Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
MIME-Version: 1.0
References: <20230713125706.2884502-1-glider@google.com> <20230713125706.2884502-4-glider@google.com>
In-Reply-To: <20230713125706.2884502-4-glider@google.com>
From:   Alexander Potapenko <glider@google.com>
Date:   Thu, 13 Jul 2023 18:37:43 +0200
Message-ID: <CAG_fn=VBUiVf5aN44SQX04ZdXOOtGu3DjbW3-ZNorzNZZ6M0CA@mail.gmail.com>
Subject: Re: [v2 3/5] arm64: mte: implement CONFIG_ARM64_MTE_COMP
To:     glider@google.com, catalin.marinas@arm.com, will@kernel.org,
        pcc@google.com, andreyknvl@gmail.com,
        andriy.shevchenko@linux.intel.com, linux@rasmusvillemoes.dk,
        yury.norov@gmail.com
Cc:     linux-kernel@vger.kernel.org, linux-arm-kernel@lists.infradead.org,
        eugenis@google.com
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

> +/* Compress the data inline. */
> +static u64 ea0_compress_inline(int len, u8 *tags, short *sizes)
> +{
> +       u64 result;
> +
> +       ea0_compress_to_buf(len, tags, sizes, (u8 *)&result, sizeof(result));
> +       result = be64_to_cpu(result);
This be64_to_cpu() is leftover from the previous compressor
implementation which treated u64 as u8[8] (i.e. big-endian).
The new implementation works with the native u64 representation, so
the conversion is not needed anymore (and actually produces invalid
handles that can't be stored in Xarray).
I will drop this call in v3.


> +bool ea0_decompress(u64 handle, u8 *tags)
> +{
> +       u8 *storage = ea0_storage(handle);
> +       int size = ea0_storage_size(handle);
> +
> +       if (size == 128) {
> +               memcpy(tags, storage, size);
> +               return true;
> +       }
> +       if (size == 8) {
> +               handle = cpu_to_be64(handle);
Ditto.