On Sun, 2025-02-16 at 15:15 +0100, Bean Huo wrote: > > External email : Please do not click links or open attachments until > you have verified the sender or the content. > > > On Fri, 2025-02-14 at 16:29 +0800, peter.wang@xxxxxxxxxxxx wrote: > > From: Peter Wang <peter.wang@xxxxxxxxxxxx> > > > > Included the ufs_hba structure as a parameter in various trace > > events > > to provide more context and improve debugging capabilities. > > > From the patch commit message, it is no clear how you can use this > change to get more infor and debug, since the content of the ftrace > output is the same after this change. Because the device name > (dev_name) is still being printed, but now it is derived dynamically > from hba->dev instead of being stored as a string in the trace event. > > I assume you mean to let bpf get more information from hba: > > > strust ufs hba *hba = ctx->hba; > > If my assumption is correct, this purpose and intent should be > prominently highlighted in patch commit message. > > > Kind regards, > Bean Hi Bean, The use of HBA varies depending on the debugging needs. When the tracepoint_probe_register is used to register a debug callback function, we can utilize the parameters of the HBA structure more efficiently to obtain the debug information we need. As for what specific debug information is required, that depends on the individual needs. It depends on how the debugger wants to use it, so the method of use is not specifically mentioned in the commit message. However, in summary, it is a patch that enhances debugging efficiency. Thanks. Peter
