* [PATCHv4 3/3] edac: altera: Add EDAC support for Altera SoC SDRAM Controller
2014-05-12 23:36 Addition of EDAC for Altera SoC SDRAM Controller tthayer
[not found] ` <1399937817-2202-1-git-send-email-tthayer-EIB2kfCEclfQT0dZR+AlfA@public.gmane.org>
2014-05-12 23:36 ` [PATCHv4 2/3] dts: socfpga: Add bindings for Altera SoC SDRAM EDAC tthayer
@ 2014-05-12 23:36 ` tthayer
2014-05-13 0:12 ` Borislav Petkov
2 siblings, 1 reply; 6+ messages in thread
From: tthayer @ 2014-05-12 23:36 UTC (permalink / raw)
To: robherring2, pawel.moll, mark.rutland, ijc+devicetree, galak, rob,
linux, dinguyen, dougthompson, grant.likely, bp
Cc: devicetree, linux-doc, linux-edac, linux-kernel, linux-arm-kernel,
tthayer.linux, tthayer
From: Thor Thayer <tthayer@altera.com>
This patch adds EDAC support for reporting ECC errors of CycloneV
and ArriaV SDRAM controllers.
- The SDRAM Controller registers are shared with the FPGA bridge so
these are accessed through the syscon interface.
- The configuration of the SDRAM memory size for the EDAC framework
is discovered from the SDRAM Controller registers.
- Documentation of the bindings in devicetree/bindings/arm/altera/
socfpga-sdram-edac.txt
- Correction & reporting of single bit errors, panic on double bit
errors.
v2: Use the SDRAM controller registers to calculate memory size
instead of the Device Tree. Update To & Cc list. Add maintainer
information.
v3: EDAC driver cleanup based on comments from Mailing list.
v4: Panic on DBE. Add macro around inject-error reads to prevent
them from being optimized out. Remove of_match_ptr since this
will always use Device Tree.
Signed-off-by: Thor Thayer <tthayer@altera.com>
---
MAINTAINERS | 5 +
drivers/edac/Kconfig | 9 +
drivers/edac/Makefile | 2 +
drivers/edac/altera_edac.c | 416 ++++++++++++++++++++++++++++++++++++++++++++
4 files changed, 432 insertions(+)
create mode 100644 drivers/edac/altera_edac.c
diff --git a/MAINTAINERS b/MAINTAINERS
index e67ea24..ecd1277 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -1290,6 +1290,11 @@ M: Dinh Nguyen <dinguyen@altera.com>
S: Maintained
F: drivers/clk/socfpga/
+ARM/SOCFPGA SDRAM EDAC SUPPORT
+M: Thor Thayer <tthayer@altera.com>
+S: Maintained
+F: drivers/edac/altera_edac.c
+
ARM/STI ARCHITECTURE
M: Srinivas Kandagatla <srinivas.kandagatla@gmail.com>
M: Maxime Coquelin <maxime.coquelin@st.com>
diff --git a/drivers/edac/Kconfig b/drivers/edac/Kconfig
index 878f090..4f4d379 100644
--- a/drivers/edac/Kconfig
+++ b/drivers/edac/Kconfig
@@ -368,4 +368,13 @@ config EDAC_OCTEON_PCI
Support for error detection and correction on the
Cavium Octeon family of SOCs.
+config EDAC_ALTERA_MC
+ bool "Altera SDRAM Memory Controller EDAC"
+ depends on EDAC_MM_EDAC && ARCH_SOCFPGA
+ help
+ Support for error detection and correction on the
+ Altera SDRAM memory controller. Note that the
+ preloader must initialize the SDRAM before loading
+ the kernel.
+
endif # EDAC
diff --git a/drivers/edac/Makefile b/drivers/edac/Makefile
index 4154ed6..9741336 100644
--- a/drivers/edac/Makefile
+++ b/drivers/edac/Makefile
@@ -64,3 +64,5 @@ obj-$(CONFIG_EDAC_OCTEON_PC) += octeon_edac-pc.o
obj-$(CONFIG_EDAC_OCTEON_L2C) += octeon_edac-l2c.o
obj-$(CONFIG_EDAC_OCTEON_LMC) += octeon_edac-lmc.o
obj-$(CONFIG_EDAC_OCTEON_PCI) += octeon_edac-pci.o
+
+obj-$(CONFIG_EDAC_ALTERA_MC) += altera_edac.o
diff --git a/drivers/edac/altera_edac.c b/drivers/edac/altera_edac.c
new file mode 100644
index 0000000..3f6e23b
--- /dev/null
+++ b/drivers/edac/altera_edac.c
@@ -0,0 +1,416 @@
+/*
+ * Copyright Altera Corporation (C) 2014. All rights reserved.
+ * Copyright 2011-2012 Calxeda, Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License. See the file "COPYING" in the main directory of this archive
+ * for more details.
+
+ *
+ * Adapted from the highbank_mc_edac driver
+ *
+ */
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/ctype.h>
+#include <linux/edac.h>
+#include <linux/interrupt.h>
+#include <linux/platform_device.h>
+#include <linux/of_platform.h>
+#include <linux/uaccess.h>
+#include <linux/mfd/syscon.h>
+#include <linux/regmap.h>
+
+#include "edac_core.h"
+#include "edac_module.h"
+
+#define EDAC_MOD_STR "altera_edac"
+#define EDAC_VERSION "1"
+
+/* SDRAM Controller CtrlCfg Register */
+#define CTLCFG 0x00
+
+/* SDRAM Controller CtrlCfg Register Bit Masks */
+#define CTLCFG_ECC_EN 0x400
+#define CTLCFG_ECC_CORR_EN 0x800
+#define CTLCFG_GEN_SB_ERR 0x2000
+#define CTLCFG_GEN_DB_ERR 0x4000
+
+#define CTLCFG_ECC_AUTO_EN (CTLCFG_ECC_EN | \
+ CTLCFG_ECC_CORR_EN)
+
+/* SDRAM Controller Address Width Register */
+#define DRAMADDRW 0x2C
+
+/* SDRAM Controller Address Widths Field Register */
+#define DRAMADDRW_COLBIT_MASK 0x001F
+#define DRAMADDRW_COLBIT_LSB 0
+#define DRAMADDRW_ROWBIT_MASK 0x03E0
+#define DRAMADDRW_ROWBIT_LSB 5
+#define DRAMADDRW_BANKBIT_MASK 0x1C00
+#define DRAMADDRW_BANKBIT_LSB 10
+#define DRAMADDRW_CSBIT_MASK 0xE000
+#define DRAMADDRW_CSBIT_LSB 13
+
+/* SDRAM Controller Interface Data Width Register */
+#define DRAMIFWIDTH 0x30
+
+/* SDRAM Controller Interface Data Width Defines */
+#define DRAMIFWIDTH_16B_ECC 24
+#define DRAMIFWIDTH_32B_ECC 40
+
+/* SDRAM Controller DRAM Status Register */
+#define DRAMSTS 0x38
+
+/* SDRAM Controller DRAM Status Register Bit Masks */
+#define DRAMSTS_SBEERR 0x04
+#define DRAMSTS_DBEERR 0x08
+#define DRAMSTS_CORR_DROP 0x10
+
+/* SDRAM Controller DRAM IRQ Register */
+#define DRAMINTR 0x3C
+
+/* SDRAM Controller DRAM IRQ Register Bit Masks */
+#define DRAMINTR_INTREN 0x01
+#define DRAMINTR_SBEMASK 0x02
+#define DRAMINTR_DBEMASK 0x04
+#define DRAMINTR_CORRDROPMASK 0x08
+#define DRAMINTR_INTRCLR 0x10
+
+/* SDRAM Controller Single Bit Error Count Register */
+#define SBECOUNT 0x40
+
+/* SDRAM Controller Single Bit Error Count Register Bit Masks */
+#define SBECOUNT_MASK 0x0F
+
+/* SDRAM Controller Double Bit Error Count Register */
+#define DBECOUNT 0x44
+
+/* SDRAM Controller Double Bit Error Count Register Bit Masks */
+#define DBECOUNT_MASK 0x0F
+
+/* SDRAM Controller ECC Error Address Register */
+#define ERRADDR 0x48
+
+/* SDRAM Controller ECC Error Address Register Bit Masks */
+#define ERRADDR_MASK 0xFFFFFFFF
+
+/* SDRAM Controller ECC Autocorrect Drop Count Register */
+#define DROPCOUNT 0x4C
+
+/* SDRAM Controller ECC Autocorrect Drop Count Register Bit Masks */
+#define DROPCOUNT_MASK 0x0F
+
+/* SDRAM Controller ECC AutoCorrect Address Register */
+#define DROPADDR 0x50
+
+/* SDRAM Controller ECC AutoCorrect Error Address Register Bit Masks */
+#define DROPADDR_MASK 0xFFFFFFFF
+
+/* Altera SDRAM Memory Controller data */
+struct altr_sdram_mc_data {
+ struct regmap *mc_vbase;
+};
+
+static irqreturn_t altr_sdram_mc_err_handler(int irq, void *dev_id)
+{
+ struct mem_ctl_info *mci = dev_id;
+ struct altr_sdram_mc_data *drvdata = mci->pvt_info;
+ u32 status, err_count, err_addr;
+
+ /* Error Address is shared by both SBE & DBE */
+ regmap_read(drvdata->mc_vbase, ERRADDR, &err_addr);
+
+ regmap_read(drvdata->mc_vbase, DRAMSTS, &status);
+
+ if (status & DRAMSTS_DBEERR) {
+ regmap_read(drvdata->mc_vbase, DBECOUNT, &err_count);
+ panic("\nEDAC: [%d Uncorrectable errors @ 0x%08X]\n",
+ err_count, err_addr);
+ }
+ if (status & DRAMSTS_SBEERR) {
+ regmap_read(drvdata->mc_vbase, SBECOUNT, &err_count);
+ edac_mc_handle_error(HW_EVENT_ERR_CORRECTED, mci, err_count,
+ err_addr >> PAGE_SHIFT,
+ err_addr & ~PAGE_MASK, 0,
+ 0, 0, -1, mci->ctl_name, "");
+ }
+
+ regmap_write(drvdata->mc_vbase, DRAMINTR,
+ (DRAMINTR_INTRCLR | DRAMINTR_INTREN));
+
+ return IRQ_HANDLED;
+}
+
+#ifdef CONFIG_EDAC_DEBUG
+static ssize_t altr_sdr_mc_err_inject_write(struct file *file,
+ const char __user *data,
+ size_t count, loff_t *ppos)
+{
+ struct mem_ctl_info *mci = file->private_data;
+ struct altr_sdram_mc_data *drvdata = mci->pvt_info;
+ u32 *ptemp;
+ dma_addr_t dma_handle;
+ u32 reg, read_reg;
+
+ ptemp = dma_alloc_coherent(mci->pdev, 16, &dma_handle, GFP_KERNEL);
+ if (IS_ERR(ptemp)) {
+ dma_free_coherent(mci->pdev, 16, ptemp, dma_handle);
+ edac_printk(KERN_ERR, EDAC_MC,
+ "Inject: Buffer Allocation error\n");
+ return -ENOMEM;
+ }
+
+ regmap_read(drvdata->mc_vbase, CTLCFG, &read_reg);
+ read_reg &= ~(CTLCFG_GEN_SB_ERR | CTLCFG_GEN_DB_ERR);
+
+ if (count == 3) {
+ edac_printk(KERN_ALERT, EDAC_MC,
+ "Inject Double bit error\n");
+ regmap_write(drvdata->mc_vbase, CTLCFG,
+ (read_reg | CTLCFG_GEN_DB_ERR));
+ } else {
+ edac_printk(KERN_ALERT, EDAC_MC,
+ "Inject Single bit error\n");
+ regmap_write(drvdata->mc_vbase, CTLCFG,
+ (read_reg | CTLCFG_GEN_SB_ERR));
+ }
+
+ ptemp[0] = 0x5A5A5A5A;
+ ptemp[1] = 0xA5A5A5A5;
+ /* Clear the error injection bits */
+ regmap_write(drvdata->mc_vbase, CTLCFG, read_reg);
+ /* Ensure it has been written out */
+ wmb();
+
+ /*
+ * To trigger the error, we need to read the data back
+ * (the data was written with errors above)
+ * The ACCESS_ONCE macros are used to prevent the
+ * compiler optimizing these reads out.
+ */
+ reg = ACCESS_ONCE(ptemp[0]);
+ read_reg = ACCESS_ONCE(ptemp[1]);
+ /* Force Read */
+ rmb();
+
+ dma_free_coherent(mci->pdev, 16, ptemp, dma_handle);
+
+ return count;
+}
+
+static const struct file_operations altr_sdr_mc_debug_inject_fops = {
+ .open = simple_open,
+ .write = altr_sdr_mc_err_inject_write,
+ .llseek = generic_file_llseek,
+};
+
+static void altr_sdr_mc_create_debugfs_nodes(struct mem_ctl_info *mci)
+{
+ if (mci->debugfs)
+ debugfs_create_file("inject_ctrl", S_IWUSR, mci->debugfs, mci,
+ &altr_sdr_mc_debug_inject_fops);
+}
+#else
+static void altr_sdr_mc_create_debugfs_nodes(struct mem_ctl_info *mci)
+{}
+#endif
+
+/* Get total memory size in bytes */
+static u32 altr_sdram_get_total_mem_size(struct regmap *mc_vbase)
+{
+ u32 size;
+ u32 read_reg, row, bank, col, cs, width;
+
+ if (regmap_read(mc_vbase, DRAMADDRW, &read_reg) < 0)
+ return 0;
+
+ if (regmap_read(mc_vbase, DRAMIFWIDTH, &width) < 0)
+ return 0;
+
+ col = (read_reg & DRAMADDRW_COLBIT_MASK) >>
+ DRAMADDRW_COLBIT_LSB;
+ row = (read_reg & DRAMADDRW_ROWBIT_MASK) >>
+ DRAMADDRW_ROWBIT_LSB;
+ bank = (read_reg & DRAMADDRW_BANKBIT_MASK) >>
+ DRAMADDRW_BANKBIT_LSB;
+ cs = (read_reg & DRAMADDRW_CSBIT_MASK) >>
+ DRAMADDRW_CSBIT_LSB;
+
+ /* Correct for ECC as its not addressible */
+ if (width == DRAMIFWIDTH_32B_ECC)
+ width = 32;
+ if (width == DRAMIFWIDTH_16B_ECC)
+ width = 16;
+
+ /* calculate the SDRAM size base on this info */
+ size = 1 << (row + bank + col);
+ size = size * cs * (width / 8);
+ return size;
+}
+
+static int altr_sdram_probe(struct platform_device *pdev)
+{
+ struct edac_mc_layer layers[2];
+ struct mem_ctl_info *mci;
+ struct altr_sdram_mc_data *drvdata;
+ struct regmap *mc_vbase;
+ struct dimm_info *dimm;
+ u32 read_reg, mem_size;
+ int irq;
+ int res = 0;
+
+ /* Validate the SDRAM controller has ECC enabled */
+ /* Grab the register values from the sdr-ctl in device tree */
+ mc_vbase = syscon_regmap_lookup_by_compatible("altr,sdr-ctl");
+ if (IS_ERR(mc_vbase)) {
+ edac_printk(KERN_ERR, EDAC_MC,
+ "regmap for altr,sdr-ctl lookup failed.\n");
+ return -ENODEV;
+ }
+
+ if (regmap_read(mc_vbase, CTLCFG, &read_reg) ||
+ ((read_reg & CTLCFG_ECC_AUTO_EN) != CTLCFG_ECC_AUTO_EN)) {
+ edac_printk(KERN_ERR, EDAC_MC,
+ "No ECC/ECC disabled [0x%08X]\n", read_reg);
+ return -ENODEV;
+ }
+
+ /* Grab memory size from device tree. */
+ mem_size = altr_sdram_get_total_mem_size(mc_vbase);
+ if (mem_size <= 0) {
+ edac_printk(KERN_ERR, EDAC_MC,
+ "Unable to calculate memory size\n");
+ return -ENODEV;
+ }
+
+ /* Ensure the SDRAM Interrupt is disabled and cleared */
+ if (regmap_write(mc_vbase, DRAMINTR, DRAMINTR_INTRCLR)) {
+ edac_printk(KERN_ERR, EDAC_MC,
+ "Error clearing SDRAM ECC IRQ\n");
+ return -ENODEV;
+ }
+
+ irq = platform_get_irq(pdev, 0);
+ if (irq < 0) {
+ edac_printk(KERN_ERR, EDAC_MC,
+ "No irq %d in DT\n", irq);
+ return -ENODEV;
+ }
+
+ layers[0].type = EDAC_MC_LAYER_CHIP_SELECT;
+ layers[0].size = 1;
+ layers[0].is_virt_csrow = true;
+ layers[1].type = EDAC_MC_LAYER_CHANNEL;
+ layers[1].size = 1;
+ layers[1].is_virt_csrow = false;
+ mci = edac_mc_alloc(0, ARRAY_SIZE(layers), layers,
+ sizeof(struct altr_sdram_mc_data));
+ if (!mci)
+ return -ENOMEM;
+
+ mci->pdev = &pdev->dev;
+ drvdata = mci->pvt_info;
+ drvdata->mc_vbase = mc_vbase;
+ platform_set_drvdata(pdev, mci);
+
+ if (!devres_open_group(&pdev->dev, NULL, GFP_KERNEL)) {
+ res = -ENOMEM;
+ goto free;
+ }
+
+ mci->mtype_cap = MEM_FLAG_DDR3;
+ mci->edac_ctl_cap = EDAC_FLAG_NONE | EDAC_FLAG_SECDED;
+ mci->edac_cap = EDAC_FLAG_SECDED;
+ mci->mod_name = EDAC_MOD_STR;
+ mci->mod_ver = EDAC_VERSION;
+ mci->ctl_name = dev_name(&pdev->dev);
+ mci->scrub_mode = SCRUB_SW_SRC;
+ mci->dev_name = dev_name(&pdev->dev);
+
+ dimm = *mci->dimms;
+ dimm->nr_pages = ((mem_size - 1) >> PAGE_SHIFT) + 1;
+ dimm->grain = 8;
+ dimm->dtype = DEV_X8;
+ dimm->mtype = MEM_DDR3;
+ dimm->edac_mode = EDAC_SECDED;
+
+ res = edac_mc_add_mc(mci);
+ if (res < 0)
+ goto err;
+
+ res = devm_request_irq(&pdev->dev, irq, altr_sdram_mc_err_handler,
+ 0, dev_name(&pdev->dev), mci);
+ if (res < 0) {
+ edac_mc_printk(mci, KERN_ERR,
+ "Unable to request irq %d\n", irq);
+ res = -ENODEV;
+ goto err2;
+ }
+
+ if (regmap_write(drvdata->mc_vbase, DRAMINTR,
+ (DRAMINTR_INTRCLR | DRAMINTR_INTREN))) {
+ edac_mc_printk(mci, KERN_ERR,
+ "Error enabling SDRAM ECC IRQ\n");
+ res = -ENODEV;
+ goto err2;
+ }
+
+ altr_sdr_mc_create_debugfs_nodes(mci);
+
+ devres_close_group(&pdev->dev, NULL);
+
+ return 0;
+
+err2:
+ edac_mc_del_mc(&pdev->dev);
+err:
+ devres_release_group(&pdev->dev, NULL);
+free:
+ edac_mc_free(mci);
+ edac_printk(KERN_ERR, EDAC_MC,
+ "EDAC Probe Failed; Error %d\n", res);
+
+ return res;
+}
+
+static int altr_sdram_remove(struct platform_device *pdev)
+{
+ struct mem_ctl_info *mci = platform_get_drvdata(pdev);
+
+ edac_mc_del_mc(&pdev->dev);
+ edac_mc_free(mci);
+ platform_set_drvdata(pdev, NULL);
+
+ return 0;
+}
+
+static const struct of_device_id altr_sdram_ctrl_of_match[] = {
+ { .compatible = "altr,sdram-edac", },
+ {},
+};
+MODULE_DEVICE_TABLE(of, altr_sdram_ctrl_of_match);
+
+static struct platform_driver altr_sdram_edac_driver = {
+ .probe = altr_sdram_probe,
+ .remove = altr_sdram_remove,
+ .driver = {
+ .name = "altr_sdram_edac",
+ .of_match_table = altr_sdram_ctrl_of_match,
+ },
+};
+
+module_platform_driver(altr_sdram_edac_driver);
+
+MODULE_LICENSE("GPL v2");
+MODULE_AUTHOR("Altera Corporation");
+MODULE_DESCRIPTION("EDAC Driver for Altera SDRAM Controller");
--
1.7.9.5
^ permalink raw reply related [flat|nested] 6+ messages in thread