On Tue, 2025-07-22 at 11:43 -0700, Viacheslav Dubeyko wrote: > On Tue, 2025-07-22 at 15:50 +0800, Yangtao Li wrote: > > Hi Slava, > > > > <skipped> > > > > > > > > Colud we rework msg to add value description? > > > > There're too much values to identify what it is. > > > > > > > > > > What do you mean by value description? > > > > For example: > > > > hfs_dbg(BNODE_MOD, "%d, %d, %d, %d, %d\n", > > be32_to_cpu(desc.next), be32_to_cpu(desc.prev), > > desc.type, desc.height, > > be16_to_cpu(desc.num_recs)); > > > > There are 5 %d. It's hard to recognize what it is. Changing it to > > following style w/ description might be a bit more clear? > > > > hfs_dbg(BNODE_MOD, "next:%d prev:%d, type:%s, > > height:%d > > num_recs:%d\n", be32_to_cpu(desc.next), be32_to_cpu(desc.prev), > > hfs_node_type(desc.type), desc.height, be16_to_cpu(desc.num_recs)); > > > > We can rework it step by step. First of all, the reworking of all > debug > messages at once is too much for one patch. Secondly, the style of > messages is history of HFS implementation. I suggest to make this > first > step and, then, we can rework the debugging messages in the > background > of bug fix. Does it make sense to you? > > > However, maybe, you are right here. I can rework debug messages in more reasonable and understandable way. It will make debugging and bug fix much easier. Thanks, Slava.
