Re: [PATCH v9 1/3] edac: Add support for Amazon's Annapurna Labs L1 EDAC

From: Hawa, Hanna
Date: Wed Mar 04 2020 - 04:03:01 EST



On 2/25/2020 8:52 PM, Borislav Petkov wrote:
On Wed, Jan 29, 2020 at 09:50:14PM +0200, Hanna Hawa wrote:

Adds support for Amazon's Annapurna Labs L1 EDAC driver to detect and
report L1 errors.

You don't add support for a driver - you either add a driver or you add
support for HW...

diff --git a/drivers/edac/al_l1_edac.c b/drivers/edac/al_l1_edac.c
new file mode 100644
index 000000000000..723b35b18f5b
--- /dev/null
+++ b/drivers/edac/al_l1_edac.c
@@ -0,0 +1,207 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright 2019 Amazon.com, Inc. or its affiliates. All Rights Reserved.
+ */
+
+#include <asm/sysreg.h>
+#include <linux/bitfield.h>
+#include <linux/of.h>
+#include <linux/platform_device.h>
+#include <linux/smp.h>

asm/ includes go after linux/ includes.

Will be fixed


+
+#include "edac_device.h"
+#include "edac_module.h"
+
+#define DRV_NAME "al_l1_edac"
+
+/* Same bit assignments of CPUMERRSR_EL1 in ARM CA57/CA72 */
+#define ARM_CA57_CPUMERRSR_EL1 sys_reg(3, 1, 15, 2, 2)
+#define ARM_CA57_CPUMERRSR_RAM_ID GENMASK(30, 24)
+#define ARM_CA57_L1_I_TAG_RAM 0x00
+#define ARM_CA57_L1_I_DATA_RAM 0x01
+#define ARM_CA57_L1_D_TAG_RAM 0x08
+#define ARM_CA57_L1_D_DATA_RAM 0x09
+#define ARM_CA57_L2_TLB_RAM 0x18
+#define ARM_CA57_CPUMERRSR_VALID BIT(31)
+#define ARM_CA57_CPUMERRSR_REPEAT GENMASK_ULL(39, 32)
+#define ARM_CA57_CPUMERRSR_OTHER GENMASK_ULL(47, 40)
+#define ARM_CA57_CPUMERRSR_FATAL BIT_ULL(63)
+
+#define AL_L1_EDAC_MSG_MAX 256
+
+static void al_l1_edac_cpumerrsr_read_status(void *arg)

This is a static function so you don't really need a function prefix of
"al_l1_edac_". Use those chars to make its name more descriptive. Ditto
for the rest of the static functions in that file.

Will be fixed.


+{
+ struct edac_device_ctl_info *edac_dev = arg;
+ int cpu, space, count;
+ u32 ramid, repeat, other, fatal;
+ u64 val;
+ char msg[AL_L1_EDAC_MSG_MAX];
+ char *p;
+ spinlock_t *lock;

Please sort function local variables declaration in a reverse christmas
tree order:

<type A> longest_variable_name;
<type B> shorter_var_name;
<type C> even_shorter;
<type D> i;

Will be fixed


+
+ val = read_sysreg_s(ARM_CA57_CPUMERRSR_EL1);
+ if (!(FIELD_GET(ARM_CA57_CPUMERRSR_VALID, val)))
+ return;
+
+ write_sysreg_s(0, ARM_CA57_CPUMERRSR_EL1);
+
+ cpu = smp_processor_id();
+ ramid = FIELD_GET(ARM_CA57_CPUMERRSR_RAM_ID, val);
+ repeat = FIELD_GET(ARM_CA57_CPUMERRSR_REPEAT, val);
+ other = FIELD_GET(ARM_CA57_CPUMERRSR_OTHER, val);
+ fatal = FIELD_GET(ARM_CA57_CPUMERRSR_FATAL, val);
+
+ space = sizeof(msg);
+ p = msg;
+ count = scnprintf(p, space, "CPU%d L1 %serror detected", cpu,
+ (fatal) ? "Fatal " : "");
+ p += count;
+ space -= count;
+
+ switch (ramid) {
+ case ARM_CA57_L1_I_TAG_RAM:
+ count = scnprintf(p, space, " RAMID='L1-I Tag RAM'");
+ break;
+ case ARM_CA57_L1_I_DATA_RAM:
+ count = scnprintf(p, space, " RAMID='L1-I Data RAM'");
+ break;
+ case ARM_CA57_L1_D_TAG_RAM:
+ count = scnprintf(p, space, " RAMID='L1-D Tag RAM'");
+ break;
+ case ARM_CA57_L1_D_DATA_RAM:
+ count = scnprintf(p, space, " RAMID='L1-D Data RAM'");
+ break;
+ case ARM_CA57_L2_TLB_RAM:
+ count = scnprintf(p, space, " RAMID='L2 TLB RAM'");
+ break;
+ default:
+ count = scnprintf(p, space, " RAMID='unknown'");
+ break;
+ }
+
+ p += count;
+ space -= count;
+ count = scnprintf(p, space,
+ " repeat=%d, other=%d (CPUMERRSR_EL1=0x%llx)",
+ repeat, other, val);
+
+ lock = edac_dev->pvt_info;

That sure looks funky - why not use a proper private struct like the
rest of the drivers?

Then the structure will include single variable..


+ spin_lock(lock);
+ if (fatal)
+ edac_device_handle_ue_count(edac_dev, repeat, 0, 0, msg);
+ else
+ edac_device_handle_ce_count(edac_dev, repeat, 0, 0, msg);
+ spin_unlock(lock);
+}
+
+static void al_l1_edac_check(struct edac_device_ctl_info *edac_dev)
+{
+ on_each_cpu(al_l1_edac_cpumerrsr_read_status, edac_dev, 1);
+}
+
+static int al_l1_edac_probe(struct platform_device *pdev)
+{
+ struct edac_device_ctl_info *edac_dev;
+ struct device *dev = &pdev->dev;
+ spinlock_t *lock;
+ int ret;
+
+ edac_dev = edac_device_alloc_ctl_info(sizeof(*lock), DRV_NAME, 1, "L",
+ 1, 1, NULL, 0,
+ edac_device_alloc_index());
+ if (!edac_dev)
+ return -ENOMEM;
+
+ edac_dev->edac_check = al_l1_edac_check;
+ edac_dev->dev = dev;
+ edac_dev->mod_name = DRV_NAME;
+ edac_dev->dev_name = dev_name(dev);
+ edac_dev->ctl_name = "L1_cache";
+ platform_set_drvdata(pdev, edac_dev);
+ lock = edac_dev->pvt_info;
+
+ spin_lock_init(lock);
+
+ ret = edac_device_add_device(edac_dev);
+ if (ret) {
+ dev_err(dev, "Failed to add L1 edac device (%d)\n", ret);
+ edac_device_free_ctl_info(edac_dev);
+ return ret;
+ }
+
+ return 0;
+}
+
+static int al_l1_edac_remove(struct platform_device *pdev)
+{
+ struct edac_device_ctl_info *edac_dev = platform_get_drvdata(pdev);
+
+ edac_device_del_device(edac_dev->dev);
+ edac_device_free_ctl_info(edac_dev);
+
+ return 0;
+}
+
+static const struct of_device_id al_l1_edac_of_match[] = {
+ /*
+ * "al,alpine-v2", and "amazon,al-alpine-v3" are machine compatible
+ * strings which have Cortex-A57/A72 configured with this support,
+ * and access to CPUMERRSR_EL1 register is enabled in firmware.
+ */
+ { .compatible = "al,alpine-v2" },

Err, checkpatch says:

WARNING: DT compatible string "al,alpine-v2" appears un-documented -- check ./Documentation/devicetree/bindings/
#236: FILE: drivers/edac/al_l1_edac.c:151:
+ { .compatible = "al,alpine-v2" },


Do you need a devicetree addition? I usually get such with the ARM EDAC
drivers...

There is missing documentation for alpine-v2, there is an open patches to add this missing doc. [1]

[1]: https://lkml.org/lkml/2020/2/25/362


+ { .compatible = "amazon,al-alpine-v3" },
+ {}
+};
+MODULE_DEVICE_TABLE(of, al_l1_edac_of_match);
+
+static struct platform_driver al_l1_edac_driver = {
+ .probe = al_l1_edac_probe,
+ .remove = al_l1_edac_remove,
+ .driver = {
+ .name = DRV_NAME,
+ },
+};
+
+static struct platform_device *edac_l1_device;

Please move to the top of the file.

Ack.


+
+static int __init al_l1_init(void)
+{
+ struct device_node *root;
+ int ret;
+
+ root = of_find_node_by_path("/");
+ if (!root) {
+ pr_debug("Can't find root node!\n");
+ return 0;
+ }
+
+ if (!of_match_node(al_l1_edac_of_match, root))
+ return 0;
+
+ ret = platform_driver_register(&al_l1_edac_driver);
+ if (ret) {
+ pr_err("Failed to register %s (%d)\n", DRV_NAME, ret);
+ return ret;
+ }
+
+ edac_l1_device = platform_device_register_simple(DRV_NAME, -1, NULL, 0);
+ if (IS_ERR(edac_l1_device)) {
+ pr_err("Failed to register EDAC AL L1 platform device\n");

platform_driver_unregister() ?

Ack


+ return PTR_ERR(edac_l1_device);
+ }
+
+ return 0;
+}
+
+static void __exit al_l1_exit(void)
+{
+ platform_device_unregister(edac_l1_device);
+ platform_driver_unregister(&al_l1_edac_driver);
+}
+
+late_initcall(al_l1_init);

Why not module_init() ?

Will be fixed.

Thanks,
Hanna