[PATCH] ipmi: Increase KCS timeouts

[PATCH] ipmi: Increase KCS timeouts

[Matthew Garrett]

We currently time out and retry KCS transactions after 1 second of waiting
for IBF or OBF. This appears to be too short for some hardware. The IPMI
spec says "All system software wait loops should include error timeouts. For
simplicity, such timeouts are not shown explicitly in the flow diagrams. A
five-second timeout or greater is recommended". Change the timeout to five
seconds to satisfy the slow hardware.

Signed-off-by: Matthew Garrett <mjg@redhat.com>
---
 drivers/char/ipmi/ipmi_kcs_sm.c |    4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/char/ipmi/ipmi_kcs_sm.c b/drivers/char/ipmi/ipmi_kcs_sm.c
index cf82fed..e53fc24 100644
--- a/drivers/char/ipmi/ipmi_kcs_sm.c
+++ b/drivers/char/ipmi/ipmi_kcs_sm.c
@@ -118,8 +118,8 @@ enum kcs_states {
 #define MAX_KCS_WRITE_SIZE IPMI_MAX_MSG_LENGTH
 
 /* Timeouts in microseconds. */
-#define IBF_RETRY_TIMEOUT 1000000
-#define OBF_RETRY_TIMEOUT 1000000
+#define IBF_RETRY_TIMEOUT 5000000
+#define OBF_RETRY_TIMEOUT 5000000
 #define MAX_ERROR_RETRIES 10
 #define ERROR0_OBF_WAIT_JIFFIES (2*HZ)
 
-- 
1.7.7.1

Re: [PATCH] ipmi: Increase KCS timeouts

[Matthew Garrett]

Ping?

On Wed, Nov 30, 2011 at 02:12:27PM -0500, Matthew Garrett wrote:
> We currently time out and retry KCS transactions after 1 second of waiting
> for IBF or OBF. This appears to be too short for some hardware. The IPMI
> spec says "All system software wait loops should include error timeouts. For
> simplicity, such timeouts are not shown explicitly in the flow diagrams. A
> five-second timeout or greater is recommended". Change the timeout to five
> seconds to satisfy the slow hardware.
> 
> Signed-off-by: Matthew Garrett <mjg@redhat.com>
> ---
>  drivers/char/ipmi/ipmi_kcs_sm.c |    4 ++--
>  1 files changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/char/ipmi/ipmi_kcs_sm.c b/drivers/char/ipmi/ipmi_kcs_sm.c
> index cf82fed..e53fc24 100644
> --- a/drivers/char/ipmi/ipmi_kcs_sm.c
> +++ b/drivers/char/ipmi/ipmi_kcs_sm.c
> @@ -118,8 +118,8 @@ enum kcs_states {
>  #define MAX_KCS_WRITE_SIZE IPMI_MAX_MSG_LENGTH
>  
>  /* Timeouts in microseconds. */
> -#define IBF_RETRY_TIMEOUT 1000000
> -#define OBF_RETRY_TIMEOUT 1000000
> +#define IBF_RETRY_TIMEOUT 5000000
> +#define OBF_RETRY_TIMEOUT 5000000
>  #define MAX_ERROR_RETRIES 10
>  #define ERROR0_OBF_WAIT_JIFFIES (2*HZ)
>  
> -- 
> 1.7.7.1
> 
> 
-- 
Matthew Garrett | mjg59@srcf.ucam.org

Re: [PATCH] ipmi: Increase KCS timeouts

[Corey Minyard]

This is fine, if it is really necessary to make slow hardware work.  
I've pulled it in, I'll submit in a bit.

-corey

On 12/14/2011 03:12 PM, Matthew Garrett wrote:
> Ping?
>
> On Wed, Nov 30, 2011 at 02:12:27PM -0500, Matthew Garrett wrote:
>> We currently time out and retry KCS transactions after 1 second of waiting
>> for IBF or OBF. This appears to be too short for some hardware. The IPMI
>> spec says "All system software wait loops should include error timeouts. For
>> simplicity, such timeouts are not shown explicitly in the flow diagrams. A
>> five-second timeout or greater is recommended". Change the timeout to five
>> seconds to satisfy the slow hardware.
>>
>> Signed-off-by: Matthew Garrett<mjg@redhat.com>
>> ---
>>   drivers/char/ipmi/ipmi_kcs_sm.c |    4 ++--
>>   1 files changed, 2 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/char/ipmi/ipmi_kcs_sm.c b/drivers/char/ipmi/ipmi_kcs_sm.c
>> index cf82fed..e53fc24 100644
>> --- a/drivers/char/ipmi/ipmi_kcs_sm.c
>> +++ b/drivers/char/ipmi/ipmi_kcs_sm.c
>> @@ -118,8 +118,8 @@ enum kcs_states {
>>   #define MAX_KCS_WRITE_SIZE IPMI_MAX_MSG_LENGTH
>>
>>   /* Timeouts in microseconds. */
>> -#define IBF_RETRY_TIMEOUT 1000000
>> -#define OBF_RETRY_TIMEOUT 1000000
>> +#define IBF_RETRY_TIMEOUT 5000000
>> +#define OBF_RETRY_TIMEOUT 5000000
>>   #define MAX_ERROR_RETRIES 10
>>   #define ERROR0_OBF_WAIT_JIFFIES (2*HZ)
>>
>> -- 
>> 1.7.7.1
>>
>>

Re: [PATCH] ipmi: Increase KCS timeouts

[Matthew Garrett]

On Thu, Dec 15, 2011 at 08:35:49AM -0600, Corey Minyard wrote:
> This is fine, if it is really necessary to make slow hardware work.
> I've pulled it in, I'll submit in a bit.

Perfect Corey, thanks!

-- 
Matthew Garrett | mjg59@srcf.ucam.org