new file mode 100644
@@ -0,0 +1,1231 @@
+// SPDX-License-Identifier: GPL-2.0-only
+// Copyright (C) 2023 Rivos Inc.
+
+#define pr_fmt(fmt) "qos: resctrl: " fmt
+
+#include <linux/slab.h>
+#include <linux/err.h>
+#include <linux/riscv_qos.h>
+#include <linux/resctrl.h>
+#include <linux/types.h>
+#include <asm/csr.h>
+#include <asm/qos.h>
+#include "internal.h"
+
+#define MAX_CONTROLLERS 6
+static struct cbqri_resctrl_res cbqri_resctrl_resources[RDT_NUM_RESOURCES];
+/*
+ * The size of the controllers array reflects that number of CBQRI controllers
+ * (3x capacity, 3x bandwidth) in the hypothetical SoC that Qemu implements for
+ * the CBQRI proof-of-concept.
+ *
+ * The proper solution will involve dynamically allocating cbqri_controller
+ * structures and adding them to a list of controllers. This avoids having to
+ * make assumptions about the system.
+ */
+static struct cbqri_controller controllers[MAX_CONTROLLERS];
+static int found_controllers;
+
+static bool exposed_alloc_capable;
+static bool exposed_mon_capable;
+/* CDP (code data prioritization) on x86 is AT (access type) on RISC-V */
+static bool exposed_cdp_l2_capable;
+static bool exposed_cdp_l3_capable;
+static bool is_cdp_l2_enabled;
+static bool is_cdp_l3_enabled;
+
+/* used by resctrl_arch_system_num_rmid_idx() */
+static u32 max_rmid;
+
+LIST_HEAD(cbqri_controllers);
+
+static int qos_wait_busy_flag(struct cbqri_controller *ctrl, int reg_offset);
+
+bool resctrl_arch_alloc_capable(void)
+{
+ return exposed_alloc_capable;
+}
+
+bool resctrl_arch_mon_capable(void)
+{
+ return exposed_mon_capable;
+}
+
+bool resctrl_arch_is_mbm_local_enabled(void)
+{
+ /*
+ * Seems to be the bandwidth between a L2 cache and L3 cache
+ * according to Intel resctrl implementation. There is no meaning
+ * in supporting such stat when we use resctrl MON_DATA feature
+ * to support CBQRI bandwidth monitoring.
+ */
+ return false;
+}
+
+bool resctrl_arch_is_mbm_total_enabled(void)
+{
+ return true;
+}
+
+bool resctrl_arch_get_cdp_enabled(enum resctrl_res_level rid)
+{
+ switch (rid) {
+ case RDT_RESOURCE_L2:
+ return is_cdp_l2_enabled;
+
+ case RDT_RESOURCE_L3:
+ return is_cdp_l3_enabled;
+
+ default:
+ return false;
+ }
+}
+
+int resctrl_arch_set_cdp_enabled(enum resctrl_res_level rid, bool enable)
+{
+ switch (rid) {
+ case RDT_RESOURCE_L2:
+ if (!exposed_cdp_l2_capable)
+ return -ENODEV;
+ is_cdp_l2_enabled = enable;
+ break;
+
+ case RDT_RESOURCE_L3:
+ if (!exposed_cdp_l3_capable)
+ return -ENODEV;
+ is_cdp_l3_enabled = enable;
+ break;
+
+ default:
+ return -ENODEV;
+ }
+
+ return 0;
+}
+
+bool resctrl_arch_is_llc_occupancy_enabled(void)
+{
+ /*
+ * There is no occupancy in CBQRI bandwidth controller (resctrl MON_DATA
+ * was the only available way to implement RISC-V memory monitoring).
+ * As we need occupancy for CBQRI capacity controller, this global
+ * setting is true, but the ll_occupancy files of CBQRI bandwidth
+ * controllers will return an error.
+ */
+ return true;
+}
+
+/*
+ * Note about terminology between x86 (Intel RDT/AMD QoS) and RISC-V:
+ * CLOSID on x86 is RCID on RISC-V
+ * RMID on x86 is MCID on RISC-V
+ */
+u32 resctrl_arch_get_num_closid(struct rdt_resource *res)
+{
+ struct cbqri_resctrl_res *hw_res;
+
+ hw_res = container_of(res, struct cbqri_resctrl_res, resctrl_res);
+
+ return hw_res->max_rcid;
+}
+
+u32 resctrl_arch_system_num_rmid_idx(void)
+{
+ return max_rmid;
+}
+
+u32 resctrl_arch_rmid_idx_encode(u32 closid, u32 rmid)
+{
+ return rmid;
+}
+
+void resctrl_arch_rmid_idx_decode(u32 idx, u32 *closid, u32 *rmid)
+{
+ *closid = ((u32)~0); /* refer to X86_RESCTRL_BAD_CLOSID */
+ *rmid = idx;
+}
+
+/* RISC-V resctrl interface does not maintain a default sqoscfg value for a given CPU */
+void resctrl_arch_set_cpu_default_closid_rmid(int cpu, u32 closid, u32 rmid) { }
+
+void resctrl_sched_in(void)
+{
+ qos_sched_in(current);
+}
+
+void resctrl_arch_sync_cpu_defaults(void *info)
+{
+ resctrl_sched_in();
+}
+
+void resctrl_arch_set_closid_rmid(struct task_struct *tsk, u32 closid, u32 rmid)
+{
+ u32 sqoscfg;
+
+ WARN_ON_ONCE((closid & SQOSCFG_RCID_MASK) != closid);
+ WARN_ON_ONCE((rmid & SQOSCFG_MCID_MASK) != rmid);
+
+ sqoscfg = rmid << SQOSCFG_MCID_SHIFT;
+ sqoscfg |= closid;
+ WRITE_ONCE(tsk->thread.sqoscfg, sqoscfg);
+}
+
+bool resctrl_arch_match_closid(struct task_struct *tsk, u32 closid)
+{
+ u32 sqoscfg;
+ bool match;
+
+ sqoscfg = READ_ONCE(tsk->thread.sqoscfg);
+ match = (sqoscfg & SQOSCFG_RCID_MASK) == closid;
+ return match;
+}
+
+bool resctrl_arch_match_rmid(struct task_struct *tsk, u32 closid, u32 rmid)
+{
+ u32 tsk_rmid;
+
+ tsk_rmid = READ_ONCE(tsk->thread.sqoscfg);
+ tsk_rmid >>= SQOSCFG_MCID_SHIFT;
+ tsk_rmid &= SQOSCFG_MCID_MASK;
+
+ return tsk_rmid == rmid;
+}
+
+struct rdt_resource *resctrl_arch_get_resource(enum resctrl_res_level l)
+{
+ if (l >= RDT_NUM_RESOURCES)
+ return NULL;
+
+ return &cbqri_resctrl_resources[l].resctrl_res;
+}
+
+struct rdt_domain *resctrl_arch_find_domain(struct rdt_resource *r, int id)
+{
+ struct rdt_domain *domain;
+ struct list_head *l;
+
+ list_for_each(l, &r->domains) {
+ domain = list_entry(l, struct rdt_domain, list);
+ if (id == domain->id)
+ return domain;
+ }
+
+ return NULL;
+}
+
+void resctrl_arch_reset_resources(void)
+{
+ /* not implemented for the RISC-V resctrl implementation */
+}
+
+int resctrl_arch_mon_ctx_alloc_no_wait(struct rdt_resource *r, int evtid)
+{
+ /* RISC-V can always read an rmid, nothing needs allocating */
+ return 0;
+}
+
+void resctrl_arch_mon_ctx_free(struct rdt_resource *r, int evtid, int ctx)
+{
+ /* not implemented for the RISC-V resctrl interface */
+}
+
+bool resctrl_arch_is_evt_configurable(enum resctrl_event_id evt)
+{
+ return false;
+}
+
+int resctrl_arch_rmid_read(struct rdt_resource *r, struct rdt_domain *d,
+ u32 closid, u32 rmid, enum resctrl_event_id eventid,
+ u64 *val, int arch_mon_ctx)
+{
+ /*
+ * The current Qemu implementation of CBQRI capacity and bandwidth
+ * controllers do not emulate the utilization of resources over
+ * time. Therefore, Qemu currently sets the invalid bit in
+ * cc_mon_ctr_val and bc_mon_ctr_val, and there is no meaningful
+ * value other than 0 to return for reading an RMID (e.g. MCID in
+ * CBQRI terminology)
+ */
+
+ return 0;
+}
+
+void resctrl_arch_reset_rmid(struct rdt_resource *r, struct rdt_domain *d,
+ u32 closid, u32 rmid, enum resctrl_event_id eventid)
+{
+ /* not implemented for the RISC-V resctrl interface */
+}
+
+void resctrl_arch_mon_event_config_read(void *info)
+{
+ /* not implemented for the RISC-V resctrl interface */
+}
+
+void resctrl_arch_mon_event_config_write(void *info)
+{
+ /* not implemented for the RISC-V resctrl interface */
+}
+
+void resctrl_arch_reset_rmid_all(struct rdt_resource *r, struct rdt_domain *d)
+{
+ /* not implemented for the RISC-V resctrl implementation */
+}
+
+/* Set capacity block mask (cc_block_mask) */
+static void cbqri_set_cbm(struct cbqri_controller *ctrl, u64 cbm)
+{
+ int reg_offset;
+ u64 reg;
+
+ reg_offset = CBQRI_CC_BLOCK_MASK_OFF;
+ reg = ioread64(ctrl->base + reg_offset);
+
+ reg = cbm;
+ iowrite64(reg, ctrl->base + reg_offset);
+
+ reg = ioread64(ctrl->base + reg_offset);
+}
+
+/* Set the Rbwb (reserved bandwidth blocks) field in bc_bw_alloc */
+static void cbqri_set_rbwb(struct cbqri_controller *ctrl, u64 rbwb)
+{
+ int reg_offset;
+ u64 reg;
+
+ reg_offset = CBQRI_BC_BW_ALLOC_OFF;
+ reg = ioread64(ctrl->base + reg_offset);
+ reg |= rbwb & CBQRI_CONTROL_REGISTERS_RBWB_MASK;
+ iowrite64(reg, ctrl->base + reg_offset);
+}
+
+/* Get the Rbwb (reserved bandwidth blocks) field in bc_bw_alloc */
+static u64 cbqri_get_rbwb(struct cbqri_controller *ctrl)
+{
+ int reg_offset;
+ u64 reg;
+
+ reg_offset = CBQRI_BC_BW_ALLOC_OFF;
+ reg = ioread64(ctrl->base + reg_offset);
+ reg &= CBQRI_CONTROL_REGISTERS_RBWB_MASK;
+
+ return reg;
+}
+
+static int qos_wait_busy_flag(struct cbqri_controller *ctrl, int reg_offset)
+{
+ unsigned long timeout = jiffies + (HZ / 10); /* Timeout after 100ms */
+ u64 reg;
+ int busy;
+
+ while (time_before(jiffies, timeout)) {
+ reg = ioread64(ctrl->base + reg_offset);
+ busy = (reg >> CBQRI_CONTROL_REGISTERS_BUSY_SHIFT) &
+ CBQRI_CONTROL_REGISTERS_BUSY_MASK;
+ if (!busy)
+ return 0;
+ }
+
+ pr_warn("%s(): busy timeout", __func__);
+ return -EIO;
+}
+
+/* Perform operation on capacity controller */
+static int do_capacity_alloc_op(struct cbqri_controller *ctrl, int operation,
+ enum resctrl_conf_type type, int rcid)
+{
+ int reg_offset = CBQRI_CC_ALLOC_CTL_OFF;
+ int status;
+ u64 reg;
+
+ reg = ioread64(ctrl->base + reg_offset);
+ reg &= ~(CBQRI_CONTROL_REGISTERS_OP_MASK << CBQRI_CONTROL_REGISTERS_OP_SHIFT);
+ reg |= (operation & CBQRI_CONTROL_REGISTERS_OP_MASK) <<
+ CBQRI_CONTROL_REGISTERS_OP_SHIFT;
+ reg &= ~(CBQRI_CONTROL_REGISTERS_RCID_MASK <<
+ CBQRI_CONTROL_REGISTERS_RCID_SHIFT);
+ reg |= (rcid & CBQRI_CONTROL_REGISTERS_RCID_MASK) <<
+ CBQRI_CONTROL_REGISTERS_RCID_SHIFT;
+
+ /* CBQRI capacity AT is only supported on caches for now */
+ if (ctrl->ctrl_info->type == CBQRI_CONTROLLER_TYPE_CAPACITY &&
+ ((ctrl->ctrl_info->cache.cache_level == 2 && is_cdp_l2_enabled) ||
+ (ctrl->ctrl_info->cache.cache_level == 3 && is_cdp_l3_enabled))) {
+ reg &= ~(CBQRI_CONTROL_REGISTERS_AT_MASK <<
+ CBQRI_CONTROL_REGISTERS_AT_SHIFT);
+ switch (type) {
+ case CDP_CODE:
+ reg |= (CBQRI_CONTROL_REGISTERS_AT_CODE &
+ CBQRI_CONTROL_REGISTERS_AT_MASK) <<
+ CBQRI_CONTROL_REGISTERS_AT_SHIFT;
+ break;
+ case CDP_DATA:
+ default:
+ reg |= (CBQRI_CONTROL_REGISTERS_AT_DATA &
+ CBQRI_CONTROL_REGISTERS_AT_MASK) <<
+ CBQRI_CONTROL_REGISTERS_AT_SHIFT;
+ break;
+ }
+ }
+
+ iowrite64(reg, ctrl->base + reg_offset);
+
+ if (qos_wait_busy_flag(ctrl, reg_offset) < 0) {
+ pr_err("%s(): BUSY timeout when executing the operation", __func__);
+ return -EIO;
+ }
+
+ reg = ioread64(ctrl->base + reg_offset);
+ status = (reg >> CBQRI_CONTROL_REGISTERS_STATUS_SHIFT) &
+ CBQRI_CONTROL_REGISTERS_STATUS_MASK;
+ if (status != 1) {
+ pr_err("%s(): operation %d failed: status=%d", __func__, operation, status);
+ return -EIO;
+ }
+
+ return 0;
+}
+
+/* Perform operation on bandwidth controller */
+static int do_bandwidth_alloc_op(struct cbqri_controller *ctrl, int operation, int rcid)
+{
+ int reg_offset = CBQRI_BC_ALLOC_CTL_OFF;
+ int status;
+ u64 reg;
+
+ reg = ioread64(ctrl->base + reg_offset);
+ reg &= ~(CBQRI_CONTROL_REGISTERS_OP_MASK << CBQRI_CONTROL_REGISTERS_OP_SHIFT);
+ reg |= (operation & CBQRI_CONTROL_REGISTERS_OP_MASK) <<
+ CBQRI_CONTROL_REGISTERS_OP_SHIFT;
+ reg &= ~(CBQRI_CONTROL_REGISTERS_RCID_MASK << CBQRI_CONTROL_REGISTERS_RCID_SHIFT);
+ reg |= (rcid & CBQRI_CONTROL_REGISTERS_RCID_MASK) <<
+ CBQRI_CONTROL_REGISTERS_RCID_SHIFT;
+ iowrite64(reg, ctrl->base + reg_offset);
+
+ if (qos_wait_busy_flag(ctrl, reg_offset) < 0) {
+ pr_err("%s(): BUSY timeout when executing the operation", __func__);
+ return -EIO;
+ }
+
+ reg = ioread64(ctrl->base + reg_offset);
+ status = (reg >> CBQRI_CONTROL_REGISTERS_STATUS_SHIFT) &
+ CBQRI_CONTROL_REGISTERS_STATUS_MASK;
+ if (status != 1) {
+ pr_err("%s(): operation %d failed with status = %d",
+ __func__, operation, status);
+ return -EIO;
+ }
+
+ return 0;
+}
+
+static int cbqri_apply_bw_config(struct cbqri_resctrl_dom *hw_dom, u32 closid,
+ enum resctrl_conf_type type, struct cbqri_config *cfg)
+{
+ struct cbqri_controller *ctrl = hw_dom->hw_ctrl;
+ int ret = 0;
+ u64 rbwb;
+
+ if (cfg->cbm != hw_dom->ctrl_val[closid]) {
+ /* Store the new rbwb in the ctrl_val array for this closid in this domain */
+ hw_dom->ctrl_val[closid] = cfg->rbwb;
+
+ /* set reserved bandwidth blocks */
+ cbqri_set_rbwb(ctrl, cfg->rbwb);
+ /* get reserved bandwidth blocks */
+ rbwb = cbqri_get_rbwb(ctrl);
+
+ /* Capacity config limit operation for RCID (closid) */
+ ret = do_capacity_alloc_op(ctrl, CBQRI_CC_ALLOC_CTL_OP_CONFIG_LIMIT, type, closid);
+ if (ret < 0) {
+ pr_err("%s(): operation failed: ret = %d", __func__, ret);
+ return ret;
+ }
+
+ /* Clear rbwb before read limit to verify op works*/
+ cbqri_set_rbwb(ctrl, 0);
+
+ /* Bandwidth allocation read limit operation for RCID (closid) to verify */
+ ret = do_capacity_alloc_op(ctrl, CBQRI_CC_ALLOC_CTL_OP_READ_LIMIT, type, closid);
+ if (ret < 0) {
+ pr_err("%s(): operation failed: ret = %d", __func__, ret);
+ return ret;
+ }
+
+ /* Read capacity block mask for RCID (closid) to verify */
+ rbwb = cbqri_get_rbwb(ctrl);
+ }
+
+ return ret;
+}
+
+static int cbqri_apply_cache_config(struct cbqri_resctrl_dom *hw_dom, u32 closid,
+ enum resctrl_conf_type type, struct cbqri_config *cfg)
+{
+ struct cbqri_controller *ctrl = hw_dom->hw_ctrl;
+ int reg_offset, err = 0;
+ u64 reg;
+
+ if (cfg->cbm != hw_dom->ctrl_val[closid]) {
+ /* store the new cbm in the ctrl_val array for this closid in this domain */
+ hw_dom->ctrl_val[closid] = cfg->cbm;
+
+ /* set capacity block mask (cc_block_mask) */
+ cbqri_set_cbm(ctrl, cfg->cbm);
+
+ /* Capacity config limit operation for RCID (closid) */
+ err = do_capacity_alloc_op(ctrl, CBQRI_CC_ALLOC_CTL_OP_CONFIG_LIMIT, type, closid);
+ if (err < 0) {
+ pr_err("%s(): operation failed: err = %d", __func__, err);
+ return err;
+ }
+
+ /* clear cc_block_mask before read limit to verify op works*/
+ cbqri_set_cbm(ctrl, 0);
+
+ /* Capacity read limit operation for RCID (closid) to verify */
+ err = do_capacity_alloc_op(ctrl, CBQRI_CC_ALLOC_CTL_OP_READ_LIMIT, type, closid);
+ if (err < 0) {
+ pr_err("%s(): operation failed: err = %d", __func__, err);
+ return err;
+ }
+
+ /* Read capacity block mask for RCID (closid) to verify */
+ reg_offset = CBQRI_CC_BLOCK_MASK_OFF;
+ reg = ioread64(ctrl->base + reg_offset);
+ }
+
+ return err;
+}
+
+int resctrl_arch_update_one(struct rdt_resource *r, struct rdt_domain *d,
+ u32 closid, enum resctrl_conf_type t, u32 cfg_val)
+{
+ struct cbqri_resctrl_res *res;
+ struct cbqri_resctrl_dom *dom;
+ struct cbqri_config cfg;
+ int err = 0;
+
+ res = container_of(r, struct cbqri_resctrl_res, resctrl_res);
+ dom = container_of(d, struct cbqri_resctrl_dom, resctrl_dom);
+
+ if (!r->alloc_capable)
+ return -EINVAL;
+
+ switch (r->rid) {
+ case RDT_RESOURCE_L2:
+ case RDT_RESOURCE_L3:
+ cfg.cbm = cfg_val;
+ err = cbqri_apply_cache_config(dom, closid, t, &cfg);
+ break;
+ case RDT_RESOURCE_MBA:
+ cfg.rbwb = cfg_val;
+ err = cbqri_apply_bw_config(dom, closid, t, &cfg);
+ break;
+ default:
+ return -EINVAL;
+ }
+
+ return err;
+}
+
+int resctrl_arch_update_domains(struct rdt_resource *r, u32 closid)
+{
+ struct resctrl_staged_config *cfg;
+ enum resctrl_conf_type t;
+ struct rdt_domain *d;
+ int err = 0;
+
+ list_for_each_entry(d, &r->domains, list) {
+ for (t = 0; t < CDP_NUM_TYPES; t++) {
+ cfg = &d->staged_config[t];
+ if (!cfg->have_new_ctrl)
+ continue;
+ err = resctrl_arch_update_one(r, d, closid, t, cfg->new_ctrl);
+ if (err) {
+ pr_debug("%s(): return err=%d", __func__, err);
+ return err;
+ }
+ }
+ }
+ return err;
+}
+
+u32 resctrl_arch_get_config(struct rdt_resource *r, struct rdt_domain *d,
+ u32 closid, enum resctrl_conf_type type)
+{
+ struct cbqri_resctrl_res *hw_res;
+ struct cbqri_resctrl_dom *hw_dom;
+ struct cbqri_controller *ctrl;
+ int reg_offset;
+ int err;
+ u64 reg;
+
+ hw_res = container_of(r, struct cbqri_resctrl_res, resctrl_res);
+ hw_dom = container_of(d, struct cbqri_resctrl_dom, resctrl_dom);
+
+ ctrl = hw_dom->hw_ctrl;
+
+ if (!r->alloc_capable)
+ return -EINVAL;
+
+ switch (r->rid) {
+ case RDT_RESOURCE_L2:
+ case RDT_RESOURCE_L3:
+
+ /* Clear cc_block_mask before read limit operation */
+ cbqri_set_cbm(ctrl, 0);
+
+ /* Capacity read limit operation for RCID (closid) */
+ err = do_capacity_alloc_op(ctrl, CBQRI_CC_ALLOC_CTL_OP_READ_LIMIT, type, closid);
+ if (err < 0) {
+ pr_err("%s(): operation failed: err = %d", __func__, err);
+ return -EIO;
+ }
+
+ /* Read capacity block mask for RCID (closid) */
+ reg_offset = CBQRI_CC_BLOCK_MASK_OFF;
+ reg = ioread64(ctrl->base + reg_offset);
+
+ /* Update the config value for the closid in this domain */
+ hw_dom->ctrl_val[closid] = reg;
+
+ return hw_dom->ctrl_val[closid];
+
+ case RDT_RESOURCE_MBA:
+
+ /* Capacity read limit operation for RCID (closid) */
+ err = do_bandwidth_alloc_op(ctrl, CBQRI_CC_ALLOC_CTL_OP_READ_LIMIT, closid);
+ if (err < 0) {
+ pr_err("%s(): operation failed: err = %d", __func__, err);
+ return -EIO;
+ }
+
+ hw_dom->ctrl_val[closid] = cbqri_get_rbwb(ctrl);
+
+ return hw_dom->ctrl_val[closid];
+
+ default:
+ return -EINVAL;
+ }
+}
+
+static int qos_discover_controller_feature(struct cbqri_controller *ctrl,
+ int reg_offset,
+ int operation,
+ int *status,
+ bool *access_type_supported)
+{
+ u64 reg, saved_reg;
+ int at;
+
+ /* Keep the initial register value to preserve the WPRI fields */
+ reg = ioread64(ctrl->base + reg_offset);
+ saved_reg = reg;
+
+ /* Execute the requested operation to find if the register is implemented */
+ reg &= ~(CBQRI_CONTROL_REGISTERS_OP_MASK << CBQRI_CONTROL_REGISTERS_OP_SHIFT);
+ reg |= (operation & CBQRI_CONTROL_REGISTERS_OP_MASK) << CBQRI_CONTROL_REGISTERS_OP_SHIFT;
+ iowrite64(reg, ctrl->base + reg_offset);
+ if (qos_wait_busy_flag(ctrl, reg_offset) < 0) {
+ pr_err("%s(): BUSY timeout when executing the operation", __func__);
+ return -EIO;
+ }
+
+ /* Get the operation status */
+ reg = ioread64(ctrl->base + reg_offset);
+ *status = (reg >> CBQRI_CONTROL_REGISTERS_STATUS_SHIFT) &
+ CBQRI_CONTROL_REGISTERS_STATUS_MASK;
+
+ /*
+ * Check for the AT support if the register is implemented
+ * (if not, the status value will remain 0)
+ */
+ if (*status != 0) {
+ /* Set the AT field to a valid value */
+ reg = saved_reg;
+ reg &= ~(CBQRI_CONTROL_REGISTERS_AT_MASK << CBQRI_CONTROL_REGISTERS_AT_SHIFT);
+ reg |= CBQRI_CONTROL_REGISTERS_AT_CODE << CBQRI_CONTROL_REGISTERS_AT_SHIFT;
+ iowrite64(reg, ctrl->base + reg_offset);
+ if (qos_wait_busy_flag(ctrl, reg_offset) < 0) {
+ pr_err("%s(): BUSY timeout when setting AT field", __func__);
+ return -EIO;
+ }
+
+ /*
+ * If the AT field value has been reset to zero,
+ * then the AT support is not present
+ */
+ reg = ioread64(ctrl->base + reg_offset);
+ at = (reg >> CBQRI_CONTROL_REGISTERS_AT_SHIFT) & CBQRI_CONTROL_REGISTERS_AT_MASK;
+ if (at == CBQRI_CONTROL_REGISTERS_AT_CODE)
+ *access_type_supported = true;
+ else
+ *access_type_supported = false;
+ }
+
+ /* Restore the original register value */
+ iowrite64(saved_reg, ctrl->base + reg_offset);
+ if (qos_wait_busy_flag(ctrl, reg_offset) < 0) {
+ pr_err("%s(): BUSY timeout when restoring the original register value", __func__);
+ return -EIO;
+ }
+
+ return 0;
+}
+
+/*
+ * Note: for the purposes of the CBQRI proof-of-concept, debug logging
+ * has been left in this function that discovers the properties of CBQRI
+ * capable controllers in the system. pr_debug calls would be removed
+ * before submitting non-RFC patches.
+ */
+static int qos_resctrl_discover_controller(struct cbqri_controller_info *ctrl_info,
+ struct cbqri_controller *ctrl)
+{
+ int err = 0, status;
+ u64 reg;
+
+ pr_debug("controller info: type=%d addr=0x%lx size=%lu max-rcid=%u max-mcid=%u",
+ ctrl_info->type, ctrl_info->addr, ctrl_info->size,
+ ctrl_info->rcid_count, ctrl_info->mcid_count);
+
+ /* max_rmid is used by resctrl_arch_system_num_rmid_idx() */
+ max_rmid = ctrl_info->mcid_count;
+
+ ctrl->ctrl_info = ctrl_info;
+
+ /* Try to access the memory-mapped CBQRI registers */
+ if (!request_mem_region(ctrl_info->addr, ctrl_info->size, "cbqri_controller")) {
+ pr_debug("%s(): return %d", __func__, err);
+ return err;
+ }
+ ctrl->base = ioremap(ctrl_info->addr, ctrl_info->size);
+ if (!ctrl->base) {
+ pr_debug("%s(): goto err_release_mem_region", __func__);
+ goto err_release_mem_region;
+ }
+
+ ctrl->alloc_capable = false;
+ ctrl->mon_capable = false;
+
+ /* Discover capacity allocation and monitoring features */
+ if (ctrl_info->type == CBQRI_CONTROLLER_TYPE_CAPACITY) {
+ pr_debug("probe capacity controller");
+
+ /* Make sure the register is implemented */
+ reg = ioread64(ctrl->base + CBQRI_CC_CAPABILITIES_OFF);
+ if (reg == 0) {
+ err = -ENODEV;
+ goto err_iounmap;
+ }
+
+ ctrl->ver_minor = reg & CBQRI_CC_CAPABILITIES_VER_MINOR_MASK;
+ ctrl->ver_major = reg & CBQRI_CC_CAPABILITIES_VER_MAJOR_MASK;
+
+ ctrl->cc.supports_alloc_op_flush_rcid = (reg >> CBQRI_CC_CAPABILITIES_FRCID_SHIFT)
+ & CBQRI_CC_CAPABILITIES_FRCID_MASK;
+
+ ctrl->cc.ncblks = (reg >> CBQRI_CC_CAPABILITIES_NCBLKS_SHIFT) &
+ CBQRI_CC_CAPABILITIES_NCBLKS_MASK;
+
+ /* Calculate size of capacity block in bytes */
+ ctrl->cc.blk_size = ctrl_info->cache.cache_size / ctrl->cc.ncblks;
+ ctrl->cc.cache_level = ctrl_info->cache.cache_level;
+
+ pr_debug("version=%d.%d ncblks=%d blk_size=%d cache_level=%d",
+ ctrl->ver_major, ctrl->ver_minor,
+ ctrl->cc.ncblks, ctrl->cc.blk_size, ctrl->cc.cache_level);
+
+ /* Discover monitoring features */
+ err = qos_discover_controller_feature(ctrl, CBQRI_CC_MON_CTL_OFF,
+ CBQRI_CC_MON_CTL_OP_READ_COUNTER,
+ &status, &ctrl->cc.supports_mon_at_code);
+ if (err) {
+ pr_err("%s() failed to discover cc_mon_ctl feature", __func__);
+ goto err_iounmap;
+ }
+
+ if (status == CBQRI_CC_MON_CTL_STATUS_SUCCESS) {
+ pr_debug("cc_mon_ctl is supported");
+ ctrl->cc.supports_mon_op_config_event = true;
+ ctrl->cc.supports_mon_op_read_counter = true;
+ ctrl->mon_capable = true;
+ } else {
+ pr_debug("cc_mon_ctl is NOT supported");
+ ctrl->cc.supports_mon_op_config_event = false;
+ ctrl->cc.supports_mon_op_read_counter = false;
+ ctrl->mon_capable = false;
+ }
+ /*
+ * AT data is "always" supported as it has the same value
+ * than when AT field is not supported.
+ */
+ ctrl->cc.supports_mon_at_data = true;
+ pr_debug("supports_mon_at_data: %d, supports_mon_at_code: %d",
+ ctrl->cc.supports_mon_at_data, ctrl->cc.supports_mon_at_code);
+
+ /* Discover allocation features */
+ err = qos_discover_controller_feature(ctrl, CBQRI_CC_ALLOC_CTL_OFF,
+ CBQRI_CC_ALLOC_CTL_OP_READ_LIMIT,
+ &status, &ctrl->cc.supports_alloc_at_code);
+ if (err) {
+ pr_err("%s() failed to discover cc_alloc_ctl feature", __func__);
+ goto err_iounmap;
+ }
+
+ if (status == CBQRI_CC_ALLOC_CTL_STATUS_SUCCESS) {
+ pr_debug("cc_alloc_ctl is supported");
+ ctrl->cc.supports_alloc_op_config_limit = true;
+ ctrl->cc.supports_alloc_op_read_limit = true;
+ ctrl->alloc_capable = true;
+ exposed_alloc_capable = true;
+ } else {
+ pr_debug("cc_alloc_ctl is NOT supported");
+ ctrl->cc.supports_alloc_op_config_limit = false;
+ ctrl->cc.supports_alloc_op_read_limit = false;
+ ctrl->alloc_capable = false;
+ }
+ /*
+ * AT data is "always" supported as it has the same value
+ * than when AT field is not supported
+ */
+ ctrl->cc.supports_alloc_at_data = true;
+ pr_debug("supports_alloc_at_data: %d, supports_alloc_at_code: %d",
+ ctrl->cc.supports_alloc_at_data,
+ ctrl->cc.supports_alloc_at_code);
+ } else if (ctrl_info->type == CBQRI_CONTROLLER_TYPE_BANDWIDTH) {
+ pr_debug("probe bandwidth controller");
+
+ /* Make sure the register is implemented */
+ reg = ioread64(ctrl->base + CBQRI_BC_CAPABILITIES_OFF);
+ if (reg == 0) {
+ err = -ENODEV;
+ goto err_iounmap;
+ }
+
+ ctrl->ver_minor = reg & CBQRI_BC_CAPABILITIES_VER_MINOR_MASK;
+ ctrl->ver_major = reg & CBQRI_BC_CAPABILITIES_VER_MAJOR_MASK;
+
+ ctrl->bc.nbwblks = (reg >> CBQRI_BC_CAPABILITIES_NBWBLKS_SHIFT) &
+ CBQRI_BC_CAPABILITIES_NBWBLKS_MASK;
+ ctrl->bc.mrbwb = (reg >> CBQRI_BC_CAPABILITIES_MRBWB_SHIFT) &
+ CBQRI_BC_CAPABILITIES_MRBWB_MASK;
+
+ pr_debug("version=%d.%d nbwblks=%d mrbwb=%d",
+ ctrl->ver_major, ctrl->ver_minor,
+ ctrl->bc.nbwblks, ctrl->bc.mrbwb);
+
+ // Discover monitoring features
+ err = qos_discover_controller_feature(ctrl, CBQRI_BC_MON_CTL_OFF,
+ CBQRI_BC_MON_CTL_OP_READ_COUNTER,
+ &status, &ctrl->bc.supports_mon_at_code);
+ if (err) {
+ pr_err("%s() failed to discover bc_mon_ctl feature", __func__);
+ goto err_iounmap;
+ }
+
+ if (status == CBQRI_BC_MON_CTL_STATUS_SUCCESS) {
+ pr_debug("bc_mon_ctl is supported");
+ ctrl->bc.supports_mon_op_config_event = true;
+ ctrl->bc.supports_mon_op_read_counter = true;
+ ctrl->mon_capable = true;
+ exposed_mon_capable = true;
+ } else {
+ pr_debug("bc_mon_ctl is NOT supported");
+ ctrl->bc.supports_mon_op_config_event = false;
+ ctrl->bc.supports_mon_op_read_counter = false;
+ ctrl->mon_capable = false;
+ }
+ // AT data is "always" supported as it has the same value
+ // than when AT field is not supported
+ ctrl->bc.supports_mon_at_data = true;
+ pr_debug("supports_mon_at_data: %d, supports_mon_at_code: %d",
+ ctrl->bc.supports_mon_at_data, ctrl->bc.supports_mon_at_code);
+
+ // Discover allocation features
+ err = qos_discover_controller_feature(ctrl, CBQRI_BC_ALLOC_CTL_OFF,
+ CBQRI_BC_ALLOC_CTL_OP_READ_LIMIT,
+ &status, &ctrl->bc.supports_alloc_at_code);
+ if (err) {
+ pr_err("%s() failed to discover bc_alloc_ctl feature", __func__);
+ goto err_iounmap;
+ }
+
+ if (status == CBQRI_BC_ALLOC_CTL_STATUS_SUCCESS) {
+ pr_debug("bc_alloc_ctl is supported");
+ ctrl->bc.supports_alloc_op_config_limit = true;
+ ctrl->bc.supports_alloc_op_read_limit = true;
+ ctrl->alloc_capable = true;
+ exposed_alloc_capable = true;
+ } else {
+ pr_debug("bc_alloc_ctl is NOT supported");
+ ctrl->bc.supports_alloc_op_config_limit = false;
+ ctrl->bc.supports_alloc_op_read_limit = false;
+ ctrl->alloc_capable = false;
+ }
+
+ /*
+ * AT data is "always" supported as it has the same value
+ * than when AT field is not supported
+ */
+ ctrl->bc.supports_alloc_at_data = true;
+ pr_debug("supports_alloc_at_data: %d, supports_alloc_at_code: %d",
+ ctrl->bc.supports_alloc_at_data, ctrl->bc.supports_alloc_at_code);
+ } else {
+ pr_err("controller type is UNKNOWN");
+ err = -ENODEV;
+ goto err_release_mem_region;
+ }
+
+ return 0;
+
+err_iounmap:
+ pr_err("%s(): err_iounmap", __func__);
+ iounmap(ctrl->base);
+
+err_release_mem_region:
+ pr_err("%s(): err_release_mem_region", __func__);
+ release_mem_region(ctrl_info->addr, ctrl_info->size);
+
+ return err;
+}
+
+static struct rdt_domain *qos_new_domain(struct cbqri_controller *ctrl)
+{
+ struct cbqri_resctrl_dom *hw_dom;
+ struct rdt_domain *domain;
+
+ hw_dom = kzalloc(sizeof(*hw_dom), GFP_KERNEL);
+ if (!hw_dom)
+ return NULL;
+
+ /* associate this cbqri_controller with the domain */
+ hw_dom->hw_ctrl = ctrl;
+
+ /* the rdt_domain struct from inside the cbqri_resctrl_dom struct */
+ domain = &hw_dom->resctrl_dom;
+
+ INIT_LIST_HEAD(&domain->list);
+
+ return domain;
+}
+
+static int qos_bw_blocks_to_percentage(struct cbqri_controller *ctrl, int blocks)
+{
+ int percentage;
+
+ if (ctrl->ctrl_info->type != CBQRI_CONTROLLER_TYPE_BANDWIDTH)
+ return -1;
+
+ blocks *= 100;
+ percentage = blocks / ctrl->bc.nbwblks;
+ if (blocks % ctrl->bc.nbwblks)
+ percentage++;
+
+ return percentage;
+}
+
+static int domain_setup_ctrlval(struct rdt_resource *r, struct rdt_domain *d)
+{
+ struct cbqri_resctrl_res *hw_res;
+ struct cbqri_resctrl_dom *hw_dom;
+ u64 *dc;
+ int i;
+
+ hw_res = container_of(r, struct cbqri_resctrl_res, resctrl_res);
+ hw_dom = container_of(d, struct cbqri_resctrl_dom, resctrl_dom);
+
+ dc = kmalloc_array(hw_res->max_rcid, sizeof(*hw_dom->ctrl_val),
+ GFP_KERNEL);
+ if (!dc)
+ return -ENOMEM;
+
+ hw_dom->ctrl_val = dc;
+
+ for (i = 0; i < hw_res->max_rcid; i++, dc++)
+ *dc = r->default_ctrl;
+
+ /*
+ * The Qemu implementation for the CBQRI proof-of-concept has
+ * known reset values for all registers. A proper solution would
+ * be to perform a CONFIG_LIMIT operation to set the default for
+ * each RCID
+ */
+ return 0;
+}
+
+static int qos_resctrl_add_controller_domain(struct cbqri_controller *ctrl, int *id)
+{
+ struct rdt_domain *domain, *mon_domain = NULL;
+ struct cbqri_resctrl_res *cbqri_res = NULL;
+ struct cbqri_resctrl_dom *hw_dom_to_free;
+ struct rdt_resource *res;
+ int internal_id = *id;
+ int err;
+
+ domain = qos_new_domain(ctrl);
+ if (!domain)
+ return -ENOSPC;
+
+ if (ctrl->ctrl_info->type == CBQRI_CONTROLLER_TYPE_CAPACITY) {
+ cpumask_copy(&domain->cpu_mask, &ctrl->ctrl_info->cache.cpu_mask);
+
+ if (ctrl->ctrl_info->cache.cache_level == 2) {
+ cbqri_res = &cbqri_resctrl_resources[RDT_RESOURCE_L2];
+ cbqri_res->max_rcid = ctrl->ctrl_info->rcid_count;
+ cbqri_res->max_mcid = ctrl->ctrl_info->mcid_count;
+ res = &cbqri_res->resctrl_res;
+ res->num_rmid = ctrl->ctrl_info->mcid_count;
+ res->fflags = RFTYPE_RES_CACHE;
+ res->rid = RDT_RESOURCE_L2;
+ res->name = "L2";
+ res->alloc_capable = ctrl->alloc_capable;
+ res->mon_capable = ctrl->mon_capable;
+ res->format_str = "%d=%0*x";
+ res->cache_level = 2;
+ res->data_width = 3;
+ res->cache.arch_has_sparse_bitmaps = false;
+ res->cache.arch_has_per_cpu_cfg = false;
+ res->cache.shareable_bits = 0x000;
+ res->cache.min_cbm_bits = 1;
+ res->cache.cbm_len = ctrl->cc.ncblks;
+ res->default_ctrl = BIT_MASK(ctrl->cc.ncblks) - 1;
+ } else if (ctrl->ctrl_info->cache.cache_level == 3) {
+ cbqri_res = &cbqri_resctrl_resources[RDT_RESOURCE_L3];
+ cbqri_res->max_rcid = ctrl->ctrl_info->rcid_count;
+ cbqri_res->max_mcid = ctrl->ctrl_info->mcid_count;
+ res = &cbqri_res->resctrl_res;
+ res->num_rmid = ctrl->ctrl_info->mcid_count;
+ res->fflags = RFTYPE_RES_CACHE;
+ res->rid = RDT_RESOURCE_L3;
+ res->name = "L3";
+ res->cache_level = 3;
+ res->alloc_capable = ctrl->alloc_capable;
+ res->mon_capable = ctrl->mon_capable;
+ res->format_str = "%d=%0*x";
+ res->data_width = 4;
+ res->cache.arch_has_sparse_bitmaps = false;
+ res->cache.arch_has_per_cpu_cfg = false;
+ res->cache.shareable_bits = 0x000;
+ res->cache.min_cbm_bits = 1;
+ res->cache.cbm_len = ctrl->cc.ncblks;
+ res->default_ctrl = BIT_MASK(ctrl->cc.ncblks) - 1;
+ } else {
+ pr_err("%s(): unknown cache level %d", __func__,
+ ctrl->ctrl_info->cache.cache_level);
+ err = -ENODEV;
+ goto err_free_domain;
+ }
+ } else if (ctrl->ctrl_info->type == CBQRI_CONTROLLER_TYPE_BANDWIDTH) {
+ if (ctrl->alloc_capable) {
+ cbqri_res = &cbqri_resctrl_resources[RDT_RESOURCE_MBA];
+ cbqri_res->max_rcid = ctrl->ctrl_info->rcid_count;
+ cbqri_res->max_mcid = ctrl->ctrl_info->mcid_count;
+ res = &cbqri_res->resctrl_res;
+ res->num_rmid = ctrl->ctrl_info->mcid_count;
+ res->fflags = RFTYPE_RES_MB;
+ res->rid = RDT_RESOURCE_MBA;
+ res->name = "MB";
+ res->alloc_capable = ctrl->alloc_capable;
+ /*
+ * MBA resource is only for allocation and
+ * monitoring can only be done with L3 resource
+ */
+ res->mon_capable = false;
+ res->format_str = "%d=%*u";
+ res->data_width = 4;
+ /*
+ * MBA schemata expects percentage so convert
+ * maximum reserved bw blocks to percentage
+ */
+ res->default_ctrl = qos_bw_blocks_to_percentage(ctrl, ctrl->bc.mrbwb);
+ /*
+ * Use values from 0 to MBA_MAX instead of power of two values,
+ * see Intel System Programming Guide manual section 18.19.7.2
+ */
+ res->membw.delay_linear = true;
+ res->membw.arch_needs_linear = true;
+ /* Disable non-linear values support */
+ res->membw.throttle_mode = THREAD_THROTTLE_UNDEFINED;
+ // The minimum percentage allowed by the CBQRI spec
+ res->membw.min_bw = 1;
+ res->membw.bw_gran = 1;
+ }
+
+ /*
+ * Monitoring is not possible with Intel MBA resource,
+ * so add a L3 domain if monitoring is supported by the CBQRI
+ * bandwidth controller.
+ */
+ if (ctrl->mon_capable) {
+ struct cbqri_resctrl_res *mon_cbqri_res;
+
+ mon_cbqri_res = &cbqri_resctrl_resources[RDT_RESOURCE_L3];
+ mon_domain = qos_new_domain(ctrl);
+ if (!domain)
+ goto err_free_mon_domain;
+
+ /*
+ * For CBQRI, any cpu (technically a hart in RISC-V terms)
+ * can access the memory-mapped registers of any CBQRI
+ * controller in the system. Thus it does not matter for
+ * RISC-V which cpu runs the resctrl code.
+ */
+ cpumask_setall(&domain->cpu_mask);
+
+ err = domain_setup_ctrlval(&mon_cbqri_res->resctrl_res, mon_domain);
+ if (err)
+ goto err_free_mon_domain;
+
+ mon_domain->id = internal_id;
+ internal_id++;
+ list_add_tail(&mon_domain->list, &mon_cbqri_res->resctrl_res.domains);
+ err = resctrl_online_domain(res, mon_domain);
+ if (err) {
+ pr_debug("%s(): BW monitoring domain online failed", __func__);
+ goto err_free_mon_domain;
+ }
+ }
+ } else {
+ pr_err("%s(): unknown resource %d", __func__, ctrl->ctrl_info->type);
+ err = -ENODEV;
+ goto err_free_domain;
+ }
+
+ domain->id = internal_id;
+ err = domain_setup_ctrlval(res, domain);
+ if (err)
+ goto err_free_mon_domain;
+
+ if (cbqri_res) {
+ list_add_tail(&domain->list, &cbqri_res->resctrl_res.domains);
+ *id = internal_id;
+ err = resctrl_online_domain(res, domain);
+ if (err) {
+ pr_debug("%s(): failed to online cbqri_res domain", __func__);
+ goto err_free_domain;
+ }
+ }
+
+ return 0;
+
+err_free_mon_domain:
+ if (!mon_domain) {
+ /*
+ * mon_domain is a struct rdt_domain which is a member of
+ * struct cbqri_resctrl_dom. That cbqri_resctrl_dom was
+ * allocated in qos_new_domain() and must be freed.
+ */
+ hw_dom_to_free = container_of(mon_domain, struct cbqri_resctrl_dom, resctrl_dom);
+ kfree(hw_dom_to_free);
+ }
+
+err_free_domain:
+ /* similar to err_free_mon_domain but the ptr is 'domain' instead */
+ hw_dom_to_free = container_of(domain, struct cbqri_resctrl_dom, resctrl_dom);
+ kfree(hw_dom_to_free);
+
+ return err;
+}
+
+static int qos_resctrl_setup_resources(void)
+{
+ struct rdt_domain *domain, *domain_temp;
+ struct cbqri_controller_info *ctrl_info;
+ struct cbqri_controller *ctrl;
+ struct cbqri_resctrl_res *res;
+ int err = 0, i, id = 0;
+
+ list_for_each_entry(ctrl_info, &cbqri_controllers, list) {
+ err = qos_resctrl_discover_controller(ctrl_info, &controllers[found_controllers]);
+ if (err) {
+ pr_err("%s(): qos_resctrl_discover_controller failed (%d)", __func__, err);
+ goto err_unmap_controllers;
+ }
+
+ found_controllers++;
+ if (found_controllers > MAX_CONTROLLERS) {
+ pr_warn("%s(): increase MAX_CONTROLLERS value", __func__);
+ break;
+ }
+ }
+
+ for (i = 0; i < RDT_NUM_RESOURCES; i++) {
+ res = &cbqri_resctrl_resources[i];
+ INIT_LIST_HEAD(&res->resctrl_res.domains);
+ INIT_LIST_HEAD(&res->resctrl_res.evt_list);
+ res->resctrl_res.rid = i;
+ }
+
+ for (i = 0; i < found_controllers; i++) {
+ ctrl = &controllers[i];
+ err = qos_resctrl_add_controller_domain(ctrl, &id);
+ if (err) {
+ pr_err("%s(): failed to add controller domain (%d)", __func__, err);
+ goto err_free_controllers_list;
+ }
+ id++;
+
+ /*
+ * CDP (code data prioritization) on x86 is similar to
+ * the AT (access type) field in CBQRI. CDP only supports
+ * caches so this must be a CBQRI capacity controller.
+ */
+ if (ctrl->ctrl_info->type == CBQRI_CONTROLLER_TYPE_CAPACITY &&
+ ctrl->cc.supports_alloc_at_code &&
+ ctrl->cc.supports_alloc_at_data) {
+ if (ctrl->ctrl_info->cache.cache_level == 2)
+ exposed_cdp_l2_capable = true;
+ else
+ exposed_cdp_l3_capable = true;
+ }
+ }
+
+ pr_debug("exposed_alloc_capable = %d", exposed_alloc_capable);
+ pr_debug("exposed_mon_capable = %d", exposed_mon_capable);
+ pr_debug("exposed_cdp_l2_capable = %d", exposed_cdp_l2_capable);
+ pr_debug("exposed_cdp_l3_capable = %d", exposed_cdp_l3_capable);
+ return 0;
+
+err_free_controllers_list:
+ for (i = 0; i < RDT_NUM_RESOURCES; i++) {
+ res = &cbqri_resctrl_resources[i];
+ list_for_each_entry_safe(domain, domain_temp, &res->resctrl_res.domains, list) {
+ kfree(domain);
+ }
+ }
+
+err_unmap_controllers:
+ for (i = 0; i < found_controllers; i++) {
+ iounmap(controllers[i].base);
+ release_mem_region(controllers[i].ctrl_info->addr, controllers[i].ctrl_info->size);
+ }
+
+ return err;
+}
+
+int qos_resctrl_setup(void)
+{
+ s32 err;
+
+ err = qos_resctrl_setup_resources();
+ if (err) {
+ pr_err("%s() failed with error %d\n", __func__, err);
+ return err;
+ }
+
+ err = resctrl_init();
+
+ return err;
+}
+
+int qos_resctrl_online_cpu(unsigned int cpu)
+{
+ return resctrl_online_cpu(cpu);
+}
+
+int qos_resctrl_offline_cpu(unsigned int cpu)
+{
+ resctrl_offline_cpu(cpu);
+ return 0;
+}