Duplicate regulator sysfs files..

Duplicate regulator sysfs files..

[Linus Torvalds]

Liam, Mark,
 appended is a report from Werner Landgraf <w.landgraf@ru.ru> who
builds a kernel with pretty much everything enabled, and forwarded
this from one of his testers.

Comments? Why does that "supply vmmc not found, using dummy regulator"
message show up twice? And apparently it tries to register it both
times, which on the second try results in a WARN_ON() from sysfs..

              Linus

---
lspci entry:

02:0b.4 Generic system peripheral [0805]: Texas Instruments
PCI6411/6421/6611/6621/7411/7421/7611/7621 Secure Digital Controller
(prog-if 01)
	Subsystem: Toshiba America Info Systems Unknown device 0001
	Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr-
Stepping- SERR- FastB2B-
	Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=medium >TAbort-
<TAbort- <MAbort- >SERR- <PERR-
	Latency: 64 (1750ns min, 1000ns max), Cache Line Size: 32 bytes
	Interrupt: pin D routed to IRQ 11
	Region 0: Memory at cee07800 (32-bit, non-prefetchable) [size=256]
	Region 1: Memory at cee07900 (32-bit, non-prefetchable) [size=256]
	Region 2: Memory at cee07a00 (32-bit, non-prefetchable) [size=256]
	Capabilities: [80] Power Management version 2
		Flags: PMEClk- DSI- D1+ D2+ AuxCurrent=0mA PME(D0+,D1+,D2+,D3hot+,D3cold-)
		Status: D0 PME-Enable- DSel=0 DScale=0 PME-
00: 4c 10 34 80 06 00 10 02 00 01 05 08 08 40 80 00
10: 00 78 e0 ce 00 79 e0 ce 00 7a e0 ce 00 00 00 00
20: 00 00 00 00 00 00 00 00 00 00 00 00 79 11 01 00
30: 00 00 00 00 80 00 00 00 00 00 00 00 ff 04 07 04

dmesg excerpt:

sdhci: Secure Digital Host Controller Interface driver
sdhci: Copyright(c) Pierre Ossman
sdhci-pci 0000:02:0b.4: SDHCI controller found [104c:8034] (rev 0)
sdhci-pci 0000:02:0b.4: enabling device (0000 -> 0002)
sdhci-pci 0000:02:0b.4: PCI INT D -> Link[LNKD] -> GSI 11 (level, low) -> IRQ 11
_regulator_get: 0000:02:0b.4 supply vmmc not found, using dummy regulator
Registered led device: mmc0::
mmc0: SDHCI controller on PCI [0000:02:0b.4] using DMA
_regulator_get: 0000:02:0b.4 supply vmmc not found, using dummy regulator
------------[ cut here ]------------
WARNING: at fs/sysfs/dir.c:455 sysfs_add_one+0x83/0x95()
Hardware name: Qosmio E10
sysfs: cannot create duplicate filename
'/devices/pci0000:00/0000:00:1e.0/0000:02:0b.4/microamps_requested_vmmc'
Modules linked in:
Pid: 1, comm: swapper Not tainted 2.6.39-rc5-git4-i486-1sys #1
Call Trace:
 [<c1128fef>] ? sysfs_add_one+0x83/0x95
 [<c103ef43>] warn_slowpath_common+0x7a/0x8f
 [<c1128fef>] ? sysfs_add_one+0x83/0x95
 [<c103efe3>] warn_slowpath_fmt+0x33/0x35
 [<c1128fef>] sysfs_add_one+0x83/0x95
 [<c11289ef>] sysfs_add_file_mode+0x4a/0x72
 [<c1128a2f>] sysfs_add_file+0x18/0x1c
 [<c1128aca>] sysfs_create_file+0x25/0x27
 [<c178a561>] device_create_file+0x18/0x1a
 [<c168c304>] _regulator_get+0x224/0x3b7
 [<c1081747>] ? request_threaded_irq+0xb9/0xe5
 [<c168c4ba>] regulator_get+0xf/0x11
 [<c1cbb273>] sdhci_add_host+0x580/0x6cb
 [<c15903ac>] ? pci_ioremap_bar+0x5b/0x5f
 [<c1ea0a43>] sdhci_pci_probe+0x320/0x3c9
 [<c15905db>] local_pci_probe+0x41/0x85
 [<c1591150>] pci_device_probe+0x48/0x6b
 [<c178c300>] driver_probe_device+0x9c/0x145
 [<c1590575>] ? pci_match_device+0x96/0x9c
 [<c1eb9041>] ? _cond_resched+0x8/0x47
 [<c178c3f1>] __driver_attach+0x48/0x64
 [<c178bc41>] bus_for_each_dev+0x42/0x6c
 [<c178c189>] driver_attach+0x1e/0x20
 [<c178c3a9>] ? driver_probe_device+0x145/0x145
 [<c178b656>] bus_add_driver+0xc6/0x20a
 [<c242b1f2>] ? parse_early_options+0x25/0x25
 [<c178c91b>] driver_register+0x98/0xf8
 [<c178c91b>] ? driver_register+0x98/0xf8
 [<c168db7e>] ? tty_register_driver+0x15e/0x1e2
 [<c242b1f2>] ? parse_early_options+0x25/0x25
 [<c159133d>] __pci_register_driver+0x3d/0x96
 [<c1eb84c8>] ? printk+0x1d/0x1f
 [<c242b1f2>] ? parse_early_options+0x25/0x25
 [<c24903fa>] sdhci_drv_init+0x14/0x16
 [<c1001174>] do_one_initcall+0x8c/0x142

Re: Duplicate regulator sysfs files..

[Mark Brown]

On Wed, May 04, 2011 at 04:39:11PM -0700, Linus Torvalds wrote:

>  appended is a report from Werner Landgraf <w.landgraf@ru.ru> who
> builds a kernel with pretty much everything enabled, and forwarded
> this from one of his testers.

> Comments? Why does that "supply vmmc not found, using dummy regulator"
> message show up twice? And apparently it tries to register it both
> times, which on the second try results in a WARN_ON() from sysfs..

Yes, this is a bug in the consumer driver - it's requesting the same
supply twice for the same device, causing the regulator API to try to
set the supply up twice.  It should only be requesting it once.  Ideally
the API should reject the second get() but the user is buggy anyway.