ie31200_register_mci() calls device_initialize() for priv->dev
unconditionally. However, in the error path, put_device() is not
called, leading to an imbalance. Similarly, in the unload path,
put_device() is missing.

Although edac_mc_free() eventually frees the memory, it does not
release the device initialized by device_initialize(). For code
readability and proper pairing of device_initialize()/put_device(),
add put_device() calls in both error and unload paths.

Found by code review.

Signed-off-by: Ma Ke <make24@iscas.ac.cn>
---
Changes in v3:
- moved put_device() from fail_free to fail_unmap to avoid using uninitialized priv variable when window allocation fails.
Changes in v2:
- modified the patch, thanks for developer's suggestions;
- removed Fixes line.
---
 drivers/edac/ie31200_edac.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/edac/ie31200_edac.c b/drivers/edac/ie31200_edac.c
index 5a080ab65476..8d4ddaa85ae8 100644
--- a/drivers/edac/ie31200_edac.c
+++ b/drivers/edac/ie31200_edac.c
@@ -526,6 +526,7 @@ static int ie31200_register_mci(struct pci_dev *pdev, struct res_config *cfg, in
 	ie31200_pvt.priv[mc] = priv;
 	return 0;
 fail_unmap:
+	put_device(&priv->dev);
 	iounmap(window);
 fail_free:
 	edac_mc_free(mci);
@@ -598,6 +599,7 @@ static void ie31200_unregister_mcis(void)
 		mci = priv->mci;
 		edac_mc_del_mc(mci->pdev);
 		iounmap(priv->window);
+		put_device(&priv->dev);
 		edac_mc_free(mci);
 	}
 }
-- 
2.17.1

On Thu, Nov 06, 2025 at 04:47:35PM +0800, Ma Ke wrote:
> ie31200_register_mci() calls device_initialize() for priv->dev
> unconditionally. However, in the error path, put_device() is not
> called, leading to an imbalance. Similarly, in the unload path,
> put_device() is missing.
> 
> Although edac_mc_free() eventually frees the memory, it does not
> release the device initialized by device_initialize(). For code
> readability and proper pairing of device_initialize()/put_device(),
> add put_device() calls in both error and unload paths.
> 
> Found by code review.
> 
> Signed-off-by: Ma Ke <make24@iscas.ac.cn>

Applied v3 version.

Thanks

-Tony

> From: Ma Ke <make24@iscas.ac.cn>
> Sent: Thursday, November 6, 2025 4:48 PM
> To: jbaron@akamai.com; bp@alien8.de; Luck, Tony <tony.luck@intel.com>;
> Zhuo, Qiuxu <qiuxu.zhuo@intel.com>
> Cc: linux-edac@vger.kernel.org; linux-kernel@vger.kernel.org; akpm@linux-
> foundation.org; Ma Ke <make24@iscas.ac.cn>
> Subject: [PATCH v3] EDAC/ie31200: Fix error handling in ie31200_register_mci
> 
> ie31200_register_mci() calls device_initialize() for priv->dev unconditionally.
> However, in the error path, put_device() is not called, leading to an
> imbalance. Similarly, in the unload path,
> put_device() is missing.
> 
> Although edac_mc_free() eventually frees the memory, it does not release the
> device initialized by device_initialize(). For code readability and proper pairing
> of device_initialize()/put_device(), add put_device() calls in both error and
> unload paths.
> 
> Found by code review.
> 
> Signed-off-by: Ma Ke <make24@iscas.ac.cn>
> ---
> Changes in v3:
> - moved put_device() from fail_free to fail_unmap to avoid using uninitialized
> priv variable when window allocation fails.

Oops ... I didn't catch that during v2 review. 

For v3:
Reviewed-by: Qiuxu Zhuo <qiuxu.zhuo@intel.com>

> [...]

On Thu, Nov 06, 2025 at 04:47:35PM +0800, Ma Ke wrote:
> ie31200_register_mci() calls device_initialize() for priv->dev
> unconditionally. However, in the error path, put_device() is not
> called, leading to an imbalance. Similarly, in the unload path,
> put_device() is missing.
> 
> Although edac_mc_free() eventually frees the memory, it does not
> release the device initialized by device_initialize(). For code
> readability and proper pairing of device_initialize()/put_device(),
> add put_device() calls in both error and unload paths.
> 
> Found by code review.
> 
> Signed-off-by: Ma Ke <make24@iscas.ac.cn>
> ---
> Changes in v3:
> - moved put_device() from fail_free to fail_unmap to avoid using uninitialized priv variable when window allocation fails.

Zapped v2 from the tree for the time being.

> Changes in v2:
> - modified the patch, thanks for developer's suggestions;
> - removed Fixes line.

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette