# Re: \[PATCH v2\] net: 8390: pcnet\_cs: release PCMCIA window on setup\_shmem\_window() error

- From: Simon Horman \<horms@kernel.org\>
- Date: 2026-09-13 14:32:23 
- List: netdev
- Page: https://ratatoskr.run/netdev/2026/09/17568952
- Thread: https://ratatoskr.run/netdev/2026/07/17350130/t.md

```text
On Thu, Sep 10, 2026 at 07:14:00PM -0400, Myeonghun Pak wrote:
> setup_shmem_window() acquires a PCMCIA memory window using
> pcmcia_request_window(). If pcmcia_map_mem_page() or the subsequent
> ioremap() fails, the function returns without releasing the requested
> window.
> 
> pcnet_config() treats shared-memory setup failure as non-fatal and falls
> back to setup_dma_config(). Probe can therefore continue while socket
> window 3 and its reserved iomem range remain unnecessarily held for the
> rest of the bound lifetime of the device. pcmcia_disable_device()
> eventually releases the window during teardown.
> 
> Route error paths after a successful request through a new release label
> that calls pcmcia_release_window(). Fold the existing buffer-verification
> cleanup into the same path, keeping iounmap() before the window release
> when a mapping exists. Leave the request failure path unchanged.
> 
> This issue was identified during our ongoing static-analysis research while
> reviewing kernel code.
> 
> Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
> Cc: <redacted>

As pointed out by the AI-Generated review of this patch at
https://netdev-ai.bots.linux.dev/sashiko/#/patchset/20260910231400.31919-1-mhun512%40gmail.com
This patch seems to be more of a cleanup than a bug fix - as noted above in
the patch description, the resources are eventually released anyway.

So I wonder if the Fixes and CC to stable should be dropped.

> Assisted-by: OpenAI:GPT-5.6
> Co-developed-by: Ijae Kim <redacted>
> Signed-off-by: Ijae Kim <redacted>
> Signed-off-by: Myeonghun Pak <redacted>
> ---
> Changes in v2:
> - Clarify that the window remains reserved only until device teardown.
> - Make the DMA fallback and continued probe description conditional.
> - Avoid claiming that every request failure leaves no resource held.
> - Restore the surrounding indentation for the new goto statements.
> 
> Link: https://lore.kernel.org/netdev/20260731161740.44955-1-mhun512@gmail.com/

The process nit above notwithstanding, this patch looks good to me.

Reviewed-by: Simon Horman <horms@kernel.org>
```
