In the Linux kernel, the following vulnerability has been resolved:<\/p>\n
ntfs3: validate split-point offset in indx_insert_into_buffer<\/p>\n
indx_insert_into_buffer() computes<\/p>\n
used = used1 – to_copy – sp_size;
\n memmove(de_t, Add2Ptr(sp, sp_size), used – le32_to_cpu(hdr1->de_off));<\/p>\n
where sp and sp_size come from hdr_find_split(). hdr_find_split()
\nwalks entries by le16_to_cpu(e->size) without validating that each
\nstep stays within hdr->used or that the size field is at least
\nsizeof(struct NTFS_DE). index_hdr_check(), the on-load gatekeeper,
\nonly validates header-level fields (used, total, de_off) and does
\nnot walk per-entry sizes.<\/p>\n
A crafted NTFS image whose leaf INDEX_HDR reports used == total but
\ncontains one interior NTFS_DE with size = 0xFFF0 therefore passes
\nvalidation, descends to indx_insert_into_buffer() through the
\nntfs_create() -> indx_insert_entry() path, and makes hdr_find_split()
\nreturn an sp whose sp_size (0xFFF0) greatly exceeds the remaining
\nbytes in the buffer. The u32 subtraction underflows and the memmove
\ncount becomes a near-4-GiB value, producing an out-of-bounds kernel
\nwrite that corrupts adjacent allocations and panics the kernel.<\/p>\n
Reproduced on 7.0.0-rc7 with UML + KASAN via a crafted image and a
\nsingle ‘touch’ inside the mounted directory; crash site resolves to
\nfs\/ntfs3\/index.c at the memmove. Trigger requires only local mount
\nof an attacker-supplied filesystem image (USB, loopback, or removable
\nmedia auto-mount).<\/p>\n
Reject the split whenever the chosen sp plus its declared size
\nalready extends past hdr1->used. This is the minimal fix; it
\npreserves the existing hdr_find_split() contract and relies on the
\nsame out: cleanup path as the pre-existing error returns.<\/p>\n
A prior OOB read in the very same indx_insert_into_buffer() memmove
\nwas fixed in commit b8c44949044e (“fs\/ntfs3: Fix OOB read in
\nindx_insert_into_buffer”) by tightening hdr_find_e(), but that fix
\ndoes not cover the split-point size field path addressed here: sp is
\nreturned by hdr_find_split(), not hdr_find_e(), and the underflow is
\ndriven by sp->size rather than hdr->used exceeding hdr->total.<\/p>\n","protected":false},"excerpt":{"rendered":"
In the Linux kernel, the following vulnerability has been resolved:<\/p>\n
ntfs3: validate split-point offset in indx_insert_into_buffer<\/p>\n
indx_insert_into_buffer() computes<\/p>\n
used = used1 – to_copy – sp_<\/p>\n","protected":false},"template":"","meta":[],"categories":[],"tags":[],"class_list":["post-6823","vuln","type-vuln","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/www.54iter.cn\/index.php?rest_route=\/wp\/v2\/vuln\/6823","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.54iter.cn\/index.php?rest_route=\/wp\/v2\/vuln"}],"about":[{"href":"https:\/\/www.54iter.cn\/index.php?rest_route=\/wp\/v2\/types\/vuln"}],"wp:attachment":[{"href":"https:\/\/www.54iter.cn\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=6823"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.54iter.cn\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=6823"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.54iter.cn\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=6823"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}