[PATCH 4/6] kexec: A new system call, kexec_file_load, for in kernel kexec

[Baoquan He]

On 12/04/13 at 09:56am, Baoquan He wrote:
> On 11/20/13 at 12:50pm, Vivek Goyal wrote:
> > + * that kexec_mutex is held.
> > + */
> 
> I think kexec_add_buffer is guaranteed to be called before allocating
> control pages, why not updating image->control_page after each time
> kexec_add_buffer is called. Then when control page is needed, effective
> address in crash_kernel region can be given. This can be a little more
> efficient.

Check again, it's not a good idea. Because buf_align is different when
kexec_add_buffer is called. Sometime the bug_align is larger than 4k.
Please ignore this comment.

> 
> > +int kexec_add_buffer(struct kimage *image, char *buffer,
> > +		unsigned long bufsz, unsigned long memsz,
> > +		unsigned long buf_align, unsigned long buf_min,
> > +		unsigned long buf_max, int top_down, unsigned long *load_addr)
> > +{
> > +
> > +	unsigned long nr_segments = image->nr_segments, new_nr_segments;
> > +	struct kexec_segment *ksegment;
> > +	struct kexec_buf *kbuf;
> > +
> > +	/* Currently adding segment this way is allowed only in file mode */
> > +	if (!image->file_mode)
> > +		return -EINVAL;
> > +
> > +	if (nr_segments >= KEXEC_SEGMENT_MAX)
> > +		return -EINVAL;
> > +
> > +	/*
> > +	 * Make sure we are not trying to add buffer after allocating
> > +	 * control pages. All segments need to be placed first before
> > +	 * any control pages are allocated. As control page allocation
> > +	 * logic goes through list of segments to make sure there are
> > +	 * no destination overlaps.
> > +	 */
> > +	WARN_ONCE(!list_empty(&image->control_pages), "Adding kexec buffer"
> > +			" after allocating control pages\n");
> > +
> > +	kbuf = kzalloc(sizeof(struct kexec_buf), GFP_KERNEL);
> > +	if (!kbuf)
> > +		return -ENOMEM;
> > +