[Kees; I'm hoping this is now good-to-go via your hardening tree? It would be good to get some linux-next testing.]
Hi All, As I reported at [1], kstack offset randomisation suffers from a couple of bugs and, on arm64 at least, the performance is poor. This series attempts to fix both; patch 1 provides back-portable fixes for the functional bugs. Patch 2 proposes a performance improvement approach. I've looked at a few different options but ultimately decided that Jeremy's original prng approach is the fastest. I made the argument that this approach is secure "enough" in the RFC [2] and the responses indicated agreement. More details in the commit logs. Performance =========== Mean and tail performance of 3 "small" syscalls was measured. syscall was made 10 million times and each individually measured and binned. These results have low noise so I'm confident that they are trustworthy. The baseline is v6.18-rc5 with stack randomization turned *off*. So I'm showing performance cost of turning it on without any changes to the implementation, then the reduced performance cost of turning it on with my changes applied. **NOTE**: The below results were generated using the RFC patches but there is no meaningful change, so the numbers are still valid. I've also rerun the tests with this version on top of v7.0-rc2 on arm64 and confirmed simialr results. arm64 (AWS Graviton3): +-----------------+--------------+-------------+---------------+ | Benchmark | Result Class | v6.18-rc5 | per-cpu-prng | | | | rndstack-on | | | | | | | +=================+==============+=============+===============+ | syscall/getpid | mean (ns) | (R) 15.62% | (R) 3.43% | | | p99 (ns) | (R) 155.01% | (R) 3.20% | | | p99.9 (ns) | (R) 156.71% | (R) 2.93% | +-----------------+--------------+-------------+---------------+ | syscall/getppid | mean (ns) | (R) 14.09% | (R) 2.12% | | | p99 (ns) | (R) 152.81% | 1.55% | | | p99.9 (ns) | (R) 153.67% | 1.77% | +-----------------+--------------+-------------+---------------+ | syscall/invalid | mean (ns) | (R) 13.89% | (R) 3.32% | | | p99 (ns) | (R) 165.82% | (R) 3.51% | | | p99.9 (ns) | (R) 168.83% | (R) 3.77% | +-----------------+--------------+-------------+---------------+ Because arm64 was previously using get_random_u16(), it was expensive when it didn't have any buffered bits and had to call into the crng. That's what caused the enormous tail latency. x86 (AWS Sapphire Rapids): +-----------------+--------------+-------------+---------------+ | Benchmark | Result Class | v6.18-rc5 | per-cpu-prng | | | | rndstack-on | | | | | | | +=================+==============+=============+===============+ | syscall/getpid | mean (ns) | (R) 13.32% | (R) 4.60% | | | p99 (ns) | (R) 13.38% | (R) 18.08% | | | p99.9 (ns) | 16.26% | (R) 19.38% | +-----------------+--------------+-------------+---------------+ | syscall/getppid | mean (ns) | (R) 11.96% | (R) 5.26% | | | p99 (ns) | (R) 11.83% | (R) 8.35% | | | p99.9 (ns) | (R) 11.42% | (R) 22.37% | +-----------------+--------------+-------------+---------------+ | syscall/invalid | mean (ns) | (R) 10.58% | (R) 2.91% | | | p99 (ns) | (R) 10.51% | (R) 4.36% | | | p99.9 (ns) | (R) 10.35% | (R) 21.97% | +-----------------+--------------+-------------+---------------+ I was surprised to see that the baseline cost on x86 is 10-12% since it is just using rdtsc. But as I say, I believe the results are accurate. Changes since v4 [5] ==================== - Moved add_random_kstack_offset() later in syscall entry code for powerpc, s390 and x86. On these platforms it was previously within noinstr sections but for some exotic Kconfigs, [get|put]_cpu_var() was calling out to instrumentable code. (reported by kernel test robot) - Removed what was previously patch 2 (inline version of prandom_u32_state()). With the above change, there is no longer an issue with calling the out-of-line version. Changes since v3 [4] ==================== - Patch 1: Fixed typo in commit log (per David L) - Patch 2: Reinstated prandom_u32_state() as out-of-line function, which forwards to inline version (per David L) - Patch 3: Added supplementary info about benefits of removing choose_random_kstack_offset() (per Mark R) Changes since v2 [3] ==================== - Moved late_initcall() to initialize kstack_rnd_state out of randomize_kstack.h and into main.c. (issue noticed by kernel test robot) Changes since v1 (RFC) [2] ========================== - Introduced patch 2 to make prandom_u32_state() __always_inline (needed since its called from noinstr code) - In patch 3, prng is now per-cpu instead of per-task (per Ard) [1] https://lore.kernel.org/all/[email protected]/ [2] https://lore.kernel.org/all/[email protected]/ [3] https://lore.kernel.org/all/[email protected]/ [4] https://lore.kernel.org/all/[email protected]/ [5] https://lore.kernel.org/all/[email protected] Thanks, Ryan Ryan Roberts (2): randomize_kstack: Maintain kstack_offset per task randomize_kstack: Unify random source across arches arch/Kconfig | 5 ++- arch/arm64/kernel/syscall.c | 11 ------ arch/loongarch/kernel/syscall.c | 11 ------ arch/powerpc/kernel/syscall.c | 16 ++------- arch/riscv/kernel/traps.c | 12 ------- arch/s390/include/asm/entry-common.h | 8 ----- arch/s390/kernel/syscall.c | 2 +- arch/x86/entry/syscall_32.c | 4 +-- arch/x86/entry/syscall_64.c | 2 +- arch/x86/include/asm/entry-common.h | 12 ------- include/linux/randomize_kstack.h | 54 +++++++++++----------------- init/main.c | 9 ++++- kernel/fork.c | 1 + 13 files changed, 37 insertions(+), 110 deletions(-) -- 2.43.0
