On Sun, Dec 20, 2020 at 03:05:40PM +0800, Dinghao Liu wrote:
> When do_ide_setup_pci_device() fails, host allocated
> by ide_host_alloc() may not have been freed, which
> leads to memleak.
>
> Signed-off-by: Dinghao Liu <dinghao....@zju.edu.cn>
> ---
> drivers/ide/setup-pci.c | 8 ++++----
> 1 file changed, 4 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/ide/setup-pci.c b/drivers/ide/setup-pci.c
> index fdc8e813170c..c7da5368fcd4 100644
> --- a/drivers/ide/setup-pci.c
> +++ b/drivers/ide/setup-pci.c
> @@ -586,7 +586,7 @@ int ide_pci_init_two(struct pci_dev *dev1, struct pci_dev
> *dev2,
> * do_ide_setup_pci_device() on the first device!
> */
> if (ret < 0)
> - goto out_free_bars;
> + goto out_free_host;
>
> /* fixup IRQ */
> if (ide_pci_is_in_compatibility_mode(pdev[i])) {
> @@ -597,11 +597,11 @@ int ide_pci_init_two(struct pci_dev *dev1, struct
> pci_dev *dev2,
> }
>
> ret = ide_host_register(host, d, hws);
> - if (ret)
> - ide_host_free(host);
> - else
> + if (!ret)
> goto out;
Maybe
if (ret)
goto out_free_host;
return 0;
would be more clear here. But this is just small nit.
>
> +out_free_host:
> + ide_host_free(host);
> out_free_bars:
> i = n_ports / 2;
> while (i--)
