The update_cpu_qos_request() function attempts to initialize the 'freq'
variable by dereferencing 'cpudata' before verifying if the 'policy'
is valid.
This issue occurs on systems booted with the "nosmt" parameter, where
all_cpu_data[cpu] is NULL for the SMT sibling threads. As a result,
any call to update_qos_requests() will result in a NULL pointer
dereference as the code will attempt to access pstate.turbo_freq using
the NULL cpudata pointer.
Also, pstate.turbo_freq may be updated by intel_pstate_get_hwp_cap()
after initializing the 'freq' variable, so it is better to defer the
'freq' until intel_pstate_get_hwp_cap() has been called.
Fix this by deferring the 'freq' assignment until after the policy and
driver_data have been validated.
Fixes: ae1bdd23b99f ("cpufreq: intel_pstate: Adjust frequency percentage computations")
Reported-by: Jirka Hladky <jhladky@redhat.com>
Closes: https://lore.kernel.org/all/CAE4VaGDfiPvz3AzrwrwM4kWB3SCkMci25nPO8W1JmTBd=xHzZg@mail.gmail.com/
Signed-off-by: David Arcari <darcari@redhat.com>
Cc: 6.18+ <stable@vger.kernel.org> # 6.18+
[ rjw: Added one paragraph to the changelog ]
Link: https://patch.msgid.link/20260224122106.228116-1-darcari@redhat.com
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
static void update_cpu_qos_request(int cpu, enum freq_qos_req_type type)
{
struct cpudata *cpudata = all_cpu_data[cpu];
- unsigned int freq = cpudata->pstate.turbo_freq;
struct freq_qos_request *req;
+ unsigned int freq;
struct cpufreq_policy *policy __free(put_cpufreq_policy) = cpufreq_cpu_get(cpu);
if (!policy)
if (hwp_active)
intel_pstate_get_hwp_cap(cpudata);
+ freq = cpudata->pstate.turbo_freq;
+
if (type == FREQ_QOS_MIN) {
freq = DIV_ROUND_UP(freq * global.min_perf_pct, 100);
} else {