| 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
| ||
|
Message-ID: <54BF8C2A.6070800@free-electrons.com>
Date: Wed, 21 Jan 2015 12:23:22 +0100
From: Gregory CLEMENT <gregory.clement@...e-electrons.com>
To: Thierry Reding <treding@...dia.com>
CC: Paul Walmsley <paul@...an.com>, Tejun Heo <tj@...nel.org>,
hdegoede@...hat.com, pwalmsley@...dia.com,
linux-tegra@...r.kernel.org, linux-arm-kernel@...ts.infradead.org,
linux-kernel@...r.kernel.org, linux-next@...r.kernel.org,
linux-ide@...r.kernel.org
Subject: Re: next-20150120 broken on Tegra by "ata: libahci: Allow using multiple
regulators"
Hi Thierry,
On 21/01/2015 11:50, Thierry Reding wrote:
> On Wed, Jan 21, 2015 at 12:03:41AM +0000, Paul Walmsley wrote:
>> (linux-ide@ added)
>>
>> Commit c7d7ddee7e24eedde6149eefbcfbfbc7125b9ff0 ("ata: libahci: Allow
>> using multiple regulators") in Linux-next 20150120 causes a panic
>> during boot on multiple Tegra boards:
>>
>> http://nvt.pwsan.com/experimental/linux-next/testlogs/test_next-20150120/20150120001539/boot/tegra30-beaver/tegra30-beaver/tegra_defconfig_log.txt
>> http://nvt.pwsan.com/experimental/linux-next/testlogs/test_next-20150120/20150120001539/boot/tegra114-dalmore-a04/tegra114-dalmore/tegra_defconfig_log.txt
>> http://nvt.pwsan.com/experimental/linux-next/testlogs/test_next-20150120/20150120001539/boot/tegra124-jetson-tk1/tegra124-jetson-tk1/tegra_defconfig_log.txt
>>
>> Taking the latter boot log as an example, the 6b6b6b9f value that the
>> kernel is trying to dereference is being passed into _regulator_put() as a
>> struct regulator *.
>>
>>
>> - Paul
>>
>
> Here's the devres log of what's happening:
>
> [ 4.949689] tegra-ahci 70027000.sata: DEVRES ADD ee1e2a00 devm_kzalloc_release (16 bytes)
> [ 4.957886] tegra-ahci 70027000.sata: DEVRES ADD ee1e29c0 devm_pinctrl_release (4 bytes)
> [ 4.965984] tegra-ahci 70027000.sata: DEVRES REM ee1e29c0 devm_pinctrl_release (4 bytes)
> [ 4.974079] tegra-ahci 70027000.sata: DEVRES REM ee1e2a00 devm_kzalloc_release (16 bytes)
> [ 4.982298] tegra-ahci 70027000.sata: DEVRES ADD ee1e29c0 grp< (0 bytes)
> [ 4.989006] tegra-ahci 70027000.sata: DEVRES ADD ee1dae00 ahci_platform_put_resources (96 bytes)
> [ 4.997784] tegra-ahci 70027000.sata: DEVRES ADD ee1e2980 devm_region_release (12 bytes)
> [ 5.005879] tegra-ahci 70027000.sata: DEVRES ADD ee1e2900 devm_ioremap_release (4 bytes)
> [ 5.014005] tegra-ahci 70027000.sata: DEVRES ADD ee1e28c0 devm_kzalloc_release (4 bytes)
> [ 5.022096] tegra-ahci 70027000.sata: DEVRES ADD ee1e2880 devm_kzalloc_release (4 bytes)
> [ 5.030209] tegra-ahci 70027000.sata: DEVRES ADD ee1e2840 devm_phy_release (4 bytes)
> [ 5.037952] tegra-ahci 70027000.sata: DEVRES REM ee1e29c0 grp< (0 bytes)
> [ 5.044660] tegra-ahci 70027000.sata: DEVRES ADD ee1dad80 devm_kzalloc_release (84 bytes)
> [ 5.052845] tegra-ahci 70027000.sata: DEVRES ADD ee1e29c0 devm_region_release (12 bytes)
> [ 5.060938] tegra-ahci 70027000.sata: DEVRES ADD ee1e2f80 devm_ioremap_release (4 bytes)
> [ 5.069039] tegra-ahci 70027000.sata: DEVRES ADD ee1e2f40 devm_reset_control_release (4 bytes)
> [ 5.077647] tegra-ahci 70027000.sata: DEVRES ADD ee1e2ec0 devm_reset_control_release (4 bytes)
> [ 5.086271] tegra-ahci 70027000.sata: DEVRES ADD ee1e2e40 devm_reset_control_release (4 bytes)
> [ 5.094890] tegra-ahci 70027000.sata: DEVRES ADD ee1e2dc0 devm_clk_release (4 bytes)
> [ 5.102659] tegra-ahci 70027000.sata: Failed to get supply 'avdd': -517
> [ 5.109276] tegra-ahci 70027000.sata: Failed to get regulators
> [ 5.115103] tegra-ahci 70027000.sata: DEVRES REL ee1e2dc0 devm_clk_release (4 bytes)
> [ 5.122850] tegra-ahci 70027000.sata: DEVRES REL ee1e2e40 devm_reset_control_release (4 bytes)
> [ 5.131462] tegra-ahci 70027000.sata: DEVRES REL ee1e2ec0 devm_reset_control_release (4 bytes)
> [ 5.140073] tegra-ahci 70027000.sata: DEVRES REL ee1e2f40 devm_reset_control_release (4 bytes)
> [ 5.148672] tegra-ahci 70027000.sata: DEVRES REL ee1e2f80 devm_ioremap_release (4 bytes)
> [ 5.156764] tegra-ahci 70027000.sata: DEVRES REL ee1e29c0 devm_region_release (12 bytes)
> [ 5.164856] tegra-ahci 70027000.sata: DEVRES REL ee1dad80 devm_kzalloc_release (84 bytes)
> [ 5.173034] tegra-ahci 70027000.sata: DEVRES REL ee1e2840 devm_phy_release (4 bytes)
> [ 5.180778] tegra-ahci 70027000.sata: DEVRES REL ee1e2880 devm_kzalloc_release (4 bytes)
> [ 5.188868] tegra-ahci 70027000.sata: DEVRES REL ee1e28c0 devm_kzalloc_release (4 bytes)
> [ 5.196948] tegra-ahci 70027000.sata: DEVRES REL ee1e2900 devm_ioremap_release (4 bytes)
> [ 5.205037] tegra-ahci 70027000.sata: DEVRES REL ee1e2980 devm_region_release (12 bytes)
> [ 5.213129] tegra-ahci 70027000.sata: DEVRES REL ee1dae00 ahci_platform_put_resources (96 bytes)
>
> What's happening here is that ahci_platform_put_resources() is added to
> the devres list before the devm_kzalloc() that allocates the target_pwrs
> array, which causes the target_pwrs array to be freed before devres gets
> to call ahci_platform_put_resources().
>
> Mixing managed and non-managed resources this way doesn't work, so I had
> to apply the attached patch to fix this.
>
> Tejun, preferably the attached patch should be squashed into commit
> c7d7ddee7e24 ("ata: libahci: Allow using multiple regulators") to avoid
> breaking bisectability. If you don't want to rewrite history, let me
> know and I can turn it into a proper patch.
Thanks having took care of it. I tested your patch and it didn't introduce
any regression on the board I tested.
Gregory
--
Gregory Clement, Free Electrons
Kernel, drivers, real-time and embedded Linux
development, consulting, training and support.
http://free-electrons.com
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists