Re: [GIT PULL v1 15/17] KVM: s390: Add map/unmap ioctl and clean mappings post-guest

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Low] The newly added `!addr` check in `kvm_s390_adapter_map` and `kvm_s390_adapter_unmap` incorrectly rejects GPA 0.
--

commit a3f46a9fac83fcdf67e052053b6923a848a2c796
Author: Douglas Freimuth <freimuth@linux.ibm.com>

KVM: s390: Add map/unmap ioctl and clean mappings post-guest

This commit adds map and unmap ioctls for s390 adapter set indicator pages.
It introduces lifecycle tracking for mapped pages using pin_user_pages_remote
and adds statistics counters to track map and unmap requests.

Message-ID: <20260604192755.203143-2-freimuth@linux.ibm.com>

> diff --git a/arch/s390/kvm/interrupt.c b/arch/s390/kvm/interrupt.c
> --- a/arch/s390/kvm/interrupt.c
> +++ b/arch/s390/kvm/interrupt.c
[ ... ]
> +static int kvm_s390_adapter_map(struct kvm *kvm, unsigned int id, __u64 addr)
> +{
> +	struct s390_io_adapter *adapter = get_io_adapter(kvm, id);
> +	struct s390_map_info *map;
> +	unsigned long flags;
> +	__u64 host_addr;
> +	int ret, idx;
> +
> +	if (!adapter || !addr)
> +		return -EINVAL;

[Severity: Low]
Does this address check incorrectly reject mapping guest physical address 0?

The ioctl argument addr represents a guest physical address. In the s390
architecture, guest physical address 0 points to the Prefix Save Area, which
is a valid memory location. Treating an address of 0 as an invalid mapping
request rejects a legitimate architectural address.

> +
> +	map = kzalloc_obj(*map, GFP_KERNEL_ACCOUNT);
[ ... ]
> +static int kvm_s390_adapter_unmap(struct kvm *kvm, unsigned int id, __u64 addr)
> +{
> +	struct s390_io_adapter *adapter = get_io_adapter(kvm, id);
> +	struct s390_map_info *map, *tmp, *map_to_free;
> +	struct page *map_page_to_put = NULL;
> +	u64 map_addr_to_mark = 0;
> +	unsigned long flags;
> +	int found = 0, idx;
> +
> +	if (!adapter || !addr)
> +		return -EINVAL;

[Severity: Low]
Similarly here, should this unmap function allow guest physical address 0?

> +
> +	spin_lock_irqsave(&adapter->maps_lock, flags);

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260615124243.187614-1-imbrenda@linux.ibm.com?part=15