From: Stephen Hemminger <stephen@networkplumber.org>
To: Tianhao Zhang <zhang.tianhao2@zte.com.cn>
Cc: wang.junlong1@zte.com.cn, yang.yonggang@zte.com.cn, dev@dpdk.org
Subject: Re: [PATCH v1 0/2] fix bug
Date: Tue, 13 Jan 2026 07:32:58 -0800 [thread overview]
Message-ID: <20260113073258.2d67ed02@phoenix.local> (raw)
In-Reply-To: <20251222102535.401064-1-zhang.tianhao2@zte.com.cn>
On Mon, 22 Dec 2025 18:25:33 +0800
Tianhao Zhang <zhang.tianhao2@zte.com.cn> wrote:
> V1:
> - Add pcie bar channel check after getting dtb space left.
> - Delete np soft resource uninstallation from NP online uninit process.
>
> Tianhao Zhang (2):
> net/zxdh: add pcie bar channel check
> net/zxdh: delete np soft resource uninstallation
>
> drivers/net/zxdh/zxdh_np.c | 11 +++++++----
> drivers/net/zxdh/zxdh_np.h | 2 ++
> 2 files changed, 9 insertions(+), 4 deletions(-)
>
I started using AI to review code in next-net.
It spotted some issues in these patches, please revise and resubmit.
Now let me read the two patches:---
## DPDK Patch Review: ZXDH PMD Patches
### Patch 1: `net/zxdh: add PCIe BAR channel check`
#### Commit Message
| Check | Status | Notes |
|-------|--------|-------|
| Subject ≤60 chars | ✓ Pass | 35 characters |
| Correct prefix | ✓ Pass | `net/zxdh:` is correct |
| Lowercase after colon | ✓ Pass | |
| Imperative mood | ✓ Pass | "add" |
| No trailing period | ✓ Pass | |
| Body ≤75 chars/line | ✓ Pass | |
| Body doesn't start with "It" | ✓ Pass | |
| Signed-off-by present | ✓ Pass | Real name and valid email |
#### Code Review
**Style Issues (Warning)**:
1. **Mixed changes**: The patch combines alignment/whitespace fixes with functional changes. The parameter alignment changes in `zxdh_np_dtb_queue_unused_item_num_get()` could be a separate cleanup patch. This makes the functional change harder to review.
2. **Log message style**: The error message `"pcie bar abnormal."` has a trailing period. While not strictly forbidden, DPDK log messages typically omit trailing periods for consistency.
**Code Quality (Info)**:
```c
if ((*p_item_num & ZXDH_DTB_SPACE_LEFT_MASK) == ZXDH_DTB_SPACE_LEFT_MASK) {
```
- The comparison is explicit (good) ✓
- Uses named constant (good) ✓
- Consider: what condition does `0x3F` represent? A brief comment explaining why this specific bit pattern indicates a BAR abnormality would aid maintainability.
**Commit Message Content (Warning)**:
The body text:
> After obtaining the DTB element register;
> check if the PCIe link is abnormal.
The semicolon makes this read awkwardly as two sentence fragments. Suggested rewrite:
> Add a check after obtaining the DTB element register to detect
> abnormal PCIe link conditions.
#### Verdict: **Acceptable with minor suggestions**
---
### Patch 2: `net/zxdh: optimize the procedure for NP online uninit`
#### Commit Message
| Check | Status | Notes |
|-------|--------|-------|
| Subject ≤60 chars | ✓ Pass | 52 characters |
| Correct prefix | ✓ Pass | `net/zxdh:` |
| Lowercase after colon | ✓ Pass | |
| Imperative mood | ✓ Pass | "optimize" |
| No trailing period | ✓ Pass | |
| Body ≤75 chars/line | ✓ Pass | All lines under 75 |
| Body doesn't start with "It" | ✓ Pass | Starts with "Remove" |
| Signed-off-by present | ✓ Pass | |
#### Issues Identified
**Warning - Misleading subject line**:
The subject says "optimize" which implies performance improvement, but the body describes removing code to "avoid duplicate resource release." This is actually a **bug fix** (preventing double-free or duplicate cleanup), not an optimization.
Suggested subject alternatives:
- `net/zxdh: fix duplicate resource release in NP online uninit`
- `net/zxdh: remove redundant resource uninstall from online uninit`
**Warning - Potentially missing `Fixes:` tag**:
If this corrects a bug where resources were being released twice, a `Fixes:` tag should reference the commit that introduced the duplicate release. The AGENTS.md guidelines state:
> `Fixes:` tag present for bug fixes with 12-char SHA and exact subject
**Info - Incomplete context**:
The commit message says the uninstallation "should be called separately within NP uninit" but doesn't specify where that happens. A reference to the function or location where `zxdh_np_soft_res_uninstall()` is properly called would help reviewers verify correctness.
#### Code Review
The code change itself is clean - a straightforward removal of 4 lines. No style issues.
#### Verdict: **Needs revision**
---
## Summary
| Patch | Status | Required Actions |
|-------|--------|------------------|
| 0007 (BAR check) | **Acceptable** | Consider separating alignment fixes; improve commit message clarity |
| 0008 (uninit optimize) | **Needs revision** | Fix subject line ("fix" not "optimize"); add `Fixes:` tag if applicable; clarify where resource uninstall is now called |
prev parent reply other threads:[~2026-01-13 15:33 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-12-09 2:16 [PATCH v1] net/zxdh: add pcie bar channel check Tianhao Zhang
2025-12-22 10:25 ` [PATCH v1 0/2] fix bug Tianhao Zhang
2025-12-22 10:25 ` [PATCH v1 1/2] net/zxdh: add pcie bar channel check Tianhao Zhang
2025-12-22 18:37 ` Stephen Hemminger
2025-12-30 8:11 ` [PATCH v2 0/2] fix npsdk bug Tianhao Zhang
2025-12-30 8:11 ` [PATCH v2 1/2] net/zxdh: add pcie bar channel check Tianhao Zhang
2025-12-30 18:29 ` Stephen Hemminger
2026-02-03 7:17 ` [PATCH v3 0/2] net/zxdh: fix npsdk bug Tianhao Zhang
2026-02-03 7:17 ` [PATCH v3 1/2] net/zxdh: add pcie bar channel check Tianhao Zhang
2026-02-03 14:35 ` Stephen Hemminger
2026-02-03 7:17 ` [PATCH v3 2/2] net/zxdh: optimize the procedure for NP online uninit Tianhao Zhang
2026-02-03 14:35 ` Stephen Hemminger
2026-02-05 22:11 ` [PATCH v3 0/2] net/zxdh: fix npsdk bug Stephen Hemminger
2025-12-30 8:11 ` [PATCH v2 2/2] net/zxdh: optimize the procedure for NP online uninit Tianhao Zhang
2025-12-22 10:25 ` [PATCH v1 2/2] net/zxdh: delete np soft resource uninstallation Tianhao Zhang
2025-12-22 18:38 ` Stephen Hemminger
2026-01-13 15:32 ` Stephen Hemminger [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20260113073258.2d67ed02@phoenix.local \
--to=stephen@networkplumber.org \
--cc=dev@dpdk.org \
--cc=wang.junlong1@zte.com.cn \
--cc=yang.yonggang@zte.com.cn \
--cc=zhang.tianhao2@zte.com.cn \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox