ALP-pool/ppc64-diag
2
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Sync from SUSE:ALP:Source:Standard:1.0 ppc64-diag revision 68babf57b28a3079f7d752c5d55dcd94

Browse Source
This commit is contained in:
Adrian Schröter 2024-03-01 16:57:40 +01:00
parent 8587cb6388
commit 09405115a2
6 changed files with 310 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

42
diag_nvme-improvements-to-status-and-err-messages.patch Normal file
View File

@ -0,0 +1,42 @@
From 3ad587d20868f154bb7ab972ee7812add1380d7a Mon Sep 17 00:00:00 2001
From: Greg Joyce <gjoyce@linux.vnet.ibm.com>
Date: Wed, 11 Oct 2023 12:10:40 -0500
Subject: [PATCH] diag_nvme: improvements to status and err messages
Signed-off-by: Greg Joyce <gjoyce@linux.vnet.ibm.com>
---
diags/diag_nvme.c | 9 ++++++---
1 file changed, 6 insertions(+), 3 deletions(-)
diff --git a/diags/diag_nvme.c b/diags/diag_nvme.c
index df191f2d1ac8..00efec212056 100644
--- a/diags/diag_nvme.c
+++ b/diags/diag_nvme.c
@@ -166,9 +166,9 @@ int main(int argc, char *argv[]) {
}
if (rc == 0)
- fprintf(stdout, "Command completed successfully\n");
+ fprintf(stdout, "NVMe diag command completed successfully\n");
else
- fprintf(stderr, "Command failed, exiting with rc %d\n", rc);
+ fprintf(stderr, "NVMe diag command failed with rc %d\n", rc);
return rc;
}
@@ -724,8 +724,11 @@ extern int get_smart_file(char *file_path, struct nvme_smart_log_page *log) {
int num_elements = 0;
struct dictionary dict[MAX_DICT_ELEMENTS];
- if ((num_elements = read_file_dict(file_path, dict, MAX_DICT_ELEMENTS)) < 0)
+ if ((num_elements = read_file_dict(file_path, dict, MAX_DICT_ELEMENTS)) < 0) {
+ fprintf(stderr, "read_file_dict failed: %s, rc % d\n",
+ file_path, num_elements);
return num_elements;
+ }
return set_smart_log_field(log, dict, num_elements);
}
--
2.43.2

76
diags-diag_nvme-call_home-command-fails-on-nvmf-driv.patch Normal file
View File

@ -0,0 +1,76 @@
From db0c6d7974d7f8909878384d77ec02457759d6df Mon Sep 17 00:00:00 2001
From: Nilay Shroff <nilay@linux.ibm.com>
Date: Tue, 16 Jan 2024 13:55:03 +0530
Subject: [PATCH] diags/diag_nvme: call_home command fails on nvmf drive
The diag_nvme command needs to retrieve the VPD log page from NVMe for
filling in the product data while generating the call-home event.
However, call-home feature is supported for directly attached NVMe
module. In the current diag_nvme implementation, if user doesn't
provide NVMe device name for diagnostics then it(diag_nvme) loops
through each NVMe moudle (directly connected to the system/LPAR as
well as discovered over fabrics) and attempt retrieving the SMART log
page as well as VPD page. Unfortunately, diag_nvme fails to retrieve
the VPD page for NVMe connected over fabrics and that causes the
diag_nvme to print "not-so-nice" failure messages on console.
Henec fixed the diag_nvme code so that for call-home event reporting,
it skips the NVMe which is connected over fabrics and prints a
"nice-message" informing the user that it's skipping diagnosting for
NVMe module connected over fabrics. In a nutshell, with this fix now
diag_nvme would only diagnose the NVMe module which is directtly
attached (over PCIe) to the system.
Signed-off-by: Nilay Shroff <nilay@linux.ibm.com>
---
diags/diag_nvme.c | 31 +++++++++++++++++++++++++++++++
1 file changed, 31 insertions(+)
diff --git a/diags/diag_nvme.c b/diags/diag_nvme.c
index c1c0a20ddf14..e86786ccdccd 100644
--- a/diags/diag_nvme.c
+++ b/diags/diag_nvme.c
@@ -375,9 +375,40 @@ static int diagnose_nvme(char *device_name, struct notify *notify, char *file_pa
char endurance_s[sizeof(vpd.endurance) + 1], capacity_s[sizeof(vpd.capacity)+1];
uint64_t event_id;
uint8_t severity;
+ FILE *fp;
+ char tr_file_path[PATH_MAX];
uint32_t raw_data_len = 0;
unsigned char *raw_data = NULL;
+ /*
+ * Skip diag test if NVMe is connected over fabric
+ */
+ snprintf(tr_file_path, sizeof(tr_file_path),
+ NVME_SYS_PATH"/%s/%s", device_name, "transport");
+ fp = fopen(tr_file_path, "r");
+ if (fp) {
+ char buf[12];
+ int n = fread(buf, 1, sizeof(buf), fp);
+
+ if (n) {
+ /*
+ * If NVMe transport is anything but pcie then skip the diag test
+ */
+ if (strncmp(buf, "pcie", 4) != 0) {
+ fprintf(stdout, "Skipping diagnostics for nvmf : %s\n",
+ device_name);
+ fclose(fp);
+ return 0;
+ }
+ }
+ fclose(fp);
+ } else {
+ fprintf(stderr, "Skipping diagnostics for %s:\n"
+ "Unable to find the nvme transport type\n",
+ device_name);
+ return -1;
+ }
+
tmp_rc = regex_controller(controller_name, device_name);
if (tmp_rc != 0)
return -1;
--
2.43.2

149
nvme_call_home-remove-d-and-f-as-visible-options.patch Normal file
View File

@ -0,0 +1,149 @@
From 11cb2a44a59b63bdc23c94e386c4e2f43ea7eb61 Mon Sep 17 00:00:00 2001
From: Greg Joyce <gjoyce@linux.vnet.ibm.com>
Date: Fri, 22 Sep 2023 15:19:34 -0500
Subject: [PATCH] nvme_call_home: remove -d and -f as visible options
The dump(-d) and file(-f) options have been removed as option in usage
and also in the man page. The options are for internal test only and
should not have been exposed. Also added some helpful status messages.
Signed-off-by: Greg Joyce <gjoyce@linux.vnet.ibm.com>
---
diags/diag_nvme.c | 23 ++++++++++++++-------
diags/man/diag_nvme.8 | 48 +++----------------------------------------
2 files changed, 19 insertions(+), 52 deletions(-)
diff --git a/diags/diag_nvme.c b/diags/diag_nvme.c
index 2a78034ecfd9..df191f2d1ac8 100644
--- a/diags/diag_nvme.c
+++ b/diags/diag_nvme.c
@@ -165,6 +165,11 @@ int main(int argc, char *argv[]) {
optind++;
}
+ if (rc == 0)
+ fprintf(stdout, "Command completed successfully\n");
+ else
+ fprintf(stderr, "Command failed, exiting with rc %d\n", rc);
+
return rc;
}
@@ -310,6 +315,7 @@ extern int dump_smart_data(char *device_name, char *dump_path) {
int fd, rc;
FILE *fp;
struct nvme_smart_log_page smart_log = { 0 };
+ char ans;
/* Read SMART data from device */
snprintf(dev_path,sizeof(dev_path), "/dev/%s", device_name);
@@ -331,8 +337,16 @@ extern int dump_smart_data(char *device_name, char *dump_path) {
}
fp = fopen(dump_path, "wx");
if (fp == NULL) {
- fprintf(stderr, "%s open failed: %s\n", dump_path, strerror(errno));
- return -1;
+ if (errno == EEXIST) {
+ fprintf(stdout, "File %s exists. Overwrite (y/n)? ", dump_path);
+ rc = scanf("%c", &ans);
+ if (ans == 'y' || ans == 'Y')
+ fp = fopen(dump_path, "w");
+ }
+ if (fp == NULL) {
+ fprintf(stderr, "%s open failed: %s\n", dump_path, strerror(errno));
+ return -1;
+ }
}
write_smart_file(fp, &smart_log);
fclose(fp);
@@ -974,11 +988,6 @@ extern int open_nvme(char *dev_path) {
static void print_usage(char *command) {
printf("Usage: %s [-h] [-d <file>] [-f <file>] [<nvme_devices>]\n"
"\t-h or --help: print this help message\n"
- "\t-d or --dump: dump SMART data to the specified path and file name <file>\n"
- "\t one <nvme_device> is expected with this option\n"
- "\t-f or --file: use SMART data from the specified path and file name <file>\n"
- "\t instead of device, one <nvme_device> is expected with\n"
- "\t this option\n"
"\t<nvme_devices>: the NVMe devices on which to operate, for\n"
"\t example nvme0; if not specified, all detected\n"
"\t nvme devices will be diagnosed\n", command);
diff --git a/diags/man/diag_nvme.8 b/diags/man/diag_nvme.8
index 45567ac26bfc..c023ed6719eb 100644
--- a/diags/man/diag_nvme.8
+++ b/diags/man/diag_nvme.8
@@ -1,7 +1,7 @@
.\"
-.\" Copyright (C) 2022 IBM Corporation
+.\" Copyright (C) 2022, 2023 IBM Corporation
.\"
-.TH "DIAG_NVME" "8" "June 2022" "Linux" "PowerLinux Diagnostic Tools"
+.TH "DIAG_NVME" "8" "September 2023" "Linux" "PowerLinux Diagnostic Tools"
.hy
.SH NAME
.PP
@@ -12,7 +12,7 @@ diag_nvme \- diagnose NVMe devices
.PD 0
.P
.PD
-\f[B]diag_nvme\f[] [\-d <\f[I]file\f[]>] [\-f <\f[I]file\f[]>]
+\f[B]diag_nvme\f[]
<\f[B]nvme\f[]\f[I]n\f[]>
.PD 0
.P
@@ -29,52 +29,10 @@ go through the diagnostics procedure.
The user can control which events will be reported through the
configuration file \f[I]/etc/ppc64\-diag/diag_nvme.config\f[]
.SH OPTIONS
-.TP
-.B \f[B]\-d\f[], \f[B]--dump\f[] \f[I]file\f[]
-Dump SMART data to the specified path and file name \f[I]file\f[].
-The SMART data is extracted from an NVMe device, so specifying one is
-mandatory if this option is selected.
-File created is in a simple key=value format.
-.RS
-.RE
-.TP
-.B \f[B]\-f\f[], \f[B]--file\f[] \f[I]file\f[]
-This option usage is for testing only.
-Use SMART data from the specified path and file name \f[I]file\f[]
-instead of device, one NVMe is mandatory if this option is selected.
-The expected format of the file is a simple key=value that is the same
-one provided with the \-d / --dump option.
-If \f[I]file\f[] is missing from the filesystem it will be treated as a
-failure to retrieve SMART data and an event will be reported.
-.RS
-.RE
-.TP
-.B \f[B]\-h\f[], \f[B]--help\f[]
-Print a help message and exit
-.RS
-.RE
-.SH EXAMPLES
-.TP
-.B \f[B]diag_nvme\f[]
-Run diagnostics in all NVMe devices detected in the system.
-.RS
-.RE
-.TP
.B \f[B]diag_nvme nvme0 nvme1\f[]
Run diagnostics only in nvme0 and nvme1 devices.
.RS
.RE
-.TP
-.B \f[B]diag_nvme \-d smart.txt nvme0\f[]
-Dump SMART data from nvme0 into file smart.txt.
-.RS
-.RE
-.TP
-.B \f[B]diag_nvme \-f smart.txt nvme0\f[]
-Read SMART data from file smart.txt and use it as health information for
-diagnostics of device nvme0.
-.RS
-.RE
.SH REPORTING BUGS
.PP
Patches and issues may be submitted at
--
2.43.2

29
nvme_call_home-remove-d-and-f-from-usage-message.patch Normal file
View File

@ -0,0 +1,29 @@
From 316d2baf5dba0d00447a3ac49f2f95339dbdf5cd Mon Sep 17 00:00:00 2001
From: Greg Joyce <gjoyce@linux.vnet.ibm.com>
Date: Wed, 15 Nov 2023 11:28:05 -0600
Subject: [PATCH] nvme_call_home: remove -d and -f from usage message
The -d and -f weren't removed from the first line of the
usage message.
Signed-off-by: Greg Joyce <gjoyce@linux.vnet.ibm.com>
---
diags/diag_nvme.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/diags/diag_nvme.c b/diags/diag_nvme.c
index 00efec212056..164afe1cadd6 100644
--- a/diags/diag_nvme.c
+++ b/diags/diag_nvme.c
@@ -989,7 +989,7 @@ extern int open_nvme(char *dev_path) {
}
static void print_usage(char *command) {
- printf("Usage: %s [-h] [-d <file>] [-f <file>] [<nvme_devices>]\n"
+ printf("Usage: %s [-h] [<nvme_devices>]\n"
"\t-h or --help: print this help message\n"
"\t<nvme_devices>: the NVMe devices on which to operate, for\n"
"\t example nvme0; if not specified, all detected\n"
--
2.43.2

9
ppc64-diag.changes
View File

@ -1,3 +1,12 @@
-------------------------------------------------------------------
Mon Feb 26 14:22:42 UTC 2024 - Michal Suchanek <msuchanek@suse.de>
- Fix NVMe diagnostics support with nvmf drive (bsc#1220345 ltc#205706).
* nvme_call_home-remove-d-and-f-as-visible-options.patch
* nvme_call_home-remove-d-and-f-from-usage-message.patch
* diag_nvme-improvements-to-status-and-err-messages.patch
* diags-diag_nvme-call_home-command-fails-on-nvmf-driv.patch
-------------------------------------------------------------------
Fri Nov 17 18:18:03 UTC 2023 - Martin Schreiner <martin.schreiner@suse.com>

6
ppc64-diag.spec
View File

@ -1,7 +1,7 @@
#
# spec file for package ppc64-diag
#
# Copyright (c) 2023 SUSE LLC
# Copyright (c) 2024 SUSE LLC
#
# All modifications and additions to the file contributed by third parties
# remain the property of their copyright owners, unless otherwise agreed
@ -36,6 +36,10 @@ Patch2: rtas_errd-Handle-multiple-platform-dumps.patch
Patch3: 0001-ppc64-diag-Move-trim_trail_space-function-to-common-.patch
Patch4: 0002-ppc64-diag-lp_diag-Utilize-trim_trail_space-function.patch
Patch5: 0003-ppc64-diag-lp_diag-Enable-light-path-diagnostics-for.patch
Patch6: nvme_call_home-remove-d-and-f-as-visible-options.patch
Patch7: nvme_call_home-remove-d-and-f-from-usage-message.patch
Patch8: diag_nvme-improvements-to-status-and-err-messages.patch
Patch9: diags-diag_nvme-call_home-command-fails-on-nvmf-driv.patch
BuildRequires: autoconf
BuildRequires: automake
BuildRequires: bison