From 22098763a10d9c1340827fcf6edab66f153b27f0 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 13 Sep 2026 12:27:00 -0700 Subject: Merge tag 'trace-v7.3-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace Pull tracing fixes from Steven Rostedt: - Don't destroy user event fields when removal fails User event fields are destroyed before the event is removed from visibility. But that can fail leaving the still visible event with no fields. Move the destroying of the fields to after the event is successfully removed from visibility. - Initialize function graph state is fork before calling copy_exec_state() For non-CLONE_VM forks, copy_exec_state() allocates a new task_exec_state. If that allocation fails, ftrace_graph_exit_task() will free the tasks ret_stack pointer. Since that pointer is still using the parent's ret_stack, it mistakenly frees the parent's pointer too. Call ftrace_graph_init() on the task first which will NULL out the new tasks's ret_stack and if the copy fails, it will not free anything. - Remove FGRAPH_MAX_INDEX The macro FGRAPH_MAX_INDEX was added but never used. Remove it. - Save ent_size in function graph printing of nested functions The function graph tracer needs to look at the next event to see if the next event is the return of the current function entry. If it is, it prints a single line: ktime_get(); Otherwise it prints it like a nested function: tick_nohz_irq_exit() { ktime_get(); kcpustat_irq_exit(); } In order to look at the next event, it must save the current event so that it has the information to print from it. It saves the event in the iterator descriptor called "ent". What it doesn't save is the ent_size of the event which is now used to know if the function graph arguments are to be printed. The peek doesn't save the size so the size used happens to be that of the size of the last event that was seen. Save the entry event size in the iterator descriptor so that the correct size is used. - Fix several errors with freeing data in the histogram code The histogram code had a lot of leaked or or incorrect accounting when failures happen. Correct them. - Fix histogram regression of .percent and .graph modifiers Up until 6.3 histogram values could have "percent" or "graph" modifiers that changed how they were printed. But a change that added restricting histograms values from being strings, stack traces and other modifiers inadvertently prevented them from using the percent and graph modifiers, which were legal use cases for values. Put back the percent and graph modifiers. - Fix various typos in the comments - Set the trace_clock before initializing a histogram with clock argument The histogram API allows the user to specific which trace clock to use via a "clock=" string. The histogram is set up first before the clock is checked. If the passed in clock is not valid, it exits without fully fixing up the histogram leaving it on the list and a use-after-free can trigger. Update the clock argument first and if it fails then exit gracefully before the histogram trigger is placed on any lists. - Restore :mod: trailer after parsing in ftrace_set_clr_event The function ftrace_set_clr_event() modifies the parse string and needs to put it back to what was passed in. It searches for ":mod:" via a strsep() but fails to put back the first ':' in the string. Add back the ':' in the passed in string. - Take trace_array reference when opening a tracer options file The options files are dynamically created and some tracers add their own options. When a tracer adds their own list of options, the trace_array holding them has an array to hold the list of options for each tracer. This array increases in size via a krealloc(), and the new entry gets a newly allocated array to hold the options of the new tracer being added. The element in each entry of the tracer's option array holds a pointer back to the trace_array, a pointer to the tracer it is associated to, a pointer to the flags of the option. The issue is that these arrays are freed when the trace_array is freed when its instance it represents is removed from the instances directory. There's a race that an open of one of these options files can happen when the instance is being removed. Add a new helper function to be called by the open function of the options file to iterate all existing trace_arrays under a lock and find the one that has the given option element in one of it's tracer arrays. If found, then update the associated trace_array's reference counter to keep it from being freed. If not found, have the open call return -ENODEV. - Disable interrupts when acquiring the lock in rb_wake_up_waiters() The function rb_wake_up_waiters() assumes it will be called in interrupt context and does not disable irqs when taking cpu_buffer->reader_lock, which can be called in hard interrupt context. The issue is in PREEMPT_RT, this function is called in thread context leaving this lock open to a deadlock. Take the lock with interrupts disabled. - Use rcu_assign_pointer() for tmp_ops filter hash The tmp_ops used in update_ftrace_direct_mod() assigns its filter_hash field directly, but that field is annotated as __rcu and sparse complains. Assign it with rcu_assign_pointer() - Fix use-after-free in enable_trigger_private_data_free() The trace_event_call is accessed through the event_trigger_data's trace_event_file pointer to put the trace_event_call on freeing. The issue is that the trace_event_file data may have been freed already causing a use-after-free. Add a field to the event_trigger_data that points directly to the trace_event_call so that it can decrement its reference directly without needing to go through the trace_event_file. - Fix accounting of buffer data remote headers trace_buffer_desc_size() and trace_remote_alloc_buffer() undercount the number of pages is needed for the asked for size as it doesn't take into account the meta data on each page. Add a helper function to do the calculation properly and use that in these functions. - Catch nr_page_va overflow in ring_buffer_desc sizing The number of pages per remote ring buffer is capped by ring_buffer_desc::nr_page_va (32 bits). A buffer_size large enough to overflow that field would silently allocate a descriptor smaller than what was asked for. - Do not resize the subbuf order if any per_cpu buffer is disabled The mmapping of ring buffers disables resizing the subbuffers, but it is done per-cpu whereas the subbuf size change is done for all the per_cpu buffers under the buffer->mutex. It could change the size of some while the mapping is happening on others. Have the resize of the subbuf order check all the per_cpu buffers under the lock to see if any of them is disabled before starting and causing an inconsistency between buffers that are being mapped. * tag 'trace-v7.3-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace: (25 commits) ring-buffer: Check resize_disabled before publishing the new subbuf order tracing/remotes: Catch nr_page_va overflow in ring_buffer_desc sizing tracing/remotes: Account for ring buffer page header in size calculation tracing: Don't dereference trace_event_file in deferred trigger free ftrace: Use rcu_assign_pointer() for tmp_ops filter hash ring-buffer: Acquire the lock with irqsave in rb_wake_up_waiters() tracing: Take trace_array reference when opening a tracer options file tracing: Fix ring_buffer_read_page_size() kernel-doc tracing: Restore :mod: trailer after parsing in ftrace_set_clr_event() tracing: Fix memory corruption from a "STACKTRACE" histogram key tracing: Fix memory corruption from the histogram stacktrace modifier tracing: Undo the registration when enabling the histogram trigger fails tracing: Take the reference before publishing the named histogram trigger tracing: Set the trace clock before registering the histogram trigger tracing: Fix typo "preceeded" in comment tracing: Fix typo "availabe" in comment tracing: Let histogram values keep the percent and graph modifiers tracing: Keep the entry count when the histogram stats allocation fails tracing: Free histogram the field rejected for a bad modifier tracing: Free histogram the var ref when its initialization fails ... --- drivers/cdx/controller/mcdi.c | 870 ++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 870 insertions(+) create mode 100644 drivers/cdx/controller/mcdi.c (limited to 'drivers/cdx/controller/mcdi.c') diff --git a/drivers/cdx/controller/mcdi.c b/drivers/cdx/controller/mcdi.c new file mode 100644 index 000000000..34a07d6f4 --- /dev/null +++ b/drivers/cdx/controller/mcdi.c @@ -0,0 +1,870 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Management-Controller-to-Driver Interface + * + * Copyright 2008-2013 Solarflare Communications Inc. + * Copyright (C) 2022-2023, Advanced Micro Devices, Inc. + */ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include "mcdid.h" + +static void cdx_mcdi_cancel_cmd(struct cdx_mcdi *cdx, struct cdx_mcdi_cmd *cmd); +static void cdx_mcdi_wait_for_cleanup(struct cdx_mcdi *cdx); +static int cdx_mcdi_rpc_async_internal(struct cdx_mcdi *cdx, + struct cdx_mcdi_cmd *cmd, + unsigned int *handle); +static void cdx_mcdi_start_or_queue(struct cdx_mcdi_iface *mcdi, + bool allow_retry); +static void cdx_mcdi_cmd_start_or_queue(struct cdx_mcdi_iface *mcdi, + struct cdx_mcdi_cmd *cmd); +static bool cdx_mcdi_complete_cmd(struct cdx_mcdi_iface *mcdi, + struct cdx_mcdi_cmd *cmd, + struct cdx_dword *outbuf, + int len, + struct list_head *cleanup_list); +static void cdx_mcdi_timeout_cmd(struct cdx_mcdi_iface *mcdi, + struct cdx_mcdi_cmd *cmd, + struct list_head *cleanup_list); +static void cdx_mcdi_cmd_work(struct work_struct *context); +static void cdx_mcdi_mode_fail(struct cdx_mcdi *cdx, struct list_head *cleanup_list); +static void _cdx_mcdi_display_error(struct cdx_mcdi *cdx, unsigned int cmd, + size_t inlen, int raw, int arg, int err_no); + +static bool cdx_cmd_cancelled(struct cdx_mcdi_cmd *cmd) +{ + return cmd->state == MCDI_STATE_RUNNING_CANCELLED; +} + +static void cdx_mcdi_cmd_release(struct kref *ref) +{ + kfree(container_of(ref, struct cdx_mcdi_cmd, ref)); +} + +static unsigned int cdx_mcdi_cmd_handle(struct cdx_mcdi_cmd *cmd) +{ + return cmd->handle; +} + +static void _cdx_mcdi_remove_cmd(struct cdx_mcdi_iface *mcdi, + struct cdx_mcdi_cmd *cmd, + struct list_head *cleanup_list) +{ + /* if cancelled, the completers have already been called */ + if (cdx_cmd_cancelled(cmd)) + return; + + if (cmd->completer) { + list_add_tail(&cmd->cleanup_list, cleanup_list); + ++mcdi->outstanding_cleanups; + kref_get(&cmd->ref); + } +} + +static void cdx_mcdi_remove_cmd(struct cdx_mcdi_iface *mcdi, + struct cdx_mcdi_cmd *cmd, + struct list_head *cleanup_list) +{ + list_del(&cmd->list); + _cdx_mcdi_remove_cmd(mcdi, cmd, cleanup_list); + cmd->state = MCDI_STATE_FINISHED; + kref_put(&cmd->ref, cdx_mcdi_cmd_release); + if (list_empty(&mcdi->cmd_list)) + wake_up(&mcdi->cmd_complete_wq); +} + +static unsigned long cdx_mcdi_rpc_timeout(struct cdx_mcdi *cdx, unsigned int cmd) +{ + if (!cdx->mcdi_ops->mcdi_rpc_timeout) + return MCDI_RPC_TIMEOUT; + else + return cdx->mcdi_ops->mcdi_rpc_timeout(cdx, cmd); +} + +/** + * cdx_mcdi_init - Initialize MCDI (Management Controller Driver Interface) state + * @cdx: Handle to the CDX MCDI structure + * + * This function allocates and initializes internal MCDI structures and resources + * for the CDX device, including the workqueue, locking primitives, and command + * tracking mechanisms. It sets the initial operating mode and prepares the device + * for MCDI operations. + * + * Return: + * * 0 - on success + * * -ENOMEM - if memory allocation or workqueue creation fails + */ +int cdx_mcdi_init(struct cdx_mcdi *cdx) +{ + struct cdx_mcdi_iface *mcdi; + int rc = -ENOMEM; + + cdx->mcdi = kzalloc_obj(*cdx->mcdi); + if (!cdx->mcdi) + goto fail; + + mcdi = cdx_mcdi_if(cdx); + mcdi->cdx = cdx; + + mcdi->workqueue = alloc_ordered_workqueue("mcdi_wq", 0); + if (!mcdi->workqueue) + goto fail2; + mutex_init(&mcdi->iface_lock); + mcdi->mode = MCDI_MODE_EVENTS; + INIT_LIST_HEAD(&mcdi->cmd_list); + init_waitqueue_head(&mcdi->cmd_complete_wq); + + mcdi->new_epoch = true; + + return 0; +fail2: + kfree(cdx->mcdi); + cdx->mcdi = NULL; +fail: + return rc; +} +EXPORT_SYMBOL_GPL(cdx_mcdi_init); + +/** + * cdx_mcdi_finish - Cleanup MCDI (Management Controller Driver Interface) state + * @cdx: Handle to the CDX MCDI structure + * + * This function is responsible for cleaning up the MCDI (Management Controller Driver Interface) + * resources associated with a cdx_mcdi structure. Also destroys the mcdi workqueue. + * + */ +void cdx_mcdi_finish(struct cdx_mcdi *cdx) +{ + struct cdx_mcdi_iface *mcdi; + + mcdi = cdx_mcdi_if(cdx); + if (!mcdi) + return; + + cdx_mcdi_wait_for_cleanup(cdx); + + destroy_workqueue(mcdi->workqueue); + kfree(cdx->mcdi); + cdx->mcdi = NULL; +} +EXPORT_SYMBOL_GPL(cdx_mcdi_finish); + +static bool cdx_mcdi_flushed(struct cdx_mcdi_iface *mcdi, bool ignore_cleanups) +{ + bool flushed; + + mutex_lock(&mcdi->iface_lock); + flushed = list_empty(&mcdi->cmd_list) && + (ignore_cleanups || !mcdi->outstanding_cleanups); + mutex_unlock(&mcdi->iface_lock); + return flushed; +} + +/* Wait for outstanding MCDI commands to complete. */ +static void cdx_mcdi_wait_for_cleanup(struct cdx_mcdi *cdx) +{ + struct cdx_mcdi_iface *mcdi = cdx_mcdi_if(cdx); + + if (!mcdi) + return; + + wait_event(mcdi->cmd_complete_wq, + cdx_mcdi_flushed(mcdi, false)); +} + +int cdx_mcdi_wait_for_quiescence(struct cdx_mcdi *cdx, + unsigned int timeout_jiffies) +{ + struct cdx_mcdi_iface *mcdi = cdx_mcdi_if(cdx); + DEFINE_WAIT_FUNC(wait, woken_wake_function); + int rc = 0; + + if (!mcdi) + return -EINVAL; + + flush_workqueue(mcdi->workqueue); + + add_wait_queue(&mcdi->cmd_complete_wq, &wait); + + while (!cdx_mcdi_flushed(mcdi, true)) { + rc = wait_woken(&wait, TASK_IDLE, timeout_jiffies); + if (rc) + continue; + break; + } + + remove_wait_queue(&mcdi->cmd_complete_wq, &wait); + + if (rc > 0) + rc = 0; + else if (rc == 0) + rc = -ETIMEDOUT; + + return rc; +} + +static u8 cdx_mcdi_payload_csum(const struct cdx_dword *hdr, size_t hdr_len, + const struct cdx_dword *sdu, size_t sdu_len) +{ + u8 *p = (u8 *)hdr; + u8 csum = 0; + int i; + + for (i = 0; i < hdr_len; i++) + csum += p[i]; + + p = (u8 *)sdu; + for (i = 0; i < sdu_len; i++) + csum += p[i]; + + return ~csum & 0xff; +} + +static void cdx_mcdi_send_request(struct cdx_mcdi *cdx, + struct cdx_mcdi_cmd *cmd) +{ + struct cdx_mcdi_iface *mcdi = cdx_mcdi_if(cdx); + const struct cdx_dword *inbuf = cmd->inbuf; + size_t inlen = cmd->inlen; + struct cdx_dword hdr[2]; + size_t hdr_len; + bool not_epoch; + u32 xflags; + + if (!mcdi) + return; + + mcdi->prev_seq = cmd->seq; + mcdi->seq_held_by[cmd->seq] = cmd; + mcdi->db_held_by = cmd; + cmd->started = jiffies; + + not_epoch = !mcdi->new_epoch; + xflags = 0; + + /* MCDI v2 */ + WARN_ON(inlen > MCDI_CTL_SDU_LEN_MAX_V2); + CDX_POPULATE_DWORD_7(hdr[0], + MCDI_HEADER_RESPONSE, 0, + MCDI_HEADER_RESYNC, 1, + MCDI_HEADER_CODE, MC_CMD_V2_EXTN, + MCDI_HEADER_DATALEN, 0, + MCDI_HEADER_SEQ, cmd->seq, + MCDI_HEADER_XFLAGS, xflags, + MCDI_HEADER_NOT_EPOCH, not_epoch); + CDX_POPULATE_DWORD_3(hdr[1], + MC_CMD_V2_EXTN_IN_EXTENDED_CMD, cmd->cmd, + MC_CMD_V2_EXTN_IN_ACTUAL_LEN, inlen, + MC_CMD_V2_EXTN_IN_MESSAGE_TYPE, + MC_CMD_V2_EXTN_IN_MCDI_MESSAGE_TYPE_PLATFORM); + hdr_len = 8; + + hdr[0].cdx_u32 |= (__force __le32)(cdx_mcdi_payload_csum(hdr, hdr_len, inbuf, inlen) << + MCDI_HEADER_XFLAGS_LBN); + + print_hex_dump_debug("MCDI REQ HEADER: ", DUMP_PREFIX_NONE, 32, 4, hdr, hdr_len, false); + print_hex_dump_debug("MCDI REQ PAYLOAD: ", DUMP_PREFIX_NONE, 32, 4, inbuf, inlen, false); + + cdx->mcdi_ops->mcdi_request(cdx, hdr, hdr_len, inbuf, inlen); + + mcdi->new_epoch = false; +} + +static int cdx_mcdi_errno(struct cdx_mcdi *cdx, unsigned int mcdi_err) +{ + switch (mcdi_err) { + case 0: + case MC_CMD_ERR_QUEUE_FULL: + return mcdi_err; + case MC_CMD_ERR_EPERM: + return -EPERM; + case MC_CMD_ERR_ENOENT: + return -ENOENT; + case MC_CMD_ERR_EINTR: + return -EINTR; + case MC_CMD_ERR_EAGAIN: + return -EAGAIN; + case MC_CMD_ERR_EACCES: + return -EACCES; + case MC_CMD_ERR_EBUSY: + return -EBUSY; + case MC_CMD_ERR_EINVAL: + return -EINVAL; + case MC_CMD_ERR_ERANGE: + return -ERANGE; + case MC_CMD_ERR_EDEADLK: + return -EDEADLK; + case MC_CMD_ERR_ENOSYS: + return -EOPNOTSUPP; + case MC_CMD_ERR_ETIME: + return -ETIME; + case MC_CMD_ERR_EALREADY: + return -EALREADY; + case MC_CMD_ERR_ENOSPC: + return -ENOSPC; + case MC_CMD_ERR_ENOMEM: + return -ENOMEM; + case MC_CMD_ERR_ENOTSUP: + return -EOPNOTSUPP; + case MC_CMD_ERR_ALLOC_FAIL: + return -ENOBUFS; + case MC_CMD_ERR_MAC_EXIST: + return -EADDRINUSE; + case MC_CMD_ERR_NO_EVB_PORT: + return -EAGAIN; + default: + return -EPROTO; + } +} + +static void cdx_mcdi_process_cleanup_list(struct cdx_mcdi *cdx, + struct list_head *cleanup_list) +{ + struct cdx_mcdi_iface *mcdi = cdx_mcdi_if(cdx); + unsigned int cleanups = 0; + + if (!mcdi) + return; + + while (!list_empty(cleanup_list)) { + struct cdx_mcdi_cmd *cmd = + list_first_entry(cleanup_list, + struct cdx_mcdi_cmd, cleanup_list); + cmd->completer(cdx, cmd->cookie, cmd->rc, + cmd->outbuf, cmd->outlen); + list_del(&cmd->cleanup_list); + kref_put(&cmd->ref, cdx_mcdi_cmd_release); + ++cleanups; + } + + if (cleanups) { + bool all_done; + + mutex_lock(&mcdi->iface_lock); + CDX_WARN_ON_PARANOID(cleanups > mcdi->outstanding_cleanups); + all_done = (mcdi->outstanding_cleanups -= cleanups) == 0; + mutex_unlock(&mcdi->iface_lock); + if (all_done) + wake_up(&mcdi->cmd_complete_wq); + } +} + +static void _cdx_mcdi_cancel_cmd(struct cdx_mcdi_iface *mcdi, + unsigned int handle, + struct list_head *cleanup_list) +{ + struct cdx_mcdi_cmd *cmd; + + list_for_each_entry(cmd, &mcdi->cmd_list, list) + if (cdx_mcdi_cmd_handle(cmd) == handle) { + switch (cmd->state) { + case MCDI_STATE_QUEUED: + case MCDI_STATE_RETRY: + pr_debug("command %#x inlen %zu cancelled in queue\n", + cmd->cmd, cmd->inlen); + /* if not yet running, properly cancel it */ + cmd->rc = -EPIPE; + cdx_mcdi_remove_cmd(mcdi, cmd, cleanup_list); + break; + case MCDI_STATE_RUNNING: + case MCDI_STATE_RUNNING_CANCELLED: + case MCDI_STATE_FINISHED: + default: + /* invalid state? */ + WARN_ON(1); + } + break; + } +} + +static void cdx_mcdi_cancel_cmd(struct cdx_mcdi *cdx, struct cdx_mcdi_cmd *cmd) +{ + struct cdx_mcdi_iface *mcdi = cdx_mcdi_if(cdx); + LIST_HEAD(cleanup_list); + + if (!mcdi) + return; + + mutex_lock(&mcdi->iface_lock); + cdx_mcdi_timeout_cmd(mcdi, cmd, &cleanup_list); + mutex_unlock(&mcdi->iface_lock); + cdx_mcdi_process_cleanup_list(cdx, &cleanup_list); +} + +struct cdx_mcdi_blocking_data { + struct kref ref; + bool done; + wait_queue_head_t wq; + int rc; + struct cdx_dword *outbuf; + size_t outlen; + size_t outlen_actual; +}; + +static void cdx_mcdi_blocking_data_release(struct kref *ref) +{ + kfree(container_of(ref, struct cdx_mcdi_blocking_data, ref)); +} + +static void cdx_mcdi_rpc_completer(struct cdx_mcdi *cdx, unsigned long cookie, + int rc, struct cdx_dword *outbuf, + size_t outlen_actual) +{ + struct cdx_mcdi_blocking_data *wait_data = + (struct cdx_mcdi_blocking_data *)cookie; + + wait_data->rc = rc; + memcpy(wait_data->outbuf, outbuf, + min(outlen_actual, wait_data->outlen)); + wait_data->outlen_actual = outlen_actual; + /* memory barrier */ + smp_wmb(); + wait_data->done = true; + wake_up(&wait_data->wq); + kref_put(&wait_data->ref, cdx_mcdi_blocking_data_release); +} + +static int cdx_mcdi_rpc_sync(struct cdx_mcdi *cdx, unsigned int cmd, + const struct cdx_dword *inbuf, size_t inlen, + struct cdx_dword *outbuf, size_t outlen, + size_t *outlen_actual, bool quiet) +{ + struct cdx_mcdi_blocking_data *wait_data; + struct cdx_mcdi_cmd *cmd_item; + unsigned int handle; + int rc; + + if (outlen_actual) + *outlen_actual = 0; + + wait_data = kmalloc_obj(*wait_data); + if (!wait_data) + return -ENOMEM; + + cmd_item = kmalloc_obj(*cmd_item); + if (!cmd_item) { + kfree(wait_data); + return -ENOMEM; + } + + kref_init(&wait_data->ref); + wait_data->done = false; + init_waitqueue_head(&wait_data->wq); + wait_data->outbuf = outbuf; + wait_data->outlen = outlen; + + kref_init(&cmd_item->ref); + cmd_item->quiet = quiet; + cmd_item->cookie = (unsigned long)wait_data; + cmd_item->completer = &cdx_mcdi_rpc_completer; + cmd_item->cmd = cmd; + cmd_item->inlen = inlen; + cmd_item->inbuf = inbuf; + + /* Claim an extra reference for the completer to put. */ + kref_get(&wait_data->ref); + rc = cdx_mcdi_rpc_async_internal(cdx, cmd_item, &handle); + if (rc) { + kref_put(&wait_data->ref, cdx_mcdi_blocking_data_release); + goto out; + } + + if (!wait_event_timeout(wait_data->wq, wait_data->done, + cdx_mcdi_rpc_timeout(cdx, cmd)) && + !wait_data->done) { + pr_err("MC command 0x%x inlen %zu timed out (sync)\n", + cmd, inlen); + + cdx_mcdi_cancel_cmd(cdx, cmd_item); + + wait_data->rc = -ETIMEDOUT; + wait_data->outlen_actual = 0; + } + + if (outlen_actual) + *outlen_actual = wait_data->outlen_actual; + rc = wait_data->rc; + +out: + kref_put(&wait_data->ref, cdx_mcdi_blocking_data_release); + + return rc; +} + +static bool cdx_mcdi_get_seq(struct cdx_mcdi_iface *mcdi, unsigned char *seq) +{ + *seq = mcdi->prev_seq; + do { + *seq = (*seq + 1) % ARRAY_SIZE(mcdi->seq_held_by); + } while (mcdi->seq_held_by[*seq] && *seq != mcdi->prev_seq); + return !mcdi->seq_held_by[*seq]; +} + +static int cdx_mcdi_rpc_async_internal(struct cdx_mcdi *cdx, + struct cdx_mcdi_cmd *cmd, + unsigned int *handle) +{ + struct cdx_mcdi_iface *mcdi = cdx_mcdi_if(cdx); + LIST_HEAD(cleanup_list); + + if (!mcdi) { + kref_put(&cmd->ref, cdx_mcdi_cmd_release); + return -ENETDOWN; + } + + if (mcdi->mode == MCDI_MODE_FAIL) { + kref_put(&cmd->ref, cdx_mcdi_cmd_release); + return -ENETDOWN; + } + + cmd->mcdi = mcdi; + INIT_WORK(&cmd->work, cdx_mcdi_cmd_work); + INIT_LIST_HEAD(&cmd->list); + INIT_LIST_HEAD(&cmd->cleanup_list); + cmd->rc = 0; + cmd->outbuf = NULL; + cmd->outlen = 0; + + queue_work(mcdi->workqueue, &cmd->work); + return 0; +} + +static void cdx_mcdi_cmd_start_or_queue(struct cdx_mcdi_iface *mcdi, + struct cdx_mcdi_cmd *cmd) +{ + struct cdx_mcdi *cdx = mcdi->cdx; + u8 seq; + + if (!mcdi->db_held_by && + cdx_mcdi_get_seq(mcdi, &seq)) { + cmd->seq = seq; + cmd->reboot_seen = false; + cdx_mcdi_send_request(cdx, cmd); + cmd->state = MCDI_STATE_RUNNING; + } else { + cmd->state = MCDI_STATE_QUEUED; + } +} + +/* try to advance other commands */ +static void cdx_mcdi_start_or_queue(struct cdx_mcdi_iface *mcdi, + bool allow_retry) +{ + struct cdx_mcdi_cmd *cmd, *tmp; + + list_for_each_entry_safe(cmd, tmp, &mcdi->cmd_list, list) + if (cmd->state == MCDI_STATE_QUEUED || + (cmd->state == MCDI_STATE_RETRY && allow_retry)) + cdx_mcdi_cmd_start_or_queue(mcdi, cmd); +} + +/** + * cdx_mcdi_process_cmd - Process an incoming MCDI response + * @cdx: Handle to the CDX MCDI structure + * @outbuf: Pointer to the response buffer received from the management controller + * @len: Length of the response buffer in bytes + * + * This function handles a response from the management controller. It locates the + * corresponding command using the sequence number embedded in the header, + * completes the command if it is still pending, and initiates any necessary cleanup. + * + * The function assumes that the response buffer is well-formed and at least one + * dword in size. + */ +void cdx_mcdi_process_cmd(struct cdx_mcdi *cdx, struct cdx_dword *outbuf, int len) +{ + struct cdx_mcdi_iface *mcdi; + struct cdx_mcdi_cmd *cmd; + LIST_HEAD(cleanup_list); + unsigned int respseq; + + if (!len || !outbuf) { + pr_err("Got empty MC response\n"); + return; + } + + mcdi = cdx_mcdi_if(cdx); + if (!mcdi) + return; + + respseq = CDX_DWORD_FIELD(outbuf[0], MCDI_HEADER_SEQ); + + mutex_lock(&mcdi->iface_lock); + cmd = mcdi->seq_held_by[respseq]; + + if (cmd) { + if (cmd->state == MCDI_STATE_FINISHED) { + mutex_unlock(&mcdi->iface_lock); + kref_put(&cmd->ref, cdx_mcdi_cmd_release); + return; + } + + cdx_mcdi_complete_cmd(mcdi, cmd, outbuf, len, &cleanup_list); + } else { + pr_err("MC response unexpected for seq : %0X\n", respseq); + } + + mutex_unlock(&mcdi->iface_lock); + + cdx_mcdi_process_cleanup_list(mcdi->cdx, &cleanup_list); +} +EXPORT_SYMBOL_GPL(cdx_mcdi_process_cmd); + +static void cdx_mcdi_cmd_work(struct work_struct *context) +{ + struct cdx_mcdi_cmd *cmd = + container_of(context, struct cdx_mcdi_cmd, work); + struct cdx_mcdi_iface *mcdi = cmd->mcdi; + + mutex_lock(&mcdi->iface_lock); + + cmd->handle = mcdi->prev_handle++; + list_add_tail(&cmd->list, &mcdi->cmd_list); + cdx_mcdi_cmd_start_or_queue(mcdi, cmd); + + mutex_unlock(&mcdi->iface_lock); +} + +/* + * Returns true if the MCDI module is finished with the command. + * (examples of false would be if the command was proxied, or it was + * rejected by the MC due to lack of resources and requeued). + */ +static bool cdx_mcdi_complete_cmd(struct cdx_mcdi_iface *mcdi, + struct cdx_mcdi_cmd *cmd, + struct cdx_dword *outbuf, + int len, + struct list_head *cleanup_list) +{ + size_t resp_hdr_len, resp_data_len; + struct cdx_mcdi *cdx = mcdi->cdx; + unsigned int respcmd, error; + bool completed = false; + int rc; + + /* ensure the command can't go away before this function returns */ + kref_get(&cmd->ref); + + respcmd = CDX_DWORD_FIELD(outbuf[0], MCDI_HEADER_CODE); + error = CDX_DWORD_FIELD(outbuf[0], MCDI_HEADER_ERROR); + + if (respcmd != MC_CMD_V2_EXTN) { + resp_hdr_len = 4; + resp_data_len = CDX_DWORD_FIELD(outbuf[0], MCDI_HEADER_DATALEN); + } else { + resp_data_len = 0; + resp_hdr_len = 8; + if (len >= 8) + resp_data_len = + CDX_DWORD_FIELD(outbuf[1], MC_CMD_V2_EXTN_IN_ACTUAL_LEN); + } + + if ((resp_hdr_len + resp_data_len) > len) { + pr_warn("Incomplete MCDI response received %d. Expected %zu\n", + len, (resp_hdr_len + resp_data_len)); + resp_data_len = 0; + } + + print_hex_dump_debug("MCDI RESP HEADER: ", DUMP_PREFIX_NONE, 32, 4, + outbuf, resp_hdr_len, false); + print_hex_dump_debug("MCDI RESP PAYLOAD: ", DUMP_PREFIX_NONE, 32, 4, + outbuf + (resp_hdr_len / 4), resp_data_len, false); + + if (error && resp_data_len == 0) { + /* MC rebooted during command */ + rc = -EIO; + } else { + if (WARN_ON_ONCE(error && resp_data_len < 4)) + resp_data_len = 4; + if (error) { + rc = CDX_DWORD_FIELD(outbuf[resp_hdr_len / 4], CDX_DWORD); + if (!cmd->quiet) { + int err_arg = 0; + + if (resp_data_len >= MC_CMD_ERR_ARG_OFST + 4) { + int offset = (resp_hdr_len + MC_CMD_ERR_ARG_OFST) / 4; + + err_arg = CDX_DWORD_VAL(outbuf[offset]); + } + + _cdx_mcdi_display_error(cdx, cmd->cmd, + cmd->inlen, rc, err_arg, + cdx_mcdi_errno(cdx, rc)); + } + rc = cdx_mcdi_errno(cdx, rc); + } else { + rc = 0; + } + } + + /* free doorbell */ + if (mcdi->db_held_by == cmd) + mcdi->db_held_by = NULL; + + if (cdx_cmd_cancelled(cmd)) { + list_del(&cmd->list); + kref_put(&cmd->ref, cdx_mcdi_cmd_release); + completed = true; + } else if (rc == MC_CMD_ERR_QUEUE_FULL) { + cmd->state = MCDI_STATE_RETRY; + } else { + cmd->rc = rc; + cmd->outbuf = outbuf + DIV_ROUND_UP(resp_hdr_len, 4); + cmd->outlen = resp_data_len; + cdx_mcdi_remove_cmd(mcdi, cmd, cleanup_list); + completed = true; + } + + /* free sequence number and buffer */ + mcdi->seq_held_by[cmd->seq] = NULL; + + cdx_mcdi_start_or_queue(mcdi, rc != MC_CMD_ERR_QUEUE_FULL); + + /* wake up anyone waiting for flush */ + wake_up(&mcdi->cmd_complete_wq); + + kref_put(&cmd->ref, cdx_mcdi_cmd_release); + + return completed; +} + +static void cdx_mcdi_timeout_cmd(struct cdx_mcdi_iface *mcdi, + struct cdx_mcdi_cmd *cmd, + struct list_head *cleanup_list) +{ + struct cdx_mcdi *cdx = mcdi->cdx; + + pr_err("MC command 0x%x inlen %zu state %d timed out after %u ms\n", + cmd->cmd, cmd->inlen, cmd->state, + jiffies_to_msecs(jiffies - cmd->started)); + + cmd->rc = -ETIMEDOUT; + cdx_mcdi_remove_cmd(mcdi, cmd, cleanup_list); + + cdx_mcdi_mode_fail(cdx, cleanup_list); +} + +/** + * cdx_mcdi_rpc - Issue an MCDI command and wait for completion + * @cdx: NIC through which to issue the command + * @cmd: Command type number + * @inbuf: Command parameters + * @inlen: Length of command parameters, in bytes. Must be a multiple + * of 4 and no greater than %MCDI_CTL_SDU_LEN_MAX_V1. + * @outbuf: Response buffer. May be %NULL if @outlen is 0. + * @outlen: Length of response buffer, in bytes. If the actual + * response is longer than @outlen & ~3, it will be truncated + * to that length. + * @outlen_actual: Pointer through which to return the actual response + * length. May be %NULL if this is not needed. + * + * This function may sleep and therefore must be called in process + * context. + * + * Return: A negative error code, or zero if successful. The error + * code may come from the MCDI response or may indicate a failure + * to communicate with the MC. In the former case, the response + * will still be copied to @outbuf and *@outlen_actual will be + * set accordingly. In the latter case, *@outlen_actual will be + * set to zero. + */ +int cdx_mcdi_rpc(struct cdx_mcdi *cdx, unsigned int cmd, + const struct cdx_dword *inbuf, size_t inlen, + struct cdx_dword *outbuf, size_t outlen, + size_t *outlen_actual) +{ + return cdx_mcdi_rpc_sync(cdx, cmd, inbuf, inlen, outbuf, outlen, + outlen_actual, false); +} +EXPORT_SYMBOL_GPL(cdx_mcdi_rpc); + +/** + * cdx_mcdi_rpc_async - Schedule an MCDI command to run asynchronously + * @cdx: NIC through which to issue the command + * @cmd: Command type number + * @inbuf: Command parameters + * @inlen: Length of command parameters, in bytes + * @complete: Function to be called on completion or cancellation. + * @cookie: Arbitrary value to be passed to @complete. + * + * This function does not sleep and therefore may be called in atomic + * context. It will fail if event queues are disabled or if MCDI + * event completions have been disabled due to an error. + * + * If it succeeds, the @complete function will be called exactly once + * in process context, when one of the following occurs: + * (a) the completion event is received (in process context) + * (b) event queues are disabled (in the process that disables them) + */ +int +cdx_mcdi_rpc_async(struct cdx_mcdi *cdx, unsigned int cmd, + const struct cdx_dword *inbuf, size_t inlen, + cdx_mcdi_async_completer *complete, unsigned long cookie) +{ + struct cdx_mcdi_cmd *cmd_item = + kmalloc(sizeof(struct cdx_mcdi_cmd) + inlen, GFP_ATOMIC); + + if (!cmd_item) + return -ENOMEM; + + kref_init(&cmd_item->ref); + cmd_item->quiet = true; + cmd_item->cookie = cookie; + cmd_item->completer = complete; + cmd_item->cmd = cmd; + cmd_item->inlen = inlen; + /* inbuf is probably not valid after return, so take a copy */ + cmd_item->inbuf = (struct cdx_dword *)(cmd_item + 1); + memcpy(cmd_item + 1, inbuf, inlen); + + return cdx_mcdi_rpc_async_internal(cdx, cmd_item, NULL); +} + +static void _cdx_mcdi_display_error(struct cdx_mcdi *cdx, unsigned int cmd, + size_t inlen, int raw, int arg, int err_no) +{ + pr_err("MC command 0x%x inlen %d failed err_no=%d (raw=%d) arg=%d\n", + cmd, (int)inlen, err_no, raw, arg); +} + +/* + * Set MCDI mode to fail to prevent any new commands, then cancel any + * outstanding commands. + * Caller must hold the mcdi iface_lock. + */ +static void cdx_mcdi_mode_fail(struct cdx_mcdi *cdx, struct list_head *cleanup_list) +{ + struct cdx_mcdi_iface *mcdi = cdx_mcdi_if(cdx); + + if (!mcdi) + return; + + mcdi->mode = MCDI_MODE_FAIL; + + while (!list_empty(&mcdi->cmd_list)) { + struct cdx_mcdi_cmd *cmd; + + cmd = list_first_entry(&mcdi->cmd_list, struct cdx_mcdi_cmd, + list); + _cdx_mcdi_cancel_cmd(mcdi, cdx_mcdi_cmd_handle(cmd), cleanup_list); + } +} -- cgit v1.3.1