Re: [PATCH v2 12/12] gpu: nova-core: mm: Add PRAMIN aperture self-tests

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Low] The `io_read!` macro invocation uses a runtime loop variable `i` with the `[build: i]` syntax, which relies on the optimizer to eliminate the bounds check and may cause a build failure.
--

--- Patch [12]: [PATCH v2 12/12] gpu: nova-core: mm: Add PRAMIN aperture self-tests ---
commit e3c26c738212001e40a3ab7fef15d2c8cf5421b5
Author: Joel Fernandes <joelagnelf@nvidia.com>

gpu: nova-core: mm: Add PRAMIN aperture self-tests

This commit adds self-tests for the PRAMIN aperture mechanism to verify
correct operation, alignment requirements, and corner cases during GPU probe.
The tests are default disabled and gated behind CONFIG_NOVA_CORE_SELFTESTS.

> diff --git a/drivers/gpu/nova-core/mm/pramin.rs b/drivers/gpu/nova-core/mm/pramin.rs
> --- a/drivers/gpu/nova-core/mm/pramin.rs
> +++ b/drivers/gpu/nova-core/mm/pramin.rs

[ ... ]

> +    fn test_u32_as_bytes(
> +        dev: &device::Device<device::Bound>,
> +        pramin: &mut Pramin<'_>,
> +        base: VramAddress,
> +    ) -> Result {
> +        let addr = base + 0x10;
> +        let val: u32 = 0xDEADBEEF;
> +        pramin.window_at::<u32>(addr)?.view().write_val(val);
> +
> +        let window = pramin.window_at::<[u8; 4]>(addr)?;
> +        for (i, &expected) in val.to_le_bytes().iter().enumerate() {
> +            selftest_assert_eq!(dev, io_read!(window.view(), [build: i]), expected);

[Severity: Low]
Will this cause a link-time build failure if the optimizer does not unroll
the loop?

The io_read! macro invocation in test_u32_as_bytes uses the runtime loop
variable i with the [build: i] syntax. This relies on LLVM optimization
heuristics to unroll the loop and resolve the value statically at build
time.

If the compiler fails to unroll this loop, the code will retain a call to
rust_build_error, which results in an undefined symbol linker error unless
CONFIG_RUST_BUILD_ASSERT_ALLOW is explicitly enabled.

Should this be rewritten to manually unroll the four reads, or use a
runtime index rather than [build: i]?

> +        }
> +        Ok(())
> +    }

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260810-pramin-split-v2-0-65a00b3c7309@nvidia.com?part=12