x86/fpu: Ensure XFD state on signal delivery
Subsystems (2)
PATCH 1/1 x86/fpu: Ensure XFD state on signal delivery
Sean reported [1] the following splat when running KVM tests:
WARNING: CPU: 232 PID: 15391 at xfd_validate_state+0x65/0x70
Call Trace:
<TASK>
fpu__clear_user_states+0x9c/0x100
arch_do_signal_or_restart+0x142/0x210
exit_to_user_mode_loop+0x55/0x100
do_syscall_64+0x205/0x2c0
entry_SYSCALL_64_after_hwframe+0x4b/0x53Chao further identified [2] a reproducible scenarios involving signal
delivery: a non-AMX task is preempted by an AMX-enabled task which modifies the XFD MSR.
When the non-AMX task resumes and reloads XSTATE with init values, a warning is triggered due to a mismatch between fpstate::xfd and the CPU's current XFD state. fpu__clear_user_states() does not currently re-synchronize the XFD state after such preemption.
Invoke xfd_update_state() which detects and corrects the mismatch if the dynamic feature is enabled.
This also benefits the sigreturn path, as fpu__restore_sig() may call fpu__clear_user_states() when the sigframe is inaccessible.
--- arch/x86/kernel/fpu/core.c | 3 +++ 1 file changed, 3 insertions(+)
arch/x86/kernel/fpu/core.c +3 -0
@@ -800,6 +800,9 @@ void fpu__clear_user_states(struct fpu *fpu) !fpregs_state_valid(fpu, smp_processor_id()))os_xrstor_supervisor (fpu->fpstate); +/* Ensure XFD state is in sync before reloading XSTATE */ +xfd_update_state (fpu->fpstate); +/* Reset user states in registers. */ restore_fpregs_from_init_fpstate (XFEATURE_MASK_USER_RESTORE );
Signature · 1 line
2.48.1
Quoted 26 lines · Chang S. Bae
Sean reported [1] the following splat when running KVM tests:
WARNING: CPU: 232 PID: 15391 at xfd_validate_state+0x65/0x70 Call Trace: <TASK> fpu__clear_user_states+0x9c/0x100 arch_do_signal_or_restart+0x142/0x210 exit_to_user_mode_loop+0x55/0x100 do_syscall_64+0x205/0x2c0 entry_SYSCALL_64_after_hwframe+0x4b/0x53Chao further identified [2] a reproducible scenarios involving signal
delivery: a non-AMX task is preempted by an AMX-enabled task which modifies the XFD MSR.When the non-AMX task resumes and reloads XSTATE with init values, a warning is triggered due to a mismatch between
fpstate::xfdand the CPU's current XFD state.fpu__clear_user_states()does not currently re-synchronize the XFD state after such preemption.Invoke
xfd_update_state()which detects and corrects the mismatch if the dynamic feature is enabled.This also benefits the sigreturn path, as
fpu__restore_sig()may callfpu__clear_user_states()when the sigframe is inaccessible.
Quoted 7 lines
Fixes 672365477ae8a ("x86/fpu: Update XFD state where required")Reported-by Sean Christopherson <[email protected]>Tested-by Chao Gao <[email protected]>Signed-off-by Chang S. Bae <[email protected]>
Thanks for looking into this issue.
Quoted 35 lines · Chao Gao
Quoted 26 lines · Chang S. Bae
Sean reported [1] the following splat when running KVM tests:
WARNING: CPU: 232 PID: 15391 at xfd_validate_state+0x65/0x70 Call Trace: <TASK> fpu__clear_user_states+0x9c/0x100 arch_do_signal_or_restart+0x142/0x210 exit_to_user_mode_loop+0x55/0x100 do_syscall_64+0x205/0x2c0 entry_SYSCALL_64_after_hwframe+0x4b/0x53Chao further identified [2] a reproducible scenarios involving signal
delivery: a non-AMX task is preempted by an AMX-enabled task which modifies the XFD MSR.When the non-AMX task resumes and reloads XSTATE with init values, a warning is triggered due to a mismatch between
fpstate::xfdand the CPU's current XFD state.fpu__clear_user_states()does not currently re-synchronize the XFD state after such preemption.Invoke
xfd_update_state()which detects and corrects the mismatch if the dynamic feature is enabled.This also benefits the sigreturn path, as
fpu__restore_sig()may callfpu__clear_user_states()when the sigframe is inaccessible.Quoted 7 lines
Fixes 672365477ae8a ("x86/fpu: Update XFD state where required")Reported-by Sean Christopherson <[email protected]>Tested-by Chao Gao <[email protected]>Signed-off-by Chang S. Bae <[email protected]>
Quoted 3 lines
Reviewed-by Chao Gao <[email protected]>Thanks for looking into this issue.
Ping. I _think_ this is still needed? AFAICT, it just fell through the cracks.