linux-kernel - Re: [PATCH v2 1/9] drm/panthor: Add 64-bit and poll register accessors

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <d9a4aa99-2f00-4da9-9f7e-26729f7f8fda@arm.com>
Date: Wed, 9 Apr 2025 14:00:54 +0100
From: Karunika Choo <karunika.choo@....com>
To: Boris Brezillon <boris.brezillon@...labora.com>
Cc: dri-devel@...ts.freedesktop.org, nd@....com,
 Steven Price <steven.price@....com>, Liviu Dudau <liviu.dudau@....com>,
 Maarten Lankhorst <maarten.lankhorst@...ux.intel.com>,
 Maxime Ripard <mripard@...nel.org>, Thomas Zimmermann <tzimmermann@...e.de>,
 David Airlie <airlied@...il.com>, Simona Vetter <simona@...ll.ch>,
 linux-kernel@...r.kernel.org
Subject: Re: [PATCH v2 1/9] drm/panthor: Add 64-bit and poll register
 accessors

On 21/03/2025 07:48, Boris Brezillon wrote:
> On Thu, 20 Mar 2025 11:17:33 +0000
> Karunika Choo <karunika.choo@....com> wrote:
>
>> This patch adds 64-bit register accessors to simplify register access in
>> Panthor. It also adds 32-bit and 64-bit variants for read_poll_timeout.
>>
>> Signed-off-by: Karunika Choo <karunika.choo@....com>
>> ---
>>  drivers/gpu/drm/panthor/panthor_regs.h | 55 ++++++++++++++++++++++++++
>>  1 file changed, 55 insertions(+)
>>
>> diff --git a/drivers/gpu/drm/panthor/panthor_regs.h
b/drivers/gpu/drm/panthor/panthor_regs.h
>> index 42dc3fedb0d4..7ec4a1d04e20 100644
>> --- a/drivers/gpu/drm/panthor/panthor_regs.h
>> +++ b/drivers/gpu/drm/panthor/panthor_regs.h
>> @@ -238,4 +238,59 @@
>>  #define gpu_read(dev, reg) \
>>  	readl((dev)->iomem + (reg))
>>
>> +#define gpu_read_relaxed(dev, reg) readl_relaxed((dev)->iomem + (reg))
>> +
>> +#define gpu_write64(dev, reg, data)                            \
>> +	do {                                                   \
>> +		u64 __val = (u64)(data);                       \
>> +		gpu_write(dev, reg, lower_32_bits(__val));     \
>> +		gpu_write(dev, reg + 4, upper_32_bits(__val)); \
>> +	} while (0)
>
> We're not doing funky name concatenation in these macros, so I'd rather
> have them defined as static inline funcs in panthor_device.h. We
> probably want to move the gpu_read/write definitions there as well if
> we do that.

I see where you're coming from, and it makes sense. I was thinking it
might be better to keep it in panthor_regs.h since that's the file we
include when accessing GPU registers. That said, we could certainly
convert them to static inline functions if you prefer.

>> +
>> +#define gpu_read64(dev, reg) \
>> +	(gpu_read(dev, reg) | ((u64)gpu_read(dev, reg + 4) << 32))
>> +
>> +#define gpu_read64_relaxed(dev, reg)  \
>> +	(gpu_read_relaxed(dev, reg) | \
>> +	 ((u64)gpu_read_relaxed(dev, reg + 4) << 32))
>> +
>> +#define gpu_read64_sync(dev, reg)                     \
>> +	({                                            \
>> +		u32 lo, hi1, hi2;                     \
>> +		do {                                  \
>> +			hi1 = gpu_read(dev, reg + 4); \
>> +			lo = gpu_read(dev, reg);      \
>> +			hi2 = gpu_read(dev, reg + 4); \
>> +		} while (hi1 != hi2);                 \
>> +		lo | ((u64)hi2 << 32);                \
>> +	})
>
> I would name that one gpu_read64_counter and make it a static inline
> function. Note that we already have panthor_gpu_read_64bit_counter()
> which does the same thing, so maybe move it there and rename it along
> the way.

Happy to rename this to gpu_read64_counter in v3, if you're okay with
us keeping the macros/functions in this file.

Kind regards,
Karunika

>> +
>> +#define gpu_read_poll_timeout(dev, reg, val, cond, delay_us,
timeout_us)    \
>> +	read_poll_timeout(gpu_read, val, cond, delay_us, timeout_us, false, \
>> +			  dev, reg)
>> +
>> +#define gpu_read_poll_timeout_atomic(dev, reg, val, cond, delay_us,
       \
>> +				     timeout_us)                            \
>> +	read_poll_timeout_atomic(gpu_read, val, cond, delay_us, timeout_us, \
>> +				 false, dev, reg)
>> +
>> +#define gpu_read64_poll_timeout(dev, reg, val, cond, delay_us,
timeout_us)    \
>> +	read_poll_timeout(gpu_read64, val, cond, delay_us, timeout_us, false, \
>> +			  dev, reg)
>> +
>> +#define gpu_read64_poll_timeout_atomic(dev, reg, val, cond,
delay_us,         \
>> +				       timeout_us)                            \
>> +	read_poll_timeout_atomic(gpu_read64, val, cond, delay_us, timeout_us, \
>> +				 false, dev, reg)
>> +
>> +#define gpu_read_relaxed_poll_timeout_atomic(dev, reg, val, cond,
delay_us, \
>> +					     timeout_us)                    \
>> +	read_poll_timeout_atomic(gpu_read_relaxed, val, cond, delay_us,     \
>> +				 timeout_us, false, dev, reg)
>> +
>> +#define gpu_read64_relaxed_poll_timeout(dev, reg, val, cond,
delay_us,         \
>> +					timeout_us)                            \
>> +	read_poll_timeout(gpu_read64_relaxed, val, cond, delay_us,
timeout_us, \
>> +			  false, dev, reg)
>> +
>>  #endif
>