public inbox for drm-ai-reviews@public-inbox.freedesktop.org
 help / color / mirror / Atom feed
From: Claude Code Review Bot <claude-review@example.com>
To: dri-devel-reviews@example.com
Subject: Claude review: drm/atomic: Return user readable error in atomic_ioctl
Date: Wed, 01 Apr 2026 07:55:28 +1000	[thread overview]
Message-ID: <review-patch5-20260331-atomic-v11-5-6a1df7ec5af8@intel.com> (raw)
In-Reply-To: <20260331-atomic-v11-5-6a1df7ec5af8@intel.com>

Patch Review

**`error_code_ptr` used uninitialized:**
```c
if (!arg->reserved)
    drm_dbg_atomic(dev, "memory not allocated...\n");
else
    error_code_ptr = (struct drm_mode_atomic_err_code __user *)
                     (unsigned long)arg->reserved;
```
When `arg->reserved == 0`, `error_code_ptr` is uninitialized. Later:
```c
if (ret < 0 && arg->reserved) {
    if (copy_to_user(error_code_ptr, ...))
```
The `arg->reserved` check guards the usage, so technically safe, but `error_code_ptr` is still declared without initialization. Set it to `NULL` at declaration.

**Debug message is backwards:** When `!arg->reserved` (i.e., userspace didn't provide a pointer), the code prints a debug message saying "memory not allocated for drm_atomic error reporting". This will fire on **every single atomic ioctl call** from any userspace that doesn't use this feature, which is essentially all current userspace. This is extremely noisy and should be removed.

**`memset` uses wrong sizeof:**
```c
memset(&state->error_code, 0, sizeof(*error_code_ptr));
```
This should be `sizeof(state->error_code)`. Using `sizeof(*error_code_ptr)` happens to be the same size but is semantically wrong and will break if `error_code_ptr`'s type ever diverges.

**`copy_to_user` on `-EDEADLK`:** The error reporting happens before the deadlock retry logic:
```c
out:
    if (ret < 0 && arg->reserved) {
        if (copy_to_user(error_code_ptr, &state->error_code, ...))
            ret = -EFAULT;
    }
    ...
    if (ret == -EDEADLK) {
        drm_atomic_state_clear(state);
        ret = drm_modeset_backoff(&ctx);
        if (!ret) goto retry;
    }
```
On `-EDEADLK`, this will copy a stale/meaningless error report to userspace, then retry. If the retry succeeds, userspace gets `ret=0` but has stale error data written to their buffer. Move the `copy_to_user` after the deadlock retry handling.

**Overwriting `-EFAULT` with `-EDEADLK` handling:** If `copy_to_user` fails and sets `ret = -EFAULT`, the deadlock check `if (ret == -EDEADLK)` won't match, so at least it won't loop, but the original error is lost.

---

---
Generated by Claude Code Patch Reviewer

  reply	other threads:[~2026-03-31 21:55 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-03-31  9:03 [PATCH v11 0/7] User readable error codes on atomic_ioctl failure Arun R Murthy
2026-03-31  9:03 ` [PATCH v11 1/7] drm: Define user readable error codes for atomic ioctl Arun R Murthy
2026-03-31 21:55   ` Claude review: " Claude Code Review Bot
2026-03-31  9:03 ` [PATCH v11 2/7] drm/atomic: Add error_code element in atomic_state Arun R Murthy
2026-03-31 21:55   ` Claude review: " Claude Code Review Bot
2026-03-31  9:03 ` [PATCH v11 3/7] drm/atomic: Call complete_signaling only if prepare_signaling is done Arun R Murthy
2026-03-31 21:55   ` Claude review: " Claude Code Review Bot
2026-03-31  9:03 ` [PATCH v11 4/7] drm/atomic: Allocate atomic_state at the beginning of atomic_ioctl Arun R Murthy
2026-03-31 21:55   ` Claude review: " Claude Code Review Bot
2026-03-31  9:03 ` [PATCH v11 5/7] drm/atomic: Return user readable error in atomic_ioctl Arun R Murthy
2026-03-31 21:55   ` Claude Code Review Bot [this message]
2026-03-31  9:03 ` [PATCH v11 6/7] drm/i915/display: Error codes for async flip failures Arun R Murthy
2026-03-31 21:55   ` Claude review: " Claude Code Review Bot
2026-03-31  9:03 ` [PATCH v11 7/7] drm: Introduce DRM_CAP_ATOMIC_ERROR_REPORTING Arun R Murthy
2026-03-31 21:55   ` Claude review: " Claude Code Review Bot
2026-03-31 21:55 ` Claude review: User readable error codes on atomic_ioctl failure Claude Code Review Bot
  -- strict thread matches above, loose matches on Subject: below --
2026-02-23  9:15 [PATCH v10 0/7] " Arun R Murthy
2026-02-23  9:16 ` [PATCH v10 5/7] drm/atomic: Return user readable error in atomic_ioctl Arun R Murthy
2026-02-24  0:27   ` Claude review: " Claude Code Review Bot
2026-02-10  9:03 [PATCH v9 0/7] User readable error codes on atomic_ioctl failure Arun R Murthy
2026-02-10  9:03 ` [PATCH v9 5/7] drm/atomic: Return user readable error in atomic_ioctl Arun R Murthy
2026-02-11  6:31   ` Claude review: " Claude Code Review Bot

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=review-patch5-20260331-atomic-v11-5-6a1df7ec5af8@intel.com \
    --to=claude-review@example.com \
    --cc=dri-devel-reviews@example.com \
    /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