CVE-2021-47117

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> ext4: fix bug on in ext4_es_cache_extent as ext4_split_extent_at failed<br /> <br /> We got follow bug_on when run fsstress with injecting IO fault:<br /> [130747.323114] kernel BUG at fs/ext4/extents_status.c:762!<br /> [130747.323117] Internal error: Oops - BUG: 0 [#1] SMP<br /> ......<br /> [130747.334329] Call trace:<br /> [130747.334553] ext4_es_cache_extent+0x150/0x168 [ext4]<br /> [130747.334975] ext4_cache_extents+0x64/0xe8 [ext4]<br /> [130747.335368] ext4_find_extent+0x300/0x330 [ext4]<br /> [130747.335759] ext4_ext_map_blocks+0x74/0x1178 [ext4]<br /> [130747.336179] ext4_map_blocks+0x2f4/0x5f0 [ext4]<br /> [130747.336567] ext4_mpage_readpages+0x4a8/0x7a8 [ext4]<br /> [130747.336995] ext4_readpage+0x54/0x100 [ext4]<br /> [130747.337359] generic_file_buffered_read+0x410/0xae8<br /> [130747.337767] generic_file_read_iter+0x114/0x190<br /> [130747.338152] ext4_file_read_iter+0x5c/0x140 [ext4]<br /> [130747.338556] __vfs_read+0x11c/0x188<br /> [130747.338851] vfs_read+0x94/0x150<br /> [130747.339110] ksys_read+0x74/0xf0<br /> <br /> This patch&amp;#39;s modification is according to Jan Kara&amp;#39;s suggestion in:<br /> https://patchwork.ozlabs.org/project/linux-ext4/patch/20210428085158.3728201-1-yebin10@huawei.com/<br /> "I see. Now I understand your patch. Honestly, seeing how fragile is trying<br /> to fix extent tree after split has failed in the middle, I would probably<br /> go even further and make sure we fix the tree properly in case of ENOSPC<br /> and EDQUOT (those are easily user triggerable). Anything else indicates a<br /> HW problem or fs corruption so I&amp;#39;d rather leave the extent tree as is and<br /> don&amp;#39;t try to fix it (which also means we will not create overlapping<br /> extents)."