[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <73c1bc90-569a-4514-93d4-e6d5e0597607@wanadoo.fr>
Date: Tue, 7 Jan 2025 19:48:28 +0100
From: Christophe JAILLET <christophe.jaillet@...adoo.fr>
To: Ma Ke <make24@...as.ac.cn>, bhelgaas@...gle.com
Cc: linux-pci@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] PCI: fix reference leak in pci_alloc_child_bus()
Le 07/01/2025 à 12:07, Ma Ke a écrit :
> When device_register(&child->dev) failed, calling put_device() to
> explicitly release child->dev. Otherwise, it could cause double free
> problem.
>
> Found by code review.
>
> Signed-off-by: Ma Ke <make24@...as.ac.cn>
> ---
> drivers/pci/probe.c | 5 ++++-
> 1 file changed, 4 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c
> index 2e81ab0f5a25..a61070ce5f88 100644
> --- a/drivers/pci/probe.c
> +++ b/drivers/pci/probe.c
> @@ -1174,7 +1174,10 @@ static struct pci_bus *pci_alloc_child_bus(struct pci_bus *parent,
> add_dev:
> pci_set_bus_msi_domain(child);
> ret = device_register(&child->dev);
> - WARN_ON(ret < 0);
> + if (WARN_ON(ret < 0)) {
> + put_device(&child->dev);
> + return ERR_PTR(ret);
Previously, the code was continuing the execution. I don't know it is
correct to return at this point.
Anyway, returning ERR_PTR(ret) just looks wrong, because
pci_add_new_bus() expect NULL in case of error.
Should ERR_PTR(ret) be returned, it is likely basd thing would happen there.
CJ
> + }
>
> pcibios_add_bus(child);
>
Powered by blists - more mailing lists