FEX-Emu/linux
1
0
Fork 0
mirror of https://github.com/FEX-Emu/linux.git synced 2024-12-20 16:30:53 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

[PATCH] ppc64: escape hatch for spinning interrupt deadlocks

Browse Source 
08-eeh-spin-counter.patch

One an EEH event is triggers, all further I/O to a device is blocked (until
reset).  Bad device drivers may end up spinning in their interrupt handlers,
trying to read an interrupt status register that will never change state.
This patch moves that spin counter to a per-device structure, and adds
some diagnostic prints to help locate the bad driver.

Signed-off-by: Linas Vepstas <linas@linas.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
This commit is contained in:
Linas Vepstas 2005-11-03 18:49:31 -06:00 committed by Paul Mackerras
parent fd761fd876
commit 5c1344e988
1 changed files with 13 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

21
arch/ppc64/kernel/eeh.c
View File

@ -78,14 +78,12 @@ DECLARE_WORK(eeh_event_wq, eeh_event_handler, NULL);
static struct notifier_block *eeh_notifier_chain;
/*
* If a device driver keeps reading an MMIO register in an interrupt
/* If a device driver keeps reading an MMIO register in an interrupt
* handler after a slot isolation event has occurred, we assume it
* is broken and panic. This sets the threshold for how many read
* attempts we allow before panicking.
*/
#define EEH_MAX_FAILS 1000
static atomic_t eeh_fail_count;
#define EEH_MAX_FAILS 100000
/* RTAS tokens */
static int ibm_set_eeh_option;
@ -521,7 +519,6 @@ static void eeh_event_handler(void *dummy)
"%s\n", event->reset_state,
pci_name(event->dev));
atomic_set(&eeh_fail_count, 0);
notifier_call_chain (&eeh_notifier_chain,
EEH_NOTIFY_FREEZE, event);
@ -657,12 +654,18 @@ int eeh_dn_check_failure(struct device_node *dn, struct pci_dev *dev)
spin_lock_irqsave(&confirm_error_lock, flags);
rc = 1;
if (pdn->eeh_mode & EEH_MODE_ISOLATED) {
atomic_inc(&eeh_fail_count);
if (atomic_read(&eeh_fail_count) >= EEH_MAX_FAILS) {
pdn->eeh_check_count ++;
if (pdn->eeh_check_count >= EEH_MAX_FAILS) {
printk (KERN_ERR "EEH: Device driver ignored %d bad reads, panicing\n",
pdn->eeh_check_count);
dump_stack();
/* re-read the slot reset state */
if (read_slot_reset_state(pdn, rets) != 0)
rets[0] = -1; /* reset state unknown */
eeh_panic(dev, rets[0]);
/* If we are here, then we hit an infinite loop. Stop. */
panic("EEH: MMIO halt (%d) on device:%s\n", rets[0], pci_name(dev));
}
goto dn_unlock;
}
@ -808,6 +811,8 @@ static void *early_enable_eeh(struct device_node *dn, void *data)
struct pci_dn *pdn = PCI_DN(dn);
pdn->eeh_mode = 0;
pdn->eeh_check_count = 0;
pdn->eeh_freeze_count = 0;
if (status && strcmp(status, "ok") != 0)
return NULL; /* ignore devices with bad status */