EDAC/ghes: Scan the system once on driver init

Change the hardware scanning and figuring out how many DIMMs a machine
has to a single, one-time thing which happens once on driver init. After
that scanning completes, struct ghes_hw_desc contains a representation
of the hardware which the driver can then use for later initialization.

Then, copy the DIMM information into the respective EDAC core
representation of those.

Get rid of ghes_edac_dimm_fill and use a struct dimm_info array
directly.

This way, hw detection and further driver initialization is nicely
and logically split. Further additions should all be added to
ghes_scan_system() and the hw representation extended as needed.

There should be no functionality change resulting from this patch.

Signed-off-by: Borislav Petkov <bp@suse.de>
This commit is contained in:
Borislav Petkov 2020-06-03 21:19:21 +02:00
parent b001694d60
commit b9cae27728
1 changed files with 165 additions and 112 deletions

View File

@ -32,6 +32,15 @@ static refcount_t ghes_refcount = REFCOUNT_INIT(0);
*/
static struct ghes_pvt *ghes_pvt;
/*
* This driver's representation of the system hardware, as collected
* from DMI.
*/
struct ghes_hw_desc {
int num_dimms;
struct dimm_info *dimms;
} ghes_hw;
/* GHES registration mutex */
static DEFINE_MUTEX(ghes_reg_mutex);
@ -72,19 +81,6 @@ struct memdev_dmi_entry {
u16 conf_mem_clk_speed;
} __attribute__((__packed__));
struct ghes_edac_dimm_fill {
struct mem_ctl_info *mci;
unsigned int count;
};
static void ghes_edac_count_dimms(const struct dmi_header *dh, void *arg)
{
int *num_dimm = arg;
if (dh->type == DMI_ENTRY_MEM_DEVICE)
(*num_dimm)++;
}
static struct dimm_info *find_dimm_by_handle(struct mem_ctl_info *mci, u16 handle)
{
struct dimm_info *dimm;
@ -108,102 +104,135 @@ static void dimm_setup_label(struct dimm_info *dimm, u16 handle)
snprintf(dimm->label, sizeof(dimm->label), "%s %s", bank, device);
}
static void ghes_edac_dmidecode(const struct dmi_header *dh, void *arg)
static void assign_dmi_dimm_info(struct dimm_info *dimm, struct memdev_dmi_entry *entry)
{
struct ghes_edac_dimm_fill *dimm_fill = arg;
struct mem_ctl_info *mci = dimm_fill->mci;
u16 rdr_mask = BIT(7) | BIT(13);
if (dh->type == DMI_ENTRY_MEM_DEVICE) {
struct memdev_dmi_entry *entry = (struct memdev_dmi_entry *)dh;
struct dimm_info *dimm = edac_get_dimm(mci, dimm_fill->count, 0, 0);
u16 rdr_mask = BIT(7) | BIT(13);
if (entry->size == 0xffff) {
pr_info("Can't get DIMM%i size\n",
dimm_fill->count);
dimm->nr_pages = MiB_TO_PAGES(32);/* Unknown */
} else if (entry->size == 0x7fff) {
dimm->nr_pages = MiB_TO_PAGES(entry->extended_size);
} else {
if (entry->size & BIT(15))
dimm->nr_pages = MiB_TO_PAGES((entry->size & 0x7fff) << 10);
else
dimm->nr_pages = MiB_TO_PAGES(entry->size);
}
switch (entry->memory_type) {
case 0x12:
if (entry->type_detail & BIT(13))
dimm->mtype = MEM_RDDR;
else
dimm->mtype = MEM_DDR;
break;
case 0x13:
if (entry->type_detail & BIT(13))
dimm->mtype = MEM_RDDR2;
else
dimm->mtype = MEM_DDR2;
break;
case 0x14:
dimm->mtype = MEM_FB_DDR2;
break;
case 0x18:
if (entry->type_detail & BIT(12))
dimm->mtype = MEM_NVDIMM;
else if (entry->type_detail & BIT(13))
dimm->mtype = MEM_RDDR3;
else
dimm->mtype = MEM_DDR3;
break;
case 0x1a:
if (entry->type_detail & BIT(12))
dimm->mtype = MEM_NVDIMM;
else if (entry->type_detail & BIT(13))
dimm->mtype = MEM_RDDR4;
else
dimm->mtype = MEM_DDR4;
break;
default:
if (entry->type_detail & BIT(6))
dimm->mtype = MEM_RMBS;
else if ((entry->type_detail & rdr_mask) == rdr_mask)
dimm->mtype = MEM_RDR;
else if (entry->type_detail & BIT(7))
dimm->mtype = MEM_SDR;
else if (entry->type_detail & BIT(9))
dimm->mtype = MEM_EDO;
else
dimm->mtype = MEM_UNKNOWN;
}
/*
* Actually, we can only detect if the memory has bits for
* checksum or not
*/
if (entry->total_width == entry->data_width)
dimm->edac_mode = EDAC_NONE;
if (entry->size == 0xffff) {
pr_info("Can't get DIMM%i size\n", dimm->idx);
dimm->nr_pages = MiB_TO_PAGES(32);/* Unknown */
} else if (entry->size == 0x7fff) {
dimm->nr_pages = MiB_TO_PAGES(entry->extended_size);
} else {
if (entry->size & BIT(15))
dimm->nr_pages = MiB_TO_PAGES((entry->size & 0x7fff) << 10);
else
dimm->edac_mode = EDAC_SECDED;
dimm->nr_pages = MiB_TO_PAGES(entry->size);
}
dimm->dtype = DEV_UNKNOWN;
dimm->grain = 128; /* Likely, worse case */
switch (entry->memory_type) {
case 0x12:
if (entry->type_detail & BIT(13))
dimm->mtype = MEM_RDDR;
else
dimm->mtype = MEM_DDR;
break;
case 0x13:
if (entry->type_detail & BIT(13))
dimm->mtype = MEM_RDDR2;
else
dimm->mtype = MEM_DDR2;
break;
case 0x14:
dimm->mtype = MEM_FB_DDR2;
break;
case 0x18:
if (entry->type_detail & BIT(12))
dimm->mtype = MEM_NVDIMM;
else if (entry->type_detail & BIT(13))
dimm->mtype = MEM_RDDR3;
else
dimm->mtype = MEM_DDR3;
break;
case 0x1a:
if (entry->type_detail & BIT(12))
dimm->mtype = MEM_NVDIMM;
else if (entry->type_detail & BIT(13))
dimm->mtype = MEM_RDDR4;
else
dimm->mtype = MEM_DDR4;
break;
default:
if (entry->type_detail & BIT(6))
dimm->mtype = MEM_RMBS;
else if ((entry->type_detail & rdr_mask) == rdr_mask)
dimm->mtype = MEM_RDR;
else if (entry->type_detail & BIT(7))
dimm->mtype = MEM_SDR;
else if (entry->type_detail & BIT(9))
dimm->mtype = MEM_EDO;
else
dimm->mtype = MEM_UNKNOWN;
}
dimm_setup_label(dimm, entry->handle);
/*
* Actually, we can only detect if the memory has bits for
* checksum or not
*/
if (entry->total_width == entry->data_width)
dimm->edac_mode = EDAC_NONE;
else
dimm->edac_mode = EDAC_SECDED;
if (dimm->nr_pages) {
edac_dbg(1, "DIMM%i: %s size = %d MB%s\n",
dimm_fill->count, edac_mem_types[dimm->mtype],
PAGES_TO_MiB(dimm->nr_pages),
(dimm->edac_mode != EDAC_NONE) ? "(ECC)" : "");
edac_dbg(2, "\ttype %d, detail 0x%02x, width %d(total %d)\n",
entry->memory_type, entry->type_detail,
entry->total_width, entry->data_width);
dimm->dtype = DEV_UNKNOWN;
dimm->grain = 128; /* Likely, worse case */
dimm_setup_label(dimm, entry->handle);
if (dimm->nr_pages) {
edac_dbg(1, "DIMM%i: %s size = %d MB%s\n",
dimm->idx, edac_mem_types[dimm->mtype],
PAGES_TO_MiB(dimm->nr_pages),
(dimm->edac_mode != EDAC_NONE) ? "(ECC)" : "");
edac_dbg(2, "\ttype %d, detail 0x%02x, width %d(total %d)\n",
entry->memory_type, entry->type_detail,
entry->total_width, entry->data_width);
}
dimm->smbios_handle = entry->handle;
}
static void enumerate_dimms(const struct dmi_header *dh, void *arg)
{
struct memdev_dmi_entry *entry = (struct memdev_dmi_entry *)dh;
struct ghes_hw_desc *hw = (struct ghes_hw_desc *)arg;
struct dimm_info *d;
if (dh->type != DMI_ENTRY_MEM_DEVICE)
return;
/* Enlarge the array with additional 16 */
if (!hw->num_dimms || !(hw->num_dimms % 16)) {
struct dimm_info *new;
new = krealloc(hw->dimms, (hw->num_dimms + 16) * sizeof(struct dimm_info),
GFP_KERNEL);
if (!new) {
WARN_ON_ONCE(1);
return;
}
dimm->smbios_handle = entry->handle;
dimm_fill->count++;
hw->dimms = new;
}
d = &hw->dimms[hw->num_dimms];
d->idx = hw->num_dimms;
assign_dmi_dimm_info(d, entry);
hw->num_dimms++;
}
static void ghes_scan_system(void)
{
static bool scanned;
if (scanned)
return;
dmi_walk(enumerate_dimms, &ghes_hw);
scanned = true;
}
void ghes_edac_report_mem_error(int sev, struct cper_sec_mem_err *mem_err)
@ -466,13 +495,12 @@ static struct acpi_platform_list plat_list[] = {
int ghes_edac_register(struct ghes *ghes, struct device *dev)
{
bool fake = false;
int rc = 0, num_dimm = 0;
struct mem_ctl_info *mci;
struct ghes_pvt *pvt;
struct edac_mc_layer layers[1];
struct ghes_edac_dimm_fill dimm_fill;
unsigned long flags;
int idx = -1;
int rc = 0;
if (IS_ENABLED(CONFIG_X86)) {
/* Check if safe to enable on this system */
@ -492,17 +520,16 @@ int ghes_edac_register(struct ghes *ghes, struct device *dev)
if (refcount_inc_not_zero(&ghes_refcount))
goto unlock;
/* Get the number of DIMMs */
dmi_walk(ghes_edac_count_dimms, &num_dimm);
ghes_scan_system();
/* Check if we've got a bogus BIOS */
if (num_dimm == 0) {
if (!ghes_hw.num_dimms) {
fake = true;
num_dimm = 1;
ghes_hw.num_dimms = 1;
}
layers[0].type = EDAC_MC_LAYER_ALL_MEM;
layers[0].size = num_dimm;
layers[0].size = ghes_hw.num_dimms;
layers[0].is_virt_csrow = true;
mci = edac_mc_alloc(0, ARRAY_SIZE(layers), layers, sizeof(struct ghes_pvt));
@ -533,13 +560,34 @@ int ghes_edac_register(struct ghes *ghes, struct device *dev)
pr_info("So, the end result of using this driver varies from vendor to vendor.\n");
pr_info("If you find incorrect reports, please contact your hardware vendor\n");
pr_info("to correct its BIOS.\n");
pr_info("This system has %d DIMM sockets.\n", num_dimm);
pr_info("This system has %d DIMM sockets.\n", ghes_hw.num_dimms);
}
if (!fake) {
dimm_fill.count = 0;
dimm_fill.mci = mci;
dmi_walk(ghes_edac_dmidecode, &dimm_fill);
struct dimm_info *src, *dst;
int i = 0;
mci_for_each_dimm(mci, dst) {
src = &ghes_hw.dimms[i];
dst->idx = src->idx;
dst->smbios_handle = src->smbios_handle;
dst->nr_pages = src->nr_pages;
dst->mtype = src->mtype;
dst->edac_mode = src->edac_mode;
dst->dtype = src->dtype;
dst->grain = src->grain;
/*
* If no src->label, preserve default label assigned
* from EDAC core.
*/
if (strlen(src->label))
memcpy(dst->label, src->label, sizeof(src->label));
i++;
}
} else {
struct dimm_info *dimm = edac_get_dimm(mci, 0, 0, 0);
@ -552,7 +600,7 @@ int ghes_edac_register(struct ghes *ghes, struct device *dev)
rc = edac_mc_add_mc(mci);
if (rc < 0) {
pr_info("Can't register at EDAC core\n");
pr_info("Can't register with the EDAC core\n");
edac_mc_free(mci);
rc = -ENODEV;
goto unlock;
@ -566,6 +614,11 @@ int ghes_edac_register(struct ghes *ghes, struct device *dev)
refcount_set(&ghes_refcount, 1);
unlock:
/* Not needed anymore */
kfree(ghes_hw.dimms);
ghes_hw.dimms = NULL;
mutex_unlock(&ghes_reg_mutex);
return rc;