Received-SPF: pass (google.com: best guess record for domain of linux-crypto-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Message-ID: <10021f43dc5b377340fde9a7716e083f6f1261c1.camel@perches.com>
Subject: Re: [PATCH 1/2] zstd: pass pointer rathen than structure to
 functions
From:   Joe Perches <joe@perches.com>
To:     Maninder Singh <maninder1.s@samsung.com>, terrelln@fb.com,
        herbert@gondor.apana.org.au, davem@davemloft.net,
        keescook@chromium.org, gustavo@embeddedor.com
Cc:     linux-crypto@vger.kernel.org, linux-kernel@vger.kernel.org,
        a.sahrawat@samsung.com, pankaj.m@samsung.com,
        Vaneet Narang <v.narang@samsung.com>
Date:   Fri, 10 May 2019 10:39:57 -0700
In-Reply-To: <1557468704-3014-1-git-send-email-maninder1.s@samsung.com>
References: <CGME20190510061311epcas5p19e9bf3d08319ac99890e03e0bd59e478@epcas5p1.samsung.com>
         <1557468704-3014-1-git-send-email-maninder1.s@samsung.com>
Content-Type: text/plain; charset="ISO-8859-1"
User-Agent: Evolution 3.30.1-1build1 
Mime-Version: 1.0
Content-Transfer-Encoding: 7bit
Sender: linux-crypto-owner@vger.kernel.org
Precedence: bulk

On Fri, 2019-05-10 at 11:41 +0530, Maninder Singh wrote:
> currently params structure is passed in all functions, which increases
> stack usage in all the function and lead to stack overflow on target like
> ARM with kernel stack size of 8 KB so better to pass pointer.
[]
> diff --git a/lib/zstd/compress.c b/lib/zstd/compress.c
[]
> @@ -206,18 +206,18 @@ ZSTD_compressionParameters ZSTD_adjustCParams(ZSTD_compressionParameters cPar, u
>  	return cPar;
>  }
>  
> -static U32 ZSTD_equivalentParams(ZSTD_parameters param1, ZSTD_parameters param2)
> +static U32 ZSTD_equivalentParams(const ZSTD_parameters *param1, const ZSTD_parameters *param2)
>  {
> -	return (param1.cParams.hashLog == param2.cParams.hashLog) & (param1.cParams.chainLog == param2.cParams.chainLog) &
> -	       (param1.cParams.strategy == param2.cParams.strategy) & ((param1.cParams.searchLength == 3) == (param2.cParams.searchLength == 3));
> +	return (param1->cParams.hashLog == param2->cParams.hashLog) & (param1->cParams.chainLog == param2->cParams.chainLog) &
> +	       (param1->cParams.strategy == param2->cParams.strategy) & ((param1->cParams.searchLength == 3) == (param2->cParams.searchLength == 3));
>  }

trivia:

Using & instead of && makes this somewhat difficult to read.
It's hard to believe this is a performance optimization.

It might be better as

	return param1->cParams.hashLog == param2->cParams.hashLog &&
	       param1->cParams.chainLog == param2->cParams.chainLog &&
	       param1->cParams.strategy == param2->cParams.strategy &&
	       param1->cParams.searchLength == 3 &&
	       param1->cParams.searchLength == param2->cParams.searchLength;