linux-kernel - Re: [PATCH 5/6] riscv: bitops: Convert to use_alternative

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: <800585ca-29ff-40ef-bddd-8cc14806dbb1@iscas.ac.cn>
Date: Wed, 20 Aug 2025 23:36:54 +0800
From: Vivian Wang <wangruikang@...as.ac.cn>
To: Yury Norov <yury.norov@...il.com>
Cc: Paul Walmsley <paul.walmsley@...ive.com>,
 Palmer Dabbelt <palmer@...belt.com>, Albert Ou <aou@...s.berkeley.edu>,
 Alexandre Ghiti <alex@...ti.fr>, Rasmus Villemoes
 <linux@...musvillemoes.dk>, Vivian Wang <uwu@...m.page>,
 linux-riscv@...ts.infradead.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 5/6] riscv: bitops: Convert to use_alternative_likely

On 8/20/25 23:04, Yury Norov wrote:

> On Wed, Aug 20, 2025 at 09:44:49PM +0800, Vivian Wang wrote:
>> Use use_alternative_likely() to check for RISCV_ISA_EXT_ZBB, replacing
>> the use of asm goto with ALTERNATIVE.
>>
>> The "likely" variant is used to match the behavior of the original
>> implementation using ALTERNATIVE("j %l[legacy]", "nop", ...).
>>
>> Signed-off-by: Vivian Wang <wangruikang@...as.ac.cn>
>> ---
>>  arch/riscv/include/asm/bitops.h | 112 ++++++++++++++++++----------------------
>>  1 file changed, 50 insertions(+), 62 deletions(-)
>>
>> diff --git a/arch/riscv/include/asm/bitops.h b/arch/riscv/include/asm/bitops.h
>> index d59310f74c2ba70caeb7b9b0e9221882117583f5..0257d547a96293909d90b017c8a48b508d0fd642 100644
>> --- a/arch/riscv/include/asm/bitops.h
>> +++ b/arch/riscv/include/asm/bitops.h
>> @@ -47,20 +47,17 @@
>>  
>>  static __always_inline unsigned long variable__ffs(unsigned long word)
>>  {
>> -	asm goto(ALTERNATIVE("j %l[legacy]", "nop", 0,
>> -				      RISCV_ISA_EXT_ZBB, 1)
>> -			  : : : : legacy);
>> -
>> -	asm volatile (".option push\n"
>> -		      ".option arch,+zbb\n"
>> -		      "ctz %0, %1\n"
>> -		      ".option pop\n"
>> -		      : "=r" (word) : "r" (word) :);
>> -
>> -	return word;
>> -
>> -legacy:
>> -	return generic___ffs(word);
>> +	if (use_alternative_likely(0, RISCV_ISA_EXT_ZBB)) {
> I don't think that 'likely' is used properly here. The likely/unlikely
> wording has a meaning of a hint to the compiler:
>
>         if (unlikely(WARN_ON(cond))
>                 goto err;
>
> In your case, it's just meaningless, because whatever is 'likely' for
> one CPU, will be always 'unlikely' for another. 

As mentioned in the reply for patch 1, I do believe that unfortunately
we currently have no way of leaving the decision up to the compiler, so
we have to pick one. The situation is similar to
static_branch_{likely,unlikely}.

So I have preserved the original choice made here: The original asm goto
uses the "likely" pattern, so I kept it as "likely".

>> +		asm volatile (".option push\n"
>> +			      ".option arch,+zbb\n"
>> +			      "ctz %0, %1\n"
>> +			      ".option pop\n"
>> +			      : "=r" (word) : "r" (word) :);
>> +
>> +		return word;
>> +	} else {
>> +		return generic___ffs(word);
>> +	}
>>  }
> This tabs wipe most of the history. Can you reorganize your patch
> such that it preserves as much history as you can?
>
>         if (use_alternative_unlikely(...))
>                 return generic___ffs();
>
>         asm volatile (".option push\n"
>                       ".option arch,+zbb\n"
>                       "ctz %0, %1\n"
>                       ".option pop\n"
>                       : "=r" (word) : "r" (word) :);
>
>         return word;
>
> And so on.

Thanks for the tip, I'll give it a go at minimizing the diff for v2.

Vivian "dramforever" Wang