From: "André Almeida" <andrealmeid@collabora.com>
To: Alistair Francis <alistair23@gmail.com>
Cc: Arnaldo Carvalho de Melo <arnaldo.melo@gmail.com>,
Alistair Francis <alistair.francis@opensource.wdc.com>,
linux-riscv <linux-riscv@lists.infradead.org>,
linux-perf-users@vger.kernel.org,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
Namhyung Kim <namhyung@kernel.org>, Jiri Olsa <jolsa@redhat.com>,
Alexander Shishkin <alexander.shishkin@linux.intel.com>,
Mark Rutland <mark.rutland@arm.com>,
Davidlohr Bueso <dave@stgolabs.net>,
Darren Hart <dvhart@infradead.org>,
Peter Zijlstra <peterz@infradead.org>,
Ingo Molnar <mingo@redhat.com>,
Thomas Gleixner <tglx@linutronix.de>,
Atish Patra <atish.patra@wdc.com>, Arnd Bergmann <arnd@arndb.de>,
Alistair Francis <alistair.francis@wdc.com>
Subject: Re: [PATCH v2 1/2] perf bench futex: Use a 64-bit time_t
Date: Tue, 19 Oct 2021 21:01:41 -0300 [thread overview]
Message-ID: <0d55b0a0-dd8c-0494-a200-7acb1976e32a@collabora.com> (raw)
In-Reply-To: <CAKmqyKP6AU-9tcYOp-ERvYFP7t6DMp4kbQgDmJxD6wb0WTUy+w@mail.gmail.com>
Às 20:17 de 19/10/21, Alistair Francis escreveu:
> On Wed, Oct 20, 2021 at 2:56 AM Arnaldo Carvalho de Melo
> <arnaldo.melo@gmail.com> wrote:
>>
>> Em Fri, Oct 15, 2021 at 10:56:33AM +1000, Alistair Francis escreveu:
>>> From: Alistair Francis <alistair.francis@wdc.com>
>>>
>>> Convert tools/perf to only use a 64-bit time_t. On 64-bit architectures
>>> this isn't a functional change. On 32-bit architectures we now only
>>> perform 64-bit time_t syscalls (__NR_futex_time64) and use a struct
>>> timespec64.
>>>
>>> This won't work on kernels before 5.1, but as perf is tied to the kernel
>>> that's ok.
>>
>> No, perf is not tied to the kernel, one can use a new perf tool on any
>> previous kernel, and an old perf tool should work on new kernels as
>> well.
>
> + André, I won't be doing this the way you requested
>
Ok, thanks anyway for your work and sorry for the trouble :)
> Ok, so back to the previous version then. I'll send the patches soon.
>
> Alistair
>
>>
>> - Arnaldo
>>
>>> This allows us to build perf for 32-bit architectures with 64-bit time_t
>>> like RISC-V 32-bit.
>>>
>>> Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
>>> ---
>>> tools/perf/bench/futex.h | 26 ++++++++++++++++++++------
>>> 1 file changed, 20 insertions(+), 6 deletions(-)
>>>
>>> diff --git a/tools/perf/bench/futex.h b/tools/perf/bench/futex.h
>>> index b3853aac3021c..b9665d43d2988 100644
>>> --- a/tools/perf/bench/futex.h
>>> +++ b/tools/perf/bench/futex.h
>>> @@ -12,6 +12,7 @@
>>> #include <sys/syscall.h>
>>> #include <sys/types.h>
>>> #include <linux/futex.h>
>>> +#include <linux/time_types.h>
>>>
>>> struct bench_futex_parameters {
>>> bool silent;
>>> @@ -27,12 +28,14 @@ struct bench_futex_parameters {
>>> unsigned int nrequeue;
>>> };
>>>
>>> +#define timespec64 __kernel_timespec
>>> +
>>> /**
>>> * futex() - SYS_futex syscall wrapper
>>> * @uaddr: address of first futex
>>> * @op: futex op code
>>> * @val: typically expected value of uaddr, but varies by op
>>> - * @timeout: typically an absolute struct timespec (except where noted
>>> + * @timeout: typically an absolute struct timespec64 (except where noted
>>> * otherwise). Overloaded by some ops
>>> * @uaddr2: address of second futex for some ops
>>> * @val3: varies by op
>>> @@ -47,15 +50,26 @@ struct bench_futex_parameters {
>>> * These argument descriptions are the defaults for all
>>> * like-named arguments in the following wrappers except where noted below.
>>> */
>>> -#define futex(uaddr, op, val, timeout, uaddr2, val3, opflags) \
>>> - syscall(SYS_futex, uaddr, op | opflags, val, timeout, uaddr2, val3)
>>> +/**
>>> + * We only support 64-bit time_t for the timeout.
>>> + * On 64-bit architectures we can use __NR_futex
>>> + * On 32-bit architectures we use __NR_futex_time64. This only works on kernel
>>> + * versions 5.1+.
>>> + */
>>> +#if __BITS_PER_LONG == 64
>>> +# define futex(uaddr, op, val, timeout, uaddr2, val3, opflags) \
>>> + syscall(__NR_futex, uaddr, op | opflags, val, timeout, uaddr2, val3)
>>> +#else
>>> +# define futex(uaddr, op, val, timeout, uaddr2, val3, opflags) \
>>> + syscall(__NR_futex_time64, uaddr, op | opflags, val, timeout, uaddr2, val3)
>>> +#endif
>>>
>>> /**
>>> * futex_wait() - block on uaddr with optional timeout
>>> * @timeout: relative timeout
>>> */
>>> static inline int
>>> -futex_wait(u_int32_t *uaddr, u_int32_t val, struct timespec *timeout, int opflags)
>>> +futex_wait(u_int32_t *uaddr, u_int32_t val, struct timespec64 *timeout, int opflags)
>>> {
>>> return futex(uaddr, FUTEX_WAIT, val, timeout, NULL, 0, opflags);
>>> }
>>> @@ -74,7 +88,7 @@ futex_wake(u_int32_t *uaddr, int nr_wake, int opflags)
>>> * futex_lock_pi() - block on uaddr as a PI mutex
>>> */
>>> static inline int
>>> -futex_lock_pi(u_int32_t *uaddr, struct timespec *timeout, int opflags)
>>> +futex_lock_pi(u_int32_t *uaddr, struct timespec64 *timeout, int opflags)
>>> {
>>> return futex(uaddr, FUTEX_LOCK_PI, 0, timeout, NULL, 0, opflags);
>>> }
>>> @@ -111,7 +125,7 @@ futex_cmp_requeue(u_int32_t *uaddr, u_int32_t val, u_int32_t *uaddr2, int nr_wak
>>> */
>>> static inline int
>>> futex_wait_requeue_pi(u_int32_t *uaddr, u_int32_t val, u_int32_t *uaddr2,
>>> - struct timespec *timeout, int opflags)
>>> + struct timespec64 *timeout, int opflags)
>>> {
>>> return futex(uaddr, FUTEX_WAIT_REQUEUE_PI, val, timeout, uaddr2, 0,
>>> opflags);
>>> --
>>> 2.31.1
>>
>> --
>>
>> - Arnaldo
next prev parent reply other threads:[~2021-10-20 0:01 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-10-15 0:56 Alistair Francis
2021-10-15 0:56 ` [PATCH v2 2/2] selftests: " Alistair Francis
2021-10-15 8:05 ` Arnd Bergmann
2021-10-15 8:04 ` [PATCH v2 1/2] perf bench " Arnd Bergmann
2021-10-19 16:56 ` Arnaldo Carvalho de Melo
2021-10-19 23:17 ` Alistair Francis
2021-10-20 0:01 ` André Almeida [this message]
2021-10-20 1:56 ` Alistair Francis
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=0d55b0a0-dd8c-0494-a200-7acb1976e32a@collabora.com \
--to=andrealmeid@collabora.com \
--cc=alexander.shishkin@linux.intel.com \
--cc=alistair.francis@opensource.wdc.com \
--cc=alistair.francis@wdc.com \
--cc=alistair23@gmail.com \
--cc=arnaldo.melo@gmail.com \
--cc=arnd@arndb.de \
--cc=atish.patra@wdc.com \
--cc=dave@stgolabs.net \
--cc=dvhart@infradead.org \
--cc=jolsa@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-perf-users@vger.kernel.org \
--cc=linux-riscv@lists.infradead.org \
--cc=mark.rutland@arm.com \
--cc=mingo@redhat.com \
--cc=namhyung@kernel.org \
--cc=peterz@infradead.org \
--cc=tglx@linutronix.de \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®