|WARNING| net/idpf: add clock read support

[dpdklab at iol.unh.edu]

Test-Label: ai-code-review
Test-Status: WARNING
https://dpdk.org/patch/166952

_AI Code Review_

=== Patch Review: series.patch (via Claude) ===
Review date: 2026-07-28

# DPDK Patch Review

## Summary
The patch adds a `read_clock` operation to the idpf driver by implementing `idpf_dev_read_clock()`. The implementation is straightforward but has one significant issue and one documentation gap.

---

## Errors

### 1. Missing error check on `idpf_ptp_read_src_clk_reg()`

**Issue:** The return value from `idpf_ptp_read_src_clk_reg()` is propagated to the caller, but there is no validation that `clock` was successfully populated. If the function returns an error code, `*clock` may contain an uninitialized or stale value.

**Why it matters:** Callers of `read_clock` will use the `clock` value for scheduling transmit times. Using garbage data from a failed read could cause packets to be scheduled incorrectly or dropped.

**Fix:**
```c
static int
idpf_dev_read_clock(struct rte_eth_dev *dev, uint64_t *clock)
{
	struct idpf_vport *vport = dev->data->dev_private;
	struct idpf_adapter *adapter = vport->adapter;
	int ret;

	ret = idpf_ptp_read_src_clk_reg(adapter, clock);
	if (ret != 0)
		*clock = 0;  /* Ensure deterministic output on error */

	return ret;
}
```

Alternatively, if the existing `idpf_ptp_read_src_clk_reg()` implementation already guarantees that `*clock` is set to zero (or another safe value) on error, this is acceptable -- but that contract should be verified by reviewing that function's implementation.

---

## Warnings

### 1. Missing release notes entry

**Issue:** Adding a new ethdev operation (`read_clock`) is a user-visible feature and should be documented in the release notes.

**Fix:** Add an entry to `doc/guides/rel_notes/release_26_03.rst` (or the appropriate current release notes file) under the idpf driver section:

```rst
* **Updated idpf driver.**

  * Added support for ``read_clock`` ethdev operation to retrieve current
    hardware time in nanoseconds for transmit time scheduling.
```

---

## Pass

- Code style conforms to DPDK formatting standards (indentation, braces, naming)
- No forbidden tokens detected
- Function signature matches the `eth_dev_ops.read_clock` prototype
- No use-after-free, resource leaks, or race conditions introduced
- No deprecated API usage
- Appropriate use of `dev->data->dev_private` pattern

---

## Final Assessment

The patch is nearly ready but requires:
1. **Error handling** around the clock read to ensure deterministic behavior on failure (Error severity)
2. **Release notes** documenting the new feature (Warning severity)