Re: [edk2-devel] [PATCH] NetworkPkg/HttpBootDxe: Correctly uninstall HttpBootCallbackProtocol

["Michael Brown" <mcb30@ipxe.org>]

On 06/04/2024 16:53, Mike Beaton wrote:
> The existing uninstall call was passing the wrong handle (parent object,
> not the correct child object) and additionally passing the address
> of a pointer to the interface to be removed rather than the pointer
> itself, so always failed with EFI_NOT_FOUND. After altering these, we
> add an ASSERT which confirms that the modified uninstall is succeeding.
> 
> Cc: Maciej Rabeda <maciej.rabeda@linux.intel.com>
> Cc: Jiaxin Wu <jiaxin.wu@intel.com>
> Cc: Siyuan Fu <siyuan.fu@intel.com>
> Signed-off-by: Mike Beaton <mjsbeaton@gmail.com>
> ---
>   NetworkPkg/HttpBootDxe/HttpBootImpl.c | 21 ++++++++++++++++-----
>   1 file changed, 16 insertions(+), 5 deletions(-)
> 
> diff --git a/NetworkPkg/HttpBootDxe/HttpBootImpl.c b/NetworkPkg/HttpBootDxe/HttpBootImpl.c
> index b4c61925b9..100b721ad4 100644
> --- a/NetworkPkg/HttpBootDxe/HttpBootImpl.c
> +++ b/NetworkPkg/HttpBootDxe/HttpBootImpl.c
> @@ -77,12 +77,23 @@ HttpBootUninstallCallback (
>     IN HTTP_BOOT_PRIVATE_DATA  *Private
>     )
>   {
> +  EFI_STATUS  Status;
> +  EFI_HANDLE  ControllerHandle;
> +
>     if (Private->HttpBootCallback == &Private->LoadFileCallback) {
> -    gBS->UninstallProtocolInterface (
> -           Private->Controller,
> -           &gEfiHttpBootCallbackProtocolGuid,
> -           &Private->HttpBootCallback
> -           );
> +    if (!Private->UsingIpv6) {
> +      ControllerHandle = Private->Ip4Nic->Controller;
> +    } else {
> +      ControllerHandle = Private->Ip6Nic->Controller;
> +    }
> +
> +    Status = gBS->UninstallProtocolInterface (
> +                    ControllerHandle,
> +                    &gEfiHttpBootCallbackProtocolGuid,
> +                    Private->HttpBootCallback
> +                    );
> +    ASSERT_EFI_ERROR (Status);

This assertion is not necessarily safe.  Uninstallation of protocol 
interfaces is allowed to fail under the UEFI model.  (This is arguably 
insane, but that's a separate discussion.)  This could therefore 
potentially result in a perfectly valid sequence of events leading to an 
ASSERT() failure.

Thanks,

Michael



-=-=-=-=-=-=-=-=-=-=-=-
Groups.io Links: You receive all messages sent to this group.
View/Reply Online (#117469): https://edk2.groups.io/g/devel/message/117469
Mute This Topic: https://groups.io/mt/105368366/7686176
Group Owner: devel+owner@edk2.groups.io
Unsubscribe: https://edk2.groups.io/g/devel/unsub [rebecca@openfw.io]
-=-=-=-=-=-=-=-=-=-=-=-