Re: [PATCH v2 2/4] btrfs: make __btrfs_dump_space_info() output better formatted

[Boris Burkov <boris@bur.io>]

On Tue, Jul 26, 2022 at 08:13:53PM +0200, David Sterba wrote:
> On Wed, Jul 20, 2022 at 06:58:16AM +0800, Qu Wenruo wrote:
> > >>> +	btrfs_info(fs_info, "  reserved:      %llu", info->bytes_reserved);
> > >>> +	btrfs_info(fs_info, "  may_use:       %llu", info->bytes_may_use);
> > >>> +	btrfs_info(fs_info, "  read_only:     %llu", info->bytes_readonly);
> > >>> +	if (btrfs_is_zoned(fs_info))
> > >>> +		btrfs_info(fs_info,
> > >>> +			    "  zone_unusable: %llu", info->bytes_zone_unusable);
> > >>
> > >> I'm (perhaps needlessly) worried about splitting this up into six/seven
> > >> messages, because of the ratelimiting rolled into btrfs_printk. The
> > >> ratelimit is 100 messages per 5 * HZ, and it seems like it would be
> > >> unfortunate if it kicked in during the middle of this dump and prevented
> > >> later info from being dumped.
> > >>
> > >> Maybe we should add a btrfs_dump_printk() helper that doesn't have a
> > >> ratelimit built in, for exceptional cases like this where we really,
> > >> really don't want anything ratelimited?
> > >
> > > Splitting the message is IMHO wrong thing, there are other subysystems
> > > writing to the log so the lines can become scattered or interleaved with
> > > the same message from other threads.
> > 
> > But that one line output is really hard to read for human beings.
> > 
> > Or do you mean that, as long as it's debug info, we should not care
> > about readability at all?
> 
> Yes we shold care about readability but kernel printk output lines can
> be interleaved, single line is much easier to grep for and all the
> values are from one event. The format where it's a series of "key=value"
> is common and I think we're used to it from tracepoints too. There are
> lines that do not put "=" between keys and values we could unify that
> eventually.

Agreed that a long line is OK, and preferable to full on splitting.

What about making some btrfs printing macros that use KERN_CONT? I think
that would do what Qu wants without splitting the lines or being bad for
ratelimiting.