lkml.org 
[lkml]   [2018]   [Jan]   [22]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
SubjectRe: [PATCH] ipmi: use dynamic memory for DMI driver override
From
Date
On 01/17/2018 11:06 AM, John Garry wrote:
> On 17/01/2018 16:49, Corey Minyard wrote:
>> On 01/17/2018 10:36 AM, John Garry wrote:
>>> Currently a crash can be seen if we reach the "err"
>>> label in dmi_add_platform_ipmi(), calling
>>> platform_device_put(), like here:
>>> [    7.270584]  (null): ipmi:dmi: Unable to add resources: -16
>>> [    7.330229] ------------[ cut here ]------------
>>> [    7.334889] kernel BUG at mm/slub.c:3894!
>>> [    7.338936] Internal error: Oops - BUG: 0 [#1] PREEMPT SMP
>>> [    7.344475] Modules linked in:
>>> [    7.347556] CPU: 1 PID: 1 Comm: swapper/0 Not tainted
>>> 4.15.0-rc2-00004-gbe9cb7b-dirty #114
>>> [    7.355907] Hardware name: Huawei Taishan 2280 /D05, BIOS Hisilicon
>>> D05 IT17 Nemo 2.0 RC0 11/29/2017
>>> [    7.365137] task: 00000000c211f6d3 task.stack: 00000000f276e9af
>>> [    7.371116] pstate: 60000005 (nZCv daif -PAN -UAO)
>>> [    7.375957] pc : kfree+0x194/0x1b4
>>> [    7.379389] lr : platform_device_release+0xcc/0xd8
>>> [    7.384225] sp : ffff0000092dba90
>>> [    7.387567] x29: ffff0000092dba90 x28: ffff000008a83000
>>> [    7.392933] x27: ffff0000092dbc10 x26: 00000000000000e6
>>> [    7.398297] x25: 0000000000000003 x24: ffff0000085b51e8
>>> [    7.403662] x23: 0000000000000100 x22: ffff7e0000234cc0
>>> [    7.409027] x21: ffff000008af3660 x20: ffff8017d21acc10
>>> [    7.414392] x19: ffff8017d21acc00 x18: 0000000000000002
>>> [    7.419757] x17: 0000000000000001 x16: 0000000000000008
>>> [    7.425121] x15: 0000000000000001 x14: 6666666678303d65
>>> [    7.430486] x13: 6469727265766f5f x12: 7265766972642e76
>>> [    7.435850] x11: 6564703e2d617020 x10: 6530326435373638
>>> [    7.441215] x9 : 3030303030303030 x8 : 3d76656420657361
>>> [    7.446580] x7 : ffff000008f59df8 x6 : ffff8017fbe0ea50
>>> [    7.451945] x5 : 0000000000000000 x4 : 0000000000000000
>>> [    7.457309] x3 : ffffffffffffffff x2 : 0000000000000000
>>> [    7.462674] x1 : 0fffc00000000800 x0 : ffff7e0000234ce0
>>> [    7.468039] Process swapper/0 (pid: 1, stack limit =
>>> 0x00000000f276e9af)
>>> [    7.474809] Call trace:
>>> [    7.477272]  kfree+0x194/0x1b4
>>> [    7.480351]  platform_device_release+0xcc/0xd8
>>> [    7.484837]  device_release+0x34/0x90
>>> [    7.488531]  kobject_put+0x70/0xcc
>>> [    7.491961]  put_device+0x14/0x1c
>>> [    7.495304]  platform_device_put+0x14/0x1c
>>> [    7.499439]  dmi_add_platform_ipmi+0x348/0x3ac
>>> [    7.503923]  scan_for_dmi_ipmi+0xfc/0x10c
>>> [    7.507970]  do_one_initcall+0x38/0x124
>>> [    7.511840]  kernel_init_freeable+0x188/0x228
>>> [    7.516238]  kernel_init+0x10/0x100
>>> [    7.519756]  ret_from_fork+0x10/0x18
>>> [    7.523362] Code: f94002c0 37780080 f94012c0 37000040 (d4210000)
>>> [    7.529552] ---[ end trace 11750e4787deef9e ]---
>>> [    7.534228] Kernel panic - not syncing: Attempted to kill init!
>>> exitcode=0x0000000b
>>> [    7.534228]
>>>
>>> This is because when the device is released in
>>> platform_device_release(), we try to free
>>> pdev.driver_override. This is a const string, hence
>>> the crash.
>>> Fix by using dynamic memory for pdev->driver_override.
>>>
>>> Signed-off-by: John Garry <john.garry@huawei.com>
>>
>> Oops, sorry about that.  Yes, queued for the next release and 4.14.x.
>>
>> -corey
>>
>>>
>>> diff --git a/drivers/char/ipmi/ipmi_dmi.c
>>> b/drivers/char/ipmi/ipmi_dmi.c
>>> index ab78b3b..c5112b1 100644
>>> --- a/drivers/char/ipmi/ipmi_dmi.c
>>> +++ b/drivers/char/ipmi/ipmi_dmi.c
>>> @@ -106,7 +106,10 @@ static void __init dmi_add_platform_ipmi(unsigned
>>> long base_addr,
>>>           pr_err("ipmi:dmi: Error allocation IPMI platform device\n");
>>>           return;
>>>       }
>>> -    pdev->driver_override = override;
>>> +    pdev->driver_override = kasprintf(GFP_KERNEL, "%s",
>>> +                      override);
>
> It would be simpler to use managed variant (to avoid the kfree()), but
> I don't think it's possible until the device probes.
>

Actually, I don't think you should do the kfree() in ipmi_si_intf.c. 
It's done in
platform_device_release(), and will cause a double free if done in the
IPMI code,
unless you set it to NULL.  I'm going to add a patch to remove that hunk.

-corey

> John
>
>>> +    if (!pdev->driver_override)
>>> +        goto err;
>>>         if (type == IPMI_DMI_TYPE_SSIF) {
>>>           set_prop_entry(p[pidx++], "i2c-addr", u16, base_addr);
>>> diff --git a/drivers/char/ipmi/ipmi_si_intf.c
>>> b/drivers/char/ipmi/ipmi_si_intf.c
>>> index 779869e..2a19aa8 100644
>>> --- a/drivers/char/ipmi/ipmi_si_intf.c
>>> +++ b/drivers/char/ipmi/ipmi_si_intf.c
>>> @@ -2426,6 +2426,7 @@ int ipmi_si_remove_by_dev(struct device *dev)
>>>   {
>>>       struct smi_info *e;
>>>       int rv = -ENOENT;
>>> +    struct platform_device *pdev = to_platform_device(dev);
>>>         mutex_lock(&smi_infos_lock);
>>>       list_for_each_entry(e, &smi_infos, link) {
>>> @@ -2436,6 +2437,7 @@ int ipmi_si_remove_by_dev(struct device *dev)
>>>           }
>>>       }
>>>       mutex_unlock(&smi_infos_lock);
>>> +    kfree(pdev->driver_override);
>>>         return rv;
>>>   }
>>
>>
>>
>> .
>>
>
>

\
 
 \ /
  Last update: 2018-01-22 14:46    [W:0.047 / U:1.424 seconds]
©2003-2018 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site