[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <ZnEzyJW9WAX0Rjsx@infradead.org>
Date: Tue, 18 Jun 2024 00:14:16 -0700
From: Christoph Hellwig <hch@...radead.org>
To: Yu Kuai <yukuai1@...weicloud.com>
Cc: axboe@...nel.dk, tj@...nel.org, gregkh@...uxfoundation.org,
bvanassche@....org, hch@...radead.org, josef@...icpanda.com,
lizefan.x@...edance.com, hannes@...xchg.org,
linux-block@...r.kernel.org, linux-kernel@...r.kernel.org,
cgroups@...r.kernel.org, yukuai3@...wei.com, yi.zhang@...wei.com,
yangerkun@...wei.com
Subject: Re: [PATCH RFC v2 3/7] block: export some API
On Tue, Jun 18, 2024 at 11:17:47AM +0800, Yu Kuai wrote:
> bool blkcg_debug_stats = false;
> +EXPORT_SYMBOL_GPL(blkcg_debug_stats);
exporting variables is lways a bad idea.a
>
> static DEFINE_RAW_SPINLOCK(blkg_stat_lock);
>
> @@ -688,6 +689,7 @@ const char *blkg_dev_name(struct blkcg_gq *blkg)
> return NULL;
> return bdi_dev_name(blkg->q->disk->bdi);
> }
> +EXPORT_SYMBOL_GPL(blkg_dev_name);
And this helper is just horribly to be honest. The bdi_dev_name
should not be used anywhere in block code, and something like this
certainly should not be exported even if we have to keep it for
legacy reasons.
> /**
> * blkcg_print_blkgs - helper for printing per-blkg data
> @@ -815,6 +817,7 @@ int blkg_conf_open_bdev(struct blkg_conf_ctx *ctx)
> ctx->bdev = bdev;
> return 0;
> }
> +EXPORT_SYMBOL_GPL(blkg_conf_open_bdev);
This again is a horrible function papeing over blk-cgroup design
mistakes. I absolutely should not be exported.
Powered by blists - more mailing lists