lkml.org 
[lkml]   [2022]   [Nov]   [14]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
Date
SubjectRe: [PATCH] RISC-V: Fix unannoted hardirqs-on in return to userspace slow-path
On Sat, Nov 12, 2022 at 6:52 PM Conor Dooley <conor@kernel.org> wrote:
>
> On Fri, Nov 11, 2022 at 05:31:08PM -0500, Andrew Bresticker wrote:
> > The return to userspace path in entry.S may enable interrupts without the
> > corresponding lockdep annotation, producing a splat[0] when DEBUG_LOCKDEP
> > is enabled. Simply calling __trace_hardirqs_on() here gets a bit messy
> > due to the use of RA to point back to ret_from_exception, so just move
> > the whole slow-path loop into C. It's more readable and it lets us use
> > local_irq_{enable,disable}(), avoiding the need for manual annotations
> > altogether.
> >
> > [0]:
> > ------------[ cut here ]------------
> > DEBUG_LOCKS_WARN_ON(!lockdep_hardirqs_enabled())
> > WARNING: CPU: 2 PID: 1 at kernel/locking/lockdep.c:5512 check_flags+0x10a/0x1e0
> > Modules linked in:
> > CPU: 2 PID: 1 Comm: init Not tainted 6.1.0-rc4-00160-gb56b6e2b4f31 #53
> > Hardware name: riscv-virtio,qemu (DT)
> > epc : check_flags+0x10a/0x1e0
> > ra : check_flags+0x10a/0x1e0
> > <snip>
> > status: 0000000200000100 badaddr: 0000000000000000 cause: 0000000000000003
> > [<ffffffff808edb90>] lock_is_held_type+0x78/0x14e
> > [<ffffffff8003dae2>] __might_resched+0x26/0x22c
> > [<ffffffff8003dd24>] __might_sleep+0x3c/0x66
> > [<ffffffff80022c60>] get_signal+0x9e/0xa70
> > [<ffffffff800054a2>] do_notify_resume+0x6e/0x422
> > [<ffffffff80003c68>] ret_from_exception+0x0/0x10
> > irq event stamp: 44512
> > hardirqs last enabled at (44511): [<ffffffff808f901c>] _raw_spin_unlock_irqrestore+0x54/0x62
> > hardirqs last disabled at (44512): [<ffffffff80008200>] __trace_hardirqs_off+0xc/0x14
> > softirqs last enabled at (44472): [<ffffffff808f9fbe>] __do_softirq+0x3de/0x51e
> > softirqs last disabled at (44467): [<ffffffff80017760>] irq_exit+0xd6/0x104
> > ---[ end trace 0000000000000000 ]---
> > possible reason: unannotated irqs-on.
> >
> > Signed-off-by: Andrew Bresticker <abrestic@rivosinc.com>
> > ---
> > This should also theoretically be fixed by the conversion to generic entry,
> > but it's not clear how far away that series is from landing.
>
> Heh.. In case this makes the cut instead - what commit does this fix?

Ah, right --

Fixes: 3c4697982982 ("riscv: Enable LOCKDEP_SUPPORT & fixup
TRACE_IRQFLAGS_SUPPORT")

-Andrew

>
> > ---
> > arch/riscv/kernel/entry.S | 18 +++++-------------
> > arch/riscv/kernel/signal.c | 34 +++++++++++++++++++++-------------
> > 2 files changed, 26 insertions(+), 26 deletions(-)
> >
> > diff --git a/arch/riscv/kernel/entry.S b/arch/riscv/kernel/entry.S
> > index b9eda3fcbd6d..58dfa8595e19 100644
> > --- a/arch/riscv/kernel/entry.S
> > +++ b/arch/riscv/kernel/entry.S
> > @@ -263,12 +263,11 @@ ret_from_exception:
> > #endif
> > bnez s0, resume_kernel
> >
> > -resume_userspace:
> > /* Interrupts must be disabled here so flags are checked atomically */
> > REG_L s0, TASK_TI_FLAGS(tp) /* current_thread_info->flags */
> > andi s1, s0, _TIF_WORK_MASK
> > - bnez s1, work_pending
> > -
> > + bnez s1, resume_userspace_slow
> > +resume_userspace:
> > #ifdef CONFIG_CONTEXT_TRACKING_USER
> > call user_enter_callable
> > #endif
> > @@ -368,19 +367,12 @@ resume_kernel:
> > j restore_all
> > #endif
> >
> > -work_pending:
> > +resume_userspace_slow:
> > /* Enter slow path for supplementary processing */
> > - la ra, ret_from_exception
> > - andi s1, s0, _TIF_NEED_RESCHED
> > - bnez s1, work_resched
> > -work_notifysig:
> > - /* Handle pending signals and notify-resume requests */
> > - csrs CSR_STATUS, SR_IE /* Enable interrupts for do_notify_resume() */
> > move a0, sp /* pt_regs */
> > move a1, s0 /* current_thread_info->flags */
> > - tail do_notify_resume
> > -work_resched:
> > - tail schedule
> > + call do_work_pending
> > + j resume_userspace
> >
> > /* Slow paths for ptrace. */
> > handle_syscall_trace_enter:
> > diff --git a/arch/riscv/kernel/signal.c b/arch/riscv/kernel/signal.c
> > index 5c591123c440..bfb2afa4135f 100644
> > --- a/arch/riscv/kernel/signal.c
> > +++ b/arch/riscv/kernel/signal.c
> > @@ -313,19 +313,27 @@ static void do_signal(struct pt_regs *regs)
> > }
> >
> > /*
> > - * notification of userspace execution resumption
> > - * - triggered by the _TIF_WORK_MASK flags
> > + * Handle any pending work on the resume-to-userspace path, as indicated by
> > + * _TIF_WORK_MASK. Entered from assembly with IRQs off.
> > */
> > -asmlinkage __visible void do_notify_resume(struct pt_regs *regs,
> > - unsigned long thread_info_flags)
> > +asmlinkage __visible void do_work_pending(struct pt_regs *regs,
> > + unsigned long thread_info_flags)
> > {
> > - if (thread_info_flags & _TIF_UPROBE)
> > - uprobe_notify_resume(regs);
> > -
> > - /* Handle pending signal delivery */
> > - if (thread_info_flags & (_TIF_SIGPENDING | _TIF_NOTIFY_SIGNAL))
> > - do_signal(regs);
> > -
> > - if (thread_info_flags & _TIF_NOTIFY_RESUME)
> > - resume_user_mode_work(regs);
> > + do {
> > + if (thread_info_flags & _TIF_NEED_RESCHED) {
> > + schedule();
> > + } else {
> > + local_irq_enable();
> > + if (thread_info_flags & _TIF_UPROBE)
> > + uprobe_notify_resume(regs);
> > + /* Handle pending signal delivery */
> > + if (thread_info_flags & (_TIF_SIGPENDING |
> > + _TIF_NOTIFY_SIGNAL))
> > + do_signal(regs);
> > + if (thread_info_flags & _TIF_NOTIFY_RESUME)
> > + resume_user_mode_work(regs);
> > + }
> > + local_irq_disable();
> > + thread_info_flags = read_thread_flags();
> > + } while (thread_info_flags & _TIF_WORK_MASK);
> > }
> > --
> > 2.25.1
> >

\
 
 \ /
  Last update: 2022-11-14 15:33    [W:0.150 / U:0.164 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site