diff options
author | Daniel Baumann <mail@daniel-baumann.ch> | 2025-06-06 10:05:23 +0000 |
---|---|---|
committer | Daniel Baumann <mail@daniel-baumann.ch> | 2025-06-06 10:05:23 +0000 |
commit | 755cc582a2473d06f3a2131d506d0311cc70e9f9 (patch) | |
tree | 3efb1ddb8d57bbb4539ac0d229b384871c57820f /hw/cxl | |
parent | Initial commit. (diff) | |
download | qemu-upstream.tar.xz qemu-upstream.zip |
Adding upstream version 1:7.2+dfsg.upstream/1%7.2+dfsgupstream
Signed-off-by: Daniel Baumann <mail@daniel-baumann.ch>
Diffstat (limited to 'hw/cxl')
-rw-r--r-- | hw/cxl/Kconfig | 3 | ||||
-rw-r--r-- | hw/cxl/cxl-cdat.c | 224 | ||||
-rw-r--r-- | hw/cxl/cxl-component-utils.c | 405 | ||||
-rw-r--r-- | hw/cxl/cxl-device-utils.c | 271 | ||||
-rw-r--r-- | hw/cxl/cxl-host-stubs.c | 15 | ||||
-rw-r--r-- | hw/cxl/cxl-host.c | 340 | ||||
-rw-r--r-- | hw/cxl/cxl-mailbox-utils.c | 478 | ||||
-rw-r--r-- | hw/cxl/meson.build | 13 |
8 files changed, 1749 insertions, 0 deletions
diff --git a/hw/cxl/Kconfig b/hw/cxl/Kconfig new file mode 100644 index 00000000..8e67519b --- /dev/null +++ b/hw/cxl/Kconfig @@ -0,0 +1,3 @@ +config CXL + bool + default y if PCI_EXPRESS diff --git a/hw/cxl/cxl-cdat.c b/hw/cxl/cxl-cdat.c new file mode 100644 index 00000000..3653aa56 --- /dev/null +++ b/hw/cxl/cxl-cdat.c @@ -0,0 +1,224 @@ +/* + * CXL CDAT Structure + * + * Copyright (C) 2021 Avery Design Systems, Inc. + * + * This work is licensed under the terms of the GNU GPL, version 2 or later. + * See the COPYING file in the top-level directory. + */ + +#include "qemu/osdep.h" +#include "hw/pci/pci.h" +#include "hw/cxl/cxl.h" +#include "qapi/error.h" +#include "qemu/error-report.h" + +static void cdat_len_check(CDATSubHeader *hdr, Error **errp) +{ + assert(hdr->length); + assert(hdr->reserved == 0); + + switch (hdr->type) { + case CDAT_TYPE_DSMAS: + assert(hdr->length == sizeof(CDATDsmas)); + break; + case CDAT_TYPE_DSLBIS: + assert(hdr->length == sizeof(CDATDslbis)); + break; + case CDAT_TYPE_DSMSCIS: + assert(hdr->length == sizeof(CDATDsmscis)); + break; + case CDAT_TYPE_DSIS: + assert(hdr->length == sizeof(CDATDsis)); + break; + case CDAT_TYPE_DSEMTS: + assert(hdr->length == sizeof(CDATDsemts)); + break; + case CDAT_TYPE_SSLBIS: + assert(hdr->length >= sizeof(CDATSslbisHeader)); + assert((hdr->length - sizeof(CDATSslbisHeader)) % + sizeof(CDATSslbe) == 0); + break; + default: + error_setg(errp, "Type %d is reserved", hdr->type); + } +} + +static void ct3_build_cdat(CDATObject *cdat, Error **errp) +{ + g_autofree CDATTableHeader *cdat_header = NULL; + g_autofree CDATEntry *cdat_st = NULL; + uint8_t sum = 0; + int ent, i; + + /* Use default table if fopen == NULL */ + assert(cdat->build_cdat_table); + + cdat_header = g_malloc0(sizeof(*cdat_header)); + if (!cdat_header) { + error_setg(errp, "Failed to allocate CDAT header"); + return; + } + + cdat->built_buf_len = cdat->build_cdat_table(&cdat->built_buf, cdat->private); + + if (!cdat->built_buf_len) { + /* Build later as not all data available yet */ + cdat->to_update = true; + return; + } + cdat->to_update = false; + + cdat_st = g_malloc0(sizeof(*cdat_st) * (cdat->built_buf_len + 1)); + if (!cdat_st) { + error_setg(errp, "Failed to allocate CDAT entry array"); + return; + } + + /* Entry 0 for CDAT header, starts with Entry 1 */ + for (ent = 1; ent < cdat->built_buf_len + 1; ent++) { + CDATSubHeader *hdr = cdat->built_buf[ent - 1]; + uint8_t *buf = (uint8_t *)cdat->built_buf[ent - 1]; + + cdat_st[ent].base = hdr; + cdat_st[ent].length = hdr->length; + + cdat_header->length += hdr->length; + for (i = 0; i < hdr->length; i++) { + sum += buf[i]; + } + } + + /* CDAT header */ + cdat_header->revision = CXL_CDAT_REV; + /* For now, no runtime updates */ + cdat_header->sequence = 0; + cdat_header->length += sizeof(CDATTableHeader); + sum += cdat_header->revision + cdat_header->sequence + + cdat_header->length; + /* Sum of all bytes including checksum must be 0 */ + cdat_header->checksum = ~sum + 1; + + cdat_st[0].base = g_steal_pointer(&cdat_header); + cdat_st[0].length = sizeof(*cdat_header); + cdat->entry_len = 1 + cdat->built_buf_len; + cdat->entry = g_steal_pointer(&cdat_st); +} + +static void ct3_load_cdat(CDATObject *cdat, Error **errp) +{ + g_autofree CDATEntry *cdat_st = NULL; + uint8_t sum = 0; + int num_ent; + int i = 0, ent = 1, file_size = 0; + CDATSubHeader *hdr; + FILE *fp = NULL; + + /* Read CDAT file and create its cache */ + fp = fopen(cdat->filename, "r"); + if (!fp) { + error_setg(errp, "CDAT: Unable to open file"); + return; + } + + fseek(fp, 0, SEEK_END); + file_size = ftell(fp); + fseek(fp, 0, SEEK_SET); + cdat->buf = g_malloc0(file_size); + + if (fread(cdat->buf, file_size, 1, fp) == 0) { + error_setg(errp, "CDAT: File read failed"); + return; + } + + fclose(fp); + + if (file_size < sizeof(CDATTableHeader)) { + error_setg(errp, "CDAT: File too short"); + return; + } + i = sizeof(CDATTableHeader); + num_ent = 1; + while (i < file_size) { + hdr = (CDATSubHeader *)(cdat->buf + i); + cdat_len_check(hdr, errp); + i += hdr->length; + num_ent++; + } + if (i != file_size) { + error_setg(errp, "CDAT: File length missmatch"); + return; + } + + cdat_st = g_malloc0(sizeof(*cdat_st) * num_ent); + if (!cdat_st) { + error_setg(errp, "CDAT: Failed to allocate entry array"); + return; + } + + /* Set CDAT header, Entry = 0 */ + cdat_st[0].base = cdat->buf; + cdat_st[0].length = sizeof(CDATTableHeader); + i = 0; + + while (i < cdat_st[0].length) { + sum += cdat->buf[i++]; + } + + /* Read CDAT structures */ + while (i < file_size) { + hdr = (CDATSubHeader *)(cdat->buf + i); + cdat_len_check(hdr, errp); + + cdat_st[ent].base = hdr; + cdat_st[ent].length = hdr->length; + + while (cdat->buf + i < + (uint8_t *)cdat_st[ent].base + cdat_st[ent].length) { + assert(i < file_size); + sum += cdat->buf[i++]; + } + + ent++; + } + + if (sum != 0) { + warn_report("CDAT: Found checksum mismatch in %s", cdat->filename); + } + cdat->entry_len = num_ent; + cdat->entry = g_steal_pointer(&cdat_st); +} + +void cxl_doe_cdat_init(CXLComponentState *cxl_cstate, Error **errp) +{ + CDATObject *cdat = &cxl_cstate->cdat; + + if (cdat->filename) { + ct3_load_cdat(cdat, errp); + } else { + ct3_build_cdat(cdat, errp); + } +} + +void cxl_doe_cdat_update(CXLComponentState *cxl_cstate, Error **errp) +{ + CDATObject *cdat = &cxl_cstate->cdat; + + if (cdat->to_update) { + ct3_build_cdat(cdat, errp); + } +} + +void cxl_doe_cdat_release(CXLComponentState *cxl_cstate) +{ + CDATObject *cdat = &cxl_cstate->cdat; + + free(cdat->entry); + if (cdat->built_buf) { + cdat->free_cdat_table(cdat->built_buf, cdat->built_buf_len, + cdat->private); + } + if (cdat->buf) { + free(cdat->buf); + } +} diff --git a/hw/cxl/cxl-component-utils.c b/hw/cxl/cxl-component-utils.c new file mode 100644 index 00000000..3edd303a --- /dev/null +++ b/hw/cxl/cxl-component-utils.c @@ -0,0 +1,405 @@ +/* + * CXL Utility library for components + * + * Copyright(C) 2020 Intel Corporation. + * + * This work is licensed under the terms of the GNU GPL, version 2. See the + * COPYING file in the top-level directory. + */ + +#include "qemu/osdep.h" +#include "qemu/log.h" +#include "qapi/error.h" +#include "hw/pci/pci.h" +#include "hw/cxl/cxl.h" + +static uint64_t cxl_cache_mem_read_reg(void *opaque, hwaddr offset, + unsigned size) +{ + CXLComponentState *cxl_cstate = opaque; + ComponentRegisters *cregs = &cxl_cstate->crb; + + if (size == 8) { + qemu_log_mask(LOG_UNIMP, + "CXL 8 byte cache mem registers not implemented\n"); + return 0; + } + + if (cregs->special_ops && cregs->special_ops->read) { + return cregs->special_ops->read(cxl_cstate, offset, size); + } else { + return cregs->cache_mem_registers[offset / sizeof(*cregs->cache_mem_registers)]; + } +} + +static void dumb_hdm_handler(CXLComponentState *cxl_cstate, hwaddr offset, + uint32_t value) +{ + ComponentRegisters *cregs = &cxl_cstate->crb; + uint32_t *cache_mem = cregs->cache_mem_registers; + bool should_commit = false; + + switch (offset) { + case A_CXL_HDM_DECODER0_CTRL: + should_commit = FIELD_EX32(value, CXL_HDM_DECODER0_CTRL, COMMIT); + break; + default: + break; + } + + memory_region_transaction_begin(); + stl_le_p((uint8_t *)cache_mem + offset, value); + if (should_commit) { + ARRAY_FIELD_DP32(cache_mem, CXL_HDM_DECODER0_CTRL, COMMIT, 0); + ARRAY_FIELD_DP32(cache_mem, CXL_HDM_DECODER0_CTRL, ERR, 0); + ARRAY_FIELD_DP32(cache_mem, CXL_HDM_DECODER0_CTRL, COMMITTED, 1); + } + memory_region_transaction_commit(); +} + +static void cxl_cache_mem_write_reg(void *opaque, hwaddr offset, uint64_t value, + unsigned size) +{ + CXLComponentState *cxl_cstate = opaque; + ComponentRegisters *cregs = &cxl_cstate->crb; + uint32_t mask; + + if (size == 8) { + qemu_log_mask(LOG_UNIMP, + "CXL 8 byte cache mem registers not implemented\n"); + return; + } + mask = cregs->cache_mem_regs_write_mask[offset / sizeof(*cregs->cache_mem_regs_write_mask)]; + value &= mask; + /* RO bits should remain constant. Done by reading existing value */ + value |= ~mask & cregs->cache_mem_registers[offset / sizeof(*cregs->cache_mem_registers)]; + if (cregs->special_ops && cregs->special_ops->write) { + cregs->special_ops->write(cxl_cstate, offset, value, size); + return; + } + + if (offset >= A_CXL_HDM_DECODER_CAPABILITY && + offset <= A_CXL_HDM_DECODER0_TARGET_LIST_HI) { + dumb_hdm_handler(cxl_cstate, offset, value); + } else { + cregs->cache_mem_registers[offset / sizeof(*cregs->cache_mem_registers)] = value; + } +} + +/* + * 8.2.3 + * The access restrictions specified in Section 8.2.2 also apply to CXL 2.0 + * Component Registers. + * + * 8.2.2 + * • A 32 bit register shall be accessed as a 4 Bytes quantity. Partial + * reads are not permitted. + * • A 64 bit register shall be accessed as a 8 Bytes quantity. Partial + * reads are not permitted. + * + * As of the spec defined today, only 4 byte registers exist. + */ +static const MemoryRegionOps cache_mem_ops = { + .read = cxl_cache_mem_read_reg, + .write = cxl_cache_mem_write_reg, + .endianness = DEVICE_LITTLE_ENDIAN, + .valid = { + .min_access_size = 4, + .max_access_size = 8, + .unaligned = false, + }, + .impl = { + .min_access_size = 4, + .max_access_size = 8, + }, +}; + +void cxl_component_register_block_init(Object *obj, + CXLComponentState *cxl_cstate, + const char *type) +{ + ComponentRegisters *cregs = &cxl_cstate->crb; + + memory_region_init(&cregs->component_registers, obj, type, + CXL2_COMPONENT_BLOCK_SIZE); + + /* io registers controls link which we don't care about in QEMU */ + memory_region_init_io(&cregs->io, obj, NULL, cregs, ".io", + CXL2_COMPONENT_IO_REGION_SIZE); + memory_region_init_io(&cregs->cache_mem, obj, &cache_mem_ops, cregs, + ".cache_mem", CXL2_COMPONENT_CM_REGION_SIZE); + + memory_region_add_subregion(&cregs->component_registers, 0, &cregs->io); + memory_region_add_subregion(&cregs->component_registers, + CXL2_COMPONENT_IO_REGION_SIZE, + &cregs->cache_mem); +} + +static void ras_init_common(uint32_t *reg_state, uint32_t *write_msk) +{ + /* + * Error status is RW1C but given bits are not yet set, it can + * be handled as RO. + */ + reg_state[R_CXL_RAS_UNC_ERR_STATUS] = 0; + /* Bits 12-13 and 17-31 reserved in CXL 2.0 */ + reg_state[R_CXL_RAS_UNC_ERR_MASK] = 0x1cfff; + write_msk[R_CXL_RAS_UNC_ERR_MASK] = 0x1cfff; + reg_state[R_CXL_RAS_UNC_ERR_SEVERITY] = 0x1cfff; + write_msk[R_CXL_RAS_UNC_ERR_SEVERITY] = 0x1cfff; + reg_state[R_CXL_RAS_COR_ERR_STATUS] = 0; + reg_state[R_CXL_RAS_COR_ERR_MASK] = 0x7f; + write_msk[R_CXL_RAS_COR_ERR_MASK] = 0x7f; + /* CXL switches and devices must set */ + reg_state[R_CXL_RAS_ERR_CAP_CTRL] = 0x00; +} + +static void hdm_init_common(uint32_t *reg_state, uint32_t *write_msk, + enum reg_type type) +{ + int decoder_count = 1; + int i; + + ARRAY_FIELD_DP32(reg_state, CXL_HDM_DECODER_CAPABILITY, DECODER_COUNT, + cxl_decoder_count_enc(decoder_count)); + ARRAY_FIELD_DP32(reg_state, CXL_HDM_DECODER_CAPABILITY, TARGET_COUNT, 1); + ARRAY_FIELD_DP32(reg_state, CXL_HDM_DECODER_CAPABILITY, INTERLEAVE_256B, 1); + ARRAY_FIELD_DP32(reg_state, CXL_HDM_DECODER_CAPABILITY, INTERLEAVE_4K, 1); + ARRAY_FIELD_DP32(reg_state, CXL_HDM_DECODER_CAPABILITY, POISON_ON_ERR_CAP, 0); + ARRAY_FIELD_DP32(reg_state, CXL_HDM_DECODER_GLOBAL_CONTROL, + HDM_DECODER_ENABLE, 0); + write_msk[R_CXL_HDM_DECODER_GLOBAL_CONTROL] = 0x3; + for (i = 0; i < decoder_count; i++) { + write_msk[R_CXL_HDM_DECODER0_BASE_LO + i * 0x20] = 0xf0000000; + write_msk[R_CXL_HDM_DECODER0_BASE_HI + i * 0x20] = 0xffffffff; + write_msk[R_CXL_HDM_DECODER0_SIZE_LO + i * 0x20] = 0xf0000000; + write_msk[R_CXL_HDM_DECODER0_SIZE_HI + i * 0x20] = 0xffffffff; + write_msk[R_CXL_HDM_DECODER0_CTRL + i * 0x20] = 0x13ff; + if (type == CXL2_DEVICE || + type == CXL2_TYPE3_DEVICE || + type == CXL2_LOGICAL_DEVICE) { + write_msk[R_CXL_HDM_DECODER0_TARGET_LIST_LO + i * 0x20] = 0xf0000000; + } else { + write_msk[R_CXL_HDM_DECODER0_TARGET_LIST_LO + i * 0x20] = 0xffffffff; + } + write_msk[R_CXL_HDM_DECODER0_TARGET_LIST_HI + i * 0x20] = 0xffffffff; + } +} + +void cxl_component_register_init_common(uint32_t *reg_state, uint32_t *write_msk, + enum reg_type type) +{ + int caps = 0; + + /* + * In CXL 2.0 the capabilities required for each CXL component are such that, + * with the ordering chosen here, a single number can be used to define + * which capabilities should be provided. + */ + switch (type) { + case CXL2_DOWNSTREAM_PORT: + case CXL2_DEVICE: + /* RAS, Link */ + caps = 2; + break; + case CXL2_UPSTREAM_PORT: + case CXL2_TYPE3_DEVICE: + case CXL2_LOGICAL_DEVICE: + /* + HDM */ + caps = 3; + break; + case CXL2_ROOT_PORT: + /* + Extended Security, + Snoop */ + caps = 5; + break; + default: + abort(); + } + + memset(reg_state, 0, CXL2_COMPONENT_CM_REGION_SIZE); + + /* CXL Capability Header Register */ + ARRAY_FIELD_DP32(reg_state, CXL_CAPABILITY_HEADER, ID, 1); + ARRAY_FIELD_DP32(reg_state, CXL_CAPABILITY_HEADER, VERSION, 1); + ARRAY_FIELD_DP32(reg_state, CXL_CAPABILITY_HEADER, CACHE_MEM_VERSION, 1); + ARRAY_FIELD_DP32(reg_state, CXL_CAPABILITY_HEADER, ARRAY_SIZE, caps); + +#define init_cap_reg(reg, id, version) \ + QEMU_BUILD_BUG_ON(CXL_##reg##_REGISTERS_OFFSET == 0); \ + do { \ + int which = R_CXL_##reg##_CAPABILITY_HEADER; \ + reg_state[which] = FIELD_DP32(reg_state[which], \ + CXL_##reg##_CAPABILITY_HEADER, ID, id); \ + reg_state[which] = \ + FIELD_DP32(reg_state[which], CXL_##reg##_CAPABILITY_HEADER, \ + VERSION, version); \ + reg_state[which] = \ + FIELD_DP32(reg_state[which], CXL_##reg##_CAPABILITY_HEADER, PTR, \ + CXL_##reg##_REGISTERS_OFFSET); \ + } while (0) + + init_cap_reg(RAS, 2, 2); + ras_init_common(reg_state, write_msk); + + init_cap_reg(LINK, 4, 2); + + if (caps < 3) { + return; + } + + init_cap_reg(HDM, 5, 1); + hdm_init_common(reg_state, write_msk, type); + + if (caps < 5) { + return; + } + + init_cap_reg(EXTSEC, 6, 1); + init_cap_reg(SNOOP, 8, 1); + +#undef init_cap_reg +} + +/* + * Helper to creates a DVSEC header for a CXL entity. The caller is responsible + * for tracking the valid offset. + * + * This function will build the DVSEC header on behalf of the caller and then + * copy in the remaining data for the vendor specific bits. + * It will also set up appropriate write masks. + */ +void cxl_component_create_dvsec(CXLComponentState *cxl, + enum reg_type cxl_dev_type, uint16_t length, + uint16_t type, uint8_t rev, uint8_t *body) +{ + PCIDevice *pdev = cxl->pdev; + uint16_t offset = cxl->dvsec_offset; + uint8_t *wmask = pdev->wmask; + + assert(offset >= PCI_CFG_SPACE_SIZE && + ((offset + length) < PCI_CFG_SPACE_EXP_SIZE)); + assert((length & 0xf000) == 0); + assert((rev & ~0xf) == 0); + + /* Create the DVSEC in the MCFG space */ + pcie_add_capability(pdev, PCI_EXT_CAP_ID_DVSEC, 1, offset, length); + pci_set_long(pdev->config + offset + PCIE_DVSEC_HEADER1_OFFSET, + (length << 20) | (rev << 16) | CXL_VENDOR_ID); + pci_set_word(pdev->config + offset + PCIE_DVSEC_ID_OFFSET, type); + memcpy(pdev->config + offset + sizeof(DVSECHeader), + body + sizeof(DVSECHeader), + length - sizeof(DVSECHeader)); + + /* Configure write masks */ + switch (type) { + case PCIE_CXL_DEVICE_DVSEC: + /* Cntrl RW Lock - so needs explicit blocking when lock is set */ + wmask[offset + offsetof(CXLDVSECDevice, ctrl)] = 0xFD; + wmask[offset + offsetof(CXLDVSECDevice, ctrl) + 1] = 0x4F; + /* Status is RW1CS */ + wmask[offset + offsetof(CXLDVSECDevice, ctrl2)] = 0x0F; + /* Lock is RW Once */ + wmask[offset + offsetof(CXLDVSECDevice, lock)] = 0x01; + /* range1/2_base_high/low is RW Lock */ + wmask[offset + offsetof(CXLDVSECDevice, range1_base_hi)] = 0xFF; + wmask[offset + offsetof(CXLDVSECDevice, range1_base_hi) + 1] = 0xFF; + wmask[offset + offsetof(CXLDVSECDevice, range1_base_hi) + 2] = 0xFF; + wmask[offset + offsetof(CXLDVSECDevice, range1_base_hi) + 3] = 0xFF; + wmask[offset + offsetof(CXLDVSECDevice, range1_base_lo) + 3] = 0xF0; + wmask[offset + offsetof(CXLDVSECDevice, range2_base_hi)] = 0xFF; + wmask[offset + offsetof(CXLDVSECDevice, range2_base_hi) + 1] = 0xFF; + wmask[offset + offsetof(CXLDVSECDevice, range2_base_hi) + 2] = 0xFF; + wmask[offset + offsetof(CXLDVSECDevice, range2_base_hi) + 3] = 0xFF; + wmask[offset + offsetof(CXLDVSECDevice, range2_base_lo) + 3] = 0xF0; + break; + case NON_CXL_FUNCTION_MAP_DVSEC: + break; /* Not yet implemented */ + case EXTENSIONS_PORT_DVSEC: + wmask[offset + offsetof(CXLDVSECPortExtensions, control)] = 0x0F; + wmask[offset + offsetof(CXLDVSECPortExtensions, control) + 1] = 0x40; + wmask[offset + offsetof(CXLDVSECPortExtensions, alt_bus_base)] = 0xFF; + wmask[offset + offsetof(CXLDVSECPortExtensions, alt_bus_limit)] = 0xFF; + wmask[offset + offsetof(CXLDVSECPortExtensions, alt_memory_base)] = 0xF0; + wmask[offset + offsetof(CXLDVSECPortExtensions, alt_memory_base) + 1] = 0xFF; + wmask[offset + offsetof(CXLDVSECPortExtensions, alt_memory_limit)] = 0xF0; + wmask[offset + offsetof(CXLDVSECPortExtensions, alt_memory_limit) + 1] = 0xFF; + wmask[offset + offsetof(CXLDVSECPortExtensions, alt_prefetch_base)] = 0xF0; + wmask[offset + offsetof(CXLDVSECPortExtensions, alt_prefetch_base) + 1] = 0xFF; + wmask[offset + offsetof(CXLDVSECPortExtensions, alt_prefetch_limit)] = 0xF0; + wmask[offset + offsetof(CXLDVSECPortExtensions, alt_prefetch_limit) + 1] = 0xFF; + wmask[offset + offsetof(CXLDVSECPortExtensions, alt_prefetch_base_high)] = 0xFF; + wmask[offset + offsetof(CXLDVSECPortExtensions, alt_prefetch_base_high) + 1] = 0xFF; + wmask[offset + offsetof(CXLDVSECPortExtensions, alt_prefetch_base_high) + 2] = 0xFF; + wmask[offset + offsetof(CXLDVSECPortExtensions, alt_prefetch_base_high) + 3] = 0xFF; + wmask[offset + offsetof(CXLDVSECPortExtensions, alt_prefetch_limit_high)] = 0xFF; + wmask[offset + offsetof(CXLDVSECPortExtensions, alt_prefetch_limit_high) + 1] = 0xFF; + wmask[offset + offsetof(CXLDVSECPortExtensions, alt_prefetch_limit_high) + 2] = 0xFF; + wmask[offset + offsetof(CXLDVSECPortExtensions, alt_prefetch_limit_high) + 3] = 0xFF; + break; + case GPF_PORT_DVSEC: + wmask[offset + offsetof(CXLDVSECPortGPF, phase1_ctrl)] = 0x0F; + wmask[offset + offsetof(CXLDVSECPortGPF, phase1_ctrl) + 1] = 0x0F; + wmask[offset + offsetof(CXLDVSECPortGPF, phase2_ctrl)] = 0x0F; + wmask[offset + offsetof(CXLDVSECPortGPF, phase2_ctrl) + 1] = 0x0F; + break; + case GPF_DEVICE_DVSEC: + wmask[offset + offsetof(CXLDVSECDeviceGPF, phase2_duration)] = 0x0F; + wmask[offset + offsetof(CXLDVSECDeviceGPF, phase2_duration) + 1] = 0x0F; + wmask[offset + offsetof(CXLDVSECDeviceGPF, phase2_power)] = 0xFF; + wmask[offset + offsetof(CXLDVSECDeviceGPF, phase2_power) + 1] = 0xFF; + wmask[offset + offsetof(CXLDVSECDeviceGPF, phase2_power) + 2] = 0xFF; + wmask[offset + offsetof(CXLDVSECDeviceGPF, phase2_power) + 3] = 0xFF; + break; + case PCIE_FLEXBUS_PORT_DVSEC: + switch (cxl_dev_type) { + case CXL2_ROOT_PORT: + /* No MLD */ + wmask[offset + offsetof(CXLDVSECPortFlexBus, ctrl)] = 0xbd; + break; + case CXL2_DOWNSTREAM_PORT: + wmask[offset + offsetof(CXLDVSECPortFlexBus, ctrl)] = 0xfd; + break; + default: /* Registers are RO for other component types */ + break; + } + /* There are rw1cs bits in the status register but never set currently */ + break; + } + + /* Update state for future DVSEC additions */ + range_init_nofail(&cxl->dvsecs[type], cxl->dvsec_offset, length); + cxl->dvsec_offset += length; +} + +uint8_t cxl_interleave_ways_enc(int iw, Error **errp) +{ + switch (iw) { + case 1: return 0x0; + case 2: return 0x1; + case 4: return 0x2; + case 8: return 0x3; + case 16: return 0x4; + case 3: return 0x8; + case 6: return 0x9; + case 12: return 0xa; + default: + error_setg(errp, "Interleave ways: %d not supported", iw); + return 0; + } +} + +uint8_t cxl_interleave_granularity_enc(uint64_t gran, Error **errp) +{ + switch (gran) { + case 256: return 0; + case 512: return 1; + case 1024: return 2; + case 2048: return 3; + case 4096: return 4; + case 8192: return 5; + case 16384: return 6; + default: + error_setg(errp, "Interleave granularity: %" PRIu64 " invalid", gran); + return 0; + } +} diff --git a/hw/cxl/cxl-device-utils.c b/hw/cxl/cxl-device-utils.c new file mode 100644 index 00000000..83ce7a82 --- /dev/null +++ b/hw/cxl/cxl-device-utils.c @@ -0,0 +1,271 @@ +/* + * CXL Utility library for devices + * + * Copyright(C) 2020 Intel Corporation. + * + * This work is licensed under the terms of the GNU GPL, version 2. See the + * COPYING file in the top-level directory. + */ + +#include "qemu/osdep.h" +#include "qemu/log.h" +#include "hw/cxl/cxl.h" + +/* + * Device registers have no restrictions per the spec, and so fall back to the + * default memory mapped register rules in 8.2: + * Software shall use CXL.io Memory Read and Write to access memory mapped + * register defined in this section. Unless otherwise specified, software + * shall restrict the accesses width based on the following: + * • A 32 bit register shall be accessed as a 1 Byte, 2 Bytes or 4 Bytes + * quantity. + * • A 64 bit register shall be accessed as a 1 Byte, 2 Bytes, 4 Bytes or 8 + * Bytes + * • The address shall be a multiple of the access width, e.g. when + * accessing a register as a 4 Byte quantity, the address shall be + * multiple of 4. + * • The accesses shall map to contiguous bytes.If these rules are not + * followed, the behavior is undefined + */ + +static uint64_t caps_reg_read(void *opaque, hwaddr offset, unsigned size) +{ + CXLDeviceState *cxl_dstate = opaque; + + if (size == 4) { + return cxl_dstate->caps_reg_state32[offset / sizeof(*cxl_dstate->caps_reg_state32)]; + } else { + return cxl_dstate->caps_reg_state64[offset / sizeof(*cxl_dstate->caps_reg_state64)]; + } +} + +static uint64_t dev_reg_read(void *opaque, hwaddr offset, unsigned size) +{ + return 0; +} + +static uint64_t mailbox_reg_read(void *opaque, hwaddr offset, unsigned size) +{ + CXLDeviceState *cxl_dstate = opaque; + + switch (size) { + case 1: + return cxl_dstate->mbox_reg_state[offset]; + case 2: + return cxl_dstate->mbox_reg_state16[offset / size]; + case 4: + return cxl_dstate->mbox_reg_state32[offset / size]; + case 8: + return cxl_dstate->mbox_reg_state64[offset / size]; + default: + g_assert_not_reached(); + } +} + +static void mailbox_mem_writel(uint32_t *reg_state, hwaddr offset, + uint64_t value) +{ + switch (offset) { + case A_CXL_DEV_MAILBOX_CTRL: + /* fallthrough */ + case A_CXL_DEV_MAILBOX_CAP: + /* RO register */ + break; + default: + qemu_log_mask(LOG_UNIMP, + "%s Unexpected 32-bit access to 0x%" PRIx64 " (WI)\n", + __func__, offset); + return; + } + + reg_state[offset / sizeof(*reg_state)] = value; +} + +static void mailbox_mem_writeq(uint64_t *reg_state, hwaddr offset, + uint64_t value) +{ + switch (offset) { + case A_CXL_DEV_MAILBOX_CMD: + break; + case A_CXL_DEV_BG_CMD_STS: + /* BG not supported */ + /* fallthrough */ + case A_CXL_DEV_MAILBOX_STS: + /* Read only register, will get updated by the state machine */ + return; + default: + qemu_log_mask(LOG_UNIMP, + "%s Unexpected 64-bit access to 0x%" PRIx64 " (WI)\n", + __func__, offset); + return; + } + + + reg_state[offset / sizeof(*reg_state)] = value; +} + +static void mailbox_reg_write(void *opaque, hwaddr offset, uint64_t value, + unsigned size) +{ + CXLDeviceState *cxl_dstate = opaque; + + if (offset >= A_CXL_DEV_CMD_PAYLOAD) { + memcpy(cxl_dstate->mbox_reg_state + offset, &value, size); + return; + } + + switch (size) { + case 4: + mailbox_mem_writel(cxl_dstate->mbox_reg_state32, offset, value); + break; + case 8: + mailbox_mem_writeq(cxl_dstate->mbox_reg_state64, offset, value); + break; + default: + g_assert_not_reached(); + } + + if (ARRAY_FIELD_EX32(cxl_dstate->mbox_reg_state32, CXL_DEV_MAILBOX_CTRL, + DOORBELL)) { + cxl_process_mailbox(cxl_dstate); + } +} + +static uint64_t mdev_reg_read(void *opaque, hwaddr offset, unsigned size) +{ + uint64_t retval = 0; + + retval = FIELD_DP64(retval, CXL_MEM_DEV_STS, MEDIA_STATUS, 1); + retval = FIELD_DP64(retval, CXL_MEM_DEV_STS, MBOX_READY, 1); + + return retval; +} + +static void ro_reg_write(void *opaque, hwaddr offset, uint64_t value, + unsigned size) +{ + /* Many register sets are read only */ +} + +static const MemoryRegionOps mdev_ops = { + .read = mdev_reg_read, + .write = ro_reg_write, + .endianness = DEVICE_LITTLE_ENDIAN, + .valid = { + .min_access_size = 1, + .max_access_size = 8, + .unaligned = false, + }, + .impl = { + .min_access_size = 8, + .max_access_size = 8, + }, +}; + +static const MemoryRegionOps mailbox_ops = { + .read = mailbox_reg_read, + .write = mailbox_reg_write, + .endianness = DEVICE_LITTLE_ENDIAN, + .valid = { + .min_access_size = 1, + .max_access_size = 8, + .unaligned = false, + }, + .impl = { + .min_access_size = 1, + .max_access_size = 8, + }, +}; + +static const MemoryRegionOps dev_ops = { + .read = dev_reg_read, + .write = ro_reg_write, + .endianness = DEVICE_LITTLE_ENDIAN, + .valid = { + .min_access_size = 1, + .max_access_size = 8, + .unaligned = false, + }, + .impl = { + .min_access_size = 1, + .max_access_size = 8, + }, +}; + +static const MemoryRegionOps caps_ops = { + .read = caps_reg_read, + .write = ro_reg_write, + .endianness = DEVICE_LITTLE_ENDIAN, + .valid = { + .min_access_size = 1, + .max_access_size = 8, + .unaligned = false, + }, + .impl = { + .min_access_size = 4, + .max_access_size = 8, + }, +}; + +void cxl_device_register_block_init(Object *obj, CXLDeviceState *cxl_dstate) +{ + /* This will be a BAR, so needs to be rounded up to pow2 for PCI spec */ + memory_region_init(&cxl_dstate->device_registers, obj, "device-registers", + pow2ceil(CXL_MMIO_SIZE)); + + memory_region_init_io(&cxl_dstate->caps, obj, &caps_ops, cxl_dstate, + "cap-array", CXL_CAPS_SIZE); + memory_region_init_io(&cxl_dstate->device, obj, &dev_ops, cxl_dstate, + "device-status", CXL_DEVICE_STATUS_REGISTERS_LENGTH); + memory_region_init_io(&cxl_dstate->mailbox, obj, &mailbox_ops, cxl_dstate, + "mailbox", CXL_MAILBOX_REGISTERS_LENGTH); + memory_region_init_io(&cxl_dstate->memory_device, obj, &mdev_ops, + cxl_dstate, "memory device caps", + CXL_MEMORY_DEVICE_REGISTERS_LENGTH); + + memory_region_add_subregion(&cxl_dstate->device_registers, 0, + &cxl_dstate->caps); + memory_region_add_subregion(&cxl_dstate->device_registers, + CXL_DEVICE_STATUS_REGISTERS_OFFSET, + &cxl_dstate->device); + memory_region_add_subregion(&cxl_dstate->device_registers, + CXL_MAILBOX_REGISTERS_OFFSET, + &cxl_dstate->mailbox); + memory_region_add_subregion(&cxl_dstate->device_registers, + CXL_MEMORY_DEVICE_REGISTERS_OFFSET, + &cxl_dstate->memory_device); +} + +static void device_reg_init_common(CXLDeviceState *cxl_dstate) { } + +static void mailbox_reg_init_common(CXLDeviceState *cxl_dstate) +{ + /* 2048 payload size, with no interrupt or background support */ + ARRAY_FIELD_DP32(cxl_dstate->mbox_reg_state32, CXL_DEV_MAILBOX_CAP, + PAYLOAD_SIZE, CXL_MAILBOX_PAYLOAD_SHIFT); + cxl_dstate->payload_size = CXL_MAILBOX_MAX_PAYLOAD_SIZE; +} + +static void memdev_reg_init_common(CXLDeviceState *cxl_dstate) { } + +void cxl_device_register_init_common(CXLDeviceState *cxl_dstate) +{ + uint64_t *cap_hdrs = cxl_dstate->caps_reg_state64; + const int cap_count = 3; + + /* CXL Device Capabilities Array Register */ + ARRAY_FIELD_DP64(cap_hdrs, CXL_DEV_CAP_ARRAY, CAP_ID, 0); + ARRAY_FIELD_DP64(cap_hdrs, CXL_DEV_CAP_ARRAY, CAP_VERSION, 1); + ARRAY_FIELD_DP64(cap_hdrs, CXL_DEV_CAP_ARRAY, CAP_COUNT, cap_count); + + cxl_device_cap_init(cxl_dstate, DEVICE_STATUS, 1); + device_reg_init_common(cxl_dstate); + + cxl_device_cap_init(cxl_dstate, MAILBOX, 2); + mailbox_reg_init_common(cxl_dstate); + + cxl_device_cap_init(cxl_dstate, MEMORY_DEVICE, 0x4000); + memdev_reg_init_common(cxl_dstate); + + assert(cxl_initialize_mailbox(cxl_dstate) == 0); +} diff --git a/hw/cxl/cxl-host-stubs.c b/hw/cxl/cxl-host-stubs.c new file mode 100644 index 00000000..cae4afcd --- /dev/null +++ b/hw/cxl/cxl-host-stubs.c @@ -0,0 +1,15 @@ +/* + * CXL host parameter parsing routine stubs + * + * Copyright (c) 2022 Huawei + */ +#include "qemu/osdep.h" +#include "qapi/error.h" +#include "hw/cxl/cxl.h" +#include "hw/cxl/cxl_host.h" + +void cxl_fmws_link_targets(CXLState *stat, Error **errp) {}; +void cxl_machine_init(Object *obj, CXLState *state) {}; +void cxl_hook_up_pxb_registers(PCIBus *bus, CXLState *state, Error **errp) {}; + +const MemoryRegionOps cfmws_ops; diff --git a/hw/cxl/cxl-host.c b/hw/cxl/cxl-host.c new file mode 100644 index 00000000..1adf6123 --- /dev/null +++ b/hw/cxl/cxl-host.c @@ -0,0 +1,340 @@ +/* + * CXL host parameter parsing routines + * + * Copyright (c) 2022 Huawei + * Modeled loosely on the NUMA options handling in hw/core/numa.c + */ + +#include "qemu/osdep.h" +#include "qemu/units.h" +#include "qemu/bitmap.h" +#include "qemu/error-report.h" +#include "qapi/error.h" +#include "sysemu/qtest.h" +#include "hw/boards.h" + +#include "qapi/qapi-visit-machine.h" +#include "hw/cxl/cxl.h" +#include "hw/cxl/cxl_host.h" +#include "hw/pci/pci_bus.h" +#include "hw/pci/pci_bridge.h" +#include "hw/pci/pci_host.h" +#include "hw/pci/pcie_port.h" +#include "hw/pci-bridge/pci_expander_bridge.h" + +static void cxl_fixed_memory_window_config(CXLState *cxl_state, + CXLFixedMemoryWindowOptions *object, + Error **errp) +{ + g_autofree CXLFixedWindow *fw = g_malloc0(sizeof(*fw)); + strList *target; + int i; + + for (target = object->targets; target; target = target->next) { + fw->num_targets++; + } + + fw->enc_int_ways = cxl_interleave_ways_enc(fw->num_targets, errp); + if (*errp) { + return; + } + + fw->targets = g_malloc0_n(fw->num_targets, sizeof(*fw->targets)); + for (i = 0, target = object->targets; target; i++, target = target->next) { + /* This link cannot be resolved yet, so stash the name for now */ + fw->targets[i] = g_strdup(target->value); + } + + if (object->size % (256 * MiB)) { + error_setg(errp, + "Size of a CXL fixed memory window must my a multiple of 256MiB"); + return; + } + fw->size = object->size; + + if (object->has_interleave_granularity) { + fw->enc_int_gran = + cxl_interleave_granularity_enc(object->interleave_granularity, + errp); + if (*errp) { + return; + } + } else { + /* Default to 256 byte interleave */ + fw->enc_int_gran = 0; + } + + cxl_state->fixed_windows = g_list_append(cxl_state->fixed_windows, + g_steal_pointer(&fw)); + + return; +} + +void cxl_fmws_link_targets(CXLState *cxl_state, Error **errp) +{ + if (cxl_state && cxl_state->fixed_windows) { + GList *it; + + for (it = cxl_state->fixed_windows; it; it = it->next) { + CXLFixedWindow *fw = it->data; + int i; + + for (i = 0; i < fw->num_targets; i++) { + Object *o; + bool ambig; + + o = object_resolve_path_type(fw->targets[i], + TYPE_PXB_CXL_DEVICE, + &ambig); + if (!o) { + error_setg(errp, "Could not resolve CXLFM target %s", + fw->targets[i]); + return; + } + fw->target_hbs[i] = PXB_CXL_DEV(o); + } + } + } +} + +/* TODO: support, multiple hdm decoders */ +static bool cxl_hdm_find_target(uint32_t *cache_mem, hwaddr addr, + uint8_t *target) +{ + uint32_t ctrl; + uint32_t ig_enc; + uint32_t iw_enc; + uint32_t target_idx; + + ctrl = cache_mem[R_CXL_HDM_DECODER0_CTRL]; + if (!FIELD_EX32(ctrl, CXL_HDM_DECODER0_CTRL, COMMITTED)) { + return false; + } + + ig_enc = FIELD_EX32(ctrl, CXL_HDM_DECODER0_CTRL, IG); + iw_enc = FIELD_EX32(ctrl, CXL_HDM_DECODER0_CTRL, IW); + target_idx = (addr / cxl_decode_ig(ig_enc)) % (1 << iw_enc); + + if (target_idx < 4) { + *target = extract32(cache_mem[R_CXL_HDM_DECODER0_TARGET_LIST_LO], + target_idx * 8, 8); + } else { + *target = extract32(cache_mem[R_CXL_HDM_DECODER0_TARGET_LIST_HI], + (target_idx - 4) * 8, 8); + } + + return true; +} + +static PCIDevice *cxl_cfmws_find_device(CXLFixedWindow *fw, hwaddr addr) +{ + CXLComponentState *hb_cstate, *usp_cstate; + PCIHostState *hb; + CXLUpstreamPort *usp; + int rb_index; + uint32_t *cache_mem; + uint8_t target; + bool target_found; + PCIDevice *rp, *d; + + /* Address is relative to memory region. Convert to HPA */ + addr += fw->base; + + rb_index = (addr / cxl_decode_ig(fw->enc_int_gran)) % fw->num_targets; + hb = PCI_HOST_BRIDGE(fw->target_hbs[rb_index]->cxl.cxl_host_bridge); + if (!hb || !hb->bus || !pci_bus_is_cxl(hb->bus)) { + return NULL; + } + + hb_cstate = cxl_get_hb_cstate(hb); + if (!hb_cstate) { + return NULL; + } + + cache_mem = hb_cstate->crb.cache_mem_registers; + + target_found = cxl_hdm_find_target(cache_mem, addr, &target); + if (!target_found) { + return NULL; + } + + rp = pcie_find_port_by_pn(hb->bus, target); + if (!rp) { + return NULL; + } + + d = pci_bridge_get_sec_bus(PCI_BRIDGE(rp))->devices[0]; + if (!d) { + return NULL; + } + + if (object_dynamic_cast(OBJECT(d), TYPE_CXL_TYPE3)) { + return d; + } + + /* + * Could also be a switch. Note only one level of switching currently + * supported. + */ + if (!object_dynamic_cast(OBJECT(d), TYPE_CXL_USP)) { + return NULL; + } + usp = CXL_USP(d); + + usp_cstate = cxl_usp_to_cstate(usp); + if (!usp_cstate) { + return NULL; + } + + cache_mem = usp_cstate->crb.cache_mem_registers; + + target_found = cxl_hdm_find_target(cache_mem, addr, &target); + if (!target_found) { + return NULL; + } + + d = pcie_find_port_by_pn(&PCI_BRIDGE(d)->sec_bus, target); + if (!d) { + return NULL; + } + + d = pci_bridge_get_sec_bus(PCI_BRIDGE(d))->devices[0]; + if (!d) { + return NULL; + } + + if (!object_dynamic_cast(OBJECT(d), TYPE_CXL_TYPE3)) { + return NULL; + } + + return d; +} + +static MemTxResult cxl_read_cfmws(void *opaque, hwaddr addr, uint64_t *data, + unsigned size, MemTxAttrs attrs) +{ + CXLFixedWindow *fw = opaque; + PCIDevice *d; + + d = cxl_cfmws_find_device(fw, addr); + if (d == NULL) { + *data = 0; + /* Reads to invalid address return poison */ + return MEMTX_ERROR; + } + + return cxl_type3_read(d, addr + fw->base, data, size, attrs); +} + +static MemTxResult cxl_write_cfmws(void *opaque, hwaddr addr, + uint64_t data, unsigned size, + MemTxAttrs attrs) +{ + CXLFixedWindow *fw = opaque; + PCIDevice *d; + + d = cxl_cfmws_find_device(fw, addr); + if (d == NULL) { + /* Writes to invalid address are silent */ + return MEMTX_OK; + } + + return cxl_type3_write(d, addr + fw->base, data, size, attrs); +} + +const MemoryRegionOps cfmws_ops = { + .read_with_attrs = cxl_read_cfmws, + .write_with_attrs = cxl_write_cfmws, + .endianness = DEVICE_LITTLE_ENDIAN, + .valid = { + .min_access_size = 1, + .max_access_size = 8, + .unaligned = true, + }, + .impl = { + .min_access_size = 1, + .max_access_size = 8, + .unaligned = true, + }, +}; + +static void machine_get_cxl(Object *obj, Visitor *v, const char *name, + void *opaque, Error **errp) +{ + CXLState *cxl_state = opaque; + bool value = cxl_state->is_enabled; + + visit_type_bool(v, name, &value, errp); +} + +static void machine_set_cxl(Object *obj, Visitor *v, const char *name, + void *opaque, Error **errp) +{ + CXLState *cxl_state = opaque; + bool value; + + if (!visit_type_bool(v, name, &value, errp)) { + return; + } + cxl_state->is_enabled = value; +} + +static void machine_get_cfmw(Object *obj, Visitor *v, const char *name, + void *opaque, Error **errp) +{ + CXLFixedMemoryWindowOptionsList **list = opaque; + + visit_type_CXLFixedMemoryWindowOptionsList(v, name, list, errp); +} + +static void machine_set_cfmw(Object *obj, Visitor *v, const char *name, + void *opaque, Error **errp) +{ + CXLState *state = opaque; + CXLFixedMemoryWindowOptionsList *cfmw_list = NULL; + CXLFixedMemoryWindowOptionsList *it; + + visit_type_CXLFixedMemoryWindowOptionsList(v, name, &cfmw_list, errp); + if (!cfmw_list) { + return; + } + + for (it = cfmw_list; it; it = it->next) { + cxl_fixed_memory_window_config(state, it->value, errp); + } + state->cfmw_list = cfmw_list; +} + +void cxl_machine_init(Object *obj, CXLState *state) +{ + object_property_add(obj, "cxl", "bool", machine_get_cxl, + machine_set_cxl, NULL, state); + object_property_set_description(obj, "cxl", + "Set on/off to enable/disable " + "CXL instantiation"); + + object_property_add(obj, "cxl-fmw", "CXLFixedMemoryWindow", + machine_get_cfmw, machine_set_cfmw, + NULL, state); + object_property_set_description(obj, "cxl-fmw", + "CXL Fixed Memory Windows (array)"); +} + +void cxl_hook_up_pxb_registers(PCIBus *bus, CXLState *state, Error **errp) +{ + /* Walk the pci busses looking for pxb busses to hook up */ + if (bus) { + QLIST_FOREACH(bus, &bus->child, sibling) { + if (!pci_bus_is_root(bus)) { + continue; + } + if (pci_bus_is_cxl(bus)) { + if (!state->is_enabled) { + error_setg(errp, "CXL host bridges present, but cxl=off"); + return; + } + pxb_cxl_hook_up_registers(state, bus, errp); + } + } + } +} diff --git a/hw/cxl/cxl-mailbox-utils.c b/hw/cxl/cxl-mailbox-utils.c new file mode 100644 index 00000000..bc1bb188 --- /dev/null +++ b/hw/cxl/cxl-mailbox-utils.c @@ -0,0 +1,478 @@ +/* + * CXL Utility library for mailbox interface + * + * Copyright(C) 2020 Intel Corporation. + * + * This work is licensed under the terms of the GNU GPL, version 2. See the + * COPYING file in the top-level directory. + */ + +#include "qemu/osdep.h" +#include "hw/cxl/cxl.h" +#include "hw/pci/pci.h" +#include "qemu/cutils.h" +#include "qemu/log.h" +#include "qemu/uuid.h" + +/* + * How to add a new command, example. The command set FOO, with cmd BAR. + * 1. Add the command set and cmd to the enum. + * FOO = 0x7f, + * #define BAR 0 + * 2. Implement the handler + * static ret_code cmd_foo_bar(struct cxl_cmd *cmd, + * CXLDeviceState *cxl_dstate, uint16_t *len) + * 3. Add the command to the cxl_cmd_set[][] + * [FOO][BAR] = { "FOO_BAR", cmd_foo_bar, x, y }, + * 4. Implement your handler + * define_mailbox_handler(FOO_BAR) { ... return CXL_MBOX_SUCCESS; } + * + * + * Writing the handler: + * The handler will provide the &struct cxl_cmd, the &CXLDeviceState, and the + * in/out length of the payload. The handler is responsible for consuming the + * payload from cmd->payload and operating upon it as necessary. It must then + * fill the output data into cmd->payload (overwriting what was there), + * setting the length, and returning a valid return code. + * + * XXX: The handler need not worry about endianess. The payload is read out of + * a register interface that already deals with it. + */ + +enum { + EVENTS = 0x01, + #define GET_RECORDS 0x0 + #define CLEAR_RECORDS 0x1 + #define GET_INTERRUPT_POLICY 0x2 + #define SET_INTERRUPT_POLICY 0x3 + FIRMWARE_UPDATE = 0x02, + #define GET_INFO 0x0 + TIMESTAMP = 0x03, + #define GET 0x0 + #define SET 0x1 + LOGS = 0x04, + #define GET_SUPPORTED 0x0 + #define GET_LOG 0x1 + IDENTIFY = 0x40, + #define MEMORY_DEVICE 0x0 + CCLS = 0x41, + #define GET_PARTITION_INFO 0x0 + #define GET_LSA 0x2 + #define SET_LSA 0x3 +}; + +/* 8.2.8.4.5.1 Command Return Codes */ +typedef enum { + CXL_MBOX_SUCCESS = 0x0, + CXL_MBOX_BG_STARTED = 0x1, + CXL_MBOX_INVALID_INPUT = 0x2, + CXL_MBOX_UNSUPPORTED = 0x3, + CXL_MBOX_INTERNAL_ERROR = 0x4, + CXL_MBOX_RETRY_REQUIRED = 0x5, + CXL_MBOX_BUSY = 0x6, + CXL_MBOX_MEDIA_DISABLED = 0x7, + CXL_MBOX_FW_XFER_IN_PROGRESS = 0x8, + CXL_MBOX_FW_XFER_OUT_OF_ORDER = 0x9, + CXL_MBOX_FW_AUTH_FAILED = 0xa, + CXL_MBOX_FW_INVALID_SLOT = 0xb, + CXL_MBOX_FW_ROLLEDBACK = 0xc, + CXL_MBOX_FW_REST_REQD = 0xd, + CXL_MBOX_INVALID_HANDLE = 0xe, + CXL_MBOX_INVALID_PA = 0xf, + CXL_MBOX_INJECT_POISON_LIMIT = 0x10, + CXL_MBOX_PERMANENT_MEDIA_FAILURE = 0x11, + CXL_MBOX_ABORTED = 0x12, + CXL_MBOX_INVALID_SECURITY_STATE = 0x13, + CXL_MBOX_INCORRECT_PASSPHRASE = 0x14, + CXL_MBOX_UNSUPPORTED_MAILBOX = 0x15, + CXL_MBOX_INVALID_PAYLOAD_LENGTH = 0x16, + CXL_MBOX_MAX = 0x17 +} ret_code; + +struct cxl_cmd; +typedef ret_code (*opcode_handler)(struct cxl_cmd *cmd, + CXLDeviceState *cxl_dstate, uint16_t *len); +struct cxl_cmd { + const char *name; + opcode_handler handler; + ssize_t in; + uint16_t effect; /* Reported in CEL */ + uint8_t *payload; +}; + +#define DEFINE_MAILBOX_HANDLER_ZEROED(name, size) \ + uint16_t __zero##name = size; \ + static ret_code cmd_##name(struct cxl_cmd *cmd, \ + CXLDeviceState *cxl_dstate, uint16_t *len) \ + { \ + *len = __zero##name; \ + memset(cmd->payload, 0, *len); \ + return CXL_MBOX_SUCCESS; \ + } +#define DEFINE_MAILBOX_HANDLER_NOP(name) \ + static ret_code cmd_##name(struct cxl_cmd *cmd, \ + CXLDeviceState *cxl_dstate, uint16_t *len) \ + { \ + return CXL_MBOX_SUCCESS; \ + } + +DEFINE_MAILBOX_HANDLER_ZEROED(events_get_records, 0x20); +DEFINE_MAILBOX_HANDLER_NOP(events_clear_records); +DEFINE_MAILBOX_HANDLER_ZEROED(events_get_interrupt_policy, 4); +DEFINE_MAILBOX_HANDLER_NOP(events_set_interrupt_policy); + +/* 8.2.9.2.1 */ +static ret_code cmd_firmware_update_get_info(struct cxl_cmd *cmd, + CXLDeviceState *cxl_dstate, + uint16_t *len) +{ + struct { + uint8_t slots_supported; + uint8_t slot_info; + uint8_t caps; + uint8_t rsvd[0xd]; + char fw_rev1[0x10]; + char fw_rev2[0x10]; + char fw_rev3[0x10]; + char fw_rev4[0x10]; + } QEMU_PACKED *fw_info; + QEMU_BUILD_BUG_ON(sizeof(*fw_info) != 0x50); + + if (cxl_dstate->pmem_size < (256 << 20)) { + return CXL_MBOX_INTERNAL_ERROR; + } + + fw_info = (void *)cmd->payload; + memset(fw_info, 0, sizeof(*fw_info)); + + fw_info->slots_supported = 2; + fw_info->slot_info = BIT(0) | BIT(3); + fw_info->caps = 0; + pstrcpy(fw_info->fw_rev1, sizeof(fw_info->fw_rev1), "BWFW VERSION 0"); + + *len = sizeof(*fw_info); + return CXL_MBOX_SUCCESS; +} + +/* 8.2.9.3.1 */ +static ret_code cmd_timestamp_get(struct cxl_cmd *cmd, + CXLDeviceState *cxl_dstate, + uint16_t *len) +{ + uint64_t time, delta; + uint64_t final_time = 0; + + if (cxl_dstate->timestamp.set) { + /* First find the delta from the last time the host set the time. */ + time = qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL); + delta = time - cxl_dstate->timestamp.last_set; + final_time = cxl_dstate->timestamp.host_set + delta; + } + + /* Then adjust the actual time */ + stq_le_p(cmd->payload, final_time); + *len = 8; + + return CXL_MBOX_SUCCESS; +} + +/* 8.2.9.3.2 */ +static ret_code cmd_timestamp_set(struct cxl_cmd *cmd, + CXLDeviceState *cxl_dstate, + uint16_t *len) +{ + cxl_dstate->timestamp.set = true; + cxl_dstate->timestamp.last_set = qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL); + + cxl_dstate->timestamp.host_set = le64_to_cpu(*(uint64_t *)cmd->payload); + + *len = 0; + return CXL_MBOX_SUCCESS; +} + +static QemuUUID cel_uuid; + +/* 8.2.9.4.1 */ +static ret_code cmd_logs_get_supported(struct cxl_cmd *cmd, + CXLDeviceState *cxl_dstate, + uint16_t *len) +{ + struct { + uint16_t entries; + uint8_t rsvd[6]; + struct { + QemuUUID uuid; + uint32_t size; + } log_entries[1]; + } QEMU_PACKED *supported_logs = (void *)cmd->payload; + QEMU_BUILD_BUG_ON(sizeof(*supported_logs) != 0x1c); + + supported_logs->entries = 1; + supported_logs->log_entries[0].uuid = cel_uuid; + supported_logs->log_entries[0].size = 4 * cxl_dstate->cel_size; + + *len = sizeof(*supported_logs); + return CXL_MBOX_SUCCESS; +} + +/* 8.2.9.4.2 */ +static ret_code cmd_logs_get_log(struct cxl_cmd *cmd, + CXLDeviceState *cxl_dstate, + uint16_t *len) +{ + struct { + QemuUUID uuid; + uint32_t offset; + uint32_t length; + } QEMU_PACKED QEMU_ALIGNED(16) *get_log = (void *)cmd->payload; + + /* + * 8.2.9.4.2 + * The device shall return Invalid Parameter if the Offset or Length + * fields attempt to access beyond the size of the log as reported by Get + * Supported Logs. + * + * XXX: Spec is wrong, "Invalid Parameter" isn't a thing. + * XXX: Spec doesn't address incorrect UUID incorrectness. + * + * The CEL buffer is large enough to fit all commands in the emulation, so + * the only possible failure would be if the mailbox itself isn't big + * enough. + */ + if (get_log->offset + get_log->length > cxl_dstate->payload_size) { + return CXL_MBOX_INVALID_INPUT; + } + + if (!qemu_uuid_is_equal(&get_log->uuid, &cel_uuid)) { + return CXL_MBOX_UNSUPPORTED; + } + + /* Store off everything to local variables so we can wipe out the payload */ + *len = get_log->length; + + memmove(cmd->payload, cxl_dstate->cel_log + get_log->offset, + get_log->length); + + return CXL_MBOX_SUCCESS; +} + +/* 8.2.9.5.1.1 */ +static ret_code cmd_identify_memory_device(struct cxl_cmd *cmd, + CXLDeviceState *cxl_dstate, + uint16_t *len) +{ + struct { + char fw_revision[0x10]; + uint64_t total_capacity; + uint64_t volatile_capacity; + uint64_t persistent_capacity; + uint64_t partition_align; + uint16_t info_event_log_size; + uint16_t warning_event_log_size; + uint16_t failure_event_log_size; + uint16_t fatal_event_log_size; + uint32_t lsa_size; + uint8_t poison_list_max_mer[3]; + uint16_t inject_poison_limit; + uint8_t poison_caps; + uint8_t qos_telemetry_caps; + } QEMU_PACKED *id; + QEMU_BUILD_BUG_ON(sizeof(*id) != 0x43); + + CXLType3Dev *ct3d = container_of(cxl_dstate, CXLType3Dev, cxl_dstate); + CXLType3Class *cvc = CXL_TYPE3_GET_CLASS(ct3d); + uint64_t size = cxl_dstate->pmem_size; + + if (!QEMU_IS_ALIGNED(size, 256 << 20)) { + return CXL_MBOX_INTERNAL_ERROR; + } + + id = (void *)cmd->payload; + memset(id, 0, sizeof(*id)); + + /* PMEM only */ + snprintf(id->fw_revision, 0x10, "BWFW VERSION %02d", 0); + + id->total_capacity = size / (256 << 20); + id->persistent_capacity = size / (256 << 20); + id->lsa_size = cvc->get_lsa_size(ct3d); + + *len = sizeof(*id); + return CXL_MBOX_SUCCESS; +} + +static ret_code cmd_ccls_get_partition_info(struct cxl_cmd *cmd, + CXLDeviceState *cxl_dstate, + uint16_t *len) +{ + struct { + uint64_t active_vmem; + uint64_t active_pmem; + uint64_t next_vmem; + uint64_t next_pmem; + } QEMU_PACKED *part_info = (void *)cmd->payload; + QEMU_BUILD_BUG_ON(sizeof(*part_info) != 0x20); + uint64_t size = cxl_dstate->pmem_size; + + if (!QEMU_IS_ALIGNED(size, 256 << 20)) { + return CXL_MBOX_INTERNAL_ERROR; + } + + /* PMEM only */ + part_info->active_vmem = 0; + part_info->next_vmem = 0; + part_info->active_pmem = size / (256 << 20); + part_info->next_pmem = 0; + + *len = sizeof(*part_info); + return CXL_MBOX_SUCCESS; +} + +static ret_code cmd_ccls_get_lsa(struct cxl_cmd *cmd, + CXLDeviceState *cxl_dstate, + uint16_t *len) +{ + struct { + uint32_t offset; + uint32_t length; + } QEMU_PACKED *get_lsa; + CXLType3Dev *ct3d = container_of(cxl_dstate, CXLType3Dev, cxl_dstate); + CXLType3Class *cvc = CXL_TYPE3_GET_CLASS(ct3d); + uint32_t offset, length; + + get_lsa = (void *)cmd->payload; + offset = get_lsa->offset; + length = get_lsa->length; + + if (offset + length > cvc->get_lsa_size(ct3d)) { + *len = 0; + return CXL_MBOX_INVALID_INPUT; + } + + *len = cvc->get_lsa(ct3d, get_lsa, length, offset); + return CXL_MBOX_SUCCESS; +} + +static ret_code cmd_ccls_set_lsa(struct cxl_cmd *cmd, + CXLDeviceState *cxl_dstate, + uint16_t *len) +{ + struct set_lsa_pl { + uint32_t offset; + uint32_t rsvd; + uint8_t data[]; + } QEMU_PACKED; + struct set_lsa_pl *set_lsa_payload = (void *)cmd->payload; + CXLType3Dev *ct3d = container_of(cxl_dstate, CXLType3Dev, cxl_dstate); + CXLType3Class *cvc = CXL_TYPE3_GET_CLASS(ct3d); + const size_t hdr_len = offsetof(struct set_lsa_pl, data); + uint16_t plen = *len; + + *len = 0; + if (!plen) { + return CXL_MBOX_SUCCESS; + } + + if (set_lsa_payload->offset + plen > cvc->get_lsa_size(ct3d) + hdr_len) { + return CXL_MBOX_INVALID_INPUT; + } + plen -= hdr_len; + + cvc->set_lsa(ct3d, set_lsa_payload->data, plen, set_lsa_payload->offset); + return CXL_MBOX_SUCCESS; +} + +#define IMMEDIATE_CONFIG_CHANGE (1 << 1) +#define IMMEDIATE_DATA_CHANGE (1 << 2) +#define IMMEDIATE_POLICY_CHANGE (1 << 3) +#define IMMEDIATE_LOG_CHANGE (1 << 4) + +static struct cxl_cmd cxl_cmd_set[256][256] = { + [EVENTS][GET_RECORDS] = { "EVENTS_GET_RECORDS", + cmd_events_get_records, 1, 0 }, + [EVENTS][CLEAR_RECORDS] = { "EVENTS_CLEAR_RECORDS", + cmd_events_clear_records, ~0, IMMEDIATE_LOG_CHANGE }, + [EVENTS][GET_INTERRUPT_POLICY] = { "EVENTS_GET_INTERRUPT_POLICY", + cmd_events_get_interrupt_policy, 0, 0 }, + [EVENTS][SET_INTERRUPT_POLICY] = { "EVENTS_SET_INTERRUPT_POLICY", + cmd_events_set_interrupt_policy, 4, IMMEDIATE_CONFIG_CHANGE }, + [FIRMWARE_UPDATE][GET_INFO] = { "FIRMWARE_UPDATE_GET_INFO", + cmd_firmware_update_get_info, 0, 0 }, + [TIMESTAMP][GET] = { "TIMESTAMP_GET", cmd_timestamp_get, 0, 0 }, + [TIMESTAMP][SET] = { "TIMESTAMP_SET", cmd_timestamp_set, 8, IMMEDIATE_POLICY_CHANGE }, + [LOGS][GET_SUPPORTED] = { "LOGS_GET_SUPPORTED", cmd_logs_get_supported, 0, 0 }, + [LOGS][GET_LOG] = { "LOGS_GET_LOG", cmd_logs_get_log, 0x18, 0 }, + [IDENTIFY][MEMORY_DEVICE] = { "IDENTIFY_MEMORY_DEVICE", + cmd_identify_memory_device, 0, 0 }, + [CCLS][GET_PARTITION_INFO] = { "CCLS_GET_PARTITION_INFO", + cmd_ccls_get_partition_info, 0, 0 }, + [CCLS][GET_LSA] = { "CCLS_GET_LSA", cmd_ccls_get_lsa, 8, 0 }, + [CCLS][SET_LSA] = { "CCLS_SET_LSA", cmd_ccls_set_lsa, + ~0, IMMEDIATE_CONFIG_CHANGE | IMMEDIATE_DATA_CHANGE }, +}; + +void cxl_process_mailbox(CXLDeviceState *cxl_dstate) +{ + uint16_t ret = CXL_MBOX_SUCCESS; + struct cxl_cmd *cxl_cmd; + uint64_t status_reg; + opcode_handler h; + uint64_t command_reg = cxl_dstate->mbox_reg_state64[R_CXL_DEV_MAILBOX_CMD]; + + uint8_t set = FIELD_EX64(command_reg, CXL_DEV_MAILBOX_CMD, COMMAND_SET); + uint8_t cmd = FIELD_EX64(command_reg, CXL_DEV_MAILBOX_CMD, COMMAND); + uint16_t len = FIELD_EX64(command_reg, CXL_DEV_MAILBOX_CMD, LENGTH); + cxl_cmd = &cxl_cmd_set[set][cmd]; + h = cxl_cmd->handler; + if (h) { + if (len == cxl_cmd->in || cxl_cmd->in == ~0) { + cxl_cmd->payload = cxl_dstate->mbox_reg_state + + A_CXL_DEV_CMD_PAYLOAD; + ret = (*h)(cxl_cmd, cxl_dstate, &len); + assert(len <= cxl_dstate->payload_size); + } else { + ret = CXL_MBOX_INVALID_PAYLOAD_LENGTH; + } + } else { + qemu_log_mask(LOG_UNIMP, "Command %04xh not implemented\n", + set << 8 | cmd); + ret = CXL_MBOX_UNSUPPORTED; + } + + /* Set the return code */ + status_reg = FIELD_DP64(0, CXL_DEV_MAILBOX_STS, ERRNO, ret); + + /* Set the return length */ + command_reg = FIELD_DP64(command_reg, CXL_DEV_MAILBOX_CMD, COMMAND_SET, 0); + command_reg = FIELD_DP64(command_reg, CXL_DEV_MAILBOX_CMD, COMMAND, 0); + command_reg = FIELD_DP64(command_reg, CXL_DEV_MAILBOX_CMD, LENGTH, len); + + cxl_dstate->mbox_reg_state64[R_CXL_DEV_MAILBOX_CMD] = command_reg; + cxl_dstate->mbox_reg_state64[R_CXL_DEV_MAILBOX_STS] = status_reg; + + /* Tell the host we're done */ + ARRAY_FIELD_DP32(cxl_dstate->mbox_reg_state32, CXL_DEV_MAILBOX_CTRL, + DOORBELL, 0); +} + +int cxl_initialize_mailbox(CXLDeviceState *cxl_dstate) +{ + /* CXL 2.0: Table 169 Get Supported Logs Log Entry */ + const char *cel_uuidstr = "0da9c0b5-bf41-4b78-8f79-96b1623b3f17"; + + for (int set = 0; set < 256; set++) { + for (int cmd = 0; cmd < 256; cmd++) { + if (cxl_cmd_set[set][cmd].handler) { + struct cxl_cmd *c = &cxl_cmd_set[set][cmd]; + struct cel_log *log = + &cxl_dstate->cel_log[cxl_dstate->cel_size]; + + log->opcode = (set << 8) | cmd; + log->effect = c->effect; + cxl_dstate->cel_size++; + } + } + } + + return qemu_uuid_parse(cel_uuidstr, &cel_uuid); +} diff --git a/hw/cxl/meson.build b/hw/cxl/meson.build new file mode 100644 index 00000000..cfa95ffd --- /dev/null +++ b/hw/cxl/meson.build @@ -0,0 +1,13 @@ +softmmu_ss.add(when: 'CONFIG_CXL', + if_true: files( + 'cxl-component-utils.c', + 'cxl-device-utils.c', + 'cxl-mailbox-utils.c', + 'cxl-host.c', + 'cxl-cdat.c', + ), + if_false: files( + 'cxl-host-stubs.c', + )) + +softmmu_ss.add(when: 'CONFIG_ALL', if_true: files('cxl-host-stubs.c')) |