cmd: usb: Prevent reset in usb tree/info command
PATCH 1/1 cmd: usb: Prevent reset in usb tree/info command
When DISTRO_DEFAULTS is not set, the default environment has bootcmd=bootflow and this will cause a UCLASS_BOOTDEV device to be added as sibling of those UCLASS_BLK devices in boot_targets, until boot succeeds from some device. If none succeeds, and usb is in boot_targets, and an usb storage device is plugged to some usb port at boot time, its UCLASS_MASS_STORAGE device will have a UCLASS_BOOTDEV device as child, besides a UCLASS_BLK child.
If once the boot fails the user enters at the U-Boot shell prompt:
usb info
or
usb tree
The code in cmd/usb.c will eventually recurse into the UCLASS_BOOTDEV and pass a null pointer to usb_device (because it has no parent_priv_). This causes a reset.
Fix it (twice) by checking for null parent_priv_ and adding UCLASS_BOOTDEV to the list of ignored class ids before the recursive call.
This prevents the current particular problem with UCLASS_BOOTDEV, even in case it ever gets some parent_priv_ struct which is not an usb_device, despite being the child of a usb_device->dev. And it also prevents possible future problems if other children are added to usb devices that don't have parent_priv_ because they are not part of the usb tree, just abstractions of functionality (like UCLASS_BLK and UCLASS_BOOTDEV are now).
---
v2: added UCLASS_BOOTDEV check (discussion of v1 dried up without much
evident consensus, so hopefully Simon Glass likes it better now)
[ https://patchwork.ozlabs.org/project/uboot/patch/[email protected]/ ]
---
Apologies to the people in Cc: for resending this. I had a typo in the email list address.
--- cmd/usb.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-)
cmd/usb.c +6 -2
@@ -421,7 +421,9 @@ static void usb_show_tree_graph(struct usb_device *dev, char *pre)* Ignore emulators and block child devices, we only want * real devices */ -if ((device_get_uclass_id (child) !=UCLASS_USB_EMUL ) && +if (udev && + (device_get_uclass_id (child) !=UCLASS_BOOTDEV ) && + (device_get_uclass_id (child) !=UCLASS_USB_EMUL ) && (device_get_uclass_id (child) !=UCLASS_BLK )) {usb_show_tree_graph (udev, pre); pre[index] =0 ;
@@ -604,10 +606,12 @@ static void usb_show_info(struct usb_device *udev)child ; device_find_next_child(&child)) {if (device_active (child) && + (device_get_uclass_id (child) !=UCLASS_BOOTDEV ) && (device_get_uclass_id (child) !=UCLASS_USB_EMUL ) && (device_get_uclass_id (child) !=UCLASS_BLK )) { udev =dev_get_parent_priv (child); -usb_show_info (udev); +if (udev) +usb_show_info (udev); } } }
Signature · 1 line
2.20.1
Quoted 35 lines · Xavier Drudis Ferran
When
DISTRO_DEFAULTSis not set, the default environment has bootcmd=bootflow and this will cause aUCLASS_BOOTDEVdevice to be added as sibling of thoseUCLASS_BLKdevices inboot_targets, until boot succeeds from some device. If none succeeds, and usb is inboot_targets, and an usb storage device is plugged to some usb port at boot time, itsUCLASS_MASS_STORAGEdevice will have aUCLASS_BOOTDEVdevice as child, besides aUCLASS_BLKchild.If once the boot fails the user enters at the U-Boot shell prompt:
usb info
or
usb tree
The code in
cmd/usb.cwill eventually recurse into theUCLASS_BOOTDEVand pass a null pointer tousb_device(because it has noparent_priv_). This causes a reset.Fix it (twice) by checking for null
parent_priv_and addingUCLASS_BOOTDEVto the list of ignored class ids before the recursive call.This prevents the current particular problem with
UCLASS_BOOTDEV, even in case it ever gets someparent_priv_struct which is not anusb_device, despite being the child of ausb_device->dev. And it also prevents possible future problems if other children are added to usb devices that don't haveparent_priv_because they are not part of the usb tree, just abstractions of functionality (likeUCLASS_BLKandUCLASS_BOOTDEVare now).Signed-off-by Xavier Drudis Ferran <[email protected]>---
Quoted 7 lines
v2: added
UCLASS_BOOTDEVcheck (discussion of v1 dried up without much
evident consensus, so hopefully Simon Glass likes it better now)
[ https://patchwork.ozlabs.org/project/uboot/patch/[email protected]/ ]
---Apologies to the people in Cc: for resending this. I had a typo in the email list address.
git send-email -v2 would help
Also you can add a CC list into the commit message below --- , then git send-email will automatically pick the CC list up. To get a list of people to CC , use get-maintainer script.