Re: [RFC][PATCH] printk: Add option to append kernel version to the dict

[Sergey Senozhatsky]

Hello Tejun,

On (05/17/16 06:51), Tejun Heo wrote:
> On Tue, May 17, 2016 at 11:42:57PM +0900, Sergey Senozhatsky wrote:
> > > > what if there is no place left for init_utsname() after
> > > > msg_print_ext_header() + msg_print_ext_body()?
> > > 
> > > It ends up being truncated, like either of the preceeding calls would.
> > 
> > well, I meant once it's truncated your matching doesn't work anymore.
> 
> Given that most of the messages are written minding typical console
> widths, whether for editing or outputting, I don't think that's a
> practical concern.  We're not talking about appending kilobytes worth
> of additional information here.

hm, I'd probably agree. any chance printk(KERN_CONT) can cause any problems?
for example,

mem_cgroup_print_oom_info():

	for_each_mem_cgroup_tree(iter, memcg) {
		pr_info("Memory cgroup stats for ");
		pr_cont_cgroup_path(iter->css.cgroup);
		pr_cont(":");

		for (i = 0; i < MEM_CGROUP_STAT_NSTATS; i++) {
			if (i == MEM_CGROUP_STAT_SWAP && !do_swap_account)
				continue;
			pr_cont(" %s:%luKB", mem_cgroup_stat_names[i],
				K(mem_cgroup_read_stat(iter, i)));
		}

		for (i = 0; i < NR_LRU_LISTS; i++)
			pr_cont(" %s:%luKB", mem_cgroup_lru_names[i],
				K(mem_cgroup_nr_lru_pages(iter, BIT(i))));

		pr_cont("\n");
	}


there are some other places that do KERN_CONT. e.g. ACPI does KERN_CONT
printk-s from acpi_error/acpi_warning/etc functions. but I've no idea if
there is anything long enough to cause the truncation.


	-ss