Re: [XenPPC] Track HID4[RM_CI] state for machine check

2006-09-08 Thread Jimi Xenidis


On Sep 8, 2006, at 10:40 AM, Hollis Blanchard wrote:


On Fri, 2006-09-08 at 02:10 +, Xen patchbot-xenppc-unstable wrote:

+static inline ulong read_scom(ulong addr)
+{
+union scomc c;
+ulong d;
+
+c.word = 0;
+c.bits.addr = addr;
+c.bits.RW = 0;


Shouldn't RW = 1 for a read request?


good point/catch, the supporting text says (11.6.3 SCOMD Format):

  "It is the source for outgoing data during an SCOM write access  
(SCOMC[RW] =‘1’ when mtscomc is issued). It is the destination for  
incoming data after an SCOMC read access (SCOMC[RW]=‘0’ after mtscomc  
completes)."


However in Table 11-15 SCOMC[48]:

  0 Write request
  1 Read request

I'll get the facts.
-JX
___
Xen-ppc-devel mailing list
Xen-ppc-devel@lists.xensource.com
http://lists.xensource.com/xen-ppc-devel


Re: [XenPPC] Track HID4[RM_CI] state for machine check

2006-09-08 Thread Hollis Blanchard
On Fri, 2006-09-08 at 02:10 +, Xen patchbot-xenppc-unstable wrote:
> +static inline ulong read_scom(ulong addr)
> +{
> +union scomc c;
> +ulong d;
> +
> +c.word = 0;
> +c.bits.addr = addr;
> +c.bits.RW = 0;

Shouldn't RW = 1 for a read request?

-- 
Hollis Blanchard
IBM Linux Technology Center


___
Xen-ppc-devel mailing list
Xen-ppc-devel@lists.xensource.com
http://lists.xensource.com/xen-ppc-devel