lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date:   Thu, 16 Jul 2020 13:58:36 +0200
From:   Marek Szyprowski <m.szyprowski@...sung.com>
To:     Minas Harutyunyan <Minas.Harutyunyan@...opsys.com>,
        "linux-usb@...r.kernel.org" <linux-usb@...r.kernel.org>
Cc:     "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        Felipe Balbi <balbi@...nel.org>,
        Bartlomiej Zolnierkiewicz <b.zolnierkie@...sung.com>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>
Subject: Re: [PATCH] usb: dwc2: Fix error path in gadget registration

Hi Minas,

On 15.07.2020 15:54, Minas Harutyunyan wrote:
> On 7/15/2020 12:42 PM, Marek Szyprowski wrote:
>> On 14.07.2020 14:32, Minas Harutyunyan wrote:
>>> On 7/14/2020 12:48 PM, Marek Szyprowski wrote:
>>>> When gadget registration fails, one should not call usb_del_gadget_udc().
>>>> Ensure this by setting gadget->udc to NULL. Also in case of a failure
>>> I was able to reproduce issue. I'm Ok with this fix.
>>>
>>>> there is no need to disable low-level hardware, so return immiedetly
>>>> instead of jumping to error_init label.
>>>>
>>> Why do you think that disable low-level hardware not required which was
>>> enabled before? Also for some platforms required to call
>>> regulator_disable() which was enabled earlier in probe function.
>>> So, I suggest to keep jump to error_init label.
>> If I keep the jump to error_init label, then there is unbalanced call to
>> dwc2_lowlevel_hw_disable(). usb_add_gadget_udc() can fail in 2 places:
>> on gadget->bind() or during udc_start(). In the first case, the HW was
>> not yet enabled, so there is no need to disable it. In the latter one,
>> the error might be returned only from the dwc2_lowlevel_hw_enable(), so
>> again there is no need to call dwc2_lowlevel_hw_disable().
>>
>> If I keep the "goto error_init;" line, I get the following errors:
>>
>> dwc2 12480000.hsotg: dwc2_check_params: Invalid parameter besl=1
>> dwc2 12480000.hsotg: dwc2_check_params: Invalid parameter
>> g_np_tx_fifo_size=1024
>> dwc2 12480000.hsotg: EPs: 16, dedicated fifos, 7808 entries in SPRAM
>> Mass Storage Function, version: 2009/09/11
>> LUN: removable file: (no medium)
>> no file given for LUN0
>> g_mass_storage 12480000.hsotg: failed to start g_mass_storage: -22
>> ------------[ cut here ]------------
>> WARNING: CPU: 3 PID: 49 at drivers/clk/clk.c:958
>> clk_core_disable+0x1e4/0x314
>> usb_device already disabled
>> Modules linked in:
>> CPU: 3 PID: 49 Comm: kworker/3:1 Not tainted
>> 5.8.0-rc5-next-20200714-00003-g105f360ba595-dirty #8758
>> Hardware name: Samsung Exynos (Flattened Device Tree)
>> Workqueue: events deferred_probe_work_func
>> [<c011184c>] (unwind_backtrace) from [<c010d250>] (show_stack+0x10/0x14)
>> [<c010d250>] (show_stack) from [<c051b8fc>] (dump_stack+0xbc/0xe8)
>> [<c051b8fc>] (dump_stack) from [<c0126ed8>] (__warn+0xf0/0x108)
>> [<c0126ed8>] (__warn) from [<c0126f64>] (warn_slowpath_fmt+0x74/0xb8)
>> [<c0126f64>] (warn_slowpath_fmt) from [<c056c338>]
>> (clk_core_disable+0x1e4/0x314)
>> [<c056c338>] (clk_core_disable) from [<c056c480>]
>> (clk_core_disable_lock+0x18/0x24)
>> [<c056c480>] (clk_core_disable_lock) from [<c073f9d4>]
>> (__dwc2_lowlevel_hw_disable+0x3c/0xa0)
>> [<c073f9d4>] (__dwc2_lowlevel_hw_disable) from [<c0740188>]
>> (dwc2_driver_probe+0x2d4/0x6ac)
>> [<c0740188>] (dwc2_driver_probe) from [<c0653640>]
>> (platform_drv_probe+0x6c/0xa4)
>> [<c0653640>] (platform_drv_probe) from [<c0650bdc>]
>> (really_probe+0x200/0x4fc)
>> [<c0650bdc>] (really_probe) from [<c0651040>]
>> (driver_probe_device+0x78/0x1fc)
>> [<c0651040>] (driver_probe_device) from [<c064ea78>]
>> (bus_for_each_drv+0x74/0xb8)
>> [<c064ea78>] (bus_for_each_drv) from [<c065093c>]
>> (__device_attach+0xd4/0x16c)
>> [<c065093c>] (__device_attach) from [<c064fa3c>]
>> (bus_probe_device+0x88/0x90)
>> [<c064fa3c>] (bus_probe_device) from [<c064ff54>]
>> (deferred_probe_work_func+0x3c/0xd0)
>> [<c064ff54>] (deferred_probe_work_func) from [<c0148a5c>]
>> (process_one_work+0x234/0x7dc)
>> [<c0148a5c>] (process_one_work) from [<c0149048>] (worker_thread+0x44/0x51c)
>> [<c0149048>] (worker_thread) from [<c015039c>] (kthread+0x158/0x1a0)
>> [<c015039c>] (kthread) from [<c0100114>] (ret_from_fork+0x14/0x20)
>> Exception stack(0xee923fb0 to 0xee923ff8)
>> ...
>> irq event stamp: 36966
>> hardirqs last  enabled at (36965): [<c02b92d8>] kfree+0x1a4/0x3f0
>> hardirqs last disabled at (36966): [<c0569edc>] clk_enable_lock+0x14/0x134
>> softirqs last  enabled at (36814): [<c010174c>] __do_softirq+0x50c/0x608
>> softirqs last disabled at (36803): [<c0130218>] irq_exit+0x168/0x16c
>> ---[ end trace f55f4b28f3080c12 ]---
>> ------------[ cut here ]------------
>> WARNING: CPU: 3 PID: 49 at drivers/clk/clk.c:817
>> clk_core_unprepare+0x33c/0x470
>> usb_device already unprepared
>> Modules linked in:
>> CPU: 3 PID: 49 Comm: kworker/3:1 Tainted: G        W
>> 5.8.0-rc5-next-20200714-00003-g105f360ba595-dirty #8758
>> Hardware name: Samsung Exynos (Flattened Device Tree)
>> Workqueue: events deferred_probe_work_func
>> [<c011184c>] (unwind_backtrace) from [<c010d250>] (show_stack+0x10/0x14)
>> [<c010d250>] (show_stack) from [<c051b8fc>] (dump_stack+0xbc/0xe8)
>> [<c051b8fc>] (dump_stack) from [<c0126ed8>] (__warn+0xf0/0x108)
>> [<c0126ed8>] (__warn) from [<c0126f64>] (warn_slowpath_fmt+0x74/0xb8)
>> [<c0126f64>] (warn_slowpath_fmt) from [<c056dcb8>]
>> (clk_core_unprepare+0x33c/0x470)
>> [<c056dcb8>] (clk_core_unprepare) from [<c056de10>]
>> (clk_unprepare+0x24/0x2c)
>> [<c056de10>] (clk_unprepare) from [<c073f9dc>]
>> (__dwc2_lowlevel_hw_disable+0x44/0xa0)
>> [<c073f9dc>] (__dwc2_lowlevel_hw_disable) from [<c0740188>]
>> (dwc2_driver_probe+0x2d4/0x6ac)
>> [<c0740188>] (dwc2_driver_probe) from [<c0653640>]
>> (platform_drv_probe+0x6c/0xa4)
>> [<c0653640>] (platform_drv_probe) from [<c0650bdc>]
>> (really_probe+0x200/0x4fc)
>> [<c0650bdc>] (really_probe) from [<c0651040>]
>> (driver_probe_device+0x78/0x1fc)
>> [<c0651040>] (driver_probe_device) from [<c064ea78>]
>> (bus_for_each_drv+0x74/0xb8)
>> [<c064ea78>] (bus_for_each_drv) from [<c065093c>]
>> (__device_attach+0xd4/0x16c)
>> [<c065093c>] (__device_attach) from [<c064fa3c>]
>> (bus_probe_device+0x88/0x90)
>> [<c064fa3c>] (bus_probe_device) from [<c064ff54>]
>> (deferred_probe_work_func+0x3c/0xd0)
>> [<c064ff54>] (deferred_probe_work_func) from [<c0148a5c>]
>> (process_one_work+0x234/0x7dc)
>> [<c0148a5c>] (process_one_work) from [<c0149048>] (worker_thread+0x44/0x51c)
>> [<c0149048>] (worker_thread) from [<c015039c>] (kthread+0x158/0x1a0)
>> [<c015039c>] (kthread) from [<c0100114>] (ret_from_fork+0x14/0x20)
>> Exception stack(0xee923fb0 to 0xee923ff8)
>> ...
>> irq event stamp: 37073
>> hardirqs last  enabled at (37081): [<c019d624>] console_unlock+0x430/0x6cc
>> hardirqs last disabled at (37108): [<c019d348>] console_unlock+0x154/0x6cc
>> softirqs last  enabled at (37124): [<c010174c>] __do_softirq+0x50c/0x608
>> softirqs last disabled at (37135): [<c0130218>] irq_exit+0x168/0x16c
>> ---[ end trace f55f4b28f3080c13 ]---
>> ------------[ cut here ]------------
>> WARNING: CPU: 3 PID: 49 at drivers/regulator/core.c:2603
>> _regulator_disable+0x180/0x1d8
>> unbalanced disables for VUOTG_3.0V
>> Modules linked in:
>> CPU: 3 PID: 49 Comm: kworker/3:1 Tainted: G        W
>> 5.8.0-rc5-next-20200714-00003-g105f360ba595-dirty #8758
>> Hardware name: Samsung Exynos (Flattened Device Tree)
>> Workqueue: events deferred_probe_work_func
>> [<c011184c>] (unwind_backtrace) from [<c010d250>] (show_stack+0x10/0x14)
>> [<c010d250>] (show_stack) from [<c051b8fc>] (dump_stack+0xbc/0xe8)
>> [<c051b8fc>] (dump_stack) from [<c0126ed8>] (__warn+0xf0/0x108)
>> [<c0126ed8>] (__warn) from [<c0126f64>] (warn_slowpath_fmt+0x74/0xb8)
>> [<c0126f64>] (warn_slowpath_fmt) from [<c0585e78>]
>> (_regulator_disable+0x180/0x1d8)
>> [<c0585e78>] (_regulator_disable) from [<c0585f04>]
>> (regulator_disable+0x34/0xe0)
>> [<c0585f04>] (regulator_disable) from [<c0587ef8>]
>> (regulator_bulk_disable+0x28/0xb0)
>> [<c0587ef8>] (regulator_bulk_disable) from [<c0740188>]
>> (dwc2_driver_probe+0x2d4/0x6ac)
>> [<c0740188>] (dwc2_driver_probe) from [<c0653640>]
>> (platform_drv_probe+0x6c/0xa4)
>> [<c0653640>] (platform_drv_probe) from [<c0650bdc>]
>> (really_probe+0x200/0x4fc)
>> [<c0650bdc>] (really_probe) from [<c0651040>]
>> (driver_probe_device+0x78/0x1fc)
>> [<c0651040>] (driver_probe_device) from [<c064ea78>]
>> (bus_for_each_drv+0x74/0xb8)
>> [<c064ea78>] (bus_for_each_drv) from [<c065093c>]
>> (__device_attach+0xd4/0x16c)
>> [<c065093c>] (__device_attach) from [<c064fa3c>]
>> (bus_probe_device+0x88/0x90)
>> [<c064fa3c>] (bus_probe_device) from [<c064ff54>]
>> (deferred_probe_work_func+0x3c/0xd0)
>> [<c064ff54>] (deferred_probe_work_func) from [<c0148a5c>]
>> (process_one_work+0x234/0x7dc)
>> [<c0148a5c>] (process_one_work) from [<c0149048>] (worker_thread+0x44/0x51c)
>> [<c0149048>] (worker_thread) from [<c015039c>] (kthread+0x158/0x1a0)
>> [<c015039c>] (kthread) from [<c0100114>] (ret_from_fork+0x14/0x20)
>> Exception stack(0xee923fb0 to 0xee923ff8)
>> ...
>> irq event stamp: 37243
>> hardirqs last  enabled at (37251): [<c019d624>] console_unlock+0x430/0x6cc
>> hardirqs last disabled at (37274): [<c0ac0790>] __schedule+0xd8/0x860
>> softirqs last  enabled at (37270): [<c010174c>] __do_softirq+0x50c/0x608
>> softirqs last disabled at (37291): [<c0130218>] irq_exit+0x168/0x16c
>> ---[ end trace f55f4b28f3080c14 ]---
>> Failed to disable vusb_a: -5
>> dwc2: probe of 12480000.hsotg failed with error -22
>> 8<--- cut here ---
>> Unable to handle kernel NULL pointer dereference at virtual address 00000004
>> pgd = (ptrval)
>> [00000004] *pgd=00000000
>> Internal error: Oops: 5 [#1] PREEMPT SMP ARM
>> Modules linked in:
>> CPU: 3 PID: 87 Comm: kworker/3:2 Tainted: G        W
>> 5.8.0-rc5-next-20200714-00003-g105f360ba595-dirty #8758
>> Hardware name: Samsung Exynos (Flattened Device Tree)
>> Workqueue:  0x0 (rcu_gp)
>> PC is at process_one_work+0x44/0x7dc
>> LR is at 0xedc226d0
>> ...
>> Process kworker/3:2 (pid: 87, stack limit = 0x(ptrval))
>> Stack: (0xeda03f00 to 0xeda04000)
>> ...
>> [<c014886c>] (process_one_work) from [<c0149048>] (worker_thread+0x44/0x51c)
>> [<c0149048>] (worker_thread) from [<c015039c>] (kthread+0x158/0x1a0)
>> [<c015039c>] (kthread) from [<c0100114>] (ret_from_fork+0x14/0x20)
>> Exception stack(0xeda03fb0 to 0xeda03ff8)
>> ...
>> ---[ end trace f55f4b28f3080c15 ]---
>> note: kworker/3:2[87] exited with preempt_count 1
>>
>> Best regards
>>
> I guess you build driver in (hsotg->dr_mode == USB_DR_MODE_PERIPHERAL)
> and this is why call of dwc2_lowlevel_hw_disable() at 'error:' label
> create issue. Could you please test on your setup follow implementation:
> ...
> error:
> 	if (hsotg->dr_mode != USB_DR_MODE_PERIPHERAL)
> 		dwc2_lowlevel_hw_disable(hsotg);
> 	return retval;
> }
> ...

Right, this will work. However, I would also change the order of 
operations - disabling usb33d regulator should be done after disabling 
lowlevel hardware to properly pair with the enable operations.

While playing with this I've also found an issue in UDC core:

https://lore.kernel.org/linux-usb/20200716115525.27519-1-m.szyprowski@samsung.com/T/#u

Best regards
-- 
Marek Szyprowski, PhD
Samsung R&D Institute Poland

Powered by blists - more mailing lists