Re: [PATCH] USB: gadget: f_mass_storage: fix in fsg_common_init() error recovery

["Michał Nazarewicz" <m.nazarewicz@samsung.com>]

On Wed, 12 May 2010 11:36:06 +0200, Viral Mehta <Viral.Mehta@lntinfotech.com> wrote:
> diff --git a/drivers/usb/gadget/f_mass_storage.c b/drivers/usb/gadget/f_mass_storage.c
> index f4911c0..7ad9a89 100644
> --- a/drivers/usb/gadget/f_mass_storage.c
> +++ b/drivers/usb/gadget/f_mass_storage.c
> @@ -2702,7 +2702,8 @@ static struct fsg_common *fsg_common_init(struct fsg_common *common,
>         if (fsg_strings[FSG_STRING_INTERFACE].id == 0) {
>                 rc = usb_string_id(cdev);
>                 if (rc < 0) {
> -                       kfree(common);
> +                       if(common->free_storage_on_release)
> +                               kfree(common);
>                         return ERR_PTR(rc);
>                 }
>                 fsg_strings[FSG_STRING_INTERFACE].id = rc;
> @@ -2713,7 +2714,8 @@ static struct fsg_common *fsg_common_init(struct fsg_common *common,
>          * LUN devices in sysfs. */
>         curlun = kzalloc(nluns * sizeof *curlun, GFP_KERNEL);
>         if (!curlun) {
> -               kfree(common);
> +               if(common->free_storage_on_release)
> +                       kfree(common);
>                 return ERR_PTR(-ENOMEM);
>         }
>         common->luns = curlun;
>
> looks correct and simple fix.

Yes, it should work fine and in fact was my first approach but as I've written in
my previous mail:

> The way I see it, it does not matter that much -- it's error recovery so we assume
> it's unlikely to happen and as such speed optimisation is not really needed here --
> it's better to optimise for space and minimise the number of possible paths.

Because of that, I the "goto error_release" to be cleaner in the sense that
there is a single error recovery path and only one place where
free_storage_on_release flag is checked and common freed.

-- 
Best regards,                                        _     _
| Humble Liege of Serenely Enlightened Majesty of  o' \,=./ `o
| Computer Science,  Michał "mina86" Nazarewicz       (o o)
+----[mina86*mina86.com]---[mina86*jabber.org]----ooO--(_)--Ooo--