On Wed, Feb 17, 2016 at 01:29:14PM -0800, Hollis Blanchard wrote:
> Knowing which CPU performed an action is essential for understanding SMP guest
> behavior.
>
> However, cpu_physical_memory_rw() may be executed by a machine init function,
> before any VCPUs are running, when there is no CPU running ('current_cpu' is
> NULL). In this case, store -1 in the trace record as the CPU index. Trace
> analysis tools may need to be aware of this special case.
>
> Signed-off-by: Hollis Blanchard <hollis_blanch...@mentor.com>
> ---
> memory.c | 48 ++++++++++++++++++++++++++++++++++++------------
> trace-events | 8 ++++----
> 2 files changed, 40 insertions(+), 16 deletions(-)
>
> diff --git a/memory.c b/memory.c
> index 2d87c21..6ae7bae 100644
> --- a/memory.c
> +++ b/memory.c
> @@ -395,13 +395,17 @@ static MemTxResult
> memory_region_oldmmio_read_accessor(MemoryRegion *mr,
> MemTxAttrs attrs)
> {
> uint64_t tmp;
> + int cpu_index = -1;
> +
> + if (current_cpu)
> + cpu_index = current_cpu->cpu_index;QEMU coding style always uses curly braces, even when the if statement body only has one line. Cases like these should be caught by scripts/checkpatch.pl. I use a git hook to run it automatically on commit: http://blog.vmsplice.net/2011/03/how-to-automatically-run-checkpatchpl.html A helper function would avoid the code duplication throughout this patch: static int get_cpu_index(void) { if (current_cpu) { return current_cpu->cpu_index; } return -1; }
signature.asc
Description: PGP signature
