Messages in this thread | | | Subject | Re: [patch 02/11] x86 architecture implementation of Hardware Breakpoint interfaces | From | Benjamin Herrenschmidt <> | Date | Sat, 14 Mar 2009 14:40:33 +1100 |
| |
On Tue, 2009-03-10 at 18:26 +0100, Ingo Molnar wrote: > > That 'arbitrarily larg number of breakpoints' worries me. It's a > pretty broken concept for a 4-items resource that cannot be > time-shared and hence cannot be overcommitted. > > Seems to me that much of the complexity of this patchset: > > 28 files changed, 2439 insertions(+), 199 deletions(-) > > Could be eliminated via a very simple exclusive reservation > mechanism. > I also have some worries about the bloat of this infrastructure, especially in the context switching code.
I would prefer the arch to be in control of the state in the task struct and just context switch the actual HW registers at that stage.
Ben.
| |