[Devel] Re: [RFC v9][PATCH 05/13] Dump memory address space

[Serge E. Hallyn]

Quoting Oren Laadan (orenl at cs.columbia.edu):
> +/**
> + * cr_fill_fname - return pathname of a given file
> + * @path: path name
> + * @root: relative root
> + * @buf: buffer for pathname
> + * @n: buffer length (in) and pathname length (out)
> + */
> +static char *
> +cr_fill_fname(struct path *path, struct path *root, char *buf, int *n)
> +{
> +	struct path tmp = *root;
> +	char *fname;
> +
> +	BUG_ON(!buf);
> +	fname = __d_path(path, &tmp, buf, *n);
> +	if (!IS_ERR(fname))
> +		*n = (buf + (*n) - fname);
> +	/*
> +	 * FIXME: if __d_path() changed these, it must have stepped out of
> +	 * init's namespace. Since currently we require a unified namespace
> +	 * within the container: simply fail.
> +	 */
> +	if (tmp.mnt != root->mnt || tmp.dentry != root->dentry)
> +		fname = ERR_PTR(-EBADF);

...

> +static int cr_ctx_checkpoint(struct cr_ctx *ctx, pid_t pid)
> +{
> +	ctx->root_pid = pid;
> +
> +	/*
> +	 * assume checkpointer is in container's root vfs
> +	 * FIXME: this works for now, but will change with real containers
> +	 */
> +	ctx->vfsroot = &current->fs->root;
> +	path_get(ctx->vfsroot);

Hi Oren,

Is there really any good reason to use current->fs->root rather
than ctx->root_task->fs->root here?

The way I'm testing, the checkpointer is in fact in a different
container, so the root passed into cr_fill_fname() is different
from the container's root, so cr_fill_fname() always returns me
-EBADF.

thanks,
-serge
_______________________________________________
Containers mailing list
Containers at lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/containers