Re: [Qemu-devel] [PATCH 4/5] sh_intc: convert interrupt controller to memory API

2011-11-17 Thread Benoît Canet
Yes, allocating the exact size would make the realloc unnecessary. But it involve more code to walk one more time through the mask_reg and prio_reg array before allocating. I made it this way to make code shorter. The freed MemoryRegion are not initialized at all. However as realloc seems to be a

Re: [Qemu-devel] [PATCH 4/5] sh_intc: convert interrupt controller to memory API

2011-11-17 Thread Avi Kivity
On 11/17/2011 02:56 PM, Peter Maydell wrote: > 2011/11/17 Benoît Canet : > > Signed-off-by: Benoit Canet > > --- a/hw/sh7750.c > > +++ b/hw/sh7750.c > > @@ -756,7 +756,7 @@ SH7750State *sh7750_init(CPUSH4State * cpu, > > MemoryRegion *sysmem) > > "cache-and-tlb", 0x08000

Re: [Qemu-devel] [PATCH 4/5] sh_intc: convert interrupt controller to memory API

2011-11-17 Thread Peter Maydell
2011/11/17 Benoît Canet : > Signed-off-by: Benoit Canet > --- a/hw/sh7750.c > +++ b/hw/sh7750.c > @@ -756,7 +756,7 @@ SH7750State *sh7750_init(CPUSH4State * cpu, MemoryRegion > *sysmem) >                           "cache-and-tlb", 0x0800); >     memory_region_add_subregion(sysmem, 0xf000,

[Qemu-devel] [PATCH 4/5] sh_intc: convert interrupt controller to memory API

2011-11-17 Thread Benoît Canet
Signed-off-by: Benoit Canet --- hw/sh7750.c |2 +- hw/sh_intc.c| 87 ++ hw/sh_intc.h|7 +++- target-sh4/helper.c |3 ++ 4 files changed, 68 insertions(+), 31 deletions(-) diff --git a/hw/sh7750.c b/hw/sh7750.c