Received-SPF: pass (google.com: domain of linux-kernel+bounces-90678-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.48.161 as permitted sender) client-ip=147.75.48.161;
Feedback-ID: i56a14606:Fastmail
User-Agent: Cyrus-JMAP/3.11.0-alpha0-205-g4dbcac4545-fm-20240301.001-g4dbcac45
Precedence: bulk
MIME-Version: 1.0
Message-Id: <61aa333f-aa13-4068-b73f-eeae615f4640@app.fastmail.com>
In-Reply-To: 
 <CAMj1kXENYqUx=esK9b_pGd4wpwE43fNaGRCUhJLK_4MQzrAQeg@mail.gmail.com>
References: 
 <CA+G9fYvG9KE15PGNoLu+SBVyShe+u5HBLQ81+kK9Zop6u=ywmw@mail.gmail.com>
 <338c89bb-a70b-4f35-b71b-f974e90e3383@app.fastmail.com>
 <20240304112441.707ded23@donnerap.manchester.arm.com>
 <1baf9a7f-b0e4-45d8-ac57-0727a213d82d@app.fastmail.com>
 <20240304114546.4e8e1e32@donnerap.manchester.arm.com>
 <badf279a-f1fa-4938-a5d2-492b89d7c27c@app.fastmail.com>
 <CAMj1kXENYqUx=esK9b_pGd4wpwE43fNaGRCUhJLK_4MQzrAQeg@mail.gmail.com>
Date: Mon, 04 Mar 2024 14:48:56 +0100
From: "Arnd Bergmann" <arnd@arndb.de>
To: "Ard Biesheuvel" <ardb@kernel.org>
Cc: "Andre Przywara" <andre.przywara@arm.com>,
 "Naresh Kamboju" <naresh.kamboju@linaro.org>,
 "open list" <linux-kernel@vger.kernel.org>,
 "Linux ARM" <linux-arm-kernel@lists.infradead.org>,
 linux-sunxi@lists.linux.dev, dri-devel@lists.freedesktop.org,
 lkft-triage@lists.linaro.org, "Maxime Ripard" <mripard@kernel.org>,
 "Dave Airlie" <airlied@redhat.com>,
 "Dan Carpenter" <dan.carpenter@linaro.org>
Subject: Re: arm: ERROR: modpost: "__aeabi_uldivmod"
 [drivers/gpu/drm/sun4i/sun4i-drm-hdmi.ko] undefined!
Content-Type: text/plain

On Mon, Mar 4, 2024, at 14:01, Ard Biesheuvel wrote:
> On Mon, 4 Mar 2024 at 13:35, Arnd Bergmann <arnd@arndb.de> wrote:
>> On Mon, Mar 4, 2024, at 12:45, Andre Przywara wrote:
>> It's not critical if this is called infrequently, and as Maxime
>> just replied, the 64-bit division is in fact required here.
>> Since we are dividing by a constant value (200), there is a good
>> chance that this will be get turned into fairly efficient
>> multiply/shift code.
>>
>
> Clang does not implement that optimization for 64-bit division. That
> is how we ended up with this error in the first place.

I meant it will use the optimization after the patch to convert
the plain '/' to div_u64().

> Perhaps it is worthwhile to make div_u64() check its divisor, e.g.,
>
> --- a/include/linux/math64.h
> +++ b/include/linux/math64.h
> @@ -127,6 +127,9 @@
>  static inline u64 div_u64(u64 dividend, u32 divisor)
>  {
>         u32 remainder;
> +
> +       if (IS_ENABLED(CONFIG_CC_IS_GCC) && __builtin_constant_p(divisor))
> +               return dividend / divisor;
>         return div_u64_rem(dividend, divisor, &remainder);
>  }

I think the div_u64()->do_div()->__div64_const32()->__arch_xprod_64()
optimization in asm-generic/div64.h already produces what we want
on both compilers. Is there something missing there?

     Arnd