RE: [PATCH] scsi: ufs: core: Ensure clk_gating.lock is used only after initialization
From: Avri Altman
Date: Thu Jan 23 2025 - 15:03:48 EST
> I just ran into the same issue on R-Car S4 (S4 Starter Kit).
>
> > --- a/drivers/ufs/core/ufshcd.c
> > +++ b/drivers/ufs/core/ufshcd.c
> > @@ -9142,7 +9142,7 @@ static int ufshcd_setup_clocks(struct ufs_hba *hba,
> bool on)
> > if (!IS_ERR_OR_NULL(clki->clk) && clki->enabled)
> > clk_disable_unprepare(clki->clk);
> > }
> > - } else if (!ret && on) {
> > + } else if (!ret && on && hba->clk_gating.is_initialized) {
> > scoped_guard(spinlock_irqsave, &hba->clk_gating.lock)
> > hba->clk_gating.state = CLKS_ON;
> > trace_ufshcd_clk_gating(dev_name(hba->dev),
>
> This looks like a very fragile solution to me...
>
> In addition, while this change does fix this particular spinlock warning, it just
> BUGs in a different place later:
>
> do_raw_spin_lock+0x34/0xb4
> _raw_spin_lock_irqsave+0x1c/0x30
> class_spinlock_irqsave_constructor+0x18/0x30
> - ufshcd_setup_clocks+0x98/0x23c
> - ufshcd_init+0x268/0xd2c
> + ufshcd_release+0x30/0x74
> + ufshcd_send_uic_cmd+0x70/0x90
> + ufshcd_link_startup.constprop.0+0x70/0x258
> + ufshcd_init+0xa38/0xd2c
> ufshcd_pltfrm_init+0x618/0x738
> ufs_renesas_probe+0x18/0x24
> platform_probe+0x68/0xb8
I don't understand how it is possible that `ufshcd_init_clk_gating(hba)` is called after `ufshcd_link_startup(hba)` in 'ufshcd_init'.
Nor how concurrency could take place in this init flow.
Evidently, this is happening.
>
> I think you should initialize all your spinlocks (and mutexes) early in
> ufshcd_init(), instead of sprinkled across various helper functions.
This is the case today. Let me suggest a different fix.
Thanks,
Avri
>
> Gr{oetje,eeting}s,
>
> Geert
>
> --
> Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-
> m68k.org
>
> In personal conversations with technical people, I call myself a hacker. But when
> I'm talking to journalists I just say "programmer" or something like that.
> -- Linus Torvalds