From a5f7a5bb3b7f28ba7e4fa246775b29a0e5537255 Mon Sep 17 00:00:00 2001 From: Jinke Han Date: Tue, 8 Sep 2026 15:37:42 +0800 Subject: x86/kprobes: Fix crash when probing CS CALL instructions When using eBPF to probe CS CALL instructions within a function, a crash can be triggered. The eBPF tool probes offset 257 of the __hrtimer_run_queues() function: <__hrtimer_run_queues+249>: nopl 0x0(%rax,%rax,1) <__hrtimer_run_queues+254>: mov %r14,%rdi <__hrtimer_run_queues+257>: cs call <__x86_indirect_thunk_r12> <__hrtimer_run_queues+263>: mov %eax,%r12d <__hrtimer_run_queues+266>: xchg %ax,%ax <__hrtimer_run_queues+268>: mov %r13,%rdi Which triggers this crash: BUG: unable to handle page fault for address: 00000000000f41c9 #PF: supervisor write access in kernel mode #PF: error_code(0x0002) - not-present page PGD 0 P4D 0 Oops: 0002 [#1] SMP NOPTI CPU: 1 PID: 0 Comm: swapper/1 Kdump: loaded Tainted: P RIP: 0010:__hrtimer_run_queues+0x106/0x230 Note that __hrtimer_run_queues+0x106 is __hrtimer_run_queues+262, which is at the 6th byte of the above CS CALL instruction. Since the CS CALL instruction occupies 6 bytes, the exception occurred in the middle of that call instruction. The root cause is that when using eBPF tools to probe in the middle of a function, a kprobe with INT3 is used as the underlying implementation. During single-step emulation of the original CALL instruction, int3_emulate_call() assumes that the probed CALL instruction is 5 bytes long. However, the actual CS-prefixed CALL instruction occupies 6 bytes, so it constructs an incorrect exception return address. When the CPU returns from the kprobe handler, the next instruction to be executed is at the address of the last byte of that CS CALL instruction. Coincidentally, starting from that address, the CPU fetches and decodes a completely different instruction, which ultimately triggers a kernel crash. Fix the issue by using the actual instruction length obtained from the instruction decoder when constructing the exception return address, rather than relying on the hardcoded CALL_INSN_SIZE macro. [ mingo: Refined the changelog ] Fixes: 6256e668b7af ("x86/kprobes: Use int3 instead of debug trap for single-step") Suggested-by: Masami Hiramatsu (Google) Signed-off-by: Jinke Han Signed-off-by: Ingo Molnar Reviewed-by: Masami Hiramatsu (Google) Acked-by: Yafang Shao Acked-by: Borislav Petkov Cc: Peter Zijlstra Link: https://patch.msgid.link/20260908073742.GA10517@didi-ThinkCentre-M920t-N000 --- Documentation/gpu/msm-preemption.rst | 99 ++++++++++++++++++++++++++++++++++++ 1 file changed, 99 insertions(+) create mode 100644 Documentation/gpu/msm-preemption.rst (limited to 'Documentation/gpu/msm-preemption.rst') diff --git a/Documentation/gpu/msm-preemption.rst b/Documentation/gpu/msm-preemption.rst new file mode 100644 index 000000000..d768ca09f --- /dev/null +++ b/Documentation/gpu/msm-preemption.rst @@ -0,0 +1,99 @@ +.. SPDX-License-Identifier: GPL-2.0 + +:orphan: + +============== +MSM Preemption +============== + +Preemption allows Adreno GPUs to switch to a higher priority ring when work is +pushed to it, reducing latency for high priority submissions. + +When preemption is enabled 4 rings are initialized, corresponding to different +priority levels. Having multiple rings is purely a software concept as the GPU +only has registers to keep track of one graphics ring. +The kernel is able to switch which ring is currently being processed by +requesting preemption. When certain conditions are met, depending on the +priority level, the GPU will save its current state in a series of buffers, +then restores state from a similar set of buffers specified by the kernel. It +then resumes execution and fires an IRQ to let the kernel know the context +switch has completed. + +This mechanism can be used by the kernel to switch between rings. Whenever a +submission occurs the kernel finds the highest priority ring which isn't empty +and preempts to it if said ring is not the one being currently executed. This is +also done whenever a submission completes to make sure execution resumes on a +lower priority ring when a higher priority ring is done. + +Preemption levels +----------------- + +Preemption can only occur at certain boundaries. The exact conditions can be +configured by changing the preemption level, this allows to compromise between +latency (ie. the time that passes between when the kernel requests preemption +and when the SQE begins saving state) and overhead (the amount of state that +needs to be saved). + +The GPU offers 3 levels: + +Level 0 + Preemption only occurs at the submission level. This requires the least amount + of state to be saved as the execution of userspace submitted IBs is never + interrupted, however it offers very little benefit compared to not enabling + preemption of any kind. + +Level 1 + Preemption occurs at either bin level, if using GMEM rendering, or draw level + in the sysmem rendering case. + +Level 2 + Preemption occurs at draw level. + +Level 1 is the mode that is used by the msm driver. + +Additionally the GPU allows to specify a `skip_save_restore` option. This +disables the saving and restoring of all registers except those relating to the +operation of the SQE itself, reducing overhead. Saving and restoring is only +skipped when using GMEM with Level 1 preemption. When enabling this userspace is +expected to set the state that isn't preserved whenever preemption occurs which +is done by specifying preamble and postambles. Those are IBs that are executed +before and after preemption. + +Preemption buffers +------------------ + +A series of buffers are necessary to store the state of rings while they are not +being executed. There are different kinds of preemption records and most of +those require one buffer per ring. This is because preemption never occurs +between submissions on the same ring, which always run in sequence when the ring +is active. This means that only one context per ring is effectively active. + +SMMU_INFO + This buffer contains info about the current SMMU configuration such as the + ttbr0 register. The SQE firmware isn't actually able to save this record. + As a result SMMU info must be saved manually from the CP to a buffer and the + SMMU record updated with info from said buffer before triggering + preemption. + +NON_SECURE + This is the main preemption record where most state is saved. It is mostly + opaque to the kernel except for the first few words that must be initialized + by the kernel. + +SECURE + This saves state related to the GPU's secure mode. + +NON_PRIV + The intended purpose of this record is unknown. The SQE firmware actually + ignores it and therefore msm doesn't handle it. + +COUNTER + This record is used to save and restore performance counters. + +Handling the permissions of those buffers is critical for security. All but the +NON_PRIV records need to be inaccessible from userspace, so they must be mapped +in the kernel address space with the MSM_BO_MAP_PRIV flag. +For example, making the NON_SECURE record accessible from userspace would allow +any process to manipulate a saved ring's RPTR which can be used to skip the +execution of some packets in a ring and execute user commands with higher +privileges. -- cgit v1.3.1