Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1189664
| From | Peter Zijlstra <peterz@infradead.org> |
|---|---|
| Newsgroups | linux.kernel |
| Subject | Re: [PATCH v3 7/8] perf: Define PMU_TXN_READ interface |
| Date | 2015-07-22 08:00 +0200 |
| Message-ID | <pOVep-6ml-1@gated-at.bofh.it> (permalink) |
| References | <pMlf3-3in-3@gated-at.bofh.it> <pMlf3-3in-13@gated-at.bofh.it> <pMZPb-3d9-1@gated-at.bofh.it> <pORu9-Tx-9@gated-at.bofh.it> |
| Organization | linux.* mail to news gateway |
On Tue, Jul 21, 2015 at 06:50:45PM -0700, Sukadev Bhattiprolu wrote: > We are trying to use the following interface: > > start_txn(pmu, PERF_PMU_TXN_READ); > > perf_event_read(leader); > list_for_each(sibling, &leader->sibling_list, group_entry) > perf_event_read(sibling) > > pmu->commit_txn(pmu); > > with the idea that the PMU driver would save the type of transaction in > ->start_txn() and use in ->read() and ->commit_txn(). > > But since ->start_txn() and the ->read() operations could happen on different > CPUs (perf_event_read() uses the event->oncpu to schedule a call), the PMU > driver cannot use a per-cpu variable to save the state in ->start_txn(). > or is there better way? I've not woken up yet, and not actually fully read the email, but can you stuff the entire above chunk inside the IPI? I think you could then actually optimize __perf_event_read() as well, because all these events should be on the same context, so no point in calling update_*time*() for every event or so. -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread
[PATCH v3 0/8] Implement group-read of events using txn interface Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com> - 2015-07-15 05:10 +0200
[PATCH v3 4/8] perf: Split perf_event_read() and perf_event_count() Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com> - 2015-07-15 05:10 +0200
[PATCH v3 7/8] perf: Define PMU_TXN_READ interface Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com> - 2015-07-15 05:10 +0200
Re: [PATCH v3 7/8] perf: Define PMU_TXN_READ interface Peter Zijlstra <peterz@infradead.org> - 2015-07-17 00:30 +0200
Re: [PATCH v3 7/8] perf: Define PMU_TXN_READ interface Peter Zijlstra <peterz@infradead.org> - 2015-07-22 08:00 +0200
Re: [PATCH v3 7/8] perf: Define PMU_TXN_READ interface Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com> - 2015-07-23 01:30 +0200
Re: [PATCH v3 7/8] perf: Define PMU_TXN_READ interface Peter Zijlstra <peterz@infradead.org> - 2015-07-23 10:10 +0200
Re: [PATCH v3 7/8] perf: Define PMU_TXN_READ interface Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com> - 2015-07-24 03:20 +0200
[PATCH v3 3/8] perf: Add a flags parameter to pmu txn interfaces Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com> - 2015-07-15 05:10 +0200
Re: [PATCH v3 3/8] perf: Add a flags parameter to pmu txn interfaces Peter Zijlstra <peterz@infradead.org> - 2015-07-16 22:20 +0200
Re: [PATCH v3 3/8] perf: Add a flags parameter to pmu txn interfaces Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com> - 2015-07-16 23:30 +0200
Re: [PATCH v3 3/8] perf: Add a flags parameter to pmu txn interfaces Peter Zijlstra <peterz@infradead.org> - 2015-07-16 22:50 +0200
[PATCH v3 5/8] perf: Split perf_event_read_value() Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com> - 2015-07-15 05:10 +0200
Re: [PATCH v3 5/8] perf: Split perf_event_read_value() Peter Zijlstra <peterz@infradead.org> - 2015-07-16 23:20 +0200
Re: [PATCH v3 5/8] perf: Split perf_event_read_value() Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com> - 2015-07-16 23:50 +0200
Re: [PATCH v3 5/8] perf: Split perf_event_read_value() Peter Zijlstra <peterz@infradead.org> - 2015-07-23 09:50 +0200
[PATCH v3 1/8] powerpc/perf/hv-24x7: Whitespace - fix parameter alignment Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com> - 2015-07-15 05:10 +0200
Re: [v3,1/8] powerpc/perf/hv-24x7: Whitespace - fix parameter alignment Michael Ellerman <mpe@ellerman.id.au> - 2015-08-03 03:40 +0200
[PATCH v3 2/8] powerpc/perf/hv-24x7: Simplify extracting counter from result buffer Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com> - 2015-07-15 05:10 +0200
Re: [v3, 2/8] powerpc/perf/hv-24x7: Simplify extracting counter from result buffer Michael Ellerman <mpe@ellerman.id.au> - 2015-08-03 03:40 +0200
csiph-web