diff options
Diffstat (limited to 'drivers/edac/edac_mc.c')
| -rw-r--r-- | drivers/edac/edac_mc.c | 1018 |
1 files changed, 409 insertions, 609 deletions
diff --git a/drivers/edac/edac_mc.c b/drivers/edac/edac_mc.c index 27e86d938262..0959320fe51c 100644 --- a/drivers/edac/edac_mc.c +++ b/drivers/edac/edac_mc.c @@ -28,16 +28,21 @@ #include <linux/ctype.h> #include <linux/edac.h> #include <linux/bitops.h> -#include <asm/uaccess.h> +#include <linux/uaccess.h> #include <asm/page.h> -#include <asm/edac.h> -#include "edac_core.h" +#include "edac_mc.h" #include "edac_module.h" - -#define CREATE_TRACE_POINTS -#define TRACE_INCLUDE_PATH ../../include/ras #include <ras/ras_event.h> +#ifdef CONFIG_EDAC_ATOMIC_SCRUB +#include <asm/edac.h> +#else +#define edac_atomic_scrub(va, size) do { } while (0) +#endif + +int edac_op_state = EDAC_OPSTATE_INVAL; +EXPORT_SYMBOL_GPL(edac_op_state); + /* lock to memory controller's control array */ static DEFINE_MUTEX(mem_ctls_mutex); static LIST_HEAD(mc_devices); @@ -46,24 +51,27 @@ static LIST_HEAD(mc_devices); * Used to lock EDAC MC to just one module, avoiding two drivers e. g. * apei/ghes and i7core_edac to be used at the same time. */ -static void const *edac_mc_owner; +static const char *edac_mc_owner; -unsigned edac_dimm_info_location(struct dimm_info *dimm, char *buf, - unsigned len) +static struct mem_ctl_info *error_desc_to_mci(struct edac_raw_error_desc *e) +{ + return container_of(e, struct mem_ctl_info, error_desc); +} + +unsigned int edac_dimm_info_location(struct dimm_info *dimm, char *buf, + unsigned int len) { struct mem_ctl_info *mci = dimm->mci; int i, n, count = 0; char *p = buf; for (i = 0; i < mci->n_layers; i++) { - n = snprintf(p, len, "%s %d ", + n = scnprintf(p, len, "%s %d ", edac_layer_name[mci->layers[i].type], dimm->location[i]); p += n; len -= n; count += n; - if (!len) - break; } return count; @@ -79,20 +87,22 @@ static void edac_mc_dump_channel(struct rank_info *chan) edac_dbg(4, " channel->dimm = %p\n", chan->dimm); } -static void edac_mc_dump_dimm(struct dimm_info *dimm, int number) +static void edac_mc_dump_dimm(struct dimm_info *dimm) { char location[80]; + if (!dimm->nr_pages) + return; + edac_dimm_info_location(dimm, location, sizeof(location)); edac_dbg(4, "%s%i: %smapped as virtual row %d, chan %d\n", dimm->mci->csbased ? "rank" : "dimm", - number, location, dimm->csrow, dimm->cschannel); + dimm->idx, location, dimm->csrow, dimm->cschannel); edac_dbg(4, " dimm = %p\n", dimm); edac_dbg(4, " dimm->label = '%s'\n", dimm->label); edac_dbg(4, " dimm->nr_pages = 0x%x\n", dimm->nr_pages); edac_dbg(4, " dimm->grain = %d\n", dimm->grain); - edac_dbg(4, " dimm->nr_pages = 0x%x\n", dimm->nr_pages); } static void edac_mc_dump_csrow(struct csrow_info *csrow) @@ -126,233 +136,97 @@ static void edac_mc_dump_mci(struct mem_ctl_info *mci) #endif /* CONFIG_EDAC_DEBUG */ -/* - * keep those in sync with the enum mem_type - */ -const char *edac_mem_types[] = { - "Empty csrow", - "Reserved csrow type", - "Unknown csrow type", - "Fast page mode RAM", - "Extended data out RAM", - "Burst Extended data out RAM", - "Single data rate SDRAM", - "Registered single data rate SDRAM", - "Double data rate SDRAM", - "Registered Double data rate SDRAM", - "Rambus DRAM", - "Unbuffered DDR2 RAM", - "Fully buffered DDR2", - "Registered DDR2 RAM", - "Rambus XDR", - "Unbuffered DDR3 RAM", - "Registered DDR3 RAM", +const char * const edac_mem_types[] = { + [MEM_EMPTY] = "Empty", + [MEM_RESERVED] = "Reserved", + [MEM_UNKNOWN] = "Unknown", + [MEM_FPM] = "FPM", + [MEM_EDO] = "EDO", + [MEM_BEDO] = "BEDO", + [MEM_SDR] = "Unbuffered-SDR", + [MEM_RDR] = "Registered-SDR", + [MEM_DDR] = "Unbuffered-DDR", + [MEM_RDDR] = "Registered-DDR", + [MEM_RMBS] = "RMBS", + [MEM_DDR2] = "Unbuffered-DDR2", + [MEM_FB_DDR2] = "FullyBuffered-DDR2", + [MEM_RDDR2] = "Registered-DDR2", + [MEM_XDR] = "XDR", + [MEM_DDR3] = "Unbuffered-DDR3", + [MEM_RDDR3] = "Registered-DDR3", + [MEM_LRDDR3] = "Load-Reduced-DDR3-RAM", + [MEM_LPDDR3] = "Low-Power-DDR3-RAM", + [MEM_DDR4] = "Unbuffered-DDR4", + [MEM_RDDR4] = "Registered-DDR4", + [MEM_LPDDR4] = "Low-Power-DDR4-RAM", + [MEM_LRDDR4] = "Load-Reduced-DDR4-RAM", + [MEM_DDR5] = "Unbuffered-DDR5", + [MEM_RDDR5] = "Registered-DDR5", + [MEM_LRDDR5] = "Load-Reduced-DDR5-RAM", + [MEM_NVDIMM] = "Non-volatile-RAM", + [MEM_WIO2] = "Wide-IO-2", + [MEM_HBM2] = "High-bandwidth-memory-Gen2", + [MEM_HBM3] = "High-bandwidth-memory-Gen3", }; EXPORT_SYMBOL_GPL(edac_mem_types); -/** - * edac_align_ptr - Prepares the pointer offsets for a single-shot allocation - * @p: pointer to a pointer with the memory offset to be used. At - * return, this will be incremented to point to the next offset - * @size: Size of the data structure to be reserved - * @n_elems: Number of elements that should be reserved - * - * If 'size' is a constant, the compiler will optimize this whole function - * down to either a no-op or the addition of a constant to the value of '*p'. - * - * The 'p' pointer is absolutely needed to keep the proper advancing - * further in memory to the proper offsets when allocating the struct along - * with its embedded structs, as edac_device_alloc_ctl_info() does it - * above, for example. - * - * At return, the pointer 'p' will be incremented to be used on a next call - * to this function. - */ -void *edac_align_ptr(void **p, unsigned size, int n_elems) +static void _edac_mc_free(struct mem_ctl_info *mci) { - unsigned align, r; - void *ptr = *p; - - *p += size * n_elems; - - /* - * 'p' can possibly be an unaligned item X such that sizeof(X) is - * 'size'. Adjust 'p' so that its alignment is at least as - * stringent as what the compiler would provide for X and return - * the aligned result. - * Here we assume that the alignment of a "long long" is the most - * stringent alignment that the compiler will ever provide by default. - * As far as I know, this is a reasonable assumption. - */ - if (size > sizeof(long)) - align = sizeof(long long); - else if (size > sizeof(int)) - align = sizeof(long); - else if (size > sizeof(short)) - align = sizeof(int); - else if (size > sizeof(char)) - align = sizeof(short); - else - return (char *)ptr; - - r = (unsigned long)p % align; - - if (r == 0) - return (char *)ptr; - - *p += align - r; - - return (void *)(((unsigned long)ptr) + align - r); + put_device(&mci->dev); } -static void _edac_mc_free(struct mem_ctl_info *mci) +static void mci_release(struct device *dev) { - int i, chn, row; + struct mem_ctl_info *mci = container_of(dev, struct mem_ctl_info, dev); struct csrow_info *csr; - const unsigned int tot_dimms = mci->tot_dimms; - const unsigned int tot_channels = mci->num_cschannel; - const unsigned int tot_csrows = mci->nr_csrows; + int i, chn, row; if (mci->dimms) { - for (i = 0; i < tot_dimms; i++) + for (i = 0; i < mci->tot_dimms; i++) kfree(mci->dimms[i]); kfree(mci->dimms); } + if (mci->csrows) { - for (row = 0; row < tot_csrows; row++) { + for (row = 0; row < mci->nr_csrows; row++) { csr = mci->csrows[row]; - if (csr) { - if (csr->channels) { - for (chn = 0; chn < tot_channels; chn++) - kfree(csr->channels[chn]); - kfree(csr->channels); - } - kfree(csr); + if (!csr) + continue; + + if (csr->channels) { + for (chn = 0; chn < mci->num_cschannel; chn++) + kfree(csr->channels[chn]); + kfree(csr->channels); } + kfree(csr); } kfree(mci->csrows); } + kfree(mci->pvt_info); + kfree(mci->layers); kfree(mci); } -/** - * edac_mc_alloc: Allocate and partially fill a struct mem_ctl_info structure - * @mc_num: Memory controller number - * @n_layers: Number of MC hierarchy layers - * layers: Describes each layer as seen by the Memory Controller - * @size_pvt: size of private storage needed - * - * - * Everything is kmalloc'ed as one big chunk - more efficient. - * Only can be used if all structures have the same lifetime - otherwise - * you have to allocate and initialize your own structures. - * - * Use edac_mc_free() to free mc structures allocated by this function. - * - * NOTE: drivers handle multi-rank memories in different ways: in some - * drivers, one multi-rank memory stick is mapped as one entry, while, in - * others, a single multi-rank memory stick would be mapped into several - * entries. Currently, this function will allocate multiple struct dimm_info - * on such scenarios, as grouping the multiple ranks require drivers change. - * - * Returns: - * On failure: NULL - * On success: struct mem_ctl_info pointer - */ -struct mem_ctl_info *edac_mc_alloc(unsigned mc_num, - unsigned n_layers, - struct edac_mc_layer *layers, - unsigned sz_pvt) +static int edac_mc_alloc_csrows(struct mem_ctl_info *mci) { - struct mem_ctl_info *mci; - struct edac_mc_layer *layer; - struct csrow_info *csr; - struct rank_info *chan; - struct dimm_info *dimm; - u32 *ce_per_layer[EDAC_MAX_LAYERS], *ue_per_layer[EDAC_MAX_LAYERS]; - unsigned pos[EDAC_MAX_LAYERS]; - unsigned size, tot_dimms = 1, count = 1; - unsigned tot_csrows = 1, tot_channels = 1, tot_errcount = 0; - void *pvt, *p, *ptr = NULL; - int i, j, row, chn, n, len, off; - bool per_rank = false; - - BUG_ON(n_layers > EDAC_MAX_LAYERS || n_layers == 0); - /* - * Calculate the total amount of dimms and csrows/cschannels while - * in the old API emulation mode - */ - for (i = 0; i < n_layers; i++) { - tot_dimms *= layers[i].size; - if (layers[i].is_virt_csrow) - tot_csrows *= layers[i].size; - else - tot_channels *= layers[i].size; - - if (layers[i].type == EDAC_MC_LAYER_CHIP_SELECT) - per_rank = true; - } - - /* Figure out the offsets of the various items from the start of an mc - * structure. We want the alignment of each item to be at least as - * stringent as what the compiler would provide if we could simply - * hardcode everything into a single struct. - */ - mci = edac_align_ptr(&ptr, sizeof(*mci), 1); - layer = edac_align_ptr(&ptr, sizeof(*layer), n_layers); - for (i = 0; i < n_layers; i++) { - count *= layers[i].size; - edac_dbg(4, "errcount layer %d size %d\n", i, count); - ce_per_layer[i] = edac_align_ptr(&ptr, sizeof(u32), count); - ue_per_layer[i] = edac_align_ptr(&ptr, sizeof(u32), count); - tot_errcount += 2 * count; - } - - edac_dbg(4, "allocating %d error counters\n", tot_errcount); - pvt = edac_align_ptr(&ptr, sz_pvt, 1); - size = ((unsigned long)pvt) + sz_pvt; - - edac_dbg(1, "allocating %u bytes for mci data (%d %s, %d csrows/channels)\n", - size, - tot_dimms, - per_rank ? "ranks" : "dimms", - tot_csrows * tot_channels); - - mci = kzalloc(size, GFP_KERNEL); - if (mci == NULL) - return NULL; - - /* Adjust pointers so they point within the memory we just allocated - * rather than an imaginary chunk of memory located at address 0. - */ - layer = (struct edac_mc_layer *)(((char *)mci) + ((unsigned long)layer)); - for (i = 0; i < n_layers; i++) { - mci->ce_per_layer[i] = (u32 *)((char *)mci + ((unsigned long)ce_per_layer[i])); - mci->ue_per_layer[i] = (u32 *)((char *)mci + ((unsigned long)ue_per_layer[i])); - } - pvt = sz_pvt ? (((char *)mci) + ((unsigned long)pvt)) : NULL; - - /* setup index and various internal pointers */ - mci->mc_idx = mc_num; - mci->tot_dimms = tot_dimms; - mci->pvt_info = pvt; - mci->n_layers = n_layers; - mci->layers = layer; - memcpy(mci->layers, layers, sizeof(*layer) * n_layers); - mci->nr_csrows = tot_csrows; - mci->num_cschannel = tot_channels; - mci->csbased = per_rank; + unsigned int tot_channels = mci->num_cschannel; + unsigned int tot_csrows = mci->nr_csrows; + unsigned int row, chn; /* - * Alocate and fill the csrow/channels structs + * Allocate and fill the csrow/channels structs */ mci->csrows = kcalloc(tot_csrows, sizeof(*mci->csrows), GFP_KERNEL); if (!mci->csrows) - goto error; + return -ENOMEM; + for (row = 0; row < tot_csrows; row++) { + struct csrow_info *csr; + csr = kzalloc(sizeof(**mci->csrows), GFP_KERNEL); if (!csr) - goto error; + return -ENOMEM; + mci->csrows[row] = csr; csr->csrow_idx = row; csr->mci = mci; @@ -360,60 +234,70 @@ struct mem_ctl_info *edac_mc_alloc(unsigned mc_num, csr->channels = kcalloc(tot_channels, sizeof(*csr->channels), GFP_KERNEL); if (!csr->channels) - goto error; + return -ENOMEM; for (chn = 0; chn < tot_channels; chn++) { + struct rank_info *chan; + chan = kzalloc(sizeof(**csr->channels), GFP_KERNEL); if (!chan) - goto error; + return -ENOMEM; + csr->channels[chn] = chan; chan->chan_idx = chn; chan->csrow = csr; } } + return 0; +} + +static int edac_mc_alloc_dimms(struct mem_ctl_info *mci) +{ + unsigned int pos[EDAC_MAX_LAYERS]; + unsigned int row, chn, idx; + int layer; + void *p; + /* * Allocate and fill the dimm structs */ - mci->dimms = kcalloc(tot_dimms, sizeof(*mci->dimms), GFP_KERNEL); + mci->dimms = kcalloc(mci->tot_dimms, sizeof(*mci->dimms), GFP_KERNEL); if (!mci->dimms) - goto error; + return -ENOMEM; memset(&pos, 0, sizeof(pos)); row = 0; chn = 0; - for (i = 0; i < tot_dimms; i++) { + for (idx = 0; idx < mci->tot_dimms; idx++) { + struct dimm_info *dimm; + struct rank_info *chan; + int n, len; + chan = mci->csrows[row]->channels[chn]; - off = EDAC_DIMM_OFF(layer, n_layers, pos[0], pos[1], pos[2]); - if (off < 0 || off >= tot_dimms) { - edac_mc_printk(mci, KERN_ERR, "EDAC core bug: EDAC_DIMM_OFF is trying to do an illegal data access\n"); - goto error; - } dimm = kzalloc(sizeof(**mci->dimms), GFP_KERNEL); if (!dimm) - goto error; - mci->dimms[off] = dimm; + return -ENOMEM; + mci->dimms[idx] = dimm; dimm->mci = mci; + dimm->idx = idx; /* * Copy DIMM location and initialize it. */ len = sizeof(dimm->label); p = dimm->label; - n = snprintf(p, len, "mc#%u", mc_num); + n = scnprintf(p, len, "mc#%u", mci->mc_idx); p += n; len -= n; - for (j = 0; j < n_layers; j++) { - n = snprintf(p, len, "%s#%u", - edac_layer_name[layers[j].type], - pos[j]); + for (layer = 0; layer < mci->n_layers; layer++) { + n = scnprintf(p, len, "%s#%u", + edac_layer_name[mci->layers[layer].type], + pos[layer]); p += n; len -= n; - dimm->location[j] = pos[j]; - - if (len <= 0) - break; + dimm->location[layer] = pos[layer]; } /* Link it to the csrows old API data */ @@ -422,29 +306,92 @@ struct mem_ctl_info *edac_mc_alloc(unsigned mc_num, dimm->cschannel = chn; /* Increment csrow location */ - if (layers[0].is_virt_csrow) { + if (mci->layers[0].is_virt_csrow) { chn++; - if (chn == tot_channels) { + if (chn == mci->num_cschannel) { chn = 0; row++; } } else { row++; - if (row == tot_csrows) { + if (row == mci->nr_csrows) { row = 0; chn++; } } /* Increment dimm location */ - for (j = n_layers - 1; j >= 0; j--) { - pos[j]++; - if (pos[j] < layers[j].size) + for (layer = mci->n_layers - 1; layer >= 0; layer--) { + pos[layer]++; + if (pos[layer] < mci->layers[layer].size) break; - pos[j] = 0; + pos[layer] = 0; } } + return 0; +} + +struct mem_ctl_info *edac_mc_alloc(unsigned int mc_num, + unsigned int n_layers, + struct edac_mc_layer *layers, + unsigned int sz_pvt) +{ + struct mem_ctl_info *mci; + struct edac_mc_layer *layer; + unsigned int idx, tot_dimms = 1; + unsigned int tot_csrows = 1, tot_channels = 1; + bool per_rank = false; + + if (WARN_ON(n_layers > EDAC_MAX_LAYERS || n_layers == 0)) + return NULL; + + /* + * Calculate the total amount of dimms and csrows/cschannels while + * in the old API emulation mode + */ + for (idx = 0; idx < n_layers; idx++) { + tot_dimms *= layers[idx].size; + + if (layers[idx].is_virt_csrow) + tot_csrows *= layers[idx].size; + else + tot_channels *= layers[idx].size; + + if (layers[idx].type == EDAC_MC_LAYER_CHIP_SELECT) + per_rank = true; + } + + mci = kzalloc(sizeof(struct mem_ctl_info), GFP_KERNEL); + if (!mci) + return NULL; + + mci->layers = kcalloc(n_layers, sizeof(struct edac_mc_layer), GFP_KERNEL); + if (!mci->layers) + goto error; + + mci->pvt_info = kzalloc(sz_pvt, GFP_KERNEL); + if (!mci->pvt_info) + goto error; + + mci->dev.release = mci_release; + device_initialize(&mci->dev); + + /* setup index and various internal pointers */ + mci->mc_idx = mc_num; + mci->tot_dimms = tot_dimms; + mci->n_layers = n_layers; + memcpy(mci->layers, layers, sizeof(*layer) * n_layers); + mci->nr_csrows = tot_csrows; + mci->num_cschannel = tot_channels; + mci->csbased = per_rank; + + if (edac_mc_alloc_csrows(mci)) + goto error; + + if (edac_mc_alloc_dimms(mci)) + goto error; + mci->op_state = OP_ALLOC; return mci; @@ -456,37 +403,30 @@ error: } EXPORT_SYMBOL_GPL(edac_mc_alloc); -/** - * edac_mc_free - * 'Free' a previously allocated 'mci' structure - * @mci: pointer to a struct mem_ctl_info structure - */ void edac_mc_free(struct mem_ctl_info *mci) { edac_dbg(1, "\n"); - /* If we're not yet registered with sysfs free only what was allocated - * in edac_mc_alloc(). - */ - if (!device_is_registered(&mci->dev)) { - _edac_mc_free(mci); - return; - } - - /* the mci instance is freed here, when the sysfs object is dropped */ - edac_unregister_sysfs(mci); + _edac_mc_free(mci); } EXPORT_SYMBOL_GPL(edac_mc_free); +bool edac_has_mcs(void) +{ + bool ret; -/** - * find_mci_by_dev - * - * scan list of controllers looking for the one that manages - * the 'dev' device - * @dev: pointer to a struct device related with the MCI - */ -struct mem_ctl_info *find_mci_by_dev(struct device *dev) + mutex_lock(&mem_ctls_mutex); + + ret = list_empty(&mc_devices); + + mutex_unlock(&mem_ctls_mutex); + + return !ret; +} +EXPORT_SYMBOL_GPL(edac_has_mcs); + +/* Caller must hold mem_ctls_mutex */ +static struct mem_ctl_info *__find_mci_by_dev(struct device *dev) { struct mem_ctl_info *mci; struct list_head *item; @@ -502,23 +442,25 @@ struct mem_ctl_info *find_mci_by_dev(struct device *dev) return NULL; } -EXPORT_SYMBOL_GPL(find_mci_by_dev); -/* - * handler for EDAC to check if NMI type handler has asserted interrupt +/** + * find_mci_by_dev + * + * scan list of controllers looking for the one that manages + * the 'dev' device + * @dev: pointer to a struct device related with the MCI */ -static int edac_mc_assert_error_check_and_clear(void) +struct mem_ctl_info *find_mci_by_dev(struct device *dev) { - int old_state; - - if (edac_op_state == EDAC_OPSTATE_POLL) - return 1; + struct mem_ctl_info *ret; - old_state = edac_err_assert; - edac_err_assert = 0; + mutex_lock(&mem_ctls_mutex); + ret = __find_mci_by_dev(dev); + mutex_unlock(&mem_ctls_mutex); - return old_state; + return ret; } +EXPORT_SYMBOL_GPL(find_mci_by_dev); /* * edac_mc_workq_function @@ -531,66 +473,18 @@ static void edac_mc_workq_function(struct work_struct *work_req) mutex_lock(&mem_ctls_mutex); - /* if this control struct has movd to offline state, we are done */ - if (mci->op_state == OP_OFFLINE) { + if (mci->op_state != OP_RUNNING_POLL) { mutex_unlock(&mem_ctls_mutex); return; } - /* Only poll controllers that are running polled and have a check */ - if (edac_mc_assert_error_check_and_clear() && (mci->edac_check != NULL)) + if (edac_op_state == EDAC_OPSTATE_POLL) mci->edac_check(mci); mutex_unlock(&mem_ctls_mutex); - /* Reschedule */ - queue_delayed_work(edac_workqueue, &mci->work, - msecs_to_jiffies(edac_mc_get_poll_msec())); -} - -/* - * edac_mc_workq_setup - * initialize a workq item for this mci - * passing in the new delay period in msec - * - * locking model: - * - * called with the mem_ctls_mutex held - */ -static void edac_mc_workq_setup(struct mem_ctl_info *mci, unsigned msec) -{ - edac_dbg(0, "\n"); - - /* if this instance is not in the POLL state, then simply return */ - if (mci->op_state != OP_RUNNING_POLL) - return; - - INIT_DELAYED_WORK(&mci->work, edac_mc_workq_function); - mod_delayed_work(edac_workqueue, &mci->work, msecs_to_jiffies(msec)); -} - -/* - * edac_mc_workq_teardown - * stop the workq processing on this mci - * - * locking model: - * - * called WITHOUT lock held - */ -static void edac_mc_workq_teardown(struct mem_ctl_info *mci) -{ - int status; - - if (mci->op_state != OP_RUNNING_POLL) - return; - - status = cancel_delayed_work(&mci->work); - if (status == 0) { - edac_dbg(0, "not canceled, flush the queue\n"); - - /* workq instance might be running, wait for it */ - flush_workqueue(edac_workqueue); - } + /* Queue ourselves again. */ + edac_queue_work(&mci->work, msecs_to_jiffies(edac_mc_get_poll_msec())); } /* @@ -599,7 +493,7 @@ static void edac_mc_workq_teardown(struct mem_ctl_info *mci) * user space has updated our poll period value, need to * reset our workq delays */ -void edac_mc_reset_delay_period(int value) +void edac_mc_reset_delay_period(unsigned long value) { struct mem_ctl_info *mci; struct list_head *item; @@ -609,9 +503,9 @@ void edac_mc_reset_delay_period(int value) list_for_each(item, &mc_devices) { mci = list_entry(item, struct mem_ctl_info, link); - edac_mc_workq_setup(mci, (unsigned long) value); + if (mci->op_state == OP_RUNNING_POLL) + edac_mod_work(&mci->work, value); } - mutex_unlock(&mem_ctls_mutex); } @@ -632,7 +526,7 @@ static int add_mc_to_global_list(struct mem_ctl_info *mci) insert_before = &mc_devices; - p = find_mci_by_dev(mci->pdev); + p = __find_mci_by_dev(mci->pdev); if (unlikely(p != NULL)) goto fail0; @@ -649,7 +543,6 @@ static int add_mc_to_global_list(struct mem_ctl_info *mci) } list_add_tail_rcu(&mci->link, insert_before); - atomic_inc(&edac_handlers); return 0; fail0: @@ -667,7 +560,6 @@ fail1: static int del_mc_from_global_list(struct mem_ctl_info *mci) { - int handlers = atomic_dec_return(&edac_handlers); list_del_rcu(&mci->link); /* these are for safe removal of devices from global list while @@ -676,49 +568,38 @@ static int del_mc_from_global_list(struct mem_ctl_info *mci) synchronize_rcu(); INIT_LIST_HEAD(&mci->link); - return handlers; + return list_empty(&mc_devices); } -/** - * edac_mc_find: Search for a mem_ctl_info structure whose index is 'idx'. - * - * If found, return a pointer to the structure. - * Else return NULL. - * - * Caller must hold mem_ctls_mutex. - */ struct mem_ctl_info *edac_mc_find(int idx) { - struct list_head *item; struct mem_ctl_info *mci; + struct list_head *item; + + mutex_lock(&mem_ctls_mutex); list_for_each(item, &mc_devices) { mci = list_entry(item, struct mem_ctl_info, link); - - if (mci->mc_idx >= idx) { - if (mci->mc_idx == idx) - return mci; - - break; - } + if (mci->mc_idx == idx) + goto unlock; } - return NULL; + mci = NULL; +unlock: + mutex_unlock(&mem_ctls_mutex); + return mci; } EXPORT_SYMBOL(edac_mc_find); -/** - * edac_mc_add_mc: Insert the 'mci' structure into the mci global list and - * create sysfs entries associated with mci structure - * @mci: pointer to the mci structure to be added to the list - * - * Return: - * 0 Success - * !0 Failure - */ +const char *edac_get_owner(void) +{ + return edac_mc_owner; +} +EXPORT_SYMBOL_GPL(edac_get_owner); /* FIXME - should a warning be printed if no error detection? correction? */ -int edac_mc_add_mc(struct mem_ctl_info *mci) +int edac_mc_add_mc_with_groups(struct mem_ctl_info *mci, + const struct attribute_group **groups) { int ret = -EINVAL; edac_dbg(0, "\n"); @@ -728,6 +609,7 @@ int edac_mc_add_mc(struct mem_ctl_info *mci) edac_mc_dump_mci(mci); if (edac_debug_level >= 4) { + struct dimm_info *dimm; int i; for (i = 0; i < mci->nr_csrows; i++) { @@ -744,9 +626,9 @@ int edac_mc_add_mc(struct mem_ctl_info *mci) if (csrow->channels[j]->dimm->nr_pages) edac_mc_dump_channel(csrow->channels[j]); } - for (i = 0; i < mci->tot_dimms; i++) - if (mci->dimms[i]->nr_pages) - edac_mc_dump_dimm(mci->dimms[i], i); + + mci_for_each_dimm(mci, dimm) + edac_mc_dump_dimm(dimm); } #endif mutex_lock(&mem_ctls_mutex); @@ -762,25 +644,29 @@ int edac_mc_add_mc(struct mem_ctl_info *mci) /* set load time so that error rate can be tracked */ mci->start_time = jiffies; - if (edac_create_sysfs_mci_device(mci)) { + mci->bus = edac_get_sysfs_subsys(); + + if (edac_create_sysfs_mci_device(mci, groups)) { edac_mc_printk(mci, KERN_WARNING, "failed to create sysfs device\n"); goto fail1; } - /* If there IS a check routine, then we are running POLLED */ - if (mci->edac_check != NULL) { - /* This instance is NOW RUNNING */ + if (mci->edac_check) { mci->op_state = OP_RUNNING_POLL; - edac_mc_workq_setup(mci, edac_mc_get_poll_msec()); + INIT_DELAYED_WORK(&mci->work, edac_mc_workq_function); + edac_queue_work(&mci->work, msecs_to_jiffies(edac_mc_get_poll_msec())); + } else { mci->op_state = OP_RUNNING_INTERRUPT; } /* Report action taken */ - edac_mc_printk(mci, KERN_INFO, "Giving out device to '%s' '%s':" - " DEV %s\n", mci->mod_name, mci->ctl_name, edac_dev_name(mci)); + edac_mc_printk(mci, KERN_INFO, + "Giving out device to module %s controller %s: DEV %s (%s)\n", + mci->mod_name, mci->ctl_name, mci->dev_name, + edac_op_state_to_string(mci->op_state)); edac_mc_owner = mci->mod_name; @@ -794,15 +680,8 @@ fail0: mutex_unlock(&mem_ctls_mutex); return ret; } -EXPORT_SYMBOL_GPL(edac_mc_add_mc); +EXPORT_SYMBOL_GPL(edac_mc_add_mc_with_groups); -/** - * edac_mc_del_mc: Remove sysfs entries for specified mci structure and - * remove mci structure from global list - * @pdev: Pointer to 'struct device' representing mci structure to remove. - * - * Return pointer to removed mci structure, or NULL if device not found. - */ struct mem_ctl_info *edac_mc_del_mc(struct device *dev) { struct mem_ctl_info *mci; @@ -812,21 +691,22 @@ struct mem_ctl_info *edac_mc_del_mc(struct device *dev) mutex_lock(&mem_ctls_mutex); /* find the requested mci struct in the global list */ - mci = find_mci_by_dev(dev); + mci = __find_mci_by_dev(dev); if (mci == NULL) { mutex_unlock(&mem_ctls_mutex); return NULL; } - if (!del_mc_from_global_list(mci)) + /* mark MCI offline: */ + mci->op_state = OP_OFFLINE; + + if (del_mc_from_global_list(mci)) edac_mc_owner = NULL; - mutex_unlock(&mem_ctls_mutex); - /* flush workq processes */ - edac_mc_workq_teardown(mci); + mutex_unlock(&mem_ctls_mutex); - /* marking MCI offline */ - mci->op_state = OP_OFFLINE; + if (mci->edac_check) + edac_stop_work(&mci->work); /* remove from sysfs */ edac_remove_sysfs_mci_device(mci); @@ -861,7 +741,7 @@ static void edac_mc_scrub_block(unsigned long page, unsigned long offset, virt_addr = kmap_atomic(pg); /* Perform architecture specific atomic scrub operation */ - atomic_scrub(virt_addr + offset, size); + edac_atomic_scrub(virt_addr + offset, size); /* Unmap and complete */ kunmap_atomic(virt_addr); @@ -921,90 +801,53 @@ const char *edac_layer_name[] = { }; EXPORT_SYMBOL_GPL(edac_layer_name); -static void edac_inc_ce_error(struct mem_ctl_info *mci, - bool enable_per_layer_report, - const int pos[EDAC_MAX_LAYERS], - const u16 count) +static void edac_inc_ce_error(struct edac_raw_error_desc *e) { - int i, index = 0; - - mci->ce_mc += count; - - if (!enable_per_layer_report) { - mci->ce_noinfo_count += count; - return; - } + int pos[EDAC_MAX_LAYERS] = { e->top_layer, e->mid_layer, e->low_layer }; + struct mem_ctl_info *mci = error_desc_to_mci(e); + struct dimm_info *dimm = edac_get_dimm(mci, pos[0], pos[1], pos[2]); - for (i = 0; i < mci->n_layers; i++) { - if (pos[i] < 0) - break; - index += pos[i]; - mci->ce_per_layer[i][index] += count; + mci->ce_mc += e->error_count; - if (i < mci->n_layers - 1) - index *= mci->layers[i + 1].size; - } + if (dimm) + dimm->ce_count += e->error_count; + else + mci->ce_noinfo_count += e->error_count; } -static void edac_inc_ue_error(struct mem_ctl_info *mci, - bool enable_per_layer_report, - const int pos[EDAC_MAX_LAYERS], - const u16 count) +static void edac_inc_ue_error(struct edac_raw_error_desc *e) { - int i, index = 0; - - mci->ue_mc += count; - - if (!enable_per_layer_report) { - mci->ce_noinfo_count += count; - return; - } + int pos[EDAC_MAX_LAYERS] = { e->top_layer, e->mid_layer, e->low_layer }; + struct mem_ctl_info *mci = error_desc_to_mci(e); + struct dimm_info *dimm = edac_get_dimm(mci, pos[0], pos[1], pos[2]); - for (i = 0; i < mci->n_layers; i++) { - if (pos[i] < 0) - break; - index += pos[i]; - mci->ue_per_layer[i][index] += count; + mci->ue_mc += e->error_count; - if (i < mci->n_layers - 1) - index *= mci->layers[i + 1].size; - } + if (dimm) + dimm->ue_count += e->error_count; + else + mci->ue_noinfo_count += e->error_count; } -static void edac_ce_error(struct mem_ctl_info *mci, - const u16 error_count, - const int pos[EDAC_MAX_LAYERS], - const char *msg, - const char *location, - const char *label, - const char *detail, - const char *other_detail, - const bool enable_per_layer_report, - const unsigned long page_frame_number, - const unsigned long offset_in_page, - long grain) +static void edac_ce_error(struct edac_raw_error_desc *e) { + struct mem_ctl_info *mci = error_desc_to_mci(e); unsigned long remapped_page; - char *msg_aux = ""; - - if (*msg) - msg_aux = " "; if (edac_mc_get_log_ce()) { - if (other_detail && *other_detail) - edac_mc_printk(mci, KERN_WARNING, - "%d CE %s%son %s (%s %s - %s)\n", - error_count, msg, msg_aux, label, - location, detail, other_detail); - else - edac_mc_printk(mci, KERN_WARNING, - "%d CE %s%son %s (%s %s)\n", - error_count, msg, msg_aux, label, - location, detail); + edac_mc_printk(mci, KERN_WARNING, + "%d CE %s%son %s (%s page:0x%lx offset:0x%lx grain:%ld syndrome:0x%lx%s%s)\n", + e->error_count, e->msg, + *e->msg ? " " : "", + e->label, e->location, e->page_frame_number, e->offset_in_page, + e->grain, e->syndrome, + *e->other_detail ? " - " : "", + e->other_detail); } - edac_inc_ce_error(mci, enable_per_layer_report, pos, error_count); - if (mci->scrub_mode & SCRUB_SW_SRC) { + edac_inc_ce_error(e); + + if (mci->scrub_mode == SCRUB_SW_SRC) { /* * Some memory controllers (called MCs below) can remap * memory so that it is still available at a different @@ -1017,113 +860,87 @@ static void edac_ce_error(struct mem_ctl_info *mci, * be scrubbed. */ remapped_page = mci->ctl_page_to_phys ? - mci->ctl_page_to_phys(mci, page_frame_number) : - page_frame_number; + mci->ctl_page_to_phys(mci, e->page_frame_number) : + e->page_frame_number; - edac_mc_scrub_block(remapped_page, - offset_in_page, grain); + edac_mc_scrub_block(remapped_page, e->offset_in_page, e->grain); } } -static void edac_ue_error(struct mem_ctl_info *mci, - const u16 error_count, - const int pos[EDAC_MAX_LAYERS], - const char *msg, - const char *location, - const char *label, - const char *detail, - const char *other_detail, - const bool enable_per_layer_report) +static void edac_ue_error(struct edac_raw_error_desc *e) { - char *msg_aux = ""; - - if (*msg) - msg_aux = " "; + struct mem_ctl_info *mci = error_desc_to_mci(e); if (edac_mc_get_log_ue()) { - if (other_detail && *other_detail) - edac_mc_printk(mci, KERN_WARNING, - "%d UE %s%son %s (%s %s - %s)\n", - error_count, msg, msg_aux, label, - location, detail, other_detail); - else - edac_mc_printk(mci, KERN_WARNING, - "%d UE %s%son %s (%s %s)\n", - error_count, msg, msg_aux, label, - location, detail); + edac_mc_printk(mci, KERN_WARNING, + "%d UE %s%son %s (%s page:0x%lx offset:0x%lx grain:%ld%s%s)\n", + e->error_count, e->msg, + *e->msg ? " " : "", + e->label, e->location, e->page_frame_number, e->offset_in_page, + e->grain, + *e->other_detail ? " - " : "", + e->other_detail); } + edac_inc_ue_error(e); + if (edac_mc_get_panic_on_ue()) { - if (other_detail && *other_detail) - panic("UE %s%son %s (%s%s - %s)\n", - msg, msg_aux, label, location, detail, other_detail); - else - panic("UE %s%son %s (%s%s)\n", - msg, msg_aux, label, location, detail); + panic("UE %s%son %s (%s page:0x%lx offset:0x%lx grain:%ld%s%s)\n", + e->msg, + *e->msg ? " " : "", + e->label, e->location, e->page_frame_number, e->offset_in_page, + e->grain, + *e->other_detail ? " - " : "", + e->other_detail); } - - edac_inc_ue_error(mci, enable_per_layer_report, pos, error_count); } -/** - * edac_raw_mc_handle_error - reports a memory event to userspace without doing - * anything to discover the error location - * - * @type: severity of the error (CE/UE/Fatal) - * @mci: a struct mem_ctl_info pointer - * @e: error description - * - * This raw function is used internally by edac_mc_handle_error(). It should - * only be called directly when the hardware error come directly from BIOS, - * like in the case of APEI GHES driver. - */ -void edac_raw_mc_handle_error(const enum hw_event_mc_err_type type, - struct mem_ctl_info *mci, - struct edac_raw_error_desc *e) +static void edac_inc_csrow(struct edac_raw_error_desc *e, int row, int chan) { - char detail[80]; - int pos[EDAC_MAX_LAYERS] = { e->top_layer, e->mid_layer, e->low_layer }; + struct mem_ctl_info *mci = error_desc_to_mci(e); + enum hw_event_mc_err_type type = e->type; + u16 count = e->error_count; + + if (row < 0) + return; + + edac_dbg(4, "csrow/channel to increment: (%d,%d)\n", row, chan); - /* Memory type dependent details about the error */ if (type == HW_EVENT_ERR_CORRECTED) { - snprintf(detail, sizeof(detail), - "page:0x%lx offset:0x%lx grain:%ld syndrome:0x%lx", - e->page_frame_number, e->offset_in_page, - e->grain, e->syndrome); - edac_ce_error(mci, e->error_count, pos, e->msg, e->location, e->label, - detail, e->other_detail, e->enable_per_layer_report, - e->page_frame_number, e->offset_in_page, e->grain); + mci->csrows[row]->ce_count += count; + if (chan >= 0) + mci->csrows[row]->channels[chan]->ce_count += count; } else { - snprintf(detail, sizeof(detail), - "page:0x%lx offset:0x%lx grain:%ld", - e->page_frame_number, e->offset_in_page, e->grain); - - edac_ue_error(mci, e->error_count, pos, e->msg, e->location, e->label, - detail, e->other_detail, e->enable_per_layer_report); + mci->csrows[row]->ue_count += count; } +} +void edac_raw_mc_handle_error(struct edac_raw_error_desc *e) +{ + struct mem_ctl_info *mci = error_desc_to_mci(e); + u8 grain_bits; + /* Sanity-check driver-supplied grain value. */ + if (WARN_ON_ONCE(!e->grain)) + e->grain = 1; + + grain_bits = fls_long(e->grain - 1); + + /* Report the error via the trace interface */ + if (IS_ENABLED(CONFIG_RAS)) + trace_mc_event(e->type, e->msg, e->label, e->error_count, + mci->mc_idx, e->top_layer, e->mid_layer, + e->low_layer, + (e->page_frame_number << PAGE_SHIFT) | e->offset_in_page, + grain_bits, e->syndrome, e->other_detail); + + if (e->type == HW_EVENT_ERR_CORRECTED) + edac_ce_error(e); + else + edac_ue_error(e); } EXPORT_SYMBOL_GPL(edac_raw_mc_handle_error); -/** - * edac_mc_handle_error - reports a memory event to userspace - * - * @type: severity of the error (CE/UE/Fatal) - * @mci: a struct mem_ctl_info pointer - * @error_count: Number of errors of the same type - * @page_frame_number: mem page where the error occurred - * @offset_in_page: offset of the error inside the page - * @syndrome: ECC syndrome - * @top_layer: Memory layer[0] position - * @mid_layer: Memory layer[1] position - * @low_layer: Memory layer[2] position - * @msg: Message meaningful to the end users that - * explains the event - * @other_detail: Technical details about the event that - * may help hardware manufacturers and - * EDAC developers to analyse the event - */ void edac_mc_handle_error(const enum hw_event_mc_err_type type, struct mem_ctl_info *mci, const u16 error_count, @@ -1136,31 +953,34 @@ void edac_mc_handle_error(const enum hw_event_mc_err_type type, const char *msg, const char *other_detail) { - char *p; + struct dimm_info *dimm; + char *p, *end; int row = -1, chan = -1; int pos[EDAC_MAX_LAYERS] = { top_layer, mid_layer, low_layer }; int i, n_labels = 0; - u8 grain_bits; struct edac_raw_error_desc *e = &mci->error_desc; + bool any_memory = true; + const char *prefix; edac_dbg(3, "MC%d\n", mci->mc_idx); /* Fills the error report buffer */ memset(e, 0, sizeof (*e)); e->error_count = error_count; + e->type = type; e->top_layer = top_layer; e->mid_layer = mid_layer; e->low_layer = low_layer; e->page_frame_number = page_frame_number; e->offset_in_page = offset_in_page; e->syndrome = syndrome; - e->msg = msg; - e->other_detail = other_detail; + /* need valid strings here for both: */ + e->msg = msg ?: ""; + e->other_detail = other_detail ?: ""; /* - * Check if the event report is consistent and if the memory - * location is known. If it is known, enable_per_layer_report will be - * true, the DIMM(s) label info will be filled and the per-layer + * Check if the event report is consistent and if the memory location is + * known. If it is, the DIMM(s) label info will be filled and the DIMM's * error counters will be incremented. */ for (i = 0; i < mci->n_layers; i++) { @@ -1179,7 +999,7 @@ void edac_mc_handle_error(const enum hw_event_mc_err_type type, pos[i] = -1; } if (pos[i] >= 0) - e->enable_per_layer_report = true; + any_memory = false; } /* @@ -1195,10 +1015,10 @@ void edac_mc_handle_error(const enum hw_event_mc_err_type type, */ p = e->label; *p = '\0'; + end = p + sizeof(e->label); + prefix = ""; - for (i = 0; i < mci->tot_dimms; i++) { - struct dimm_info *dimm = mci->dimms[i]; - + mci_for_each_dimm(mci, dimm) { if (top_layer >= 0 && top_layer != dimm->location[0]) continue; if (mid_layer >= 0 && mid_layer != dimm->location[1]) @@ -1212,81 +1032,61 @@ void edac_mc_handle_error(const enum hw_event_mc_err_type type, /* * If the error is memory-controller wide, there's no need to - * seek for the affected DIMMs because the whole - * channel/memory controller/... may be affected. - * Also, don't show errors for empty DIMM slots. + * seek for the affected DIMMs because the whole channel/memory + * controller/... may be affected. Also, don't show errors for + * empty DIMM slots. */ - if (e->enable_per_layer_report && dimm->nr_pages) { - if (n_labels >= EDAC_MAX_LABELS) { - e->enable_per_layer_report = false; - break; - } - n_labels++; - if (p != e->label) { - strcpy(p, OTHER_LABEL); - p += strlen(OTHER_LABEL); - } - strcpy(p, dimm->label); - p += strlen(p); - *p = '\0'; + if (!dimm->nr_pages) + continue; - /* - * get csrow/channel of the DIMM, in order to allow - * incrementing the compat API counters - */ - edac_dbg(4, "%s csrows map: (%d,%d)\n", - mci->csbased ? "rank" : "dimm", - dimm->csrow, dimm->cschannel); - if (row == -1) - row = dimm->csrow; - else if (row >= 0 && row != dimm->csrow) - row = -2; - - if (chan == -1) - chan = dimm->cschannel; - else if (chan >= 0 && chan != dimm->cschannel) - chan = -2; + n_labels++; + if (n_labels > EDAC_MAX_LABELS) { + p = e->label; + *p = '\0'; + } else { + p += scnprintf(p, end - p, "%s%s", prefix, dimm->label); + prefix = OTHER_LABEL; } - } - if (!e->enable_per_layer_report) { - strcpy(e->label, "any memory"); - } else { - edac_dbg(4, "csrow/channel to increment: (%d,%d)\n", row, chan); - if (p == e->label) - strcpy(e->label, "unknown memory"); - if (type == HW_EVENT_ERR_CORRECTED) { - if (row >= 0) { - mci->csrows[row]->ce_count += error_count; - if (chan >= 0) - mci->csrows[row]->channels[chan]->ce_count += error_count; - } - } else - if (row >= 0) - mci->csrows[row]->ue_count += error_count; + /* + * get csrow/channel of the DIMM, in order to allow + * incrementing the compat API counters + */ + edac_dbg(4, "%s csrows map: (%d,%d)\n", + mci->csbased ? "rank" : "dimm", + dimm->csrow, dimm->cschannel); + if (row == -1) + row = dimm->csrow; + else if (row >= 0 && row != dimm->csrow) + row = -2; + + if (chan == -1) + chan = dimm->cschannel; + else if (chan >= 0 && chan != dimm->cschannel) + chan = -2; } + if (any_memory) + strscpy(e->label, "any memory", sizeof(e->label)); + else if (!*e->label) + strscpy(e->label, "unknown memory", sizeof(e->label)); + + edac_inc_csrow(e, row, chan); + /* Fill the RAM location data */ p = e->location; + end = p + sizeof(e->location); + prefix = ""; for (i = 0; i < mci->n_layers; i++) { if (pos[i] < 0) continue; - p += sprintf(p, "%s:%d ", - edac_layer_name[mci->layers[i].type], - pos[i]); + p += scnprintf(p, end - p, "%s%s:%d", prefix, + edac_layer_name[mci->layers[i].type], pos[i]); + prefix = " "; } - if (p > e->location) - *(p - 1) = '\0'; - - /* Report the error via the trace interface */ - grain_bits = fls_long(e->grain) + 1; - trace_mc_event(type, e->msg, e->label, e->error_count, - mci->mc_idx, e->top_layer, e->mid_layer, e->low_layer, - PAGES_TO_MiB(e->page_frame_number) | e->offset_in_page, - grain_bits, e->syndrome, e->other_detail); - edac_raw_mc_handle_error(type, mci, e); + edac_raw_mc_handle_error(e); } EXPORT_SYMBOL_GPL(edac_mc_handle_error); |
