Re: [PATCH] powerpc/le: enable RTAS events support

2014-03-31 Thread Stewart Smith
Benjamin Herrenschmidt writes: > On Mon, 2014-03-31 at 09:27 +1100, Stewart Smith wrote: >> Greg Kurz writes: >> > struct rtas_error_log { >> > +#ifdef __BIG_ENDIAN__ >> > + /* Byte 0 */ >> >unsigned long version:8;/* Architectural version */ >> > + /* Byte 1 */ >> >> I th

Re: [PATCH] powerpc/le: enable RTAS events support

2014-03-31 Thread Stewart Smith
Greg Kurz writes: > On Mon, 31 Mar 2014 09:27:16 +1100 > Stewart Smith wrote: >> Greg Kurz writes: >> > struct rtas_error_log { >> > +#ifdef __BIG_ENDIAN__ >> > + /* Byte 0 */ >> >unsigned long version:8;/* Architectural version */ >> > + /* Byte 1 */ >> >> I think it wou

Re: [PATCH] powerpc/le: enable RTAS events support

2014-03-31 Thread Benjamin Herrenschmidt
On Mon, 2014-03-31 at 09:27 +1100, Stewart Smith wrote: > Greg Kurz writes: > > struct rtas_error_log { > > +#ifdef __BIG_ENDIAN__ > > + /* Byte 0 */ > > unsigned long version:8;/* Architectural version */ > > + /* Byte 1 */ > > I think it would be great if we got rid of

Re: [PATCH] powerpc/le: enable RTAS events support

2014-03-31 Thread Greg Kurz
On Mon, 31 Mar 2014 09:27:16 +1100 Stewart Smith wrote: > Greg Kurz writes: > > struct rtas_error_log { > > +#ifdef __BIG_ENDIAN__ > > + /* Byte 0 */ > > unsigned long version:8;/* Architectural version */ > > + /* Byte 1 */ > > I think it would be great if we got rid of

Re: [PATCH] powerpc/le: enable RTAS events support

2014-03-30 Thread Stewart Smith
Greg Kurz writes: > struct rtas_error_log { > +#ifdef __BIG_ENDIAN__ > + /* Byte 0 */ > unsigned long version:8;/* Architectural version */ > + /* Byte 1 */ I think it would be great if we got rid of the usage of bitfields. As soon as the mood of the compiler change

Re: [PATCH] powerpc/le: enable RTAS events support

2014-03-28 Thread Greg Kurz
On Fri, 28 Mar 2014 09:57:03 -0500 Nathan Fontenot wrote: > Greg, > > There is one more place that needs fixing up, in mobility_rtas_call(), > and handle_rtas_event() in arch/powerpc/platforms/pseries/mobility.c. > > This relates to rtas event handling for PRRN notifications, we need to > conver

Re: [PATCH] powerpc/le: enable RTAS events support

2014-03-28 Thread Nathan Fontenot
Greg, There is one more place that needs fixing up, in mobility_rtas_call(), and handle_rtas_event() in arch/powerpc/platforms/pseries/mobility.c. This relates to rtas event handling for PRRN notifications, we need to convert the scope variable (PRRN notifications re-use the extended log length f

Re: [PATCH] powerpc/le: enable RTAS events support

2014-03-28 Thread Greg Kurz
On Fri, 28 Mar 2014 09:29:22 +0100 Laurent Dufour wrote: > On 28/03/2014 08:33, Greg Kurz wrote: > > The current kernel code assumes big endian and parses RTAS events all > > wrong. The most visible effect is that we cannot honor EPOW events, > > meaning, for example, we cannot shut down a guest p

Re: [PATCH] powerpc/le: enable RTAS events support

2014-03-28 Thread Laurent Dufour
On 28/03/2014 08:33, Greg Kurz wrote: > The current kernel code assumes big endian and parses RTAS events all > wrong. The most visible effect is that we cannot honor EPOW events, > meaning, for example, we cannot shut down a guest properly from the > hypervisor. > > This patch fixes that. > > Si