Re: [PATCH] perf/x86/intel: Make cpuc allocations consistent

From: Peter Zijlstra
Date: Fri Mar 15 2019 - 04:14:23 EST


On Thu, Mar 14, 2019 at 02:10:14PM -0700, Guenter Roeck wrote:
> On Tue, Mar 05, 2019 at 10:23:15PM +0100, Peter Zijlstra (Intel) wrote:
> > The cpuc data structure allocation is different between fake and real
> > cpuc's; use the same code to init/free both.
> >
> > Signed-off-by: Peter Zijlstra (Intel) <peterz@xxxxxxxxxxxxx>
> > Signed-off-by: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
>
> With allmodconfig-CONFIG_CPU_SUP_INTEL, this patch results in:
>
> In file included from arch/x86/events/amd/core.c:8:0:
> arch/x86/events/amd/../perf_event.h:1036:45: warning: âstruct cpu_hw_eventâ declared inside parameter list will not be visible outside of this definition or declaration
> static inline int intel_cpuc_prepare(struct cpu_hw_event *cpuc, int cpu)
> ^~~~~~~~~~~~
> arch/x86/events/amd/../perf_event.h:1041:45: warning: âstruct cpu_hw_eventâ declared inside parameter list will not be visible outside of this definition or declaration
> static inline void intel_cpuc_finish(struct cpu_hw_event *cpuc)
> ^~~~~~~~~~~~
>
> With -Werror, this is fatal, and I think it may be buggy (should it be
> cpu_hw_events) ?

Yep it should be. Won't result in a bug though, an unused pointer is an
unused poiner, no matter what known or unknown type it points to.

> The patch has been applied to stable releases. Any idea, anyone, why
> this is the case ? It doesn't look like a bug fix to me, and reverting
> it from v4.14.106 didn't seem to have a negative impact.

Reverting just this one patch will make your kernel go boom when you run
on a skylake or more recent Intel cpu, have updated your microcode and
do:

echo 0 > /sys/bus/event_source/devices/cpu/allow_tsx_force_abort

And then try and create perf events. See also commit:

400816f60c54 ("perf/x86/intel: Implement support for TSX Force Abort")

---
Subject: perf/x86: Fixup typo in stub functions

Guenter reported a build warning for CONFIG_CPU_SUP_INTEL=n. While
harmless (an unsed pointer is an unused pointer, no matter the type) it
needs fixing.

Reported-by: Guenter Roeck <linux@xxxxxxxxxxxx>
Fixes: d01b1f96a82e ("perf/x86/intel: Make cpuc allocations consistent")
Signed-off-by: Peter Zijlstra (Intel) <peterz@xxxxxxxxxxxxx>
---
arch/x86/events/perf_event.h | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/x86/events/perf_event.h b/arch/x86/events/perf_event.h
index b04ae6c8775e..a75955741c50 100644
--- a/arch/x86/events/perf_event.h
+++ b/arch/x86/events/perf_event.h
@@ -1033,12 +1033,12 @@ static inline int intel_pmu_init(void)
return 0;
}

-static inline int intel_cpuc_prepare(struct cpu_hw_event *cpuc, int cpu)
+static inline int intel_cpuc_prepare(struct cpu_hw_events *cpuc, int cpu)
{
return 0;
}

-static inline void intel_cpuc_finish(struct cpu_hw_event *cpuc)
+static inline void intel_cpuc_finish(struct cpu_hw_events *cpuc)
{
}