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;
Message-ID: <b552e551-6961-b95a-1af3-0c9114447528@linux.alibaba.com>
Date:   Sun, 30 Jul 2023 22:06:14 +0800
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.15; rv:102.0)
 Gecko/20100101 Thunderbird/102.13.0
Subject: Re: [PATCH v4] erofs: DEFLATE compression support
From:   Gao Xiang <hsiangkao@linux.alibaba.com>
To:     Chao Yu <chao@kernel.org>, linux-erofs@lists.ozlabs.org
Cc:     LKML <linux-kernel@vger.kernel.org>
References: <20230713001441.30462-1-hsiangkao@linux.alibaba.com>
 <20230716091940.40899-1-hsiangkao@linux.alibaba.com>
 <9570176a-02c2-c500-14c6-b8ca220cfb96@kernel.org>
 <07c85385-be34-8fdc-876c-dc97639c5643@linux.alibaba.com>
 <f6210979-4322-feb9-9f22-76f4cb1a33da@kernel.org>
 <a8b82109-fde6-3b6f-a911-981722223197@linux.alibaba.com>
In-Reply-To: <a8b82109-fde6-3b6f-a911-981722223197@linux.alibaba.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit
Precedence: bulk


On 2023/7/30 22:05, Gao Xiang wrote:
> 
> 
> On 2023/7/30 22:01, Chao Yu wrote:
>> On 2023/7/30 21:49, Gao Xiang wrote:
>>>
>>>
>>> On 2023/7/30 20:53, Chao Yu wrote:
>>>> On 2023/7/16 17:19, Gao Xiang wrote:
>>>>> Add DEFLATE compression as the 3rd supported algorithm.
>>>>>
>>>>> DEFLATE is a popular generic-purpose compression algorithm for quite
>>>>> long time (many advanced formats like gzip, zlib, zip, png are all
>>>>> based on that) as Apple documentation written "If you require
>>>>> interoperability with non-Apple devices, use COMPRESSION_ZLIB. [1]".
>>>>>
>>>>> Due to its popularity, there are several hardware on-market DEFLATE
>>>>> accelerators, such as (s390) DFLTCC, (Intel) IAA/QAT, (HiSilicon) ZIP
>>>>> accelerator, etc.  In addition, there are also several high-performence
>>>>> IP cores and even open-source FPGA approches available for DEFLATE.
>>>>> Therefore, it's useful to support DEFLATE compression in order to find
>>>>> a way to utilize these accelerators for asynchronous I/Os and get
>>>>> benefits from these later.
>>>>>
>>>>> Besides, it's a good choice to trade off between compression ratios
>>>>> and performance compared to LZ4 and LZMA.  The DEFLATE core format is
>>>>> simple as well as easy to understand, therefore the code size of its
>>>>> decompressor is small even for the bootloader use cases.  The runtime
>>>>> memory consumption is quite limited too (e.g. 32K + ~7K for each zlib
>>>>> stream).  As usual, EROFS ourperforms similar approaches too.
>>>>>
>>>>> Alternatively, DEFLATE could still be used for some specific files
>>>>> since EROFS supports multiple compression algorithms in one image.
>>>>>
>>>>> [1] https://developer.apple.com/documentation/compression/compression_algorithm
>>>>> ---
>>>>> changes since v3:
>>>>>    - fix 'insz' mis-calculation, which leads to failure on 4k pclusters.
>>>>>
>>>>>    fs/erofs/Kconfig                |  15 ++
>>>>>    fs/erofs/Makefile               |   1 +
>>>>>    fs/erofs/compress.h             |   2 +
>>>>>    fs/erofs/decompressor.c         |   6 +
>>>>>    fs/erofs/decompressor_deflate.c | 250 ++++++++++++++++++++++++++++++++
>>>>>    fs/erofs/erofs_fs.h             |   7 +
>>>>>    fs/erofs/internal.h             |  20 +++
>>>>>    fs/erofs/super.c                |  10 ++
>>>>>    fs/erofs/zmap.c                 |   5 +-
>>>>>    9 files changed, 314 insertions(+), 2 deletions(-)
>>>>>    create mode 100644 fs/erofs/decompressor_deflate.c
>>>>>
>>>>> diff --git a/fs/erofs/Kconfig b/fs/erofs/Kconfig
>>>>> index f259d92c9720..56a99ba8ce22 100644
>>>>> --- a/fs/erofs/Kconfig
>>>>> +++ b/fs/erofs/Kconfig
>>>>> @@ -99,6 +99,21 @@ config EROFS_FS_ZIP_LZMA
>>>>>          If unsure, say N.
>>>>> +config EROFS_FS_ZIP_DEFLATE
>>>>> +    bool "EROFS DEFLATE compressed data support"
>>>>> +    depends on EROFS_FS_ZIP
>>>>> +    select ZLIB_INFLATE
>>>>> +    help
>>>>> +      Saying Y here includes support for reading EROFS file systems
>>>>> +      containing DEFLATE compressed data.  It gives better compression
>>>>> +      ratios than the default LZ4 format, while it costs more CPU
>>>>> +      overhead.
>>>>> +
>>>>> +      DEFLATE support is an experimental feature for now and so most
>>>>> +      file systems will be readable without selecting this option.
>>>>> +
>>>>> +      If unsure, say N.
>>>>> +
>>>>>    config EROFS_FS_ONDEMAND
>>>>>        bool "EROFS fscache-based on-demand read support"
>>>>>        depends on CACHEFILES_ONDEMAND && (EROFS_FS=m && FSCACHE || EROFS_FS=y && FSCACHE=y)
>>>>> diff --git a/fs/erofs/Makefile b/fs/erofs/Makefile
>>>>> index a3a98fc3e481..994d0b9deddf 100644
>>>>> --- a/fs/erofs/Makefile
>>>>> +++ b/fs/erofs/Makefile
>>>>> @@ -5,4 +5,5 @@ erofs-objs := super.o inode.o data.o namei.o dir.o utils.o sysfs.o
>>>>>    erofs-$(CONFIG_EROFS_FS_XATTR) += xattr.o
>>>>>    erofs-$(CONFIG_EROFS_FS_ZIP) += decompressor.o zmap.o zdata.o pcpubuf.o
>>>>>    erofs-$(CONFIG_EROFS_FS_ZIP_LZMA) += decompressor_lzma.o
>>>>> +erofs-$(CONFIG_EROFS_FS_ZIP_DEFLATE) += decompressor_deflate.o
>>>>>    erofs-$(CONFIG_EROFS_FS_ONDEMAND) += fscache.o
>>>>> diff --git a/fs/erofs/compress.h b/fs/erofs/compress.h
>>>>> index b1b846504027..349c3316ae6b 100644
>>>>> --- a/fs/erofs/compress.h
>>>>> +++ b/fs/erofs/compress.h
>>>>> @@ -94,4 +94,6 @@ extern const struct z_erofs_decompressor erofs_decompressors[];
>>>>>    /* prototypes for specific algorithms */
>>>>>    int z_erofs_lzma_decompress(struct z_erofs_decompress_req *rq,
>>>>>                    struct page **pagepool);
>>>>> +int z_erofs_deflate_decompress(struct z_erofs_decompress_req *rq,
>>>>> +                   struct page **pagepool);
>>>>>    #endif
>>>>> diff --git a/fs/erofs/decompressor.c b/fs/erofs/decompressor.c
>>>>> index cfad1eac7fd9..332ec5f74002 100644
>>>>> --- a/fs/erofs/decompressor.c
>>>>> +++ b/fs/erofs/decompressor.c
>>>>> @@ -379,4 +379,10 @@ const struct z_erofs_decompressor erofs_decompressors[] = {
>>>>>            .name = "lzma"
>>>>>        },
>>>>>    #endif
>>>>> +#ifdef CONFIG_EROFS_FS_ZIP_DEFLATE
>>>>> +    [Z_EROFS_COMPRESSION_DEFLATE] = {
>>>>> +        .decompress = z_erofs_deflate_decompress,
>>>>> +        .name = "deflate"
>>>>> +    },
>>>>> +#endif
>>>>>    };
>>>>> diff --git a/fs/erofs/decompressor_deflate.c b/fs/erofs/decompressor_deflate.c
>>>>> new file mode 100644
>>>>> index 000000000000..c34e29b15465
>>>>> --- /dev/null
>>>>> +++ b/fs/erofs/decompressor_deflate.c
>>>>> @@ -0,0 +1,250 @@
>>>>> +// SPDX-License-Identifier: GPL-2.0-or-later
>>>>> +#include <linux/module.h>
>>>>> +#include <linux/zlib.h>
>>>>> +#include "compress.h"
>>>>> +
>>>>> +struct z_erofs_deflate {
>>>>> +    struct z_erofs_deflate *next;
>>>>> +    struct z_stream_s z;
>>>>> +    u8 bounce[PAGE_SIZE];
>>>>> +};
>>>>> +
>>>>> +static DEFINE_SPINLOCK(z_erofs_deflate_lock);
>>>>> +static unsigned int z_erofs_deflate_nstrms, z_erofs_deflate_avail_strms;
>>>>> +static struct z_erofs_deflate *z_erofs_deflate_head;
>>>>> +static DECLARE_WAIT_QUEUE_HEAD(z_erofs_deflate_wq);
>>>>> +
>>>>> +module_param_named(deflate_streams, z_erofs_deflate_nstrms, uint, 0444);
>>>>> +
>>>>> +void z_erofs_deflate_exit(void)
>>>>> +{
>>>>> +    /* there should be no running fs instance */
>>>>> +    while (z_erofs_deflate_avail_strms) {
>>>>> +        struct z_erofs_deflate *strm;
>>>>> +
>>>>> +        spin_lock(&z_erofs_deflate_lock);
>>>>> +        strm = z_erofs_deflate_head;
>>>>> +        if (!strm) {
>>>>> +            spin_unlock(&z_erofs_deflate_lock);
>>>>> +            DBG_BUGON(1);
>>>>> +            return;
>>>>> +        }
>>>>> +        z_erofs_deflate_head = NULL;
>>>>> +        spin_unlock(&z_erofs_deflate_lock);
>>>>> +
>>>>> +        while (strm) {
>>>>> +            struct z_erofs_deflate *n = strm->next;
>>>>> +
>>>>> +            vfree(strm->z.workspace);
>>>>> +            kfree(strm);
>>>>> +            --z_erofs_deflate_avail_strms;
>>>>> +            strm = n;
>>>>> +        }
>>>>> +    }
>>>>> +}
>>>>> +
>>>>> +int __init z_erofs_deflate_init(void)
>>>>> +{
>>>>> +    /* by default, use # of possible CPUs instead */
>>>>> +    if (!z_erofs_deflate_nstrms)
>>>>> +        z_erofs_deflate_nstrms = num_possible_cpus();
>>>>> +
>>>>> +    for (; z_erofs_deflate_avail_strms < z_erofs_deflate_nstrms;
>>>>> +         ++z_erofs_deflate_avail_strms) {
>>>>> +        struct z_erofs_deflate *strm;
>>>>> +
>>>>> +        strm = kzalloc(sizeof(*strm), GFP_KERNEL);
>>>>> +        if (!strm)
>>>>> +            goto out_failed;
>>>>> +
>>>>> +        /* XXX: in-kernel zlib cannot shrink windowbits currently */
>>>>> +        strm->z.workspace = vmalloc(zlib_inflate_workspacesize());
>>>>> +        if (!strm->z.workspace)
>>>>
>>>> kfree(strm)?
>>>
>>> z_erofs_deflate_exit() below will handle this.
>>
>> strm is not in z_erofs_deflate_head list yet?
>>
>> Thanks
>>
>>>
>>> Thanks,
>>> Gao Xiang
>>>
>>>>
>>>> Thanks,
>>>>
>>>>> +            goto out_failed;
>>>>> +
>>>>> +        spin_lock(&z_erofs_deflate_lock);
>>>>> +        strm->next = z_erofs_deflate_head;
>>>>> +        z_erofs_deflate_head = strm;
> 
> z_erofs_deflate_head will be added in each turn here.

Or you are right, thanks for pointing out, I will
make a version soon.

Thanks,
Gao Xiang