Date: Thu, 12 Mar 2015 21:05:19 +1100 (EST) From: Bruce Evans <brde@optusnet.com.au> To: Gleb Smirnoff <glebius@freebsd.org> Cc: Konstantin Belousov <kostikbel@gmail.com>, svn-src-head@freebsd.org, svn-src-all@freebsd.org, src-committers@freebsd.org Subject: Re: svn commit: r279764 - head/sys/vm Message-ID: <20150312201342.W1351@besplex.bde.org> In-Reply-To: <20150310103046.GV17947@FreeBSD.org> References: <201503080213.t282DlXj012465@svn.freebsd.org> <20150310100141.GS17947@FreeBSD.org> <20150310101812.GS2379@kib.kiev.ua> <20150310103046.GV17947@FreeBSD.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, 10 Mar 2015, Gleb Smirnoff wrote: > On Tue, Mar 10, 2015 at 12:18:13PM +0200, Konstantin Belousov wrote: > K> On Tue, Mar 10, 2015 at 01:01:41PM +0300, Gleb Smirnoff wrote: > K> > On Sun, Mar 08, 2015 at 02:13:47AM +0000, Konstantin Belousov wrote: > K> > K> Author: kib > K> > K> Date: Sun Mar 8 02:13:46 2015 > K> > K> New Revision: 279764 > K> > K> URL: https://svnweb.freebsd.org/changeset/base/279764 > K> > K> > K> > K> Log: > K> > K> Fix function name in the panic message. > K> > > K> > Why not use "%s, __func__" always and never encounter this problem > K> > in future? > K> > K> Because you cannot grep for the panic string when __func__ is used. It would also be an an obfuscation. > Grepping for panic string doesn't work in general. Yes it does. The string just needs to be reasonably unique. Even a mispelled function name is usually unique enough. Only function names like f or printf would give too many matches. __func__ is not unique enough. Neither are short format strings like "%d, %d, %d". Longer format strings might be unique enough, but are harder to type. > A panic message > can report pointers or numbers, which make text not unique. Actually, Uniqueness is not needed. Even for a function name you would probably only type a part of the name that you hope is unique, then examine the grep output to see if more context is needed. > the messages that do report extra information are more useful. Also, I consider them as usually just bloat. Use a debugger to find more info. Unfortunately, not all users can run debuggers, and optimization is now excessive so it breaks finding variable values. > if panic string resides in the source code under several levels of > indentation, it is likely to be split into two lines. That would be another obfuscation. Much worse that using __func__. > But you can always grep for the function name and locate the panic > or KASSERT in the function manually, which isn't a big deal. And > if %s, __func__ is used, you will never get to a wrong function. Better yet, spell all function names as __func__ or "this" in comments so that they are write-only there too ;-). Even better yet, spell all function names as __func__ in calls too. Something like __func__ would work for determining the function to call only for recursive calls. __func__ itself doesn't work for that since it is a string. Determining the name of a different function is more difficult. If I knew C++, I might be able to give an example using "this". In C I don't see a better obfuscation (that can be easily written) than using macros to change hard-to-type function names like printf to that_there_func_0. The name of the current function is of course this_here_func ;-). To actually be easier to type, change to names like _0, _1, _2, etc. Bruce
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20150312201342.W1351>