Re: kernel oops and panic in acpi_atomic_read under 2.6.39.3. calltrace included

From: Huang Ying
Date: Thu Aug 25 2011 - 20:34:17 EST


On 08/25/2011 11:47 PM, rick@xxxxxxxxxxxx wrote:
> Hi Huang,
>
> My new setup reproduced the panic. However I do not have any gar accessed
> messages on it. The gar mapped messages are in my previous email. Here
> is the latest call trace. There is no GHES output prior to it:
>

That is wired. Can you try the patch attached? If my guessing is
correct, there will be no panic, but something as follow will be in dmesg:

ACPI atomic read mem: addr 0xxxxx mapped to 0

Best Regards,
Huang Ying
---
drivers/acpi/apei/ghes.c | 6 ++++++
drivers/acpi/atomicio.c | 4 ++++
2 files changed, 10 insertions(+)

--- a/drivers/acpi/apei/ghes.c
+++ b/drivers/acpi/apei/ghes.c
@@ -299,6 +299,9 @@ static struct ghes *ghes_new(struct acpi
return ERR_PTR(-ENOMEM);
ghes->generic = generic;
rc = acpi_pre_map_gar(&generic->error_status_address);
+ pr_info(GHES_PFX "gar mapped: %d, 0x%llx\n",
+ generic->error_status_address.space_id,
+ generic->error_status_address.address);
if (rc)
goto err_free;
error_block_length = generic->error_block_length;
@@ -398,6 +401,9 @@ static int ghes_read_estatus(struct ghes
u32 len;
int rc;

+ pr_err(GHES_PFX "gar accessed: %d, 0x%llx\n",
+ g->error_status_address.space_id,
+ g->error_status_address.address);
rc = acpi_atomic_read(&buf_paddr, &g->error_status_address);
if (rc) {
if (!silent && printk_ratelimit())
--- a/drivers/acpi/atomicio.c
+++ b/drivers/acpi/atomicio.c
@@ -270,6 +270,10 @@ static int acpi_atomic_read_mem(u64 padd

rcu_read_lock();
addr = __acpi_ioremap_fast(paddr, width);
+ pr_err("ACPI atomic read mem: addr 0x%llx mapped to %p\n",
+ paddr, addr);
+ if (!addr)
+ return -EIO;
switch (width) {
case 8:
*val = readb(addr);