Re: Clarifying confusion of our variable placement rules caused by cleanup.h

[Steven Rostedt]

On Tue, 18 Nov 2025 11:22:27 -0800
Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx> wrote:

> But again: I don't want to make this some kind of hard rule, and I
> think it should be done judiciously and with taste, not some kind of
> crazy conversion thing.

For the few places that do what that example shows, I may update them, as I
think it does make the code look better.

I do have several places that have something like this:

        struct ring_buffer_per_cpu *cpu_buffer __free(kfree) = NULL;
        struct ring_buffer_cpu_meta *meta;
        struct buffer_page *bpage;
        struct page *page;
        int ret;

        cpu_buffer = kzalloc_node(ALIGN(sizeof(*cpu_buffer), cache_line_size()),
                                  GFP_KERNEL, cpu_to_node(cpu));
        if (!cpu_buffer)
                return NULL;

Where the allocation happens right after the declaration. I think I did it
this way because the full line goes over 80 characters, and breaks up the
declaration.

        struct ring_buffer_per_cpu *cpu_buffer __free(kfree) = 
				kzalloc_node(ALIGN(sizeof(*cpu_buffer), cache_line_size()),
                                		  GFP_KERNEL, cpu_to_node(cpu));
        struct ring_buffer_cpu_meta *meta;
        struct buffer_page *bpage;
        struct page *page;
        int ret;

        if (!cpu_buffer)
                return NULL;

Doesn't look nice. I wonder since its the first allocation, if doing:

        struct ring_buffer_cpu_meta *meta;
        struct buffer_page *bpage;
        struct page *page;
        int ret;

        struct ring_buffer_per_cpu *cpu_buffer __free(kfree) = 
		 kzalloc_node(ALIGN(sizeof(*cpu_buffer), cache_line_size()),
				    GFP_KERNEL, cpu_to_node(cpu));
        if (!cpu_buffer)
                return NULL;

Would be acceptable? The "cpu_buffer" is declared right after the declaration,
but the space after the declaration also makes it easier to read than:

        struct ring_buffer_cpu_meta *meta;
        struct buffer_page *bpage;
        struct page *page;
        int ret;
        struct ring_buffer_per_cpu *cpu_buffer __free(kfree) = 
		 kzalloc_node(ALIGN(sizeof(*cpu_buffer), cache_line_size()),
				    GFP_KERNEL, cpu_to_node(cpu));

        if (!cpu_buffer)
                return NULL;


-- Steve