Messages in this thread Patch in this message | | | Subject | Re: [RFC][PATCH 09/11] perf_counter: revamp syscall input ABI | From | Peter Zijlstra <> | Date | Thu, 19 Mar 2009 12:41:07 +0100 |
| |
On Thu, 2009-03-19 at 09:15 +1100, Paul Mackerras wrote: > Peter Zijlstra writes: > > > Ah, I thought we should keep a pure 64 bit raw value. You never know > > what hardware will do. > > Oh I see, you use hw_event->raw_event if hw_event->raw is set. I > missed that before. > > Still, you're putting that into hwc->config along with other bits like > ARCH_PERFMON_EVENTSEL_USR and ARCH_PERFMON_EVENTSEL_OS, so I would > think we could spare two bits for the type, leaving 62 bits for the > raw event code. And if that isn't enough, there's the > hw_event.extra_config_len field, which allows userspace to pass in > arbitrary amounts of extra PMU configuration data.
Good point, overflow interrupt, and usr/os/hv event filter and enable bits are usually in the config word.
OK, how about the below? I didn't cut it to 2 bits, as that would already exhaust the TYPE space -- then again, 60 does feel cramped a bit..
Index: linux-2.6/include/linux/perf_counter.h =================================================================== --- linux-2.6.orig/include/linux/perf_counter.h +++ linux-2.6/include/linux/perf_counter.h @@ -27,6 +27,12 @@ enum perf_event_types { PERF_TYPE_HARDWARE = 0, PERF_TYPE_SOFTWARE = 1, PERF_TYPE_TRACEPOINT = 2, + + /* + * available TYPE space, raw is the max value. + */ + + PERF_TYPE_RAW = 15, }; /* @@ -79,14 +85,8 @@ enum perf_counter_record_type { * Hardware event to monitor via a performance monitoring counter: */ struct perf_counter_hw_event { - union { - __u64 raw_event; - struct { - __u64 event_id : 32, - type : 8, - __reserved_0 : 24; - }; - }; + __u64 event_id : 60, + type : 4; __u64 irq_period; __u64 record_type; @@ -94,7 +94,6 @@ struct perf_counter_hw_event { __u64 disabled : 1, /* off by default */ nmi : 1, /* NMI sampling */ - raw : 1, /* raw event type */ inherit : 1, /* children inherit it */ pinned : 1, /* must always be on PMU */ exclusive : 1, /* only group on PMU */ @@ -103,7 +102,7 @@ struct perf_counter_hw_event { exclude_hv : 1, /* ditto hypervisor */ exclude_idle : 1, /* don't count when idle */ - __reserved_1 : 54; + __reserved_1 : 55; __u32 extra_config_len; __u32 __reserved_4;
| |