From: Gavin Shan <shangw@linux.vnet.ibm.com>
To: linuxppc-dev@ozlabs.org
Cc: Gavin Shan <shangw@linux.vnet.ibm.com>
Subject: [PATCH 12/21] ppc/eeh: trace error based on PE from beginning
Date: Thu, 28 Jun 2012 00:01:42 +0800 [thread overview]
Message-ID: <1340812911-6793-13-git-send-email-shangw@linux.vnet.ibm.com> (raw)
In-Reply-To: <1340812911-6793-1-git-send-email-shangw@linux.vnet.ibm.com>
There're 2 conditions to trigger EEH error detection: invalid value
returned from reading I/O or config space. On each case, the function
eeh_dn_check_failure will be called to initialize EEH event and put
it into the poll for further processing.
The patch changes the function for a little bit so that the EEH error
will be traced based on PE instead of EEH device any more. Also, the
function eeh_find_device_pe() has been removed since the eeh device
is tracing the PE by struct eeh_dev::pe.
Signed-off-by: Gavin Shan <shangw@linux.vnet.ibm.com>
---
arch/powerpc/include/asm/ppc-pci.h | 1 -
arch/powerpc/platforms/pseries/eeh.c | 51 +++++++++++++---------------------
arch/powerpc/platforms/pseries/msi.c | 6 +++-
3 files changed, 25 insertions(+), 33 deletions(-)
diff --git a/arch/powerpc/include/asm/ppc-pci.h b/arch/powerpc/include/asm/ppc-pci.h
index c7e5bd6..3e301b1 100644
--- a/arch/powerpc/include/asm/ppc-pci.h
+++ b/arch/powerpc/include/asm/ppc-pci.h
@@ -59,7 +59,6 @@ int rtas_write_config(struct pci_dn *, int where, int size, u32 val);
int rtas_read_config(struct pci_dn *, int where, int size, u32 *val);
void eeh_pe_state_mark(struct eeh_pe *pe, int state);
void eeh_pe_state_clear(struct eeh_pe *pe, int state);
-struct device_node *eeh_find_device_pe(struct device_node *dn);
void eeh_sysfs_add_device(struct pci_dev *pdev);
void eeh_sysfs_remove_device(struct pci_dev *pdev);
diff --git a/arch/powerpc/platforms/pseries/eeh.c b/arch/powerpc/platforms/pseries/eeh.c
index c527c46..341ba1a 100644
--- a/arch/powerpc/platforms/pseries/eeh.c
+++ b/arch/powerpc/platforms/pseries/eeh.c
@@ -264,21 +264,6 @@ static inline unsigned long eeh_token_to_phys(unsigned long token)
}
/**
- * eeh_find_device_pe - Retrieve the PE for the given device
- * @dn: device node
- *
- * Return the PE under which this device lies
- */
-struct device_node *eeh_find_device_pe(struct device_node *dn)
-{
- while (dn->parent && of_node_to_eeh_dev(dn->parent) &&
- (of_node_to_eeh_dev(dn->parent)->mode & EEH_MODE_SUPPORTED)) {
- dn = dn->parent;
- }
- return dn;
-}
-
-/**
* eeh_dn_check_failure - Check if all 1's data is due to EEH slot freeze
* @dn: device node
* @dev: pci device, if known
@@ -297,6 +282,7 @@ int eeh_dn_check_failure(struct device_node *dn, struct pci_dev *dev)
{
int ret;
unsigned long flags;
+ struct eeh_pe *pe;
struct eeh_dev *edev;
int rc = 0;
const char *location;
@@ -306,23 +292,26 @@ int eeh_dn_check_failure(struct device_node *dn, struct pci_dev *dev)
if (!eeh_subsystem_enabled)
return 0;
- if (!dn) {
+ if (dn) {
+ edev = of_node_to_eeh_dev(dn);
+ } else if (dev) {
+ edev = pci_dev_to_eeh_dev(dev);
+ dn = pci_device_to_OF_node(dev);
+ } else {
eeh_stats.no_dn++;
return 0;
}
- dn = eeh_find_device_pe(dn);
- edev = of_node_to_eeh_dev(dn);
+ pe = edev->pe;
/* Access to IO BARs might get this far and still not want checking. */
- if (!(edev->mode & EEH_MODE_SUPPORTED) ||
- edev->mode & EEH_MODE_NOCHECK) {
+ if (!pe) {
eeh_stats.ignored_check++;
- pr_debug("EEH: Ignored check (%x) for %s %s\n",
- edev->mode, eeh_pci_name(dev), dn->full_name);
+ pr_debug("EEH: Ignored check for %s %s\n",
+ eeh_pci_name(dev), dn->full_name);
return 0;
}
- if (!edev->config_addr && !edev->pe_config_addr) {
+ if (!pe->addr && !pe->config_addr) {
eeh_stats.no_cfg_addr++;
return 0;
}
@@ -335,13 +324,13 @@ int eeh_dn_check_failure(struct device_node *dn, struct pci_dev *dev)
*/
raw_spin_lock_irqsave(&confirm_error_lock, flags);
rc = 1;
- if (edev->mode & EEH_MODE_ISOLATED) {
- edev->check_count++;
- if (edev->check_count % EEH_MAX_FAILS == 0) {
+ if (pe->state & EEH_PE_ISOLATED) {
+ pe->check_count++;
+ if (pe->check_count % EEH_MAX_FAILS == 0) {
location = of_get_property(dn, "ibm,loc-code", NULL);
printk(KERN_ERR "EEH: %d reads ignored for recovering device at "
"location=%s driver=%s pci addr=%s\n",
- edev->check_count, location,
+ pe->check_count, location,
eeh_driver_name(dev), eeh_pci_name(dev));
printk(KERN_ERR "EEH: Might be infinite loop in %s driver\n",
eeh_driver_name(dev));
@@ -357,7 +346,7 @@ int eeh_dn_check_failure(struct device_node *dn, struct pci_dev *dev)
* function zero of a multi-function device.
* In any case they must share a common PHB.
*/
- ret = eeh_ops->get_state(dn, NULL);
+ ret = eeh_ops->get_state(pe, NULL);
/* Note that config-io to empty slots may fail;
* they are empty when they don't have children.
@@ -370,7 +359,7 @@ int eeh_dn_check_failure(struct device_node *dn, struct pci_dev *dev)
(ret & (EEH_STATE_MMIO_ACTIVE | EEH_STATE_DMA_ACTIVE)) ==
(EEH_STATE_MMIO_ACTIVE | EEH_STATE_DMA_ACTIVE)) {
eeh_stats.false_positives++;
- edev->false_positives ++;
+ pe->false_positives++;
rc = 0;
goto dn_unlock;
}
@@ -381,10 +370,10 @@ int eeh_dn_check_failure(struct device_node *dn, struct pci_dev *dev)
* with other functions on this device, and functions under
* bridges.
*/
- eeh_mark_slot(dn, EEH_MODE_ISOLATED);
+ eeh_pe_state_mark(pe, EEH_PE_ISOLATED);
raw_spin_unlock_irqrestore(&confirm_error_lock, flags);
- eeh_send_failure_event(edev);
+ eeh_send_failure_event(pe);
/* Most EEH events are due to device driver bugs. Having
* a stack trace will help the device-driver authors figure
diff --git a/arch/powerpc/platforms/pseries/msi.c b/arch/powerpc/platforms/pseries/msi.c
index 109fdb7..c8534fa 100644
--- a/arch/powerpc/platforms/pseries/msi.c
+++ b/arch/powerpc/platforms/pseries/msi.c
@@ -210,6 +210,7 @@ static struct device_node *find_pe_total_msi(struct pci_dev *dev, int *total)
static struct device_node *find_pe_dn(struct pci_dev *dev, int *total)
{
struct device_node *dn;
+ struct eeh_dev *edev;
/* Found our PE and assume 8 at that point. */
@@ -217,7 +218,10 @@ static struct device_node *find_pe_dn(struct pci_dev *dev, int *total)
if (!dn)
return NULL;
- dn = eeh_find_device_pe(dn);
+ /* Get the top level device in the PE */
+ edev = of_node_to_eeh_dev(dn);
+ edev = list_first_entry(&edev->pe->edevs, struct eeh_dev, list);
+ dn = eeh_dev_to_of_node(edev);
if (!dn)
return NULL;
--
1.7.9.5
next prev parent reply other threads:[~2012-06-27 17:38 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-06-27 16:01 [PATCH V2 00/16] powerpc/eeh: PE support Gavin Shan
2012-06-27 16:01 ` [PATCH 01/21] ppc/eeh: move EEH initialization around Gavin Shan
2012-06-27 16:01 ` [PATCH 02/21] ppc/eeh: use slab to allocate eeh devices Gavin Shan
2012-06-27 16:01 ` [PATCH 03/21] ppc/eeh: more logs for EEH initialization Gavin Shan
2012-06-27 23:45 ` Michael Ellerman
2012-06-28 2:40 ` Gavin Shan
2012-06-27 16:01 ` [PATCH 04/21] ppc/eeh: Introduce eeh_pe struct Gavin Shan
2012-06-27 16:01 ` [PATCH 05/21] ppc/eeh: introduce global mutex Gavin Shan
2012-06-27 16:01 ` [PATCH 06/21] ppc/eeh: Create PEs for PHBs Gavin Shan
2012-06-27 16:01 ` [PATCH 07/21] ppc/eeh: Search PE based on requirement Gavin Shan
2012-06-27 16:01 ` [PATCH 08/21] ppc/eeh: create PEs duing EEH initialization Gavin Shan
2012-06-27 16:01 ` [PATCH 09/21] ppc/eeh: remove PE at appropriate time Gavin Shan
2012-06-27 16:01 ` [PATCH 10/21] ppc/eeh: build EEH event based on PE Gavin Shan
2012-06-27 16:01 ` [PATCH 11/21] ppc/eeh: trace EEH state " Gavin Shan
2012-06-27 16:01 ` Gavin Shan [this message]
2012-06-27 16:01 ` [PATCH 13/21] ppc/eeh: eeh options " Gavin Shan
2012-06-27 16:01 ` [PATCH 14/21] ppc/eeh: device bars restore " Gavin Shan
2012-06-27 16:01 ` [PATCH 15/21] ppc/eeh: I/O enable and log retrival " Gavin Shan
2012-06-27 16:01 ` [PATCH 16/21] ppc/eeh: do reset " Gavin Shan
2012-06-27 16:01 ` [PATCH 17/21] ppc/eeh: make EEH handler PE sensitive Gavin Shan
2012-06-27 16:01 ` [PATCH 18/21] ppc/eeh: handle EEH error based on PE Gavin Shan
2012-06-27 16:01 ` [PATCH 19/21] ppc/eeh: move stats to PE Gavin Shan
2012-06-27 16:01 ` [PATCH 20/21] ppc/eeh: probe mode support Gavin Shan
2012-06-27 16:01 ` [PATCH 21/21] ppc/eeh: trace eeh device from I/O cache Gavin Shan
2012-06-27 16:05 ` [PATCH V2 00/16] powerpc/eeh: PE support Gavin Shan
2012-09-05 6:14 [PATCH 00/21 V3] " Gavin Shan
2012-09-05 6:14 ` [PATCH 12/21] ppc/eeh: trace error based on PE from beginning Gavin Shan
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1340812911-6793-13-git-send-email-shangw@linux.vnet.ibm.com \
--to=shangw@linux.vnet.ibm.com \
--cc=linuxppc-dev@ozlabs.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).