x86: perf_event_intel_pt.c: use arch_initcall to hook in enabling
authorPaul Gortmaker <paul.gortmaker@windriver.com>
Sat, 2 May 2015 01:57:34 +0000 (21:57 -0400)
committerPaul Gortmaker <paul.gortmaker@windriver.com>
Tue, 16 Jun 2015 18:12:35 +0000 (14:12 -0400)
commit5b00c1eb94e5936e5bf5cdd9ad1ddfbed0c39159
treeab895fd5e580ac7d6e15838550221cc0bb901a9e
parentca41d24cf56458a699b44e918c5a19b7077df422
x86: perf_event_intel_pt.c: use arch_initcall to hook in enabling

This was using module_init, but the current Kconfig situation is
as follows:

In arch/x86/kernel/cpu/Makefile:

  obj-$(CONFIG_CPU_SUP_INTEL)    += perf_event_intel_pt.o perf_event_intel_bts.o

and in arch/x86/Kconfig.cpu:

  config CPU_SUP_INTEL
        default y
        bool "Support Intel processors" if PROCESSOR_SELECT

So currently, the end user can not build this code into a module.
If in the future, there is desire for this to be modular, then
it can be changed to include <linux/module.h> and use module_init.

But currently, in the non-modular case, a module_init becomes a
device_initcall.  But this really isn't a device, so we should
choose a more appropriate initcall bucket to put it in.

The obvious choice here seems to be arch_initcall, but that does
make it earlier than it was currently through device_initcall.
As long as perf_pmu_register() is functional, we should be OK.

Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: x86@kernel.org
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
arch/x86/kernel/cpu/perf_event_intel_pt.c
This page took 0.025679 seconds and 5 git commands to generate.