[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <87msav9wm6.fsf@intel.com>
Date: Thu, 29 May 2025 10:04:33 -0700
From: Vinicius Costa Gomes <vinicius.gomes@...el.com>
To: Yi Sun <yi.sun@...el.com>, dave.jiang@...el.com,
dmaengine@...r.kernel.org, linux-kernel@...r.kernel.org
Cc: yi.sun@...el.com, xueshuai@...ux.alibaba.com, gordon.jin@...el.com
Subject: Re: [PATCH 2/2] dmaengine: idxd: Fix refcount underflow on module
unload
Yi Sun <yi.sun@...el.com> writes:
> A recent refactor introduced a misplaced put_device() call, resulting in
> reference count underflow when the module is unloaded.
>
> Expand the idxd_cleanup() function to handle proper cleanup, and remove
> idxd_cleanup_internals() as it was not part of the driver unload path.
>
'idxd_cleanup_internals()' frees a bunch of stuff. I would expect an
explanation of when those things are being free'd now that removed that
call.
> Fixes: a409e919ca32 ("dmaengine: idxd: Refactor remove call with idxd_cleanup() helper")
> Signed-off-by: Yi Sun <yi.sun@...el.com>
>
> diff --git a/drivers/dma/idxd/init.c b/drivers/dma/idxd/init.c
> index 504aca0fd597..a5eabeb6a8bd 100644
> --- a/drivers/dma/idxd/init.c
> +++ b/drivers/dma/idxd/init.c
> @@ -1321,7 +1321,12 @@ static void idxd_remove(struct pci_dev *pdev)
> device_unregister(idxd_confdev(idxd));
> idxd_shutdown(pdev);
> idxd_device_remove_debugfs(idxd);
> - idxd_cleanup(idxd);
> + perfmon_pmu_remove(idxd);
> + idxd_cleanup_interrupts(idxd);
> + if (device_pasid_enabled(idxd))
> + idxd_disable_system_pasid(idxd);
> + if (device_user_pasid_enabled(idxd))
> + idxd_disable_sva(idxd->pdev);
> pci_iounmap(pdev, idxd->reg_base);
> put_device(idxd_confdev(idxd));
> pci_disable_device(pdev);
> --
> 2.43.0
>
--
Vinicius
Powered by blists - more mailing lists