在 2012-12-04二的 12:49 +0000，Peter Maydell写道： > On 4 December 2012 08:11, liguang <lig.f...@cn.fujitsu.com> wrote: > > Signed-off-by: liguang <lig.f...@cn.fujitsu.com> > > --- > > target-i386/cpu.h | 7 +++++++ > > 1 files changed, 7 insertions(+), 0 deletions(-) > > > > diff --git a/target-i386/cpu.h b/target-i386/cpu.h > > index 90ef1ff..9abec3e 100644 > > --- a/target-i386/cpu.h > > +++ b/target-i386/cpu.h > > @@ -231,6 +231,13 @@ > > #define DR7_TYPE_SHIFT 16 > > #define DR7_LEN_SHIFT 18 > > #define DR7_FIXED_1 0x00000400 > > +#define DR7_LOCAL_BP_MASK 0x55 > > +#define DR7_MAX_BP 4 > > +#define DR7_BP_INST 0x0 > > +#define DR7_DATA_WR 0x1 > > +#define DR7_IO_RW 0x2 > > +#define DR7_DATA_RW 0x3 > > I still think these last four should be DR7_TYPE_BP_INST etc to > indicate that they're values for the TYPE field, not direct > specifications of bits in DR7.
hmm, is it necessary? you know, the use of these names is after calling 'hw_breakpoint_type' function, so, it's obvious for dr7's type field. > > -- PMM -- regards! li guang
