File s390-tools-sles15sp2-01-zpcictl-Initiate-recover-after-reset.patch of Package s390-tools.15659
Subject: [PATCH] [BZ 184174] zpcictl: Initiate recover after reset
From: Jan Hoeppner <hoeppner@linux.ibm.com>
Description: zpcictl: Initiate recover after reset
Symptom: If a PCI function is reset using zpcictl --reset, the function
is in an error state.
Problem: zpcictl --reset only issues a SCLP reset and leaves the PCI
function in an error state.
Solution: Initiate an OS level recovery by calling
/sys/bus/devices/<dev>/recover after the SCLP reset.
Reproduction: Call zpcictl --reset <dev>
Under z/VM check the state of the function with 'vmcp q pcif'
Upstream-ID: bc0d40c5803d4c5426b17b6d59aa0f1e46a2aacc
Problem-ID: 184174
Upstream-Description:
zpcictl: Initiate recover after reset
After a zpcitctl --reset the PCI function is currently left in an error
state. This seems unexpected, so follow the SCLP reset with an OS level
recovery using /sys/bus/devices/<dev>/recover.
Signed-off-by: Niklas Schnelle <schnelle@linux.ibm.com>
Reviewed-by: Jan Hoeppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Hoeppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Hoeppner <hoeppner@linux.ibm.com>
---
zpcictl/zpcictl.c | 21 +++++++++++++++++++++
1 file changed, 21 insertions(+)
--- a/zpcictl/zpcictl.c
+++ b/zpcictl/zpcictl.c
@@ -163,6 +163,26 @@ static unsigned int sysfs_read_value(str
return val;
}
+static void sysfs_write_value(struct zpci_device *pdev, const char *attr,
+ unsigned int val)
+{
+ char *path;
+ FILE *fp;
+
+ path = util_path_sysfs("bus/pci/devices/%s/%s", pdev->slot, attr);
+ fp = fopen(path, "w");
+ if (!fp)
+ fopen_err(path);
+ if (fprintf(fp, "%x", val) < 0) {
+ fclose(fp);
+ warnx("Could not write to file %s: %s", path, strerror(errno));
+ free(path);
+ exit(EXIT_FAILURE);
+ }
+ fclose(fp);
+ free(path);
+}
+
static void sysfs_write_data(struct zpci_report_error *report, char *slot)
{
size_t r_size;
@@ -297,6 +317,7 @@ static void sclp_issue_action(struct zpc
static void sclp_reset_device(struct zpci_device *pdev)
{
sclp_issue_action(pdev, SCLP_ERRNOTIFY_AQ_RESET);
+ sysfs_write_value(pdev, "recover", 1);
}
/*
