lkml.org 
[lkml]   [2020]   [Sep]   [30]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH] mfd: sun4i-gpadc: Interrupt numbers should start from 1
On Thu, Sep 17, 2020 at 04:01:17PM +0200, Ondřej Jirman wrote:
> Hello Maxime,
>
> On Thu, Sep 17, 2020 at 03:19:04PM +0200, Maxime Ripard wrote:
> > Hi,
> >
> > On Sat, Sep 12, 2020 at 01:22:00PM +0200, Ondrej Jirman wrote:
> > > mfd: sun4i-gpadc: Interrupt numbers should start from 1
> >
> > Why? An hwirq with 0 is totally fine
> >
> > > This avoids a warning:
> > >
> > > [ 2.891592] ------------[ cut here ]------------
> > > [ 2.895052] WARNING: CPU: 0 PID: 149 at drivers/base/platform.c:317 __platform_get_irq_byname+0x7c/0x8c
> > > [ 2.903212] usb 1-1: new high-speed USB device number 2 using ehci-platform
> > > [ 2.908930] 0 is an invalid IRQ number
> > > [ 2.911425] Modules linked in: sun4i_gpadc_iio(+) r8188eu(C) lib80211 ohci_platform ohci_hcd ehci_platform ehci_hcd cyttsp4_i2c cyttsp_i2c_common cyttsp4_core g_cdc usb_f_acm u_serial usb_f_ecm u_ether libcomposite sunxi phy_generic musb_hdrc udc_core usbcore sun5ieink
> > > [ 2.934048] CPU: 0 PID: 149 Comm: tablet-init Tainted: G C 5.8.0-rc2-00316-gc6a5213fdeba-dirty #8
> > > [ 2.943027] Hardware name: Allwinner sun4i/sun5i Families
> > > [ 2.947204] [<c010c080>] (unwind_backtrace) from [<c0109ee4>] (show_stack+0x10/0x14)
> > > [ 2.953714] [<c0109ee4>] (show_stack) from [<c012cd04>] (__warn+0xc0/0xd8)
> > > [ 2.959364] [<c012cd04>] (__warn) from [<c012cda0>] (warn_slowpath_fmt+0x84/0x94)
> > > [ 2.965599] [<c012cda0>] (warn_slowpath_fmt) from [<c0592a60>] (__platform_get_irq_byname+0x7c/0x8c)
> > > [ 2.973480] [<c0592a60>] (__platform_get_irq_byname) from [<c0592a80>] (platform_get_irq_byname+0x10/0x48)
> > > [ 2.981896] [<c0592a80>] (platform_get_irq_byname) from [<bf1142f0>] (sun4i_irq_init+0x38/0xe0 [sun4i_gpadc_iio])
> > > [ 2.990923] [<bf1142f0>] (sun4i_irq_init [sun4i_gpadc_iio]) from [<bf1145cc>] (sun4i_gpadc_probe+0x234/0x308 [sun4i_gpadc_iio])
> > > [ 3.001152] [<bf1145cc>] (sun4i_gpadc_probe [sun4i_gpadc_iio]) from [<c0592edc>] (platform_drv_probe+0x48/0x98)
> > > [ 3.010051] [<c0592edc>] (platform_drv_probe) from [<c0591074>] (really_probe+0x1e0/0x348)
> > > [ 3.017152] [<c0591074>] (really_probe) from [<c05912e8>] (driver_probe_device+0x5c/0xb4)
> > > [ 3.024081] [<c05912e8>] (driver_probe_device) from [<c05914e0>] (device_driver_attach+0x58/0x60)
> > > [ 3.031696] [<c05914e0>] (device_driver_attach) from [<c0591540>] (__driver_attach+0x58/0xcc)
> > > [ 3.038966] [<c0591540>] (__driver_attach) from [<c058f418>] (bus_for_each_dev+0x64/0x90)
> > > [ 3.045886] [<c058f418>] (bus_for_each_dev) from [<c0590490>] (bus_add_driver+0x15c/0x1e0)
> > > [ 3.052892] [<c0590490>] (bus_add_driver) from [<c0591dc4>] (driver_register+0x7c/0x114)
> > > [ 3.059731] [<c0591dc4>] (driver_register) from [<c0101674>] (do_one_initcall+0x44/0x194)
> > > [ 3.066696] [<c0101674>] (do_one_initcall) from [<c01946e8>] (do_init_module+0x5c/0x220)
> > > [ 3.073568] [<c01946e8>] (do_init_module) from [<c0196a3c>] (load_module+0x20ec/0x2380)
> > > [ 3.080340] [<c0196a3c>] (load_module) from [<c0196e04>] (sys_init_module+0x134/0x154)
> > > [ 3.087020] [<c0196e04>] (sys_init_module) from [<c0100060>] (ret_fast_syscall+0x0/0x54)
> > > [ 3.093852] Exception stack(0xc8ea7fa8 to 0xc8ea7ff0)
> > > [ 3.097649] 7fa0: 00000000 00000000 b6fe2000 00003b14 00031284 00000005
> > > [ 3.104596] 7fc0: 00000000 00000000 b6fe2000 00000080 00052220 be9fd940 00052246 0000002e
> > > [ 3.111626] 7fe0: be9fd79c be9fd780 00015bd0 0001dcf4
> > > [ 3.115468] ---[ end trace df4dd47fb61bf5a4 ]---
> >
> > In which situation?
>
> During boot.
>
> It's a new check added in Linux 5.8. https://elixir.bootlin.com/linux/latest/source/drivers/base/platform.c#L317
>
> > > Signed-off-by: Ondrej Jirman <megous@megous.com>
> > > ---
> > > include/linux/mfd/sun4i-gpadc.h | 4 ++--
> > > 1 file changed, 2 insertions(+), 2 deletions(-)
> > >
> > > diff --git a/include/linux/mfd/sun4i-gpadc.h b/include/linux/mfd/sun4i-gpadc.h
> > > index ea0ccf33a459..021f820f9d52 100644
> > > --- a/include/linux/mfd/sun4i-gpadc.h
> > > +++ b/include/linux/mfd/sun4i-gpadc.h
> > > @@ -81,8 +81,8 @@
> > > #define SUN4I_GPADC_TEMP_DATA 0x20
> > > #define SUN4I_GPADC_DATA 0x24
> > >
> > > -#define SUN4I_GPADC_IRQ_FIFO_DATA 0
> > > -#define SUN4I_GPADC_IRQ_TEMP_DATA 1
> > > +#define SUN4I_GPADC_IRQ_FIFO_DATA 1
> > > +#define SUN4I_GPADC_IRQ_TEMP_DATA 2
> >
> > Where is it coming from, and why is it the proper fix?
>
> Actual numbers seem irrelevant. It's just an index into this array:
>
> https://elixir.bootlin.com/linux/latest/source/drivers/mfd/sun4i-gpadc.c#L27
>
> The array will get sparse, but that doesn't seem like an issue to me,
> because the irq code handles the holes in the list if mask is 0.
>
> Not sure it's the best fix.

Explaining all that in the commit log would be great, especially if
you're not sure.

Maxime
[unhandled content-type:application/pgp-signature]
\
 
 \ /
  Last update: 2020-09-30 16:34    [W:0.042 / U:0.640 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site