Description

In the Linux kernel, the following vulnerability has been resolved: svcrdma: Fix offset arithmetic in read_chunk_range svc_rdma_read_chunk_range() walks a Read chunk's segment list to build a sub-range starting at byte offset and spanning length bytes for a Position-Zero or Call chunk. Two arithmetic defects in the per-segment loop produce wrong DMA lengths and a u32 underflow: pcl_for_each_segment(segment, chunk) { if (offset > segment->rs_length) { offset -= segment->rs_length; continue; } dummy.rs_handle = segment->rs_handle; dummy.rs_length = min_t(u32, length, segment->rs_length) - offset; dummy.rs_offset = segment->rs_offset + offset; First, the skip predicate uses '>' instead of '>='. When offset equals the segment's full rs_length, the segment is fully consumed and should be skipped, but the loop falls through into the body. The resulting dummy.rs_length is min_t(u32, length, rs_length) - rs_length, which underflows to a near-UINT_MAX u32 when length is smaller than rs_length, or is zero otherwise. Second, the length formula subtracts offset from the min_t() result rather than from segment->rs_length before the cap. For offset > 0 the segment's residual is rs_length - offset, not rs_length, so the cap must be applied to the residual. With the current bracketing, whenever length is smaller than rs_length - offset the per-segment length becomes length - offset instead of length, silently dropping offset bytes from the rebuilt chunk. Combined with the boundary case above it also enables the u32 underflow path, which propagates a huge nr_bvec into svc_rdma_build_read_segment() and a multi-MiB kmalloc_array_node() in svc_rdma_get_rw_ctxt(). Additionally, svc_rdma_read_call_chunk() can invoke this function with length == 0 when the last Read chunk ends exactly at the end of the Call chunk. With the corrected >= predicate, every segment is skipped and the function returns the initial -EINVAL, rejecting a valid request. Return success immediately when length is zero. Also break out of the loop once length is fully consumed to avoid passing zero-length segments to svc_rdma_build_read_segment(). Fix by using '>=' so a fully-consumed segment is skipped, by moving '- offset' inside min_t() so the cap is applied to the segment's residual length, by returning success for zero-length requests, and by stopping iteration when the requested range has been consumed.

Executive Summary

Technical Details

Mitigations

References

Additional Metadata

{
  "score": "9.8",
  "pubDate": "2026-09-11T20:19:36.250Z",
  "pubdate": "2026-09-11T20:19:36.250Z",
  "executiveSummary": "The Linux kernel svcrdma component contains an integer underflow vulnerability within the svc_rdma_read_chunk_range function.\nThe flaw stems from improper offset arithmetic and segment boundary handling when processing Read chunk segment lists.\nAn unauthenticated remote attacker can trigger this vulnerability by crafting specific Read chunk requests during an RDMA-based RPC operation.\nSuccessful exploitation results in the calculation of an invalidly large length value, leading to an integer underflow.\nThis underflow propagates to downstream functions such as svc_rdma_get_rw_ctxt, potentially causing massive memory allocation requests via kmalloc_array_node, which can lead to a kernel panic or denial-of-service (DoS) condition.\nThe issue affects the svcrdma subsystem, which is integral to NFS-over-RDMA operations, making systems providing network storage services particularly vulnerable.",
  "technicalDetails": "The vulnerability resides in the loop logic of svc_rdma_read_chunk_range within the Linux kernel's svcrdma implementation. This function is responsible for parsing segment lists in Read chunks to identify sub-ranges defined by a byte offset and a specific length.\nRoot Cause: Two primary defects in the segment iteration logic drive the vulnerability. First, the skip predicate uses a strict 'greater than' (>) operator instead of 'greater than or equal to' (>=). When the provided offset exactly matches the segment's length (rs_length), the code fails to skip the segment, leading to an incorrect arithmetic operation. Second, the calculation of the segment's remaining length is flawed; the code subtracts the offset from the result of a min_t() call rather than from the segment's original length before applying the cap. This leads to an incorrect residual length calculation when the offset is greater than zero.\nExploitation Flow: An attacker initiates an RPC call involving a specially crafted Read chunk. When svc_rdma_read_chunk_range processes this, the faulty skip predicate causes it to treat a fully consumed segment as active. The subsequent length calculation performs an unsigned integer subtraction: min_t(u32, length, rs_length) - offset. When 'length' is smaller than 'rs_length', this result underflows, producing a near-UINT_MAX value.\nImpact of Underflow: This underflowed value is treated as the 'length' for the segment, which is subsequently used to calculate 'nr_bvec' (the number of buffer vectors). A massive, incorrect 'nr_bvec' value is then passed into svc_rdma_build_read_segment. This triggers an allocation attempt via kmalloc_array_node in svc_rdma_get_rw_ctxt. Because the requested size is multi-MiB or potentially larger, it can lead to memory exhaustion, failed allocations, or kernel stability issues.\nBoundary Conditions: The logic fails to handle cases where the requested length is zero, potentially causing the function to reject valid requests or incorrectly calculate segments. The lack of an early break condition once the requested range is satisfied also allows for potential processing of zero-length segments, exacerbating the risks associated with the faulty arithmetic logic."
}