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 --- drivers/cpufreq/freq_table.c | 370 +++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 370 insertions(+) create mode 100644 drivers/cpufreq/freq_table.c (limited to 'drivers/cpufreq/freq_table.c') diff --git a/drivers/cpufreq/freq_table.c b/drivers/cpufreq/freq_table.c new file mode 100644 index 000000000..ea994647a --- /dev/null +++ b/drivers/cpufreq/freq_table.c @@ -0,0 +1,370 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * linux/drivers/cpufreq/freq_table.c + * + * Copyright (C) 2002 - 2003 Dominik Brodowski + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include +#include + +/********************************************************************* + * FREQUENCY TABLE HELPERS * + *********************************************************************/ + +static bool policy_has_boost_freq(struct cpufreq_policy *policy) +{ + struct cpufreq_frequency_table *pos, *table = policy->freq_table; + + if (!table) + return false; + + cpufreq_for_each_valid_entry(pos, table) + if (pos->flags & CPUFREQ_BOOST_FREQ) + return true; + + return false; +} + +int cpufreq_frequency_table_cpuinfo(struct cpufreq_policy *policy) +{ + struct cpufreq_frequency_table *pos, *table = policy->freq_table; + unsigned int min_freq = ~0; + unsigned int max_freq = 0; + unsigned int freq, i; + + cpufreq_for_each_valid_entry_idx(pos, table, i) { + freq = pos->frequency; + + if ((!cpufreq_boost_enabled() || !policy->boost_enabled) + && (pos->flags & CPUFREQ_BOOST_FREQ)) + continue; + + pr_debug("table entry %u: %u kHz\n", i, freq); + if (freq < min_freq) + min_freq = freq; + if (freq > max_freq) + max_freq = freq; + } + + policy->cpuinfo.min_freq = min_freq; + /* + * If the driver has set its own cpuinfo.max_freq above max_freq, leave + * it as is. + */ + if (policy->cpuinfo.max_freq < max_freq) + policy->cpuinfo.max_freq = max_freq; + + if (min_freq == ~0) + return -EINVAL; + else + return 0; +} + +int cpufreq_frequency_table_verify(struct cpufreq_policy_data *policy) +{ + struct cpufreq_frequency_table *pos, *table = policy->freq_table; + unsigned int freq, prev_smaller = 0; + bool found = false; + + pr_debug("request for verification of policy (%u - %u kHz) for cpu %u\n", + policy->min, policy->max, policy->cpu); + + cpufreq_verify_within_cpu_limits(policy); + + cpufreq_for_each_valid_entry(pos, table) { + freq = pos->frequency; + + if ((freq >= policy->min) && (freq <= policy->max)) { + found = true; + break; + } + + if ((prev_smaller < freq) && (freq <= policy->max)) + prev_smaller = freq; + } + + if (!found) { + policy->max = prev_smaller; + cpufreq_verify_within_cpu_limits(policy); + } + + pr_debug("verification lead to (%u - %u kHz) for cpu %u\n", + policy->min, policy->max, policy->cpu); + + return 0; +} +EXPORT_SYMBOL_GPL(cpufreq_frequency_table_verify); + +/* + * Generic routine to verify policy & frequency table, requires driver to set + * policy->freq_table prior to it. + */ +int cpufreq_generic_frequency_table_verify(struct cpufreq_policy_data *policy) +{ + if (!policy->freq_table) + return -ENODEV; + + return cpufreq_frequency_table_verify(policy); +} +EXPORT_SYMBOL_GPL(cpufreq_generic_frequency_table_verify); + +int cpufreq_table_index_unsorted(struct cpufreq_policy *policy, + unsigned int target_freq, unsigned int min, + unsigned int max, unsigned int relation) +{ + struct cpufreq_frequency_table optimal = { + .driver_data = ~0, + .frequency = 0, + }; + struct cpufreq_frequency_table suboptimal = { + .driver_data = ~0, + .frequency = 0, + }; + struct cpufreq_frequency_table *pos; + struct cpufreq_frequency_table *table = policy->freq_table; + unsigned int freq, diff, i; + int index; + + pr_debug("request for target %u kHz (relation: %u) for cpu %u\n", + target_freq, relation, policy->cpu); + + switch (relation) { + case CPUFREQ_RELATION_H: + suboptimal.frequency = ~0; + break; + case CPUFREQ_RELATION_L: + case CPUFREQ_RELATION_C: + optimal.frequency = ~0; + break; + } + + cpufreq_for_each_valid_entry_idx(pos, table, i) { + freq = pos->frequency; + + if (freq < min || freq > max) + continue; + if (freq == target_freq) { + optimal.driver_data = i; + break; + } + switch (relation) { + case CPUFREQ_RELATION_H: + if (freq < target_freq) { + if (freq >= optimal.frequency) { + optimal.frequency = freq; + optimal.driver_data = i; + } + } else { + if (freq <= suboptimal.frequency) { + suboptimal.frequency = freq; + suboptimal.driver_data = i; + } + } + break; + case CPUFREQ_RELATION_L: + if (freq > target_freq) { + if (freq <= optimal.frequency) { + optimal.frequency = freq; + optimal.driver_data = i; + } + } else { + if (freq >= suboptimal.frequency) { + suboptimal.frequency = freq; + suboptimal.driver_data = i; + } + } + break; + case CPUFREQ_RELATION_C: + diff = abs(freq - target_freq); + if (diff < optimal.frequency || + (diff == optimal.frequency && + freq > table[optimal.driver_data].frequency)) { + optimal.frequency = diff; + optimal.driver_data = i; + } + break; + } + } + if (optimal.driver_data > i) { + if (suboptimal.driver_data > i) { + WARN(1, "Invalid frequency table: %u\n", policy->cpu); + return 0; + } + + index = suboptimal.driver_data; + } else + index = optimal.driver_data; + + pr_debug("target index is %u, freq is:%u kHz\n", index, + table[index].frequency); + return index; +} +EXPORT_SYMBOL_GPL(cpufreq_table_index_unsorted); + +int cpufreq_frequency_table_get_index(struct cpufreq_policy *policy, + unsigned int freq) +{ + struct cpufreq_frequency_table *pos, *table = policy->freq_table; + int idx; + + if (unlikely(!table)) { + pr_debug("%s: Unable to find frequency table\n", __func__); + return -ENOENT; + } + + cpufreq_for_each_valid_entry_idx(pos, table, idx) + if (pos->frequency == freq) + return idx; + + return -EINVAL; +} +EXPORT_SYMBOL_GPL(cpufreq_frequency_table_get_index); + +/* + * show_available_freqs - show available frequencies for the specified CPU + */ +static ssize_t show_available_freqs(struct cpufreq_policy *policy, char *buf, + bool show_boost) +{ + ssize_t count = 0; + struct cpufreq_frequency_table *pos, *table = policy->freq_table; + + if (!table) + return -ENODEV; + + cpufreq_for_each_valid_entry(pos, table) { + /* + * show_boost = true and driver_data = BOOST freq + * display BOOST freqs + * + * show_boost = false and driver_data = BOOST freq + * show_boost = true and driver_data != BOOST freq + * continue - do not display anything + * + * show_boost = false and driver_data != BOOST freq + * display NON BOOST freqs + */ + if (show_boost ^ (pos->flags & CPUFREQ_BOOST_FREQ)) + continue; + + count += sprintf(&buf[count], "%u ", pos->frequency); + } + count += sprintf(&buf[count], "\n"); + + return count; + +} + +#define cpufreq_attr_available_freq(_name) \ +struct freq_attr cpufreq_freq_attr_##_name##_freqs = \ +__ATTR_RO(_name##_frequencies) + +/* + * scaling_available_frequencies_show - show available normal frequencies for + * the specified CPU + */ +static ssize_t scaling_available_frequencies_show(struct cpufreq_policy *policy, + char *buf) +{ + return show_available_freqs(policy, buf, false); +} +cpufreq_attr_available_freq(scaling_available); + +/* + * scaling_boost_frequencies_show - show available boost frequencies for + * the specified CPU + */ +static ssize_t scaling_boost_frequencies_show(struct cpufreq_policy *policy, + char *buf) +{ + return show_available_freqs(policy, buf, true); +} +cpufreq_attr_available_freq(scaling_boost); + +static int set_freq_table_sorted(struct cpufreq_policy *policy) +{ + struct cpufreq_frequency_table *pos, *table = policy->freq_table; + struct cpufreq_frequency_table *prev = NULL; + int ascending = 0; + + policy->freq_table_sorted = CPUFREQ_TABLE_UNSORTED; + + cpufreq_for_each_valid_entry(pos, table) { + if (!prev) { + prev = pos; + continue; + } + + if (pos->frequency == prev->frequency) { + pr_warn("Duplicate freq-table entries: %u\n", + pos->frequency); + return -EINVAL; + } + + /* Frequency increased from prev to pos */ + if (pos->frequency > prev->frequency) { + /* But frequency was decreasing earlier */ + if (ascending < 0) { + pr_debug("Freq table is unsorted\n"); + return 0; + } + + ascending++; + } else { + /* Frequency decreased from prev to pos */ + + /* But frequency was increasing earlier */ + if (ascending > 0) { + pr_debug("Freq table is unsorted\n"); + return 0; + } + + ascending--; + } + + prev = pos; + } + + if (ascending > 0) + policy->freq_table_sorted = CPUFREQ_TABLE_SORTED_ASCENDING; + else + policy->freq_table_sorted = CPUFREQ_TABLE_SORTED_DESCENDING; + + pr_debug("Freq table is sorted in %s order\n", + ascending > 0 ? "ascending" : "descending"); + + return 0; +} + +int cpufreq_table_validate_and_sort(struct cpufreq_policy *policy) +{ + int ret; + + if (!policy->freq_table) { + /* Freq table must be passed by drivers with target_index() */ + if (has_target_index()) + return -EINVAL; + + return 0; + } + + ret = cpufreq_frequency_table_cpuinfo(policy); + if (ret) + return ret; + + /* Driver's may have set this field already */ + if (policy_has_boost_freq(policy)) + policy->boost_supported = true; + + if (policy->freq_table_sorted == CPUFREQ_TABLE_SORTED_ASCENDING || + policy->freq_table_sorted == CPUFREQ_TABLE_SORTED_DESCENDING) + return 0; + + return set_freq_table_sorted(policy); +} + +MODULE_AUTHOR("Dominik Brodowski "); +MODULE_DESCRIPTION("CPUfreq frequency table helpers"); -- cgit v1.3.1