From 6c13d7ff81e6d2f01f62ccbfa49d1b8d87f274d0 Mon Sep 17 00:00:00 2001
From: Borislav Petkov <bp@suse.de>
Date: Sat, 12 Dec 2020 15:20:28 +0100
Subject: EDAC/amd64: Do not load on family 0x15, model 0x13
Those were only laptops and are very very unlikely to have ECC memory.
Currently, when the driver attempts to load, it issues:
EDAC amd64: Error: F1 not found: device 0x1601 (broken BIOS?)
because the PCI device is the wrong one (it uses the F15h default one).
So do not load the driver on them as that is pointless.
Reported-by: Don Curtis <bugrprt21882@online.de>
Signed-off-by: Borislav Petkov <bp@suse.de>
Tested-by: Don Curtis <bugrprt21882@online.de>
Link: http://bugzilla.opensuse.org/show_bug.cgi?id=1179763
Link: https://lkml.kernel.org/r/20201218160622.20146-1-bp@alien8.de
---
drivers/edac/amd64_edac.c | 10 +++++++---
1 file changed, 7 insertions(+), 3 deletions(-)
diff --git a/drivers/edac/amd64_edac.c b/drivers/edac/amd64_edac.c
index f7087ddddb90..5754f429a8d2 100644
--- a/drivers/edac/amd64_edac.c
+++ b/drivers/edac/amd64_edac.c
@@ -3342,10 +3342,13 @@ static struct amd64_family_type *per_family_init(struct amd64_pvt *pvt)
fam_type = &family_types[F15_M60H_CPUS];
pvt->ops = &family_types[F15_M60H_CPUS].ops;
break;
+ /* Richland is only client */
+ } else if (pvt->model == 0x13) {
+ return NULL;
+ } else {
+ fam_type = &family_types[F15_CPUS];
+ pvt->ops = &family_types[F15_CPUS].ops;
}
-
- fam_type = &family_types[F15_CPUS];
- pvt->ops = &family_types[F15_CPUS].ops;
break;
case 0x16:
@@ -3539,6 +3542,7 @@ static int probe_one_instance(unsigned int nid)
pvt->mc_node_id = nid;
pvt->F3 = F3;
+ ret = -ENODEV;
fam_type = per_family_init(pvt);
if (!fam_type)
goto err_enable;
--
cgit v1.2.3
From 6a4afe38787d7556948b026652a6770e46ce0f6d Mon Sep 17 00:00:00 2001
From: Yazen Ghannam <yazen.ghannam@amd.com>
Date: Tue, 15 Dec 2020 17:01:31 +0000
Subject: EDAC/amd64: Tone down messages about missing PCI IDs
Give these messages a debug severity as they are really only useful to
the module developers.
Also, drop the "(broken BIOS?)" phrase, since this can cause churn for
BIOS folks. The PCI IDs needed by the module, at least on modern systems,
are fixed in hardware.
Signed-off-by: Yazen Ghannam <yazen.ghannam@amd.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Link: https://lkml.kernel.org/r/20201215170131.8496-1-Yazen.Ghannam@amd.com
---
drivers/edac/amd64_edac.c | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/drivers/edac/amd64_edac.c b/drivers/edac/amd64_edac.c
index 5754f429a8d2..543221e3fe9b 100644
--- a/drivers/edac/amd64_edac.c
+++ b/drivers/edac/amd64_edac.c
@@ -2665,7 +2665,7 @@ reserve_mc_sibling_devs(struct amd64_pvt *pvt, u16 pci_id1, u16 pci_id2)
if (pvt->umc) {
pvt->F0 = pci_get_related_function(pvt->F3->vendor, pci_id1, pvt->F3);
if (!pvt->F0) {
- amd64_err("F0 not found, device 0x%x (broken BIOS?)\n", pci_id1);
+ edac_dbg(1, "F0 not found, device 0x%x\n", pci_id1);
return -ENODEV;
}
@@ -2674,7 +2674,7 @@ reserve_mc_sibling_devs(struct amd64_pvt *pvt, u16 pci_id1, u16 pci_id2)
pci_dev_put(pvt->F0);
pvt->F0 = NULL;
- amd64_err("F6 not found: device 0x%x (broken BIOS?)\n", pci_id2);
+ edac_dbg(1, "F6 not found: device 0x%x\n", pci_id2);
return -ENODEV;
}
@@ -2691,7 +2691,7 @@ reserve_mc_sibling_devs(struct amd64_pvt *pvt, u16 pci_id1, u16 pci_id2)
/* Reserve the ADDRESS MAP Device */
pvt->F1 = pci_get_related_function(pvt->F3->vendor, pci_id1, pvt->F3);
if (!pvt->F1) {
- amd64_err("F1 not found: device 0x%x (broken BIOS?)\n", pci_id1);
+ edac_dbg(1, "F1 not found: device 0x%x\n", pci_id1);
return -ENODEV;
}
@@ -2701,7 +2701,7 @@ reserve_mc_sibling_devs(struct amd64_pvt *pvt, u16 pci_id1, u16 pci_id2)
pci_dev_put(pvt->F1);
pvt->F1 = NULL;
- amd64_err("F2 not found: device 0x%x (broken BIOS?)\n", pci_id2);
+ edac_dbg(1, "F2 not found: device 0x%x\n", pci_id2);
return -ENODEV;
}
--
cgit v1.2.3
From 2a28ceef00bac65d6bb1757002f742806837e100 Mon Sep 17 00:00:00 2001
From: Borislav Petkov <bp@suse.de>
Date: Mon, 14 Dec 2020 20:47:11 +0100
Subject: EDAC/amd64: Merge sysfs debugging attributes setup code
There's no need for them to be in a separate file so merge them into the
main driver compilation unit like the other EDAC drivers do.
Drop now-unneeded function export, make the function static and shorten
static function names.
No functional changes.
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Yazen Ghannam <yazen.ghannam@amd.com>
Link: https://lkml.kernel.org/r/20201215110517.5215-1-bp@alien8.de
---
drivers/edac/Makefile | 1 -
drivers/edac/amd64_edac.c | 65 +++++++++++++++++++++++++++++++++++++++----
drivers/edac/amd64_edac.h | 7 -----
drivers/edac/amd64_edac_dbg.c | 55 ------------------------------------
4 files changed, 59 insertions(+), 69 deletions(-)
delete mode 100644 drivers/edac/amd64_edac_dbg.c
diff --git a/drivers/edac/Makefile b/drivers/edac/Makefile
index 464d3d8d850a..1c70cdcf2b15 100644
--- a/drivers/edac/Makefile
+++ b/drivers/edac/Makefile
@@ -46,7 +46,6 @@ obj-$(CONFIG_EDAC_I82860) += i82860_edac.o
obj-$(CONFIG_EDAC_R82600) += r82600_edac.o
amd64_edac_mod-y := amd64_edac.o
-amd64_edac_mod-$(CONFIG_EDAC_DEBUG) += amd64_edac_dbg.o
amd64_edac_mod-$(CONFIG_EDAC_AMD64_ERROR_INJECTION) += amd64_edac_inj.o
obj-$(CONFIG_EDAC_AMD64) += amd64_edac_mod.o
diff --git a/drivers/edac/amd64_edac.c b/drivers/edac/amd64_edac.c
index 543221e3fe9b..b00dea78541d 100644
--- a/drivers/edac/amd64_edac.c
+++ b/drivers/edac/amd64_edac.c
@@ -500,8 +500,8 @@ static int input_addr_to_csrow(struct mem_ctl_info *mci, u64 input_addr)
* complete 32-bit values despite the fact that the bitfields in the DHAR
* only represent bits 31-24 of the base and offset values.
*/
-int amd64_get_dram_hole_info(struct mem_ctl_info *mci, u64 *hole_base,
- u64 *hole_offset, u64 *hole_size)
+static int get_dram_hole_info(struct mem_ctl_info *mci, u64 *hole_base,
+ u64 *hole_offset, u64 *hole_size)
{
struct amd64_pvt *pvt = mci->pvt_info;
@@ -554,7 +554,61 @@ int amd64_get_dram_hole_info(struct mem_ctl_info *mci, u64 *hole_base,
return 0;
}
-EXPORT_SYMBOL_GPL(amd64_get_dram_hole_info);
+
+#ifdef CONFIG_EDAC_DEBUG
+#define EDAC_DCT_ATTR_SHOW(reg) \
+static ssize_t reg##_show(struct device *dev, \
+ struct device_attribute *mattr, char *data) \
+{ \
+ struct mem_ctl_info *mci = to_mci(dev); \
+ struct amd64_pvt *pvt = mci->pvt_info; \
+ \
+ return sprintf(data, "0x%016llx\n", (u64)pvt->reg); \
+}
+
+EDAC_DCT_ATTR_SHOW(dhar);
+EDAC_DCT_ATTR_SHOW(dbam0);
+EDAC_DCT_ATTR_SHOW(top_mem);
+EDAC_DCT_ATTR_SHOW(top_mem2);
+
+static ssize_t hole_show(struct device *dev, struct device_attribute *mattr,
+ char *data)
+{
+ struct mem_ctl_info *mci = to_mci(dev);
+
+ u64 hole_base = 0;
+ u64 hole_offset = 0;
+ u64 hole_size = 0;
+
+ get_dram_hole_info(mci, &hole_base, &hole_offset, &hole_size);
+
+ return sprintf(data, "%llx %llx %llx\n", hole_base, hole_offset,
+ hole_size);
+}
+
+/*
+ * update NUM_DBG_ATTRS in case you add new members
+ */
+static DEVICE_ATTR(dhar, S_IRUGO, dhar_show, NULL);
+static DEVICE_ATTR(dbam, S_IRUGO, dbam0_show, NULL);
+static DEVICE_ATTR(topmem, S_IRUGO, top_mem_show, NULL);
+static DEVICE_ATTR(topmem2, S_IRUGO, top_mem2_show, NULL);
+static DEVICE_ATTR(dram_hole, S_IRUGO, hole_show, NULL);
+
+static struct attribute *dbg_attrs[] = {
+ &dev_attr_dhar.attr,
+ &dev_attr_dbam.attr,
+ &dev_attr_topmem.attr,
+ &dev_attr_topmem2.attr,
+ &dev_attr_dram_hole.attr,
+ NULL
+};
+
+static const struct attribute_group dbg_group = {
+ .attrs = dbg_attrs,
+};
+#endif /* CONFIG_EDAC_DEBUG */
+
/*
* Return the DramAddr that the SysAddr given by @sys_addr maps to. It is
@@ -593,8 +647,7 @@ static u64 sys_addr_to_dram_addr(struct mem_ctl_info *mci, u64 sys_addr)
dram_base = get_dram_base(pvt, pvt->mc_node_id);
- ret = amd64_get_dram_hole_info(mci, &hole_base, &hole_offset,
- &hole_size);
+ ret = get_dram_hole_info(mci, &hole_base, &hole_offset, &hole_size);
if (!ret) {
if ((sys_addr >= (1ULL << 32)) &&
(sys_addr < ((1ULL << 32) + hole_size))) {
@@ -3415,7 +3468,7 @@ static struct amd64_family_type *per_family_init(struct amd64_pvt *pvt)
static const struct attribute_group *amd64_edac_attr_groups[] = {
#ifdef CONFIG_EDAC_DEBUG
- &amd64_edac_dbg_group,
+ &dbg_group,
#endif
#ifdef CONFIG_EDAC_AMD64_ERROR_INJECTION
&amd64_edac_inj_group,
diff --git a/drivers/edac/amd64_edac.h b/drivers/edac/amd64_edac.h
index 52b5d03eeba0..c072ccd3e7e2 100644
--- a/drivers/edac/amd64_edac.h
+++ b/drivers/edac/amd64_edac.h
@@ -462,10 +462,6 @@ struct ecc_settings {
} flags;
};
-#ifdef CONFIG_EDAC_DEBUG
-extern const struct attribute_group amd64_edac_dbg_group;
-#endif
-
#ifdef CONFIG_EDAC_AMD64_ERROR_INJECTION
extern const struct attribute_group amd64_edac_inj_group;
#endif
@@ -501,9 +497,6 @@ int __amd64_write_pci_cfg_dword(struct pci_dev *pdev, int offset,
#define amd64_write_pci_cfg(pdev, offset, val) \
__amd64_write_pci_cfg_dword(pdev, offset, val, __func__)
-int amd64_get_dram_hole_info(struct mem_ctl_info *mci, u64 *hole_base,
- u64 *hole_offset, u64 *hole_size);
-
#define to_mci(k) container_of(k, struct mem_ctl_info, dev)
/* Injection helpers */
diff --git a/drivers/edac/amd64_edac_dbg.c b/drivers/edac/amd64_edac_dbg.c
deleted file mode 100644
index 393be3351493..000000000000
--- a/drivers/edac/amd64_edac_dbg.c
+++ /dev/null
@@ -1,55 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0
-#include "amd64_edac.h"
-
-#define EDAC_DCT_ATTR_SHOW(reg) \
-static ssize_t amd64_##reg##_show(struct device *dev, \
- struct device_attribute *mattr, \
- char *data) \
-{ \
- struct mem_ctl_info *mci = to_mci(dev); \
- struct amd64_pvt *pvt = mci->pvt_info; \
- return sprintf(data, "0x%016llx\n", (u64)pvt->reg); \
-}
-
-EDAC_DCT_ATTR_SHOW(dhar);
-EDAC_DCT_ATTR_SHOW(dbam0);
-EDAC_DCT_ATTR_SHOW(top_mem);
-EDAC_DCT_ATTR_SHOW(top_mem2);
-
-static ssize_t amd64_hole_show(struct device *dev,
- struct device_attribute *mattr,
- char *data)
-{
- struct mem_ctl_info *mci = to_mci(dev);
-
- u64 hole_base = 0;
- u64 hole_offset = 0;
- u64 hole_size = 0;
-
- amd64_get_dram_hole_info(mci, &hole_base, &hole_offset, &hole_size);
-
- return sprintf(data, "%llx %llx %llx\n", hole_base, hole_offset,
- hole_size);
-}
-
-/*
- * update NUM_DBG_ATTRS in case you add new members
- */
-static DEVICE_ATTR(dhar, S_IRUGO, amd64_dhar_show, NULL);
-static DEVICE_ATTR(dbam, S_IRUGO, amd64_dbam0_show, NULL);
-static DEVICE_ATTR(topmem, S_IRUGO, amd64_top_mem_show, NULL);
-static DEVICE_ATTR(topmem2, S_IRUGO, amd64_top_mem2_show, NULL);
-static DEVICE_ATTR(dram_hole, S_IRUGO, amd64_hole_show, NULL);
-
-static struct attribute *amd64_edac_dbg_attrs[] = {
- &dev_attr_dhar.attr,
- &dev_attr_dbam.attr,
- &dev_attr_topmem.attr,
- &dev_attr_topmem2.attr,
- &dev_attr_dram_hole.attr,
- NULL
-};
-
-const struct attribute_group amd64_edac_dbg_group = {
- .attrs = amd64_edac_dbg_attrs,
-};
--
cgit v1.2.3
From 61810096de3c3ec977c71dbb7e00447d70045163 Mon Sep 17 00:00:00 2001
From: Borislav Petkov <bp@suse.de>
Date: Tue, 15 Dec 2020 09:18:44 +0100
Subject: EDAC/amd64: Merge error injection sysfs facilities
Merge them into the main driver and put them inside an EDAC_DEBUG
ifdeffery to simplify the driver and have all debugging/injection stuff
behind a debug build-time switch.
No functional changes.
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Yazen Ghannam <yazen.ghannam@amd.com>
Link: https://lkml.kernel.org/r/20201215110517.5215-2-bp@alien8.de
---
drivers/edac/Kconfig | 7 +-
drivers/edac/Makefile | 6 +-
drivers/edac/amd64_edac.c | 235 +++++++++++++++++++++++++++++++++++++++++-
drivers/edac/amd64_edac.h | 4 -
drivers/edac/amd64_edac_inj.c | 235 ------------------------------------------
5 files changed, 235 insertions(+), 252 deletions(-)
delete mode 100644 drivers/edac/amd64_edac_inj.c
diff --git a/drivers/edac/Kconfig b/drivers/edac/Kconfig
index 81c42664f21b..47953b06d6c8 100644
--- a/drivers/edac/Kconfig
+++ b/drivers/edac/Kconfig
@@ -81,10 +81,9 @@ config EDAC_AMD64
Support for error detection and correction of DRAM ECC errors on
the AMD64 families (>= K8) of memory controllers.
-config EDAC_AMD64_ERROR_INJECTION
- bool "Sysfs HW Error injection facilities"
- depends on EDAC_AMD64
- help
+ When EDAC_DEBUG is enabled, hardware error injection facilities
+ through sysfs are available:
+
Recent Opterons (Family 10h and later) provide for Memory Error
Injection into the ECC detection circuits. The amd64_edac module
allows the operator/user to inject Uncorrectable and Correctable
diff --git a/drivers/edac/Makefile b/drivers/edac/Makefile
index 1c70cdcf2b15..2d1641a27a28 100644
--- a/drivers/edac/Makefile
+++ b/drivers/edac/Makefile
@@ -44,11 +44,7 @@ obj-$(CONFIG_EDAC_IE31200) += ie31200_edac.o
obj-$(CONFIG_EDAC_X38) += x38_edac.o
obj-$(CONFIG_EDAC_I82860) += i82860_edac.o
obj-$(CONFIG_EDAC_R82600) += r82600_edac.o
-
-amd64_edac_mod-y := amd64_edac.o
-amd64_edac_mod-$(CONFIG_EDAC_AMD64_ERROR_INJECTION) += amd64_edac_inj.o
-
-obj-$(CONFIG_EDAC_AMD64) += amd64_edac_mod.o
+obj-$(CONFIG_EDAC_AMD64) += amd64_edac.o
obj-$(CONFIG_EDAC_PASEMI) += pasemi_edac.o
diff --git a/drivers/edac/amd64_edac.c b/drivers/edac/amd64_edac.c
index b00dea78541d..d55f8ef2240c 100644
--- a/drivers/edac/amd64_edac.c
+++ b/drivers/edac/amd64_edac.c
@@ -607,8 +607,237 @@ static struct attribute *dbg_attrs[] = {
static const struct attribute_group dbg_group = {
.attrs = dbg_attrs,
};
-#endif /* CONFIG_EDAC_DEBUG */
+static ssize_t inject_section_show(struct device *dev,
+ struct device_attribute *mattr, char *buf)
+{
+ struct mem_ctl_info *mci = to_mci(dev);
+ struct amd64_pvt *pvt = mci->pvt_info;
+ return sprintf(buf, "0x%x\n", pvt->injection.section);
+}
+
+/*
+ * store error injection section value which refers to one of 4 16-byte sections
+ * within a 64-byte cacheline
+ *
+ * range: 0..3
+ */
+static ssize_t inject_section_store(struct device *dev,
+ struct device_attribute *mattr,
+ const char *data, size_t count)
+{
+ struct mem_ctl_info *mci = to_mci(dev);
+ struct amd64_pvt *pvt = mci->pvt_info;
+ unsigned long value;
+ int ret;
+
+ ret = kstrtoul(data, 10, &value);
+ if (ret < 0)
+ return ret;
+
+ if (value > 3) {
+ amd64_warn("%s: invalid section 0x%lx\n", __func__, value);
+ return -EINVAL;
+ }
+
+ pvt->injection.section = (u32) value;
+ return count;
+}
+
+static ssize_t inject_word_show(struct device *dev,
+ struct device_attribute *mattr, char *buf)
+{
+ struct mem_ctl_info *mci = to_mci(dev);
+ struct amd64_pvt *pvt = mci->pvt_info;
+ return sprintf(buf, "0x%x\n", pvt->injection.word);
+}
+
+/*
+ * store error injection word value which refers to one of 9 16-bit word of the
+ * 16-byte (128-bit + ECC bits) section
+ *
+ * range: 0..8
+ */
+static ssize_t inject_word_store(struct device *dev,
+ struct device_attribute *mattr,
+ const char *data, size_t count)
+{
+ struct mem_ctl_info *mci = to_mci(dev);
+ struct amd64_pvt *pvt = mci->pvt_info;
+ unsigned long value;
+ int ret;
+
+ ret = kstrtoul(data, 10, &value);
+ if (ret < 0)
+ return ret;
+
+ if (value > 8) {
+ amd64_warn("%s: invalid word 0x%lx\n", __func__, value);
+ return -EINVAL;
+ }
+
+ pvt->injection.word = (u32) value;
+ return count;
+}
+
+static ssize_t inject_ecc_vector_show(struct device *dev,
+ struct device_attribute *mattr,
+ char *buf)
+{
+ struct mem_ctl_info *mci = to_mci(dev);
+ struct amd64_pvt *pvt = mci->pvt_info;
+ return sprintf(buf, "0x%x\n", pvt->injection.bit_map);
+}
+
+/*
+ * store 16 bit error injection vector which enables injecting errors to the
+ * corresponding bit within the error injection word above. When used during a
+ * DRAM ECC read, it holds the contents of the of the DRAM ECC bits.
+ */
+static ssize_t inject_ecc_vector_store(struct device *dev,
+ struct device_attribute *mattr,
+ const char *data, size_t count)
+{
+ struct mem_ctl_info *mci = to_mci(dev);
+ struct amd64_pvt *pvt = mci->pvt_info;
+ unsigned long value;
+ int ret;
+
+ ret = kstrtoul(data, 16, &value);
+ if (ret < 0)
+ return ret;
+
+ if (value & 0xFFFF0000) {
+ amd64_warn("%s: invalid EccVector: 0x%lx\n", __func__, value);
+ return -EINVAL;
+ }
+
+ pvt->injection.bit_map = (u32) value;
+ return count;
+}
+
+/*
+ * Do a DRAM ECC read. Assemble staged values in the pvt area, format into
+ * fields needed by the injection registers and read the NB Array Data Port.
+ */
+static ssize_t inject_read_store(struct device *dev,
+ struct device_attribute *mattr,
+ const char *data, size_t count)
+{
+ struct mem_ctl_info *mci = to_mci(dev);
+ struct amd64_pvt *pvt = mci->pvt_info;
+ unsigned long value;
+ u32 section, word_bits;
+ int ret;
+
+ ret = kstrtoul(data, 10, &value);
+ if (ret < 0)
+ return ret;
+
+ /* Form value to choose 16-byte section of cacheline */
+ section = F10_NB_ARRAY_DRAM | SET_NB_ARRAY_ADDR(pvt->injection.section);
+
+ amd64_write_pci_cfg(pvt->F3, F10_NB_ARRAY_ADDR, section);
+
+ word_bits = SET_NB_DRAM_INJECTION_READ(pvt->injection);
+
+ /* Issue 'word' and 'bit' along with the READ request */
+ amd64_write_pci_cfg(pvt->F3, F10_NB_ARRAY_DATA, word_bits);
+
+ edac_dbg(0, "section=0x%x word_bits=0x%x\n", section, word_bits);
+
+ return count;
+}
+
+/*
+ * Do a DRAM ECC write. Assemble staged values in the pvt area and format into
+ * fields needed by the injection registers.
+ */
+static ssize_t inject_write_store(struct device *dev,
+ struct device_attribute *mattr,
+ const char *data, size_t count)
+{
+ struct mem_ctl_info *mci = to_mci(dev);
+ struct amd64_pvt *pvt = mci->pvt_info;
+ u32 section, word_bits, tmp;
+ unsigned long value;
+ int ret;
+
+ ret = kstrtoul(data, 10, &value);
+ if (ret < 0)
+ return ret;
+
+ /* Form value to choose 16-byte section of cacheline */
+ section = F10_NB_ARRAY_DRAM | SET_NB_ARRAY_ADDR(pvt->injection.section);
+
+ amd64_write_pci_cfg(pvt->F3, F10_NB_ARRAY_ADDR, section);
+
+ word_bits = SET_NB_DRAM_INJECTION_WRITE(pvt->injection);
+
+ pr_notice_once("Don't forget to decrease MCE polling interval in\n"
+ "/sys/bus/machinecheck/devices/machinecheck<CPUNUM>/check_interval\n"
+ "so that you can get the error report faster.\n");
+
+ on_each_cpu(disable_caches, NULL, 1);
+
+ /* Issue 'word' and 'bit' along with the READ request */
+ amd64_write_pci_cfg(pvt->F3, F10_NB_ARRAY_DATA, word_bits);
+
+ retry:
+ /* wait until injection happens */
+ amd64_read_pci_cfg(pvt->F3, F10_NB_ARRAY_DATA, &tmp);
+ if (tmp & F10_NB_ARR_ECC_WR_REQ) {
+ cpu_relax();
+ goto retry;
+ }
+
+ on_each_cpu(enable_caches, NULL, 1);
+
+ edac_dbg(0, "section=0x%x word_bits=0x%x\n", section, word_bits);
+
+ return count;
+}
+
+/*
+ * update NUM_INJ_ATTRS in case you add new members
+ */
+
+static DEVICE_ATTR(inject_section, S_IRUGO | S_IWUSR,
+ inject_section_show, inject_section_store);
+static DEVICE_ATTR(inject_word, S_IRUGO | S_IWUSR,
+ inject_word_show, inject_word_store);
+static DEVICE_ATTR(inject_ecc_vector, S_IRUGO | S_IWUSR,
+ inject_ecc_vector_show, inject_ecc_vector_store);
+static DEVICE_ATTR(inject_write, S_IWUSR,
+ NULL, inject_write_store);
+static DEVICE_ATTR(inject_read, S_IWUSR,
+ NULL, inject_read_store);
+
+static struct attribute *inj_attrs[] = {
+ &dev_attr_inject_section.attr,
+ &dev_attr_inject_word.attr,
+ &dev_attr_inject_ecc_vector.attr,
+ &dev_attr_inject_write.attr,
+ &dev_attr_inject_read.attr,
+ NULL
+};
+
+static umode_t inj_is_visible(struct kobject *kobj, struct attribute *attr, int idx)
+{
+ struct device *dev = kobj_to_dev(kobj);
+ struct mem_ctl_info *mci = container_of(dev, struct mem_ctl_info, dev);
+ struct amd64_pvt *pvt = mci->pvt_info;
+
+ if (pvt->fam < 0x10)
+ return 0;
+ return attr->mode;
+}
+
+static const struct attribute_group inj_group = {
+ .attrs = inj_attrs,
+ .is_visible = inj_is_visible,
+};
+#endif /* CONFIG_EDAC_DEBUG */
/*
* Return the DramAddr that the SysAddr given by @sys_addr maps to. It is
@@ -3469,9 +3698,7 @@ static struct amd64_family_type *per_family_init(struct amd64_pvt *pvt)
static const struct attribute_group *amd64_edac_attr_groups[] = {
#ifdef CONFIG_EDAC_DEBUG
&dbg_group,
-#endif
-#ifdef CONFIG_EDAC_AMD64_ERROR_INJECTION
- &amd64_edac_inj_group,
+ &inj_group,
#endif
NULL
};
diff --git a/drivers/edac/amd64_edac.h b/drivers/edac/amd64_edac.h
index c072ccd3e7e2..85aa820bc165 100644
--- a/drivers/edac/amd64_edac.h
+++ b/drivers/edac/amd64_edac.h
@@ -462,10 +462,6 @@ struct ecc_settings {
} flags;
};
-#ifdef CONFIG_EDAC_AMD64_ERROR_INJECTION
-extern const struct attribute_group amd64_edac_inj_group;
-#endif
-
/*
* Each of the PCI Device IDs types have their own set of hardware accessor
* functions and per device encoding/decoding logic.
diff --git a/drivers/edac/amd64_edac_inj.c b/drivers/edac/amd64_edac_inj.c
deleted file mode 100644
index d96d6116f0fb..000000000000
--- a/drivers/edac/amd64_edac_inj.c
+++ /dev/null
@@ -1,235 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0
-#include "amd64_edac.h"
-
-static ssize_t amd64_inject_section_show(struct device *dev,
- struct device_attribute *mattr,
- char *buf)
-{
- struct mem_ctl_info *mci = to_mci(dev);
- struct amd64_pvt *pvt = mci->pvt_info;
- return sprintf(buf, "0x%x\n", pvt->injection.section);
-}
-
-/*
- * store error injection section value which refers to one of 4 16-byte sections
- * within a 64-byte cacheline
- *
- * range: 0..3
- */
-static ssize_t amd64_inject_section_store(struct device *dev,
- struct device_attribute *mattr,
- const char *data, size_t count)
-{
- struct mem_ctl_info *mci = to_mci(dev);
- struct amd64_pvt *pvt = mci->pvt_info;
- unsigned long value;
- int ret;
-
- ret = kstrtoul(data, 10, &value);
- if (ret < 0)
- return ret;
-
- if (value > 3) {
- amd64_warn("%s: invalid section 0x%lx\n", __func__, value);
- return -EINVAL;
- }
-
- pvt->injection.section = (u32) value;
- return count;
-}
-
-static ssize_t amd64_inject_word_show(struct device *dev,
- struct device_attribute *mattr,
- char *buf)
-{
- struct mem_ctl_info *mci = to_mci(dev);
- struct amd64_pvt *pvt = mci->pvt_info;
- return sprintf(buf, "0x%x\n", pvt->injection.word);
-}
-
-/*
- * store error injection word value which refers to one of 9 16-bit word of the
- * 16-byte (128-bit + ECC bits) section
- *
- * range: 0..8
- */
-static ssize_t amd64_inject_word_store(struct device *dev,
- struct device_attribute *mattr,
- const char *data, size_t count)
-{
- struct mem_ctl_info *mci = to_mci(dev);
- struct amd64_pvt *pvt = mci->pvt_info;
- unsigned long value;
- int ret;
-
- ret = kstrtoul(data, 10, &value);
- if (ret < 0)
- return ret;
-
- if (value > 8) {
- amd64_warn("%s: invalid word 0x%lx\n", __func__, value);
- return -EINVAL;
- }
-
- pvt->injection.word = (u32) value;
- return count;
-}
-
-static ssize_t amd64_inject_ecc_vector_show(struct device *dev,
- struct device_attribute *mattr,
- char *buf)
-{
- struct mem_ctl_info *mci = to_mci(dev);
- struct amd64_pvt *pvt = mci->pvt_info;
- return sprintf(buf, "0x%x\n", pvt->injection.bit_map);
-}
-
-/*
- * store 16 bit error injection vector which enables injecting errors to the
- * corresponding bit within the error injection word above. When used during a
- * DRAM ECC read, it holds the contents of the of the DRAM ECC bits.
- */
-static ssize_t amd64_inject_ecc_vector_store(struct device *dev,
- struct device_attribute *mattr,
- const char *data, size_t count)
-{
- struct mem_ctl_info *mci = to_mci(dev);
- struct amd64_pvt *pvt = mci->pvt_info;
- unsigned long value;
- int ret;
-
- ret = kstrtoul(data, 16, &value);
- if (ret < 0)
- return ret;
-
- if (value & 0xFFFF0000) {
- amd64_warn("%s: invalid EccVector: 0x%lx\n", __func__, value);
- return -EINVAL;
- }
-
- pvt->injection.bit_map = (u32) value;
- return count;
-}
-
-/*
- * Do a DRAM ECC read. Assemble staged values in the pvt area, format into
- * fields needed by the injection registers and read the NB Array Data Port.
- */
-static ssize_t amd64_inject_read_store(struct device *dev,
- struct device_attribute *mattr,
- const char *data, size_t count)
-{
- struct mem_ctl_info *mci = to_mci(dev);
- struct amd64_pvt *pvt = mci->pvt_info;
- unsigned long value;
- u32 section, word_bits;
- int ret;
-
- ret = kstrtoul(data, 10, &value);
- if (ret < 0)
- return ret;
-
- /* Form value to choose 16-byte section of cacheline */
- section = F10_NB_ARRAY_DRAM | SET_NB_ARRAY_ADDR(pvt->injection.section);
-
- amd64_write_pci_cfg(pvt->F3, F10_NB_ARRAY_ADDR, section);
-
- word_bits = SET_NB_DRAM_INJECTION_READ(pvt->injection);
-
- /* Issue 'word' and 'bit' along with the READ request */
- amd64_write_pci_cfg(pvt->F3, F10_NB_ARRAY_DATA, word_bits);
-
- edac_dbg(0, "section=0x%x word_bits=0x%x\n", section, word_bits);
-
- return count;
-}
-
-/*
- * Do a DRAM ECC write. Assemble staged values in the pvt area and format into
- * fields needed by the injection registers.
- */
-static ssize_t amd64_inject_write_store(struct device *dev,
- struct device_attribute *mattr,
- const char *data, size_t count)
-{
- struct mem_ctl_info *mci = to_mci(dev);
- struct amd64_pvt *pvt = mci->pvt_info;
- u32 section, word_bits, tmp;
- unsigned long value;
- int ret;
-
- ret = kstrtoul(data, 10, &value);
- if (ret < 0)
- return ret;
-
- /* Form value to choose 16-byte section of cacheline */
- section = F10_NB_ARRAY_DRAM | SET_NB_ARRAY_ADDR(pvt->injection.section);
-
- amd64_write_pci_cfg(pvt->F3, F10_NB_ARRAY_ADDR, section);
-
- word_bits = SET_NB_DRAM_INJECTION_WRITE(pvt->injection);
-
- pr_notice_once("Don't forget to decrease MCE polling interval in\n"
- "/sys/bus/machinecheck/devices/machinecheck<CPUNUM>/check_interval\n"
- "so that you can get the error report faster.\n");
-
- on_each_cpu(disable_caches, NULL, 1);
-
- /* Issue 'word' and 'bit' along with the READ request */
- amd64_write_pci_cfg(pvt->F3, F10_NB_ARRAY_DATA, word_bits);
-
- retry:
- /* wait until injection happens */
- amd64_read_pci_cfg(pvt->F3, F10_NB_ARRAY_DATA, &tmp);
- if (tmp & F10_NB_ARR_ECC_WR_REQ) {
- cpu_relax();
- goto retry;
- }
-
- on_each_cpu(enable_caches, NULL, 1);
-
- edac_dbg(0, "section=0x%x word_bits=0x%x\n", section, word_bits);
-
- return count;
-}
-
-/*
- * update NUM_INJ_ATTRS in case you add new members
- */
-
-static DEVICE_ATTR(inject_section, S_IRUGO | S_IWUSR,
- amd64_inject_section_show, amd64_inject_section_store);
-static DEVICE_ATTR(inject_word, S_IRUGO | S_IWUSR,
- amd64_inject_word_show, amd64_inject_word_store);
-static DEVICE_ATTR(inject_ecc_vector, S_IRUGO | S_IWUSR,
- amd64_inject_ecc_vector_show, amd64_inject_ecc_vector_store);
-static DEVICE_ATTR(inject_write, S_IWUSR,
- NULL, amd64_inject_write_store);
-static DEVICE_ATTR(inject_read, S_IWUSR,
- NULL, amd64_inject_read_store);
-
-static struct attribute *amd64_edac_inj_attrs[] = {
- &dev_attr_inject_section.attr,
- &dev_attr_inject_word.attr,
- &dev_attr_inject_ecc_vector.attr,
- &dev_attr_inject_write.attr,
- &dev_attr_inject_read.attr,
- NULL
-};
-
-static umode_t amd64_edac_inj_is_visible(struct kobject *kobj,
- struct attribute *attr, int idx)
-{
- struct device *dev = kobj_to_dev(kobj);
- struct mem_ctl_info *mci = container_of(dev, struct mem_ctl_info, dev);
- struct amd64_pvt *pvt = mci->pvt_info;
-
- if (pvt->fam < 0x10)
- return 0;
- return attr->mode;
-}
-
-const struct attribute_group amd64_edac_inj_group = {
- .attrs = amd64_edac_inj_attrs,
- .is_visible = amd64_edac_inj_is_visible,
-};
--
cgit v1.2.3
From 1865bc71a869ede69098b1f3e65857b4330f0607 Mon Sep 17 00:00:00 2001
From: Borislav Petkov <bp@suse.de>
Date: Tue, 22 Dec 2020 18:55:06 +0100
Subject: EDAC/amd64: Limit error injection functionality to supported hw
Families up to and including 0x16 allow access to the injection
hardware. Starting with family 0x17, access to those registers is
blocked by security policy.
Limit that only on the families which support it.
Suggested-by: Yazen Ghannam <yazen.ghannam@amd.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Link: https://lkml.kernel.org/r/20201222180013.GD13463@zn.tnic
---
drivers/edac/Kconfig | 8 ++++----
drivers/edac/amd64_edac.c | 8 +++++---
2 files changed, 9 insertions(+), 7 deletions(-)
diff --git a/drivers/edac/Kconfig b/drivers/edac/Kconfig
index 47953b06d6c8..27d0c4cdc58d 100644
--- a/drivers/edac/Kconfig
+++ b/drivers/edac/Kconfig
@@ -84,10 +84,10 @@ config EDAC_AMD64
When EDAC_DEBUG is enabled, hardware error injection facilities
through sysfs are available:
- Recent Opterons (Family 10h and later) provide for Memory Error
- Injection into the ECC detection circuits. The amd64_edac module
- allows the operator/user to inject Uncorrectable and Correctable
- errors into DRAM.
+ AMD CPUs up to and excluding family 0x17 provide for Memory
+ Error Injection into the ECC detection circuits. The amd64_edac
+ module allows the operator/user to inject Uncorrectable and
+ Correctable errors into DRAM.
When enabled, in each of the respective memory controller directories
(/sys/devices/system/edac/mc/mcX), there are 3 input files:
diff --git a/drivers/edac/amd64_edac.c b/drivers/edac/amd64_edac.c
index d55f8ef2240c..9868f95a5622 100644
--- a/drivers/edac/amd64_edac.c
+++ b/drivers/edac/amd64_edac.c
@@ -828,9 +828,11 @@ static umode_t inj_is_visible(struct kobject *kobj, struct attribute *attr, int
struct mem_ctl_info *mci = container_of(dev, struct mem_ctl_info, dev);
struct amd64_pvt *pvt = mci->pvt_info;
- if (pvt->fam < 0x10)
- return 0;
- return attr->mode;
+ /* Families which have that injection hw */
+ if (pvt->fam >= 0x10 && pvt->fam <= 0x16)
+ return attr->mode;
+
+ return 0;
}
static const struct attribute_group inj_group = {
--
cgit v1.2.3
From e0e0427412d0f374461a5294efc161e00df4be53 Mon Sep 17 00:00:00 2001
From: Zheng Yongjun <zhengyongjun3@huawei.com>
Date: Wed, 16 Dec 2020 21:18:46 +0800
Subject: EDAC/ppc4xx: Convert comma to semicolon
Replace a comma between expression statements with a semicolon.
Signed-off-by: Zheng Yongjun <zhengyongjun3@huawei.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Link: https://lkml.kernel.org/r/20201216131846.14937-1-zhengyongjun3@huawei.com
---
drivers/edac/ppc4xx_edac.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/edac/ppc4xx_edac.c b/drivers/edac/ppc4xx_edac.c
index 677095769182..6793f6d799e7 100644
--- a/drivers/edac/ppc4xx_edac.c
+++ b/drivers/edac/ppc4xx_edac.c
@@ -1058,7 +1058,7 @@ static int ppc4xx_edac_mc_init(struct mem_ctl_info *mci,
/* Initialize strings */
mci->mod_name = PPC4XX_EDAC_MODULE_NAME;
- mci->ctl_name = ppc4xx_edac_match->compatible,
+ mci->ctl_name = ppc4xx_edac_match->compatible;
mci->dev_name = np->full_name;
/* Initialize callbacks */
--
cgit v1.2.3
From e26124cd5f7099949109608845bba9e9bf96599c Mon Sep 17 00:00:00 2001
From: Menglong Dong <dong.menglong@zte.com.cn>
Date: Tue, 12 Jan 2021 02:35:40 -0800
Subject: EDAC/xgene: Do not print a failure message to get an IRQ twice
Coccinelle reports a redundant error print in xgene_edac_probe() because
platform_get_irq() will already print an error message when it is unable
to get an IRQ.
Use platform_get_irq_optional() instead which avoids the error message
and keep the driver-specific one.
[ bp: Sanitize commit message. ]
Signed-off-by: Menglong Dong <dong.menglong@zte.com.cn>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Robert Richter <rric@kernel.org>
Link: https://lkml.kernel.org/r/20210112103540.7818-1-dong.menglong@zte.com.cn
---
drivers/edac/xgene_edac.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/edac/xgene_edac.c b/drivers/edac/xgene_edac.c
index 1d2c27a00a4a..2ccd1db5e98f 100644
--- a/drivers/edac/xgene_edac.c
+++ b/drivers/edac/xgene_edac.c
@@ -1916,7 +1916,7 @@ static int xgene_edac_probe(struct platform_device *pdev)
int i;
for (i = 0; i < 3; i++) {
- irq = platform_get_irq(pdev, i);
+ irq = platform_get_irq_optional(pdev, i);
if (irq < 0) {
dev_err(&pdev->dev, "No IRQ resource\n");
rc = -EINVAL;
--
cgit v1.2.3
From 4cbcb73b1c7a73a13b336e0c048601d6e8eecaa8 Mon Sep 17 00:00:00 2001
From: Borislav Petkov <bp@suse.de>
Date: Wed, 13 Jan 2021 20:13:30 +0100
Subject: EDAC/amd64: Issue probing messages only on properly detected hardware
amd64_edac was converted to CPU family autoprobing (from PCI device
IDs) to not have to add a new PCI device ID each time a new platform is
shipped but to support the whole family out-of-the-box.
However, this caused a lot of noise in dmesg even when the machine
doesn't have ECC DIMMs or ECC has been disabled in the BIOS:
EDAC MC: Ver: 3.0.0
EDAC amd64: F17h detected (node 0).
EDAC amd64: Node 0: DRAM ECC disabled.
EDAC amd64: F17h detected (node 1).
EDAC amd64: Node 1: DRAM ECC disabled.
EDAC amd64: F17h detected (node 2).
EDAC amd64: Node 2: DRAM ECC disabled.
EDAC amd64: F17h detected (node 3).
EDAC amd64: Node 3: DRAM ECC disabled.
EDAC amd64: F17h detected (node 4).
EDAC amd64: Node 4: DRAM ECC disabled.
EDAC amd64: F17h detected (node 5).
EDAC amd64: Node 5: DRAM ECC disabled.
EDAC amd64: F17h detected (node 6).
EDAC amd64: Node 6: DRAM ECC disabled.
EDAC amd64: F17h detected (node 7).
EDAC amd64: Node 7: DRAM ECC disabled.
or even
$ grep EDAC dmesg.log | sed 's/\[.*\] //' | sort | uniq -c
128 EDAC amd64: F17h detected (node 0).
128 EDAC amd64: Node 0: DRAM ECC disabled.
1 EDAC MC: Ver: 3.0.0
on a big machine. Yap, that's once per CPU for 128 of them.
So move the init messages after all probing has succeeded to avoid
unnecessary spew in dmesg.
Signed-off-by: Borislav Petkov <bp@suse.de>
Link: https://lkml.kernel.org/r/20210119164141.17417-1-bp@alien8.de
---
drivers/edac/amd64_edac.c | 14 +++++++-------
1 file changed, 7 insertions(+), 7 deletions(-)
diff --git a/drivers/edac/amd64_edac.c b/drivers/edac/amd64_edac.c
index 9868f95a5622..9fa4dfc6ebee 100644
--- a/drivers/edac/amd64_edac.c
+++ b/drivers/edac/amd64_edac.c
@@ -3528,8 +3528,7 @@ static bool ecc_enabled(struct amd64_pvt *pvt)
MSR_IA32_MCG_CTL, nid);
}
- amd64_info("Node %d: DRAM ECC %s.\n",
- nid, (ecc_en ? "enabled" : "disabled"));
+ edac_dbg(3, "Node %d: DRAM ECC %s.\n", nid, (ecc_en ? "enabled" : "disabled"));
if (!ecc_en || !nb_mce_en)
return false;
@@ -3689,11 +3688,6 @@ static struct amd64_family_type *per_family_init(struct amd64_pvt *pvt)
return NULL;
}
- amd64_info("%s %sdetected (node %d).\n", fam_type->ctl_name,
- (pvt->fam == 0xf ?
- (pvt->ext_model >= K8_REV_F ? "revF or later "
- : "revE or earlier ")
- : ""), pvt->mc_node_id);
return fam_type;
}
@@ -3865,6 +3859,12 @@ static int probe_one_instance(unsigned int nid)
goto err_enable;
}
+ amd64_info("%s %sdetected (node %d).\n", fam_type->ctl_name,
+ (pvt->fam == 0xf ?
+ (pvt->ext_model >= K8_REV_F ? "revF or later "
+ : "revE or earlier ")
+ : ""), pvt->mc_node_id);
+
dump_misc_regs(pvt);
return ret;
--
cgit v1.2.3