概要

In the Linux kernel, the following vulnerability has been resolved:bpf: Fix overrunning reservations in ringbufThe BPF ring buffer internally is implemented as a power-of-2 sized circularbuffer, with two logical and ever-increasing counters: consumer_pos is theconsumer counter to show which logical position the consumer consumed thedata, and producer_pos which is the producer counter denoting the amount ofdata reserved by all producers.Each time a record is reserved, the producer that owns the record willsuccessfully advance producer counter. In user space each time a record isread, the consumer of the data advanced the consumer counter once it finishedprocessing. Both counters are stored in separate pages so that from userspace, the producer counter is read-only and the consumer counter is read-write.One aspect that simplifies and thus speeds up the implementation of bothproducers and consumers is how the data area is mapped twice contiguouslyback-to-back in the virtual memory, allowing to not take any special measuresfor samples that have to wrap around at the end of the circular buffer dataarea, because the next page after the last data page would be first data pageagain, and thus the sample will still appear completely contiguous in virtualmemory.Each record has a struct bpf_ringbuf_hdr { u32 len; u32 pg_off; } header forbook-keeping the length and offset, and is inaccessible to the BPF program.Helpers like bpf_ringbuf_reserve() return `(void *)hdr + BPF_RINGBUF_HDR_SZ`for the BPF program to use. Bing-Jhong and Muhammad reported that it is howeverpossible to make a second allocated memory chunk overlapping with the firstchunk and as a result, the BPF program is now able to edit first chunk sheader.For example, consider the creation of a BPF_MAP_TYPE_RINGBUF map with sizeof 0x4000. Next, the consumer_pos is modified to 0x3000 /before/ a call tobpf_ringbuf_reserve() is made. This will allocate a chunk A, which is in[0x0,0x3008], and the BPF program is able to edit [0x8,0x3008]. Now, letsallocate a chunk B with size 0x3000. This will succeed because consumer_poswas edited ahead of time to pass the `new_prod_pos - cons_pos > rb->mask`check. Chunk B will be in range [0x3008,0x6010], and the BPF program is ableto edit [0x3010,0x6010]. Due to the ring buffer memory layout mentionedearlier, the ranges [0x0,0x4000] and [0x4000,0x8000] point to the same datapages. This means that chunk B at [0x4000,0x4008] is chunk A s header.bpf_ringbuf_submit() / bpf_ringbuf_discard() use the header s pg_off to thenlocate the bpf_ringbuf itself via bpf_ringbuf_restore_from_rec(). Once chunkB modified chunk A s header, then bpf_ringbuf_commit() refers to the wrongpage and could cause a crash.Fix it by calculating the oldest pending_pos and check whether the rangefrom the oldest outstanding record to the newest would span beyond the ringbuffer size. If that is the case, then reject the request. We ve tested withthe ring buffer benchmark in BPF selftests (./benchs/run_bench_ringbufs.sh)before/after the fix and while it seems a bit slower on some benchmarks, itis still not significantly enough to matter.

CVSS v3 指标

安全公告

影响产品