("ip discovery uninitialized\n");
+ return -EINVAL;
+ }
+
+ bhdr = (struct binary_header *)(adev->discovery + PSP_HEADER_SIZE);
+ ihdr = (struct ip_discovery_header *)(adev->discovery +
+ le16_to_cpu(bhdr->table_list[IP_DISCOVERY].offset));
+ num_dies = le16_to_cpu(ihdr->num_dies);
+
+ for (i = 0; i < num_dies; i++) {
+ die_offset = le16_to_cpu(ihdr->die_info[i].die_offset);
+ dhdr = (struct die_header *)(adev->discovery + die_offset);
+ num_ips = le16_to_cpu(dhdr->num_ips);
+ ip_offset = die_offset + sizeof(*dhdr);
+
+ for (j = 0; j < num_ips; j++) {
+ ip = (struct ip *)(adev->discovery + ip_offset);
+
+ if (le16_to_cpu(ip->hw_id) == hw_id) {
+ if (major)
+ *major = ip->major;
+ if (minor)
+ *minor = ip->minor;
+ return 0;
+ }
+ ip_offset += sizeof(*ip) + 4 * (ip->num_base_address - 1);
+ }
+ }
+
+ return -EINVAL;
+}
+
+int amdgpu_discovery_get_gfx_info(struct amdgpu_device *adev)
+{
+ struct binary_header *bhdr;
+ struct gc_info_v1_0 *gc_info;
+
+ if (!adev->discovery) {
+ DRM_ERROR("ip discovery uninitialized\n");
+ return -EINVAL;
+ }
+
+ bhdr = (struct binary_header *)(adev->discovery + PSP_HEADER_SIZE);
+ gc_info = (struct gc_info_v1_0 *)(adev->discovery +
+ le16_to_cpu(bhdr->table_list[GC].offset));
+
+ adev->gfx.config.max_shader_engines = le32_to_cpu(gc_info->gc_num_se);
+ adev->gfx.config.max_cu_per_sh = le32_to_cpu(gc_info->gc_num_wgp0_per_sa);
+ adev->gfx.config.max_sh_per_se = le32_to_cpu(gc_info->gc_num_wgp1_per_sa);
+ adev->gfx.config.max_backends_per_se = le32_to_cpu(gc_info->gc_num_rb_per_se);
+ adev->gfx.config.max_texture_channel_caches = le32_to_cpu(gc_info->gc_num_gl2c);
+ adev->gfx.config.max_gprs = le32_to_cpu(gc_info->gc_num_gprs);
+ adev->gfx.config.max_gs_threads = le32_to_cpu(gc_info->gc_num_max_gs_thds);
+ adev->gfx.config.gs_vgt_table_depth = le32_to_cpu(gc_info->gc_gs_table_depth);
+ adev->gfx.config.gs_prim_buffer_depth = le32_to_cpu(gc_info->gc_gsprim_buff_depth);
+ adev->gfx.config.double_offchip_lds_buf = le32_to_cpu(gc_info->gc_double_offchip_lds_buffer);
+ adev->gfx.cu_info.wave_front_size = le32_to_cpu(gc_info->gc_wave_size);
+ adev->gfx.cu_info.max_waves_per_simd = le32_to_cpu(gc_info->gc_max_waves_per_simd);
+ adev->gfx.cu_info.max_scratch_slots_per_cu = le32_to_cpu(gc_info->gc_max_scratch_slots_per_cu);
+ adev->gfx.cu_info.lds_size = le32_to_cpu(gc_info->gc_lds_size);
+ adev->gfx.config.num_sc_per_sh = le32_to_cpu(gc_info->gc_num_sc_per_se);
+ adev->gfx.config.num_packer_per_sc = le32_to_cpu(gc_info->gc_num_sa_per_se);
+
+ return 0;
+}
--- /dev/null
+/*
+ * Copyright 2018 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+
+#ifndef _DISCOVERY_H_
+#define _DISCOVERY_H_
+
+#define PSP_HEADER_SIZE 256
+#define BINARY_MAX_SIZE (64 << 10)
+#define BINARY_SIGNATURE 0x28211407
+#define DISCOVERY_TABLE_SIGNATURE 0x53445049
+
+typedef enum
+{
+ IP_DISCOVERY = 0,
+ GC,
+ TABLE_3,
+ TABLE_4,
+ RESERVED_1,
+ RESERVED_2,
+ TOTAL_TABLES = 6
+} table;
+
+#pragma pack(1)
+
+typedef struct table_info
+{
+ uint16_t offset; /* Byte offset */
+ uint16_t checksum; /* Byte sum of the table */
+} table_info;
+
+typedef struct binary_header
+{
+ /* psp structure should go at the top of this structure */
+ uint32_t binary_signature; /* 0x7, 0x14, 0x21, 0x28 */
+ uint16_t binary_checksum; /* Byte sum of the binary after this field */
+ uint16_t binary_size; /* Binary Size*/
+ table_info table_list[TOTAL_TABLES];
+} binary_header;
+
+typedef struct die_info
+{
+ uint16_t die_id;
+ uint16_t die_offset; /* Points to the corresponding die_header structure */
+} die_info;
+
+
+typedef struct ip_discovery_header
+{
+ uint32_t signature; /* Table Signature */
+ uint16_t version; /* Table Version */
+ uint16_t size; /* Table Size */
+ uint32_t id; /* Table ID */
+ uint16_t num_dies; /* Number of Dies */
+ die_info die_info[16]; /* list die information for up to 16 dies */
+ uint16_t padding[1]; /* padding */
+} ip_discovery_header;
+
+typedef struct ip
+{
+ uint16_t hw_id; /* Hardware ID */
+ uint8_t number_instance; /* instance of the IP */
+ uint8_t num_base_address; /* Number of Base Addresses */
+ uint8_t major; /* HCID Major */
+ uint8_t minor; /* HCID Minor */
+ uint8_t revision; /* HCID Revision */
+#if defined(__BIG_ENDIAN)
+ uint8_t reserved : 4; /* Placeholder field */
+ uint8_t harvest : 4; /* Harvest */
+#else
+ uint8_t harvest : 4; /* Harvest */
+ uint8_t reserved : 4; /* Placeholder field */
+#endif
+ uint32_t base_address[1]; /* variable number of Addresses */
+} ip;
+
+typedef struct die_header
+{
+ uint32_t die_id;
+ uint32_t num_ips;
+} die_header;
+
+typedef struct ip_structure
+{
+ ip_discovery_header* header;
+ struct die
+ {
+ die_header *die_header;
+ ip *ip_list;
+ } die;
+} ip_structure;
+
+struct gpu_info_header {
+ uint32_t table_id; /* table ID */
+ uint16_t version_major; /* table version */
+ uint16_t version_minor; /* table version */
+ uint32_t size; /* size of the entire header+data in bytes */
+};
+
+struct gc_info_v1_0 {
+ struct gpu_info_header header;
+
+ uint32_t gc_num_se;
+ uint32_t gc_num_wgp0_per_sa;
+ uint32_t gc_num_wgp1_per_sa;
+ uint32_t gc_num_rb_per_se;
+ uint32_t gc_num_gl2c;
+ uint32_t gc_num_gprs;
+ uint32_t gc_num_max_gs_thds;
+ uint32_t gc_gs_table_depth;
+ uint32_t gc_gsprim_buff_depth;
+ uint32_t gc_parameter_cache_depth;
+ uint32_t gc_double_offchip_lds_buffer;
+ uint32_t gc_wave_size;
+ uint32_t gc_max_waves_per_simd;
+ uint32_t gc_max_scratch_slots_per_cu;
+ uint32_t gc_lds_size;
+ uint32_t gc_num_sc_per_se;
+ uint32_t gc_num_sa_per_se;
+ uint32_t gc_num_packer_per_sc;
+ uint32_t gc_num_gl2a;
+};
+
+#pragma pack()
+
+#endif