Generic Linux architectural discussions
 help / color / mirror / Atom feed
From: Vincenzo Frascino <vincenzo.frascino@arm.com>
To: Arnd Bergmann <arnd@arndb.de>
Cc: linux-arch <linux-arch@vger.kernel.org>,
	Catalin Marinas <catalin.marinas@arm.com>,
	Daniel Lezcano <daniel.lezcano@linaro.org>,
	Will Deacon <will.deacon@arm.com>,
	Russell King - ARM Linux <linux@armlinux.org.uk>,
	Ralf Baechle <ralf@linux-mips.org>,
	salyzyn@android.com, Paul Burton <paul.burton@mips.com>,
	Thomas Gleixner <tglx@linutronix.de>,
	pcc@google.com, Linux ARM <linux-arm-kernel@lists.infradead.org>
Subject: Re: [PATCH 07/27] arm64: Substitute gettimeofday with C implementation
Date: Tue, 13 Nov 2018 10:58:31 +0000	[thread overview]
Message-ID: <8e306faf-6261-8363-3f18-70e754fa3a9d@arm.com> (raw)
In-Reply-To: <CAK8P3a1w=GBcYk2OaVNzfagMMeD_zdBK3G16BSYjFANKdSAezQ@mail.gmail.com>

On 09/11/2018 16:13, Arnd Bergmann wrote:
>> +
>> +extern struct vdso_data _vdso_data;
>> +
>> +static __always_inline notrace int gettimeofday_fallback(struct timeval *_tv,
>> +                                                        struct timezone *_tz)
> 
> I'm trying to get rid of the last users of 'struct timeval' in the kernel so
> we can remove the definition (it will clash with future glibc implementations
> in the uapi headers). Could you change this to use __kernel_old_timeval
> instead?
>

Ok, will update all the data types in v2.

... 

>> diff --git a/arch/arm64/kernel/vdso/vdso.lds.S b/arch/arm64/kernel/vdso/vdso.lds.S
>> index beca249bc2f3..9de0ffc369c5 100644
>> --- a/arch/arm64/kernel/vdso/vdso.lds.S
>> +++ b/arch/arm64/kernel/vdso/vdso.lds.S
>> @@ -88,6 +88,7 @@ VERSION
>>                 __kernel_gettimeofday;
>>                 __kernel_clock_gettime;
>>                 __kernel_clock_getres;
>> +               __kernel_time;
>>         local: *;
>>         };
>>  }
> 
> I would  prefer to not add any deprecated interfaces in the VDSO. If we
> have the 64-bit version of clock_gettime, we don't need the 32-bit version
> of it, and we don't need gettimeofday() or time() either. The C library
> can easily implement those by calling into clock_gettime.
> 

I like the idea, this would make the vdso lib code more simple and more maintainable.

In this patchset I tried to cover the widest possible scenario making things configurable: each architecture can select and enable exactly what it needs from the vdso common code. 

Based on what you are proposing, once the C library will implement things in this way, it will be easy to deprecate and remove the unused code.

I am not familiar with the development plans of the various C libraries, but looking at bionic libc currently seems using all the vdso exposed functions [1].

[1] https://github.com/aosp-mirror/platform_bionic/blob/master/libc/bionic/vdso.cpp

...

> 
>         Arnd
> 

-- 
Regards,
Vincenzo

WARNING: multiple messages have this Message-ID (diff)
From: Vincenzo Frascino <vincenzo.frascino@arm.com>
To: Arnd Bergmann <arnd@arndb.de>
Cc: linux-arch <linux-arch@vger.kernel.org>,
	Linux ARM <linux-arm-kernel@lists.infradead.org>,
	Catalin Marinas <catalin.marinas@arm.com>,
	Will Deacon <will.deacon@arm.com>,
	Russell King - ARM Linux <linux@armlinux.org.uk>,
	Ralf Baechle <ralf@linux-mips.org>,
	Paul Burton <paul.burton@mips.com>,
	Daniel Lezcano <daniel.lezcano@linaro.org>,
	Thomas Gleixner <tglx@linutronix.de>,
	salyzyn@android.com, pcc@google.com
Subject: Re: [PATCH 07/27] arm64: Substitute gettimeofday with C implementation
Date: Tue, 13 Nov 2018 10:58:31 +0000	[thread overview]
Message-ID: <8e306faf-6261-8363-3f18-70e754fa3a9d@arm.com> (raw)
Message-ID: <20181113105831.tgZQtDF9NkTDjqhHvcmis0quRhJgRB0qlJqPsLkLTVw@z> (raw)
In-Reply-To: <CAK8P3a1w=GBcYk2OaVNzfagMMeD_zdBK3G16BSYjFANKdSAezQ@mail.gmail.com>

On 09/11/2018 16:13, Arnd Bergmann wrote:
>> +
>> +extern struct vdso_data _vdso_data;
>> +
>> +static __always_inline notrace int gettimeofday_fallback(struct timeval *_tv,
>> +                                                        struct timezone *_tz)
> 
> I'm trying to get rid of the last users of 'struct timeval' in the kernel so
> we can remove the definition (it will clash with future glibc implementations
> in the uapi headers). Could you change this to use __kernel_old_timeval
> instead?
>

Ok, will update all the data types in v2.

... 

>> diff --git a/arch/arm64/kernel/vdso/vdso.lds.S b/arch/arm64/kernel/vdso/vdso.lds.S
>> index beca249bc2f3..9de0ffc369c5 100644
>> --- a/arch/arm64/kernel/vdso/vdso.lds.S
>> +++ b/arch/arm64/kernel/vdso/vdso.lds.S
>> @@ -88,6 +88,7 @@ VERSION
>>                 __kernel_gettimeofday;
>>                 __kernel_clock_gettime;
>>                 __kernel_clock_getres;
>> +               __kernel_time;
>>         local: *;
>>         };
>>  }
> 
> I would  prefer to not add any deprecated interfaces in the VDSO. If we
> have the 64-bit version of clock_gettime, we don't need the 32-bit version
> of it, and we don't need gettimeofday() or time() either. The C library
> can easily implement those by calling into clock_gettime.
> 

I like the idea, this would make the vdso lib code more simple and more maintainable.

In this patchset I tried to cover the widest possible scenario making things configurable: each architecture can select and enable exactly what it needs from the vdso common code. 

Based on what you are proposing, once the C library will implement things in this way, it will be easy to deprecate and remove the unused code.

I am not familiar with the development plans of the various C libraries, but looking at bionic libc currently seems using all the vdso exposed functions [1].

[1] https://github.com/aosp-mirror/platform_bionic/blob/master/libc/bionic/vdso.cpp

...

> 
>         Arnd
> 

-- 
Regards,
Vincenzo

  parent reply	other threads:[~2018-11-13 10:58 UTC|newest]

Thread overview: 96+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-11-09 12:37 [PATCH 00/27] Unify vDSOs across more architectures Vincenzo Frascino
2018-11-09 12:37 ` Vincenzo Frascino
2018-11-09 12:37 ` [PATCH 01/27] kernel: Standardize vdso_datapage Vincenzo Frascino
2018-11-09 12:37   ` Vincenzo Frascino
2018-11-09 12:37 ` [PATCH 02/27] kernel: Add Monotonic boot time support Vincenzo Frascino
2018-11-09 12:37   ` Vincenzo Frascino
2018-11-09 12:37 ` [PATCH 03/27] kernel: Add International Atomic Time support Vincenzo Frascino
2018-11-09 12:37   ` Vincenzo Frascino
2018-11-09 12:37 ` [PATCH 04/27] kernel: Add masks support for Raw and NTP time Vincenzo Frascino
2018-11-09 12:37   ` Vincenzo Frascino
2018-11-09 12:37 ` [PATCH 05/27] kernel: Add clock_mode support Vincenzo Frascino
2018-11-09 12:37   ` Vincenzo Frascino
2018-11-09 12:37 ` [PATCH 06/27] kernel: Define gettimeofday vdso common code Vincenzo Frascino
2018-11-09 12:37   ` Vincenzo Frascino
2018-11-09 16:05   ` Arnd Bergmann
2018-11-09 16:05     ` Arnd Bergmann
2018-11-13 10:21     ` Vincenzo Frascino
2018-11-13 10:21       ` Vincenzo Frascino
2018-11-10 16:18   ` Arnd Bergmann
2018-11-10 16:18     ` Arnd Bergmann
2018-11-13 10:24     ` Vincenzo Frascino
2018-11-13 10:24       ` Vincenzo Frascino
2018-11-09 12:37 ` [PATCH 07/27] arm64: Substitute gettimeofday with C implementation Vincenzo Frascino
2018-11-09 12:37   ` Vincenzo Frascino
2018-11-09 16:13   ` Arnd Bergmann
2018-11-09 16:13     ` Arnd Bergmann
2018-11-09 17:23     ` Mark Salyzyn
2018-11-09 17:23       ` Mark Salyzyn
2018-11-09 21:13       ` Arnd Bergmann
2018-11-09 21:13         ` Arnd Bergmann
2018-11-13 10:58     ` Vincenzo Frascino [this message]
2018-11-13 10:58       ` Vincenzo Frascino
2018-11-14 17:47       ` Arnd Bergmann
2018-11-14 17:47         ` Arnd Bergmann
2018-11-23 10:49         ` Vincenzo Frascino
2018-11-23 10:49           ` Vincenzo Frascino
2018-11-23 11:14           ` Arnd Bergmann
2018-11-23 11:14             ` Arnd Bergmann
2018-11-09 18:09   ` Peter Collingbourne
2018-11-09 18:09     ` Peter Collingbourne
2018-11-13 10:36     ` Vincenzo Frascino
2018-11-13 10:36       ` Vincenzo Frascino
2018-11-10  0:29   ` David Miller
2018-11-10  0:29     ` David Miller
2018-11-12 18:21     ` Mark Salyzyn
2018-11-12 18:21       ` Mark Salyzyn
2018-11-14 10:22     ` Vincenzo Frascino
2018-11-14 10:22       ` Vincenzo Frascino
2018-11-09 12:37 ` [PATCH 08/27] arm64: compat: Alloc separate pages for vectors and sigpage Vincenzo Frascino
2018-11-09 12:37   ` Vincenzo Frascino
2018-11-09 12:37 ` [PATCH 09/27] arm64: compat: Split kuser32 Vincenzo Frascino
2018-11-09 12:37   ` Vincenzo Frascino
2018-11-09 12:37 ` [PATCH 10/27] arm64: compat: Refactor aarch32_alloc_vdso_pages() Vincenzo Frascino
2018-11-09 12:37   ` Vincenzo Frascino
2018-11-09 12:37 ` [PATCH 11/27] arm64: compat: Add KUSER_HELPERS config option Vincenzo Frascino
2018-11-09 12:37   ` Vincenzo Frascino
2018-11-09 12:37 ` [PATCH 12/27] arm64: compat: Add missing syscall numbers Vincenzo Frascino
2018-11-09 12:37   ` Vincenzo Frascino
2018-11-09 12:37 ` [PATCH 13/27] arm64: compat: Expose signal related structures Vincenzo Frascino
2018-11-09 12:37   ` Vincenzo Frascino
2018-11-09 12:37 ` [PATCH 14/27] arm64: compat: Generate asm offsets for signals Vincenzo Frascino
2018-11-09 12:37   ` Vincenzo Frascino
2018-11-09 12:37 ` [PATCH 15/27] lib: vdso: Add compat support Vincenzo Frascino
2018-11-09 12:37   ` Vincenzo Frascino
2018-11-09 12:37 ` [PATCH 16/27] arm64: compat: Add vDSO Vincenzo Frascino
2018-11-09 12:37   ` Vincenzo Frascino
2018-11-09 16:23   ` Arnd Bergmann
2018-11-09 16:23     ` Arnd Bergmann
2018-11-09 12:37 ` [PATCH 17/27] arm64: Refactor vDSO code Vincenzo Frascino
2018-11-09 12:37   ` Vincenzo Frascino
2018-11-09 12:37 ` [PATCH 18/27] arm64: compat: vDSO setup for compat layer Vincenzo Frascino
2018-11-09 12:37   ` Vincenzo Frascino
2018-11-09 12:37 ` [PATCH 19/27] arm64: elf: vDSO code page discovery Vincenzo Frascino
2018-11-09 12:37   ` Vincenzo Frascino
2018-11-09 12:37 ` [PATCH 20/27] arm64: compat: Get sigreturn trampolines from vDSO Vincenzo Frascino
2018-11-09 12:37   ` Vincenzo Frascino
2018-11-09 12:37 ` [PATCH 21/27] arm64: Add vDSO compat support Vincenzo Frascino
2018-11-09 12:37   ` Vincenzo Frascino
2018-11-09 12:37 ` [PATCH 22/27] arm64: Enable compat vDSO support Vincenzo Frascino
2018-11-09 12:37   ` Vincenzo Frascino
2018-11-09 12:37 ` [PATCH 23/27] arm: Add support for generic vDSO Vincenzo Frascino
2018-11-09 12:37   ` Vincenzo Frascino
2018-11-09 16:27   ` Arnd Bergmann
2018-11-09 16:27     ` Arnd Bergmann
2018-11-09 12:37 ` [PATCH 24/27] mips: Introduce vdso_direct Vincenzo Frascino
2018-11-09 12:37   ` Vincenzo Frascino
2018-11-09 12:37 ` [PATCH 25/27] clock: csrc-4k: Add support for vdso_direct Vincenzo Frascino
2018-11-09 12:37   ` Vincenzo Frascino
2018-11-09 12:37 ` [PATCH 26/27] clock: gic-timer: " Vincenzo Frascino
2018-11-09 12:37   ` Vincenzo Frascino
2018-11-09 12:37 ` [PATCH 27/27] mips: Add support for generic vDSO Vincenzo Frascino
2018-11-09 12:37   ` Vincenzo Frascino
2018-11-10 16:33 ` [PATCH 00/27] Unify vDSOs across more architectures Arnd Bergmann
2018-11-10 16:33   ` Arnd Bergmann
2018-11-23 11:05   ` Vincenzo Frascino
2018-11-23 11:05     ` Vincenzo Frascino

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=8e306faf-6261-8363-3f18-70e754fa3a9d@arm.com \
    --to=vincenzo.frascino@arm.com \
    --cc=arnd@arndb.de \
    --cc=catalin.marinas@arm.com \
    --cc=daniel.lezcano@linaro.org \
    --cc=linux-arch@vger.kernel.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux@armlinux.org.uk \
    --cc=paul.burton@mips.com \
    --cc=pcc@google.com \
    --cc=ralf@linux-mips.org \
    --cc=salyzyn@android.com \
    --cc=tglx@linutronix.de \
    --cc=will.deacon@arm.com \
    /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