|
|
| Prev | Home | Next |
| Up |
If you type an Emacs command you did not intend, the results are often mysterious. This chapter tells what you can do to cancel your mistake or recover from a mysterious situation. Emacs bugs and system crashes are also considered. Quitting and Aborting
C-g C-BREAK (MS-DOS) Quit. Cancel running or partially typed command. C-] Abort innermost recursive editing level and cancel the command which invoked it (abort-recursive-edit). ESC ESC ESC Either quit or abort, whichever makes sense (keyboard-escape-quit). M-x top-level Abort all recursive editing levels that are currently executing. C-x u Cancel a previously made change in the buffer contents (undo). There are two ways of canceling commands which are not finished executing: quitting with C-g, and aborting with C-] or M-x top-level. Quitting cancels a partially typed command or one which is already running. Aborting exits a recursive editing level and cancels the command that invoked the recursive edit. (See section Recursive Editing Levels.) Quitting with C-g is used for getting rid of a partially typed command, or a numeric argument that you don't want. It also stops a running command in the middle in a relatively safe way, so you can use it if you accidentally give a command which takes a long time. In particular, it is safe to quit out of killing; either your text will all still be in the buffer, or it will all be in the kill ring (or maybe both). Quitting an incremental search does special things documented under searching; in general, it may take two successive C-g characters to get out of a search (see section Incremental Search). On MS-DOS, the character C-BREAK serves as a quit character like C-g. The reason is that it is not feasible, on MS-DOS, to recognize C-g while a command is running, between interactions with the user. By contrast, it is feasible to recognize C-BREAK at all times. See section Keyboard and Mouse on MS-DOS. C-g works by setting the variable quit-flag to t the instant C-g is typed; Emacs Lisp checks this variable frequently and quits if it is non-nil. C-g is only actually executed as a command if you type it while Emacs is waiting for input. If you quit with C-g a second time before the first C-g is recognized, you activate the "emergency escape" feature and return to the shell. See section Emergency Escape. There may be times when you cannot quit. When Emacs is waiting for the operating system to do something, quitting is impossible unless special pains are taken for the particular system call within Emacs where the waiting occurs. We have done this for the system calls that users are likely to want to quit from, but it's possible you will find another. In one very common case--waiting for file input or output using NFS--Emacs itself knows how to quit, but most NFS implementations simply do not allow user programs to stop waiting for NFS when the NFS server is hung. Aborting with C-] (abort-recursive-edit) is used to get out of a recursive editing level and cancel the command which invoked it. Quitting with C-g does not do this, and could not do this, because it is used to cancel a partially typed command within the recursive editing level. Both operations are useful. For example, if you are in a recursive edit and type C-u 8 to enter a numeric argument, you can cancel that argument with C-g and remain in the recursive edit. The command ESC ESC ESC (keyboard-escape-quit) can either quit or abort. This key was defined because ESC is used to "get out" in many PC programs. It can cancel a prefix argument, clear a selected region, or get out of a Query Replace, like C-g. It can get out of the minibuffer or a recursive edit, like C-]. It can also get out of splitting the frame into multiple windows, like C-x 1. One thing it cannot do, however, is stop a command that is running. That's because it executes as an ordinary command, and Emacs doesn't notice it until it is ready for a command. The command M-x top-level is equivalent to "enough" C-] commands to get you out of all the levels of recursive edits that you are in. C-] gets you out one level at a time, but M-x top-level goes out all levels at once. Both C-] and M-x top-level are like all other commands, and unlike C-g, in that they take effect only when Emacs is ready for a command. C-] is an ordinary key and has its meaning only because of its binding in the keymap. See section Recursive Editing Levels. C-x u (undo) is not strictly speaking a way of canceling a command, but you can think of it as canceling a command that already finished executing. See section Undoing Changes. Dealing with Emacs Trouble
This section describes various conditions in which Emacs fails to work normally, and how to recognize them and correct them. If DEL Fails to Delete If you find that DEL enters Help like Control-h instead of deleting a character, your terminal is sending the wrong code for DEL. You can work around this problem by changing the keyboard translation table (see section Keyboard Translations). Recursive Editing Levels Recursive editing levels are important and useful features of Emacs, but they can seem like malfunctions to the user who does not understand them. If the mode line has square brackets `[...]' around the parentheses that contain the names of the major and minor modes, you have entered a recursive editing level. If you did not do this on purpose, or if you don't understand what that means, you should just get out of the recursive editing level. To do so, type M-x top-level. This is called getting back to top level. See section Recursive Editing Levels. Garbage on the Screen If the data on the screen looks wrong, the first thing to do is see whether the text is really wrong. Type C-l, to redisplay the entire screen. If the screen appears correct after this, the problem was entirely in the previous screen update. (Otherwise, see section Garbage in the Text.) Display updating problems often result from an incorrect termcap entry for the terminal you are using. The file `etc/TERMS' in the Emacs distribution gives the fixes for known problems of this sort. `INSTALL' contains general advice for these problems in one of its sections. Very likely there is simply insufficient padding for certain display operations. To investigate the possibility that you have this sort of problem, try Emacs on another terminal made by a different manufacturer. If problems happen frequently on one kind of terminal but not another kind, it is likely to be a bad termcap entry, though it could also be due to a bug in Emacs that appears for terminals that have or that lack specific features. Garbage in the Text If C-l shows that the text is wrong, try undoing the changes to it using C-x u until it gets back to a state you consider correct. Also try C-h l to find out what command you typed to produce the observed results. If a large portion of text appears to be missing at the beginning or end of the buffer, check for the word `Narrow' in the mode line. If it appears, the text you don't see is probably still present, but temporarily off-limits. To make it accessible again, type C-x n w. See section Narrowing. Spontaneous Entry to Incremental Search If Emacs spontaneously displays `I-search:' at the bottom of the screen, it means that the terminal is sending C-s and C-q according to the poorly designed xon/xoff "flow control" protocol. If this happens to you, your best recourse is to put the terminal in a mode where it will not use flow control, or give it so much padding that it will never send a C-s. (One way to increase the amount of padding is to set the variable baud-rate to a larger value. Its value is the terminal output speed, measured in the conventional units of baud.) If you don't succeed in turning off flow control, the next best thing is to tell Emacs to cope with it. To do this, call the function enable-flow-control. Typically there are particular terminal types with which you must use flow control. You can conveniently ask for flow control on those terminal types only, using enable-flow-control-on. For example, if you find you must use flow control on VT-100 and H19 terminals, put the following in your `.emacs' file: (enable-flow-control-on "vt100" "h19")
When flow control is enabled, you must type C-\ to get the effect of a C-s, and type C-^ to get the effect of a C-q. (These aliases work by means of keyboard translations; see section Keyboard Translations.) Running out of Memory If you get the error message `Virtual memory exceeded', save your modified buffers with C-x s. This method of saving them has the smallest need for additional memory. Emacs keeps a reserve of memory which it makes available when this error happens; that should be enough to enable C-x s to complete its work. Once you have saved your modified buffers, you can exit this Emacs job and start another, or you can use M-x kill-some-buffers to free space in the current Emacs job. If you kill buffers containing a substantial amount of text, you can safely go on editing. Emacs refills its memory reserve automatically when it sees sufficient free space available, in case you run out of memory another time. Do not use M-x buffer-menu to save or kill buffers when you run out of memory, because the buffer menu needs a fair amount memory itself, and the reserve supply may not be enough. Recovery After a Crash If Emacs or the computer crashes, you can recover the files you were editing at the time of the crash from their auto-save files. To do this, start Emacs again and type the command M-x recover-session. This command initially displays a buffer which lists interrupted session files, each with its date. You must choose which session to recover from. Typically the one you want is the most recent one. Move point to the one you choose, and type C-c C-c. Then recover-session asks about each of the files that you were editing during that session; it asks whether to recover that file. If you answer y for a file, it shows the dates of that file and its auto-save file, then asks once again whether to recover that file. For the second question, you must confirm with yes. If you do, Emacs visits the file but gets the text from the auto-save file. When recover-session is done, the files you've chosen to recover are present in Emacs buffers. You should then save them. Only this--saving them--updates the files themselves. Emergency Escape Because at times there have been bugs causing Emacs to loop without checking quit-flag, a special feature causes Emacs to be suspended immediately if you type a second C-g while the flag is already set, so you can always get out of GNU Emacs. Normally Emacs recognizes and clears quit-flag (and quits!) quickly enough to prevent this from happening. (On MS-DOS and compatible systems, type C-BREAK twice.) When you resume Emacs after a suspension caused by multiple C-g, it asks two questions before going back to what it had been doing: Auto-save? (y or n) Abort (and dump core)? (y or n)
Answer each one with y or n followed by RET. Saying y to `Auto-save?' causes immediate auto-saving of all modified buffers in which auto-saving is enabled. Saying y to `Abort (and dump core)?' causes an illegal instruction to be executed, dumping core. This is to enable a wizard to figure out why Emacs was failing to quit in the first place. Execution does not continue after a core dump. If you answer n, execution does continue. With luck, GNU Emacs will ultimately check quit-flag and quit normally. If not, and you type another C-g, it is suspended again. If Emacs is not really hung, just slow, you may invoke the double C-g feature without really meaning to. Then just resume and answer n to both questions, and you will arrive at your former state. Presumably the quit you requested will happen soon. The double-C-g feature is turned off when Emacs is running under the X Window System, since you can use the window manager to kill Emacs or to create another window and run another program. On MS-DOS and compatible systems, the emergency escape feature is sometimes unavailable, even if you press C-BREAK twice, when some system call (MS-DOS or BIOS) hangs, or when Emacs is stuck in a very tight endless loop (in C code, not in Lisp code). Help for Total Frustration If using Emacs (or something else) becomes terribly frustrating and none of the techniques described above solve the problem, Emacs can still help you. First, if the Emacs you are using is not responding to commands, type C-g C-g to get out of it and then start a new one. Second, type M-x doctor RET. The doctor will help you feel better. Each time you say something to the doctor, you must end it by typing RET RET. This lets the doctor know you are finished. Reporting Bugs
Sometimes you will encounter a bug in Emacs. Although we cannot promise we can or will fix the bug, and we might not even agree that it is a bug, we want to hear about problems you encounter. Often we agree they are bugs and want to fix them. To make it possible for us to fix a bug, you must report it. In order to do so effectively, you must know when and how to do it. When Is There a Bug If Emacs executes an illegal instruction, or dies with an operating system error message that indicates a problem in the program (as opposed to something like "disk full"), then it is certainly a bug. If Emacs updates the display in a way that does not correspond to what is in the buffer, then it is certainly a bug. If a command seems to do the wrong thing but the problem corrects itself if you type C-l, it is a case of incorrect display updating. Taking forever to complete a command can be a bug, but you must make certain that it was really Emacs's fault. Some commands simply take a long time. Type C-g (C-BREAK on MS-DOS) and then C-h l to see whether the input Emacs received was what you intended to type; if the input was such that you know it should have been processed quickly, report a bug. If you don't know whether the command should take a long time, find out by looking in the manual or by asking for assistance. If a command you are familiar with causes an Emacs error message in a case where its usual definition ought to be reasonable, it is probably a bug. If a command does the wrong thing, that is a bug. But be sure you know for certain what it ought to have done. If you aren't familiar with the command, or don't know for certain how the command is supposed to work, then it might actually be working right. Rather than jumping to conclusions, show the problem to someone who knows for certain. Finally, a command's intended definition may not be best for editing with. This is a very important sort of problem, but it is also a matter of judgment. Also, it is easy to come to such a conclusion out of ignorance of some of the existing features. It is probably best not to complain about such a problem until you have checked the documentation in the usual ways, feel confident that you understand it, and know for certain that what you want is not available. If you are not sure what the command is supposed to do after a careful reading of the manual, check the index and glossary for any terms that may be unclear. If after careful rereading of the manual you still do not understand what the command should do, that indicates a bug in the manual, which you should report. The manual's job is to make everything clear to people who are not Emacs experts--including you. It is just as important to report documentation bugs as program bugs. If the on-line documentation string of a function or variable disagrees with the manual, one of them must be wrong; that is a bug. Understanding Bug Reporting When you decide that there is a bug, it is important to report it and to report it in a way which is useful. What is most useful is an exact description of what commands you type, starting with the shell command to run Emacs, until the problem happens. The most important principle in reporting a bug is to report facts. Hypotheses and verbal descriptions are no substitute for the detailed raw data. Reporting the facts is straightforward, but many people strain to posit explanations and report them instead of the facts. If the explanations are based on guesses about how Emacs is implemented, they will be useless; meanwhile, lacking the facts, we will have no real information about the bug. For example, suppose that you type C-x C-f /glorp/baz.ugh RET, visiting a file which (you know) happens to be rather large, and Emacs displayed `I feel pretty today'. The best way to report the bug is with a sentence like the preceding one, because it gives all the facts. A bad way would be to assume that the problem is due to the size of the file and say, "I visited a large file, and Emacs displayed `I feel pretty today'." This is what we mean by "guessing explanations." The problem is just as likely to be due to the fact that there is a `z' in the file name. If this is so, then when we got your report, we would try out the problem with some "large file," probably with no `z' in its name, and not see any problem. There is no way in the world that we could guess that we should try visiting a file with a `z' in its name. Alternatively, the problem might be due to the fact that the file starts with exactly 25 spaces. For this reason, you should make sure that you inform us of the exact contents of any file that is needed to reproduce the bug. What if the problem only occurs when you have typed the C-x C-a command previously? This is why we ask you to give the exact sequence of characters you typed since starting the Emacs session. You should not even say "visit a file" instead of C-x C-f unless you know that it makes no difference which visiting command is used. Similarly, rather than saying "if I have three characters on the line," say "after I type RET A B C RET C-p," if that is the way you entered the text. So please don't guess any explanations when you report a bug. If you want to actually debug the problem, and report explanations that are more than guesses, that is useful--but please include the facts as well. Checklist for Bug Reports The best way to send a bug report is to mail it electronically to the Emacs maintainers at `bug-gnu-emacs@gnu.org'. (If you want to suggest a change as an improvement, use the same address.) If you'd like to read the bug reports, you can find them on the newsgroup `gnu.emacs.bug'; keep in mind, however, that as a spectator you should not criticize anything about what you see there. The purpose of bug reports is to give information to the Emacs maintainers. Spectators are welcome only as long as they do not interfere with this. In particular, some bug reports contain large amounts of data; spectators should not complain about this. Please do not post bug reports using netnews; mail is more reliable than netnews about reporting your correct address, which we may need in order to ask you for more information. If you can't send electronic mail, then mail the bug report on paper or machine-readable media to this address: GNU Emacs Bugs Free Software Foundation 59 Temple Place, Suite 330 Boston, MA 02111-1307 USA
We do not promise to fix the bug; but if the bug is serious, or ugly, or easy to fix, chances are we will want to. A convenient way to send a bug report for Emacs is to use the command M-x report-emacs-bug. This sets up a mail buffer (see section Sending Mail) and automatically inserts some of the essential information. However, it cannot supply all the necessary information; you should still read and follow the guidelines below, so you can enter the other crucial information by hand before you send the message. To enable maintainers to investigate a bug, your report should include all these things: The version number of Emacs. Without this, we won't know whether there is any point in looking for the bug in the current version of GNU Emacs. You can get the version number by typing M-x emacs-version RET. If that command does not work, you probably have something other than GNU Emacs, so you will have to report the bug somewhere else. The type of machine you are using, and the operating system name and version number. M-x emacs-version RET provides this information too. Copy its output from the `Messages' buffer, so that you get it all and get it accurately. The operands given to the configure command when Emacs was installed. A complete list of any modifications you have made to the Emacs source. (We may not have time to investigate the bug unless it happens in an unmodified Emacs. But if you've made modifications and you don't tell us, you are sending us on a wild goose chase.) Be precise about these changes. A description in English is not enough--send a context diff for them. Adding files of your own, or porting to another machine, is a modification of the source. Details of any other deviations from the standard procedure for installing GNU Emacs. The complete text of any files needed to reproduce the bug. If you can tell us a way to cause the problem without visiting any files, please do so. This makes it much easier to debug. If you do need files, make sure you arrange for us to see their exact contents. For example, it can often matter whether there are spaces at the ends of lines, or a newline after the last line in the buffer (nothing ought to care whether the last line is terminated, but try telling the bugs that). The precise commands we need to type to reproduce the bug. The easy way to record the input to Emacs precisely is to write a dribble file. To start the file, execute the Lisp expression (open-dribble-file "~/dribble") using M-: or from the `scratch' buffer just after starting Emacs. From then on, Emacs copies all your input to the specified dribble file until the Emacs process is killed. For possible display bugs, the terminal type (the value of environment variable TERM), the complete termcap entry for the terminal from `/etc/termcap' (since that file is not identical on all machines), and the output that Emacs actually sent to the terminal. The way to collect the terminal output is to execute the Lisp expression (open-termscript "~/termscript") using M-: or from the `scratch' buffer just after starting Emacs. From then on, Emacs copies all terminal output to the specified termscript file as well, until the Emacs process is killed. If the problem happens when Emacs starts up, put this expression into your `.emacs' file so that the termscript file will be open when Emacs displays the screen for the first time. Be warned: it is often difficult, and sometimes impossible, to fix a terminal-dependent bug without access to a terminal of the type that stimulates the bug. A description of what behavior you observe that you believe is incorrect. For example, "The Emacs process gets a fatal signal," or, "The resulting text is as follows, which I think is wrong." Of course, if the bug is that Emacs gets a fatal signal, then one can't miss it. But if the bug is incorrect text, the maintainer might fail to notice what is wrong. Why leave it to chance? Even if the problem you experience is a fatal signal, you should still say so explicitly. Suppose something strange is going on, such as, your copy of the source is out of sync, or you have encountered a bug in the C library on your system. (This has happened!) Your copy might crash and the copy here might not. If you said to expect a crash, then when Emacs here fails to crash, we would know that the bug was not happening. If you don't say to expect a crash, then we would not know whether the bug was happening--we would not be able to draw any conclusion from our observations. If the manifestation of the bug is an Emacs error message, it is important to report the precise text of the error message, and a backtrace showing how the Lisp program in Emacs arrived at the error. To get the error message text accurately, copy it from the `Messages' buffer into the bug report. Copy all of it, not just part. To make a backtrace for the error, evaluate the Lisp expression (setq debug-on-error t) before the error happens (that is to say, you must execute that expression and then make the bug happen). This causes the error to run the Lisp debugger, which shows you a backtrace. Copy the text of the debugger's backtrace into the bug report. This use of the debugger is possible only if you know how to make the bug happen again. If you can't make it happen again, at least copy the whole error message. Check whether any programs you have loaded into the Lisp world, including your `.emacs' file, set any variables that may affect the functioning of Emacs. Also, see whether the problem happens in a freshly started Emacs without loading your `.emacs' file (start Emacs with the -q switch to prevent loading the init file). If the problem does not occur then, you must report the precise contents of any programs that you must load into the Lisp world in order to cause the problem to occur. If the problem does depend on an init file or other Lisp programs that are not part of the standard Emacs system, then you should make sure it is not a bug in those programs by complaining to their maintainers first. After they verify that they are using Emacs in a way that is supposed to work, they should report the bug. If you wish to mention something in the GNU Emacs source, show the line of code with a few lines of context. Don't just give a line number. The line numbers in the development sources don't match those in your sources. It would take extra work for the maintainers to determine what code is in your version at a given line number, and we could not be certain. Additional information from a C debugger such as GDB might enable someone to find a problem on a machine which he does not have available. If you don't know how to use GDB, please read the GDB manual--it is not very long, and using GDB is easy. You can find the GDB distribution, including the GDB manual in online form, in most of the same places you can find the Emacs distribution. However, you need to think when you collect the additional information if you want it to show what causes the bug. For example, many people send just a backtrace, but that is not very useful by itself. A simple backtrace with arguments often conveys little about what is happening inside GNU Emacs, because most of the arguments listed in the backtrace are pointers to Lisp objects. The numeric values of these pointers have no significance whatever; all that matters is the contents of the objects they point to (and most of the contents are themselves pointers). To provide useful information, you need to show the values of Lisp objects in Lisp notation. Do this for each variable which is a Lisp object, in several stack frames near the bottom of the stack. Look at the source to see which variables are Lisp objects, because the debugger thinks of them as integers. To show a variable's value in Lisp syntax, first print its value, then use the user-defined GDB command pr to print the Lisp object in Lisp syntax. (If you must use another debugger, call the function debug_print with the object as an argument.) The pr command is defined by the file `.gdbinit' in the `src' subdirectory, and it works only if you are debugging a running process (not with a core dump). To make Lisp errors stop Emacs and return to GDB, put a breakpoint at Fsignal. To find out which Lisp functions are running, using GDB, move up the stack, and each time you get to a frame for the function Ffuncall, type these GDB commands: p args pr To print the first argument that the function received, use these commands: p args[1] pr You can print the other arguments likewise. The argument nargs of Ffuncall says how many arguments Ffuncall received; these include the Lisp function itself and the arguments for that function. The file `src/.gdbinit' includes several other commands that are useful for examining the data types and contents of Lisp objects. These commands work at a lower level than pr, and are less convenient, but they may work even when pr does not. If the symptom of the bug is that Emacs fails to respond, don't assume Emacs is "hung"---it may instead be in an infinite loop. To find out which, make the problem happen under GDB and stop Emacs once it is not responding. (If Emacs is using X Windows directly, you can stop Emacs by typing C-c at the GDB job.) Then try stepping with `step'. If Emacs is hung, the `step' command won't return. If it is looping, `step' will return. If this shows Emacs is hung in a system call, stop it again and examine the arguments of the call. In your bug report, state exactly where in the source the system call is, and what the arguments are. If Emacs is in an infinite loop, please determine where the loop starts and ends. The easiest way to do this is to use the GDB command `finish'. Each time you use it, Emacs resumes execution until it exits one stack frame. Keep typing `finish' until it doesn't return--that means the infinite loop is in the stack frame which you just tried to finish. Stop Emacs again, and use `finish' repeatedly again until you get back to that frame. Then use `next' to step through that frame. By stepping, you will see where the loop starts and ends. Also please examine the data being used in the loop and try to determine why the loop does not exit when it should. Include all of this information in your bug report. Here are some things that are not necessary in a bug report: A description of the envelope of the bug--this is not necessary for a reproducible bug. Often people who encounter a bug spend a lot of time investigating which changes to the input file will make the bug go away and which changes will not affect it. This is often time-consuming and not very useful, because the way we will find the bug is by running a single example under the debugger with breakpoints, not by pure deduction from a series of examples. You might as well save time by not searching for additional examples. Of course, if you can find a simpler example to report instead of the original one, that is a convenience. Errors in the output will be easier to spot, running under the debugger will take less time, etc. However, simplification is not vital; if you can't do this or don't have time to try, please report the bug with your original test case. A system-call trace of Emacs execution. System-call traces are very useful for certain special kinds of debugging, but in most cases they give little useful information. It is therefore strange that many people seem to think that the way to report information about a crash is to send a system-call trace. Perhaps this is a habit formed from experience debugging programs that don't have source code or debugging symbols. In most programs, a backtrace is normally far, far more informative than a system-call trace. Even in Emacs, a simple backtrace is generally more informative, though to give full information you should supplement the backtrace by displaying variable values and printing them as Lisp objects with pr (see above). A patch for the bug. A patch for the bug is useful if it is a good one. But don't omit the other information that a bug report needs, such as the test case, on the assumption that a patch is sufficient. We might see problems with your patch and decide to fix the problem another way, or we might not understand it at all. And if we can't understand what bug you are trying to fix, or why your patch should be an improvement, we mustn't install it. A guess about what the bug is or what it depends on. Such guesses are usually wrong. Even experts can't guess right about such things without first using the debugger to find the facts.
| Prev | Home | Next |
| Up |