linux-kernel - Re: [PATCH] soc/fsl/qe: fix err handling of ucc_of_parse

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CADRPPNQ97rNJPPL8f0yqhEqTRJw9FnR0U4YCCLxrx5nDB4Ydkg@mail.gmail.com>
Date:   Wed, 5 Dec 2018 14:09:35 -0600
From:   Li Yang <leoyang.li@....com>
To:     wang.yi59@....com.cn
Cc:     Zhao Qiang <qiang.zhao@....com>, zhong.weidong@....com.cn,
        lkml <linux-kernel@...r.kernel.org>, julia.lawall@...6.fr,
        linuxppc-dev <linuxppc-dev@...ts.ozlabs.org>,
        wen.yang99@....com.cn,
        "moderated list:ARM/FREESCALE IMX / MXC ARM ARCHITECTURE" 
        <linux-arm-kernel@...ts.infradead.org>
Subject: Re: [PATCH] soc/fsl/qe: fix err handling of ucc_of_parse_tdm

On Thu, Nov 22, 2018 at 2:42 PM Yi Wang <wang.yi59@....com.cn> wrote:
>
> From: Wen Yang <wen.yang99@....com.cn>
>
> Currently there are 2 problems with the ucc_of_parse_tdm function:
> 1,a possible null pointer dereference in ucc_of_parse_tdm,
> detected by the semantic patch deref_null.cocci,
> with the following warning:
> drivers/soc/fsl/qe/qe_tdm.c:177:21-24: ERROR: pdev is NULL but dereferenced.
> 2,dev gets modified, so in any case that devm_iounmap() will fail even when
> the new pdev is valid, because the iomap was done with a different pdev.
> This patch fixes them.

While we are at this, I think this logic need more serious fixing.  I
see there is no driver bind with the "fsl,t1040-qe-si" or
"fsl,t1040-qe-siram" device.  So allocating resources using devm_*()
with these devices won't provide a cleanup path for these resources
when the caller fails.  I think we should probably allocate resource
under device of caller (e.g. ucc-hdlc), so that when caller probe
fails or is removed it will trigger the cleanup.

>
> Suggested-by: Christophe LEROY <christophe.leroy@....fr>
> Signed-off-by: Wen Yang <wen.yang99@....com.cn>
> CC: Julia Lawall <julia.lawall@...6.fr>
> CC: Zhao Qiang <qiang.zhao@....com>
> ---
>  drivers/soc/fsl/qe/qe_tdm.c | 20 ++++++++++----------
>  1 file changed, 10 insertions(+), 10 deletions(-)
>
> diff --git a/drivers/soc/fsl/qe/qe_tdm.c b/drivers/soc/fsl/qe/qe_tdm.c
> index f78c346..9a29f0b 100644
> --- a/drivers/soc/fsl/qe/qe_tdm.c
> +++ b/drivers/soc/fsl/qe/qe_tdm.c
> @@ -47,7 +47,7 @@ int ucc_of_parse_tdm(struct device_node *np, struct ucc_tdm *utdm,
>         struct resource *res;
>         struct device_node *np2;
>         static int siram_init_flag;
> -       struct platform_device *pdev;
> +       struct platform_device *pdev_si, *pdev_siram;
>
>         sprop = of_get_property(np, "fsl,rx-sync-clock", NULL);
>         if (sprop) {
> @@ -129,16 +129,16 @@ int ucc_of_parse_tdm(struct device_node *np, struct ucc_tdm *utdm,
>         if (!np2)
>                 return -EINVAL;
>
> -       pdev = of_find_device_by_node(np2);
> -       if (!pdev) {
> +       pdev_si = of_find_device_by_node(np2);
> +       if (!pdev_si) {
>                 pr_err("%pOFn: failed to lookup pdev\n", np2);
>                 of_node_put(np2);
>                 return -EINVAL;
>         }
>
>         of_node_put(np2);
> -       res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
> -       utdm->si_regs = devm_ioremap_resource(&pdev->dev, res);
> +       res = platform_get_resource(pdev_si, IORESOURCE_MEM, 0);
> +       utdm->si_regs = devm_ioremap_resource(&pdev_si->dev, res);
>         if (IS_ERR(utdm->si_regs)) {
>                 ret = PTR_ERR(utdm->si_regs);
>                 goto err_miss_siram_property;
> @@ -150,8 +150,8 @@ int ucc_of_parse_tdm(struct device_node *np, struct ucc_tdm *utdm,
>                 goto err_miss_siram_property;
>         }
>
> -       pdev = of_find_device_by_node(np2);
> -       if (!pdev) {
> +       pdev_siram = of_find_device_by_node(np2);
> +       if (!pdev_siram) {
>                 ret = -EINVAL;
>                 pr_err("%pOFn: failed to lookup pdev\n", np2);
>                 of_node_put(np2);
> @@ -159,8 +159,8 @@ int ucc_of_parse_tdm(struct device_node *np, struct ucc_tdm *utdm,
>         }
>
>         of_node_put(np2);
> -       res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
> -       utdm->siram = devm_ioremap_resource(&pdev->dev, res);
> +       res = platform_get_resource(pdev_siram, IORESOURCE_MEM, 0);
> +       utdm->siram = devm_ioremap_resource(&pdev_siram->dev, res);
>         if (IS_ERR(utdm->siram)) {
>                 ret = PTR_ERR(utdm->siram);
>                 goto err_miss_siram_property;
> @@ -174,7 +174,7 @@ int ucc_of_parse_tdm(struct device_node *np, struct ucc_tdm *utdm,
>         return ret;
>
>  err_miss_siram_property:
> -       devm_iounmap(&pdev->dev, utdm->si_regs);
> +       devm_iounmap(&pdev_si->dev, utdm->si_regs);
>         return ret;
>  }
>  EXPORT_SYMBOL(ucc_of_parse_tdm);
> --
> 2.9.5
>