[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <YH62o3A6DXQPiylT@kernel.org>
Date: Tue, 20 Apr 2021 08:10:27 -0300
From: Arnaldo Carvalho de Melo <acme@...nel.org>
To: Rob Herring <robh@...nel.org>
Cc: Peter Zijlstra <peterz@...radead.org>,
Ingo Molnar <mingo@...hat.com>, Jiri Olsa <jolsa@...hat.com>,
linux-kernel@...r.kernel.org, Will Deacon <will@...nel.org>,
Catalin Marinas <catalin.marinas@....com>,
Mark Rutland <mark.rutland@....com>,
Namhyung Kim <namhyung@...nel.org>,
Itaru Kitayama <itaru.kitayama@...il.com>
Subject: Re: [PATCH v8 4/4] libperf: Add support for user space counter access
Em Wed, Apr 14, 2021 at 11:07:39AM -0500, Rob Herring escreveu:
> x86 and arm64 can both support direct access of event counters in
> userspace. The access sequence is less than trivial and currently exists
> in perf test code (tools/perf/arch/x86/tests/rdpmc.c) with copies in
> projects such as PAPI and libpfm4.
>
> In order to support usersapce access, an event must be mmapped first
> with perf_evsel__mmap(). Then subsequent calls to perf_evsel__read()
> will use the fast path (assuming the arch supports it).
Had to apply this to fix the build on the other arches:
> +#if defined(__i386__) || defined(__x86_64__)
> +static u64 read_perf_counter(unsigned int counter)
> +{
> + unsigned int low, high;
> +
> + asm volatile("rdpmc" : "=a" (low), "=d" (high) : "c" (counter));
> +
> + return low | ((u64)high) << 32;
> +}
> +
> +static u64 read_timestamp(void)
> +{
> + unsigned int low, high;
> +
> + asm volatile("rdtsc" : "=a" (low), "=d" (high));
> +
> + return low | ((u64)high) << 32;
> +}
> +#else
> +static u64 read_perf_counter(unsigned int counter) { return 0; }
> +static u64 read_timestamp(void) { return 0; }
> +#endif
diff --git a/tools/lib/perf/mmap.c b/tools/lib/perf/mmap.c
index 915469f00cf4c3fb..c89dfa5f67b3a408 100644
--- a/tools/lib/perf/mmap.c
+++ b/tools/lib/perf/mmap.c
@@ -295,7 +295,7 @@ static u64 read_timestamp(void)
return low | ((u64)high) << 32;
}
#else
-static u64 read_perf_counter(unsigned int counter) { return 0; }
+static u64 read_perf_counter(unsigned int counter __maybe_unused) { return 0; }
static u64 read_timestamp(void) { return 0; }
#endif
Powered by blists - more mailing lists