The overhead was significant enough, the fast case this time was 132 seconds.
The following slow case had 209 seconds (still slower but not more by factors).
While the overhead might have influenced the absolute timing of this too much.
It still might help indicate the size/frequency distribution of these calls that we wanted to know.
(systemtap) "vfio_iommu_ type1") .function( "vfio_iommu_ type1_ioctl" ) { type1_ioctl\ n"); stopwatch( "vfioioctl" ); "vfio_iommu_ type1") .function( "vfio_iommu_ type1_ioctl" ).return { read_stopwatch_ ns("vfioioctl" ) "Completed vfio_iommu_ type1_ioctl: %d\n", timer); stopwatch( "vfioioctl" ); "vfio_iommu_ type1") .function( "vfio_pin_ pages_remote" ) { read_stopwatch_ ns("vfioioctl" )
probe module(
printf("New vfio_iommu_
start_
}
probe module(
timer=
printf(
stop_
}
probe module(
timer=
printf("%ld: %s\n", timer, $$parms);
}
The overhead was significant enough, the fast case this time was 132 seconds.
The following slow case had 209 seconds (still slower but not more by factors).
While the overhead might have influenced the absolute timing of this too much.
It still might help indicate the size/frequency distribution of these calls that we wanted to know.