|SUCCESS| net/bnxt/tf_ulp: use 64-bit shift in burst size calculation

[dpdklab at iol.unh.edu]

Test-Label: ai-code-review
Test-Status: SUCCESS
https://dpdk.org/patch/169574

_AI Code Review_

=== Patch Review: series.patch (via Claude) ===
Review date: 2026-09-15

# DPDK Patch Review

## Patch Summary
The patch fixes a potential left-shift undefined behavior in the bnxt PMD meter code by changing `1 << n` to `1ULL << n` when calculating burst size parameters.

---

## Findings

### **Errors**

None.

### **Warnings**

None.

### **Info**

1. **Second shift operand also needs 64-bit literal**

   The fix correctly changes the first `1 << (e - 8)` to `1ULL << (e - 8)`, but the second shift `1 << 7` should also be changed to `1ULL << 7` for consistency and to ensure the entire expression is evaluated in 64-bit arithmetic.

   The current fix:
   ```c
   m = xbs / (1ULL << (e - 8)) - (1 << 7);
   ```

   More consistent form:
   ```c
   m = xbs / (1ULL << (e - 8)) - (1ULL << 7);
   ```

   While `1 << 7` is safe (shift count is 7, well below 32), using `1ULL` consistently makes the intent clearer that this is 64-bit arithmetic, since `xbs` is `int64_t` and the division result will be 64-bit.

---

## Summary

The patch correctly addresses the primary issue (undefined behavior when `e - 8 >= 32`). The suggested change to use `1ULL << 7` is a consistency improvement, not a correctness bug, since the shift count is constant and safe. The patch is acceptable as-is, but using `1ULL` for both shifts would be more consistent.