| 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: <87d0ib732z.fsf@FE-laptop>
Date: Mon, 15 Jul 2019 17:03:00 +0200
From: Gregory CLEMENT <gregory.clement@...tlin.com>
To: Auger Eric <eric.auger@...hat.com>,
Alex Williamson <alex.williamson@...hat.com>,
linux-kernel@...r.kernel.org, kvm@...r.kernel.org
Cc: Thomas Petazzoni <thomas.petazzoni@...tlin.com>,
Antoine Tenart <antoine.tenart@...tlin.com>,
Miquèl Raynal <miquel.raynal@...tlin.com>,
Maxime Chevallier <maxime.chevallier@...tlin.com>,
Nadav Haklai <nadavh@...vell.com>
Subject: Re: [PATCH] vfio: platform: reset: add support for XHCI reset hook
Hi Eric,
> Hi Gregory,
>
> On 7/11/19 4:31 PM, Gregory CLEMENT wrote:
>> The VFIO reset hook is called every time a platform device is passed
>> to a guest or removed from a guest.
>>
>> When the XHCI device is unbound from the host, the host driver
>> disables the XHCI clocks/phys/regulators so when the device is passed
>> to the guest it becomes dis-functional.
>>
>> This initial implementation uses the VFIO reset hook to enable the
>> XHCI clocks/phys on behalf of the guest.
>
> the platform reset module must also make sure there are no more DMA
> requests and interrupts that can be sent by the device anymore.
OK I will take care of it too.
>>
>> Ported from Marvell LSP code originally written by Yehuda Yitschak
>>
>> Signed-off-by: Gregory CLEMENT <gregory.clement@...tlin.com>
>> ---
>> drivers/vfio/platform/reset/Kconfig | 8 +++
>> drivers/vfio/platform/reset/Makefile | 2 +
>> .../vfio/platform/reset/vfio_platform_xhci.c | 60 +++++++++++++++++++
>> 3 files changed, 70 insertions(+)
>> create mode 100644 drivers/vfio/platform/reset/vfio_platform_xhci.c
>>
>> diff --git a/drivers/vfio/platform/reset/Kconfig b/drivers/vfio/platform/reset/Kconfig
>> index 392e3c09def0..14f620fd250d 100644
>> --- a/drivers/vfio/platform/reset/Kconfig
>> +++ b/drivers/vfio/platform/reset/Kconfig
>> @@ -22,3 +22,11 @@ config VFIO_PLATFORM_BCMFLEXRM_RESET
>> Enables the VFIO platform driver to handle reset for Broadcom FlexRM
>>
>> If you don't know what to do here, say N.
>> +
>> +config VFIO_PLATFORM_XHCI_RESET
>> + tristate "VFIO support for USB XHCI reset"
>> + depends on VFIO_PLATFORM
>> + help
>> + Enables the VFIO platform driver to handle reset for USB XHCI
>> +
>> + If you don't know what to do here, say N.
>> diff --git a/drivers/vfio/platform/reset/Makefile b/drivers/vfio/platform/reset/Makefile
>> index 7294c5ea122e..d84c4d3dc041 100644
>> --- a/drivers/vfio/platform/reset/Makefile
>> +++ b/drivers/vfio/platform/reset/Makefile
>> @@ -1,7 +1,9 @@
>> # SPDX-License-Identifier: GPL-2.0
>> vfio-platform-calxedaxgmac-y := vfio_platform_calxedaxgmac.o
>> vfio-platform-amdxgbe-y := vfio_platform_amdxgbe.o
>> +vfio-platform-xhci-y := vfio_platform_xhci.o
>>
>> obj-$(CONFIG_VFIO_PLATFORM_CALXEDAXGMAC_RESET) += vfio-platform-calxedaxgmac.o
>> obj-$(CONFIG_VFIO_PLATFORM_AMDXGBE_RESET) += vfio-platform-amdxgbe.o
>> obj-$(CONFIG_VFIO_PLATFORM_BCMFLEXRM_RESET) += vfio_platform_bcmflexrm.o
>> +obj-$(CONFIG_VFIO_PLATFORM_XHCI_RESET) += vfio-platform-xhci.o
>> diff --git a/drivers/vfio/platform/reset/vfio_platform_xhci.c b/drivers/vfio/platform/reset/vfio_platform_xhci.c
>> new file mode 100644
>> index 000000000000..7b75a04402ee
>> --- /dev/null
>> +++ b/drivers/vfio/platform/reset/vfio_platform_xhci.c
>> @@ -0,0 +1,60 @@
>> +// SPDX-License-Identifier: GPL-2.0-or-later
>> +/*
>> + * VFIO platform driver specialized for XHCI reset
>> + *
>> + * Copyright 2016 Marvell Semiconductors, Inc.
>> + *
>> + */
>> +
>> +#include <linux/clk.h>
>> +#include <linux/device.h>
>> +#include <linux/init.h>
>> +#include <linux/io.h>
>> +#include <linux/kernel.h>
> io, init, kernel should be removable (noticed init and kernel.h also are
> in other reset modules though)
OK
>> +#include <linux/module.h>
>> +#include <linux/of.h>
>> +#include <linux/phy/phy.h>
>> +#include <linux/usb/phy.h>
>> +
>> +#include "../vfio_platform_private.h"
>> +
>> +#define MAX_XHCI_CLOCKS 4
> Where does this number come from?
>
> From Documentation/devicetree/bindings/usb/usb-xhci.txt I understand
> there are max 2 clocks, "core" and "reg" (I don't have any specific
> knowledge on the device though).
Right, I guess the intent was to be future proof if there is more clocks
needed, but as we don't know, it's better to use the number of clokck we
know.
>
>> +#define MAX_XHCI_PHYS 2
> not used
Right!
>> +
>> +int vfio_platform_xhci_reset(struct vfio_platform_device *vdev)
>> +{
>> + struct device *dev = vdev->device;
>> + struct device_node *np = dev->of_node;
>> + struct usb_phy *usb_phy;
>> + struct clk *clk;
>> + int ret, i;
>> +
>> + /*
>> + * Compared to the native driver, no need to handle the
>> + * deferred case, because the resources are already
>> + * there
>> + */
>> + for (i = 0; i < MAX_XHCI_CLOCKS; i++) {
>> + clk = of_clk_get(np, i);
>> + if (!IS_ERR(clk)) {
>> + ret = clk_prepare_enable(clk);
>> + if (ret)
>> + return -ENODEV;
> return ret?
OK
>> + }
>> + }
>> +
>> + usb_phy = devm_usb_get_phy_by_phandle(dev, "usb-phy", 0);
>> + if (!IS_ERR(usb_phy)) {
>> + ret = usb_phy_init(usb_phy);
>> + if (ret)
>> + return -ENODEV;
> return ret?
OK
Thanks,
Gregory
>> + }
>
>> +
>> + return 0;
>> +}
>> +
>> +module_vfio_reset_handler("generic-xhci", vfio_platform_xhci_reset);
>> +
>> +MODULE_AUTHOR("Yehuda Yitschak");
>> +MODULE_DESCRIPTION("Reset support for XHCI vfio platform device");
>> +MODULE_LICENSE("GPL");
>>
> Thanks
>
> Eric
--
Gregory Clement, Bootlin
Embedded Linux and Kernel engineering
http://bootlin.com
Powered by blists - more mailing lists