Re: [Qemu-devel] [PATCH 2/5] trace: [all] Add "guest_vmem" event

2016-03-21 Thread Lluís Vilanova
Peter Maydell writes: > On 20 March 2016 at 18:09, Lluís Vilanova wrote: >> Oh! Yes, I seem to remember that code path now, I checked it a >> really long time ago. I was assuming that whenever this event is >> enabled at compile time, I would have to modify QEMU's TLB to

Re: [Qemu-devel] [PATCH 2/5] trace: [all] Add "guest_vmem" event

2016-03-20 Thread Peter Maydell
On 20 March 2016 at 18:09, Lluís Vilanova wrote: > Oh! Yes, I seem to remember that code path now, I checked it a > really long time ago. I was assuming that whenever this event is > enabled at compile time, I would have to modify QEMU's TLB to store > the guest physical

Re: [Qemu-devel] [PATCH 2/5] trace: [all] Add "guest_vmem" event

2016-03-20 Thread Lluís Vilanova
Peter Maydell writes: > On 18 March 2016 at 18:50, Lluís Vilanova wrote: >> Peter Maydell writes: >>> Trying to trace physaddrs is very tricky -- in the case of >>> a TLB hit there is no guarantee you can still identify the >>> physaddr of what you're accessing (the guest

Re: [Qemu-devel] [PATCH 2/5] trace: [all] Add "guest_vmem" event

2016-03-19 Thread Peter Maydell
On 23 February 2016 at 18:22, Lluís Vilanova wrote: > Signed-off-by: Lluís Vilanova > +### Guest events, keep at bottom > + > +# @vaddr: Access' virtual address. > +# @size : Access' size (bytes). > +# @store: Whether the access is a store. > +# > +#

Re: [Qemu-devel] [PATCH 2/5] trace: [all] Add "guest_vmem" event

2016-03-19 Thread Peter Maydell
On 18 March 2016 at 18:50, Lluís Vilanova wrote: > Peter Maydell writes: >> Trying to trace physaddrs is very tricky -- in the case of >> a TLB hit there is no guarantee you can still identify the >> physaddr of what you're accessing (the guest might have >> changed the page

Re: [Qemu-devel] [PATCH 2/5] trace: [all] Add "guest_vmem" event

2016-03-19 Thread Peter Maydell
On 17 March 2016 at 19:22, Lluís Vilanova wrote: > M, the endianness seems more of a vCPU property than one of the memory > access. A separate event could be added for that (e.g., at vCPU > initalization/hot-plug and whenever it is dynamically changed like in ARM). We've

Re: [Qemu-devel] [PATCH 2/5] trace: [all] Add "guest_vmem" event

2016-03-19 Thread Lluís Vilanova
Peter Maydell writes: > On 17 March 2016 at 19:22, Lluís Vilanova wrote: >> M, the endianness seems more of a vCPU property than one of the memory >> access. A separate event could be added for that (e.g., at vCPU >> initalization/hot-plug and whenever it is dynamically

Re: [Qemu-devel] [PATCH 2/5] trace: [all] Add "guest_vmem" event

2016-03-19 Thread Richard Henderson
On 03/17/2016 12:22 PM, Lluís Vilanova wrote: > M, the endianness seems more of a vCPU property than one of the memory > access. On the contrary. Plenty of cpus have endian-swapping load/store insns: x86 (haswell's movbe), powerpc, sparcv9, s390. Maybe others I've forgotten. r~

Re: [Qemu-devel] [PATCH 2/5] trace: [all] Add "guest_vmem" event

2016-03-18 Thread Lluís Vilanova
Peter Maydell writes: > On 23 February 2016 at 18:22, Lluís Vilanova wrote: >> Signed-off-by: Lluís Vilanova >> +### Guest events, keep at bottom >> + >> +# @vaddr: Access' virtual address. >> +# @size : Access' size (bytes). >> +# @store: Whether the

[Qemu-devel] [PATCH 2/5] trace: [all] Add "guest_vmem" event

2016-02-23 Thread Lluís Vilanova
Signed-off-by: Lluís Vilanova --- include/exec/cpu_ldst_template.h | 17 +++ include/exec/cpu_ldst_useronly_template.h | 14 tcg/tcg-op.c | 34 ++--- trace-events