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] [day] [month] [year] [list]
Message-ID: <4ce7de89-fafb-3796-ae90-8132d2c819cf@huawei.com>
Date:   Fri, 23 Aug 2019 09:10:22 +0800
From:   Chao Yu <yuchao0@...wei.com>
To:     Jaegeuk Kim <jaegeuk@...nel.org>
CC:     <linux-kernel@...r.kernel.org>,
        <linux-f2fs-devel@...ts.sourceforge.net>
Subject: Re: [f2fs-dev] [PATCH v2] f2fs: allocate memory in batch in
 build_sit_info()

On 2019/8/23 1:47, Jaegeuk Kim wrote:
> On 08/20, Jaegeuk Kim wrote:
>> On 08/20, Chao Yu wrote:
>>> On 2019/8/20 4:20, Jaegeuk Kim wrote:
>>>> On 07/26, Chao Yu wrote:
>>>>> build_sit_info() allocate all bitmaps for each segment one by one,
>>>>> it's quite low efficiency, this pach changes to allocate large
>>>>> continuous memory at a time, and divide it and assign for each bitmaps
>>>>> of segment. For large size image, it can expect improving its mount
>>>>> speed.
>>>>
>>>> Hmm, I hit a kernel panic when mounting a partition during fault injection test:
>>>>
>>>> 726 #ifdef CONFIG_F2FS_CHECK_FS
>>>> 727         if (f2fs_test_bit(offset, sit_i->sit_bitmap) !=
>>>> 728                         f2fs_test_bit(offset, sit_i->sit_bitmap_mir))
>>>> 729                 f2fs_bug_on(sbi, 1);
>>>> 730 #endif
>>>
>>> We didn't change anything about sit_i->sit_bitmap{_mir,}, it's so wired we panic
>>> here... :(
>>>
>>> I double check the change, but find nothing suspicious, btw, my fault injection
>>> testcase shows normal.
>>>
>>> Jaegeuk, do you have any idea about this issue?
>>
>> I'm bisecting. :P
> 
> It was caused by wrong bitmap size in "f2fs: Fix indefinite loop in f2fs_gc()".
> Fixed by:

Good catch! This alternative one looks good to me.

Thanks,

> ---
>  fs/f2fs/segment.c | 18 ++++++++++--------
>  1 file changed, 10 insertions(+), 8 deletions(-)
> 
> diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
> index 9b20ce3b87cc..cc230fc829e1 100644
> --- a/fs/f2fs/segment.c
> +++ b/fs/f2fs/segment.c
> @@ -3950,7 +3950,7 @@ static int build_sit_info(struct f2fs_sb_info *sbi)
>  	struct sit_info *sit_i;
>  	unsigned int sit_segs, start;
>  	char *src_bitmap, *bitmap;
> -	unsigned int bitmap_size;
> +	unsigned int bitmap_size, main_bitmap_size, sit_bitmap_size;
>  
>  	/* allocate memory for SIT information */
>  	sit_i = f2fs_kzalloc(sbi, sizeof(struct sit_info), GFP_KERNEL);
> @@ -3966,8 +3966,8 @@ static int build_sit_info(struct f2fs_sb_info *sbi)
>  	if (!sit_i->sentries)
>  		return -ENOMEM;
>  
> -	bitmap_size = f2fs_bitmap_size(MAIN_SEGS(sbi));
> -	sit_i->dirty_sentries_bitmap = f2fs_kvzalloc(sbi, bitmap_size,
> +	main_bitmap_size = f2fs_bitmap_size(MAIN_SEGS(sbi));
> +	sit_i->dirty_sentries_bitmap = f2fs_kvzalloc(sbi, main_bitmap_size,
>  								GFP_KERNEL);
>  	if (!sit_i->dirty_sentries_bitmap)
>  		return -ENOMEM;
> @@ -4016,19 +4016,21 @@ static int build_sit_info(struct f2fs_sb_info *sbi)
>  	sit_segs = le32_to_cpu(raw_super->segment_count_sit) >> 1;
>  
>  	/* setup SIT bitmap from ckeckpoint pack */
> -	bitmap_size = __bitmap_size(sbi, SIT_BITMAP);
> +	sit_bitmap_size = __bitmap_size(sbi, SIT_BITMAP);
>  	src_bitmap = __bitmap_ptr(sbi, SIT_BITMAP);
>  
> -	sit_i->sit_bitmap = kmemdup(src_bitmap, bitmap_size, GFP_KERNEL);
> +	sit_i->sit_bitmap = kmemdup(src_bitmap, sit_bitmap_size, GFP_KERNEL);
>  	if (!sit_i->sit_bitmap)
>  		return -ENOMEM;
>  
>  #ifdef CONFIG_F2FS_CHECK_FS
> -	sit_i->sit_bitmap_mir = kmemdup(src_bitmap, bitmap_size, GFP_KERNEL);
> +	sit_i->sit_bitmap_mir = kmemdup(src_bitmap,
> +					sit_bitmap_size, GFP_KERNEL);
>  	if (!sit_i->sit_bitmap_mir)
>  		return -ENOMEM;
>  
> -	sit_i->invalid_segmap = f2fs_kvzalloc(sbi, bitmap_size, GFP_KERNEL);
> +	sit_i->invalid_segmap = f2fs_kvzalloc(sbi,
> +					main_bitmap_size, GFP_KERNEL);
>  	if (!sit_i->invalid_segmap)
>  		return -ENOMEM;
>  #endif
> @@ -4039,7 +4041,7 @@ static int build_sit_info(struct f2fs_sb_info *sbi)
>  	sit_i->sit_base_addr = le32_to_cpu(raw_super->sit_blkaddr);
>  	sit_i->sit_blocks = sit_segs << sbi->log_blocks_per_seg;
>  	sit_i->written_valid_blocks = 0;
> -	sit_i->bitmap_size = bitmap_size;
> +	sit_i->bitmap_size = sit_bitmap_size;
>  	sit_i->dirty_sentries = 0;
>  	sit_i->sents_per_block = SIT_ENTRY_PER_BLOCK;
>  	sit_i->elapsed_time = le64_to_cpu(sbi->ckpt->elapsed_time);
> 

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ