* bug#8564: 24.0.50; bootstrap on windows 7: `gl-stamp' target failed @ 2011-04-27 5:20 William Xu 2011-04-27 10:29 ` Eli Zaretskii 0 siblings, 1 reply; 6+ messages in thread From: William Xu @ 2011-04-27 5:20 UTC (permalink / raw) To: 8564 Here is the last build log: mkdir "oo-spd" mkdir "oo-spd/i386" echo oo-spd/i386 > stamp_BLD rm gl-tmp "../lib-src/oo-spd/i386/make-docfile" -d . -g dosfns.o msdos.o xterm.o xfns.o xmenu.o xselect.o xrdb.o xsmfns.o dbusbind.o dosfns.o msdos.o xterm.o xfns.o xmenu.o xselect.o xrdb.o xsmfns.o fringe.o image.o fontset.o menu.o dbusbind.o w32.o w32console.o w32fns.o w32heap.o w32inevt.o w32menu.o w32proc.o w32reg.o w32select.o w32term.o w32xfns.o font.o w32font.o w32uniscribe.o dispnew.o frame.o scroll.o xdisp.o window.o bidi.o charset.o coding.o category.o ccl.o character.o chartab.o cm.o term.o terminal.o xfaces.o emacs.o keyboard.o macros.o keymap.o sysdep.o buffer.o filelock.o insdel.o marker.o minibuf.o fileio.o dired.o cmds.o casetab.o casefiddle.o indent.o search.o regex.o undo.o alloc.o data.o doc.o editfns.o callint.o eval.o floatfns.o fns.o print.o lread.o syntax.o bytecode.o proces s.o callproc.o unexw32.o region-cache.o sound.o atimer.o doprnt.o intervals.o textprop.o composite.o > gl-tmp cmd /c "fc /b gl-tmp globals.h >nul 2>&1 || cp -f gl-tmp globals.h" Microsoft Windows [Version 6.1.7600] Copyright (c) 2009 Microsoft Corporation. All rights reserved. l:\bin\emacs\src> Executing: cmd /c "fc /b gl-tmp globals.h >nul 2>&1 || cp -f gl-tmp globals.h" directly on cmd.exe seems OK. I don't know why it fails in bootstrapping? I simply replace that line with: cp -f gl-tmp globals.h # Wouldn't this be simpler & good enough? Now it bootstraps. -- William http://xwl.appspot.com ^ permalink raw reply [flat|nested] 6+ messages in thread
* bug#8564: 24.0.50; bootstrap on windows 7: `gl-stamp' target failed 2011-04-27 5:20 bug#8564: 24.0.50; bootstrap on windows 7: `gl-stamp' target failed William Xu @ 2011-04-27 10:29 ` Eli Zaretskii 2011-04-28 2:24 ` William Xu 0 siblings, 1 reply; 6+ messages in thread From: Eli Zaretskii @ 2011-04-27 10:29 UTC (permalink / raw) To: William Xu; +Cc: 8564 > From: William Xu <william.xwl@gmail.com> > Date: Wed, 27 Apr 2011 13:20:17 +0800 > > Here is the last build log: > > mkdir "oo-spd" > mkdir "oo-spd/i386" > echo oo-spd/i386 > stamp_BLD > rm gl-tmp > "../lib-src/oo-spd/i386/make-docfile" -d . -g dosfns.o msdos.o xterm.o xfns.o xmenu.o xselect.o xrdb.o xsmfns.o dbusbind.o dosfns.o msdos.o xterm.o xfns.o xmenu.o xselect.o xrdb.o xsmfns.o fringe.o image.o fontset.o menu.o dbusbind.o w32.o w32console.o w32fns.o w32heap.o w32inevt.o w32menu.o w32proc.o w32reg.o w32select.o w32term.o w32xfns.o font.o w32font.o w32uniscribe.o dispnew.o frame.o scroll.o xdisp.o window.o bidi.o charset.o coding.o category.o ccl.o character.o chartab.o cm.o term.o terminal.o xfaces.o emacs.o keyboard.o macros.o keymap.o sysdep.o buffer.o filelock.o insdel.o marker.o minibuf.o fileio.o dired.o cmds.o casetab.o casefiddle.o indent.o search.o regex.o undo.o alloc.o data.o doc.o editfns.o callint.o eval.o floatfns.o fns.o print.o lread.o syntax.o bytecode.o proc ess.o callproc.o unexw32.o region-cache.o sound.o atimer.o doprnt.o intervals.o textprop.o composite.o > gl-tmp > cmd /c "fc /b gl-tmp globals.h >nul 2>&1 || cp -f gl-tmp globals.h" > Microsoft Windows [Version 6.1.7600] > Copyright (c) 2009 Microsoft Corporation. All rights reserved. > > l:\bin\emacs\src> Is this something new? Did you ever bootstrap on that machine, or on Windows 7 in general? After you bootstrap, does the same command work for a normal (i.e., non-bootstrap) build, or does it fail in the same way? Finally, what version of Make did you use for that bootstrap? What does "make --version" display? Also, did you use any kind of Unixy shell (Make announces at the very beginning of the build process whioch shell it uses)? > cp -f gl-tmp globals.h # Wouldn't this be simpler & good enough? No, it's not good enough. `cp' updates the time stamp of globals.h each time you build, and therefore all the C files will be recompiled, because they all include globals.h, even though the contents of globals.h rarely changes. That kinda makes Make redundant, since its main goal is to avoid unnecessary compilations... The command that invokes `fc' only updates globals.h if the new version has different contents. IOW, this is the Windows way of saying move-if-change. ^ permalink raw reply [flat|nested] 6+ messages in thread
* bug#8564: 24.0.50; bootstrap on windows 7: `gl-stamp' target failed 2011-04-27 10:29 ` Eli Zaretskii @ 2011-04-28 2:24 ` William Xu 2011-04-28 4:53 ` Eli Zaretskii 0 siblings, 1 reply; 6+ messages in thread From: William Xu @ 2011-04-28 2:24 UTC (permalink / raw) To: bug-gnu-emacs Eli Zaretskii <eliz@gnu.org> writes: Is this something new? Did you ever bootstrap on that machine, or on Windows 7 in general? This is the first time i tried to compile emacs on Windows 7. After you bootstrap, does the same command work for a normal (i.e., non-bootstrap) build, or does it fail in the same way? Yes, if i revert my change. I modified some c files, and run following command: ,----[ make gl-stamp ] | cmd /c "fc /b gl-tmp globals.h >nul 2>&1 || cp -f gl-tmp globals.h" | Microsoft Windows [Version 6.1.7600] | Copyright (c) 2009 Microsoft Corporation. All rights reserved. | | L:\bin\emacs\src>mingw32-make: *** [gl-stamp] Error 66048 `---- Finally, what version of Make did you use for that bootstrap? What does "make --version" display? Also, did you use any kind of Unixy shell (Make announces at the very beginning of the build process whioch shell it uses)? ,----[ mingw32-make --version ] | GNU Make 3.82 | Built for i386-pc-mingw32 | Copyright (C) 2010 Free Software Foundation, Inc. | License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html> | This is free software: you are free to change and redistribute it. | There is NO WARRANTY, to the extent permitted by law. `---- I was compiling in on cmd.exe(i have some unix tools on %path%, though) The beginning: ,---- | L:\bin\emacs\src>mingw32-make gl-stamp | [Please ignore a syntax error on the next line - it is intentional] | /bin/sh: -c: line 0: unexpected EOF while looking for matching `"' | /bin/sh: -c: line 1: syntax error: unexpected end of file | rm gl-tmp `---- > cp -f gl-tmp globals.h # Wouldn't this be simpler & good enough? No, it's not good enough. `cp' updates the time stamp of globals.h each time you build, and therefore all the C files will be recompiled, because they all include globals.h, even though the contents of globals.h rarely changes. That kinda makes Make redundant, since its main goal is to avoid unnecessary compilations... The command that invokes `fc' only updates globals.h if the new version has different contents. IOW, this is the Windows way of saying move-if-change. How about using `diff' here? (considering commands like `cp' is already dependent? ) diff gl-tmp globals.h >nul 2>&1 || $(CP) gl-tmp globals.h -- William http://xwl.appspot.com ^ permalink raw reply [flat|nested] 6+ messages in thread
* bug#8564: 24.0.50; bootstrap on windows 7: `gl-stamp' target failed 2011-04-28 2:24 ` William Xu @ 2011-04-28 4:53 ` Eli Zaretskii 2011-04-28 6:19 ` William Xu 0 siblings, 1 reply; 6+ messages in thread From: Eli Zaretskii @ 2011-04-28 4:53 UTC (permalink / raw) To: William Xu; +Cc: bug-gnu-emacs > From: William Xu <william.xwl@gmail.com> > Date: Thu, 28 Apr 2011 10:24:26 +0800 > > After you bootstrap, does the same command work for a normal (i.e., > non-bootstrap) build, or does it fail in the same way? > > Yes, if i revert my change. I modified some c files, and run following > command: > > ,----[ make gl-stamp ] > | cmd /c "fc /b gl-tmp globals.h >nul 2>&1 || cp -f gl-tmp globals.h" > | Microsoft Windows [Version 6.1.7600] > | Copyright (c) 2009 Microsoft Corporation. All rights reserved. > | > | L:\bin\emacs\src>mingw32-make: *** [gl-stamp] Error 66048 > `---- So it fails in non-bootstrap case as well. Good, this makes things simpler. > I was compiling in on cmd.exe(i have some unix tools on %path%, though) > The beginning: > > ,---- > | L:\bin\emacs\src>mingw32-make gl-stamp > | [Please ignore a syntax error on the next line - it is intentional] > | /bin/sh: -c: line 0: unexpected EOF while looking for matching `"' > | /bin/sh: -c: line 1: syntax error: unexpected end of file > | rm gl-tmp > `---- Uh-oh! Is this the MSYS port of Bash, by any chance? Or maybe the Cygwin Bash? If so, can you remove sh.exe from PATH and try again? At least the MSYS Bash has a nasty habit of munging the command-line arguments if it "thinks" that those arguments are Unix-style /foo/bar file names. So my crystal ball says that it "thinks" "/c" in "cmd /c" is such a file name and changes it to something like "cmd L:/c", which of course totally confuses cmd.exe. You will see that nt/INSTALL advises against the MSYS shell: MSYS sh.exe also appears to cause various problems. If you have MSYS installed, try "make SHELL=cmd.exe" to force the use of cmd.exe instead of sh.exe. > > cp -f gl-tmp globals.h # Wouldn't this be simpler & good enough? > > No, it's not good enough. `cp' updates the time stamp of globals.h > each time you build, and therefore all the C files will be recompiled, > because they all include globals.h, even though the contents of > globals.h rarely changes. That kinda makes Make redundant, since its > main goal is to avoid unnecessary compilations... The command that > invokes `fc' only updates globals.h if the new version has different > contents. IOW, this is the Windows way of saying move-if-change. > > How about using `diff' here? (considering commands like `cp' is already > dependent? ) > diff gl-tmp globals.h >nul 2>&1 || $(CP) gl-tmp globals.h Using Diff means we ask the Windows users to have yet another non-standard package installed as a prerequisite to building Emacs. I would like to avoid that if possible; `fc' is a standard program on every Windows machine, and it does this particular job just fine. It is also unsafe to use this command without going through "cmd /c", because ported Unixy shells generally don't know about "NUL" being the Windows null device, while cmd.exe doesn't know about "/dev/null". Anyway, I think your problem is not with `fc', it's with "cmd /c". So using Diff won't solve it. ^ permalink raw reply [flat|nested] 6+ messages in thread
* bug#8564: 24.0.50; bootstrap on windows 7: `gl-stamp' target failed 2011-04-28 4:53 ` Eli Zaretskii @ 2011-04-28 6:19 ` William Xu 2011-04-28 17:26 ` Eli Zaretskii 0 siblings, 1 reply; 6+ messages in thread From: William Xu @ 2011-04-28 6:19 UTC (permalink / raw) To: bug-gnu-emacs Eli Zaretskii <eliz@gnu.org> writes: Uh-oh! Is this the MSYS port of Bash, by any chance? Or maybe the Cygwin Bash? If so, can you remove sh.exe from PATH and try again? At least the MSYS Bash has a nasty habit of munging the command-line arguments if it "thinks" that those arguments are Unix-style /foo/bar file names. So my crystal ball says that it "thinks" "/c" in "cmd /c" is such a file name and changes it to something like "cmd L:/c", which of course totally confuses cmd.exe. You will see that nt/INSTALL advises against the MSYS shell: MSYS sh.exe also appears to cause various problems. If you have MSYS installed, try "make SHELL=cmd.exe" to force the use of cmd.exe instead of sh.exe. OK. That is the problem! It is a MSYS bash, included in msys-git. Thanks. It seems kind of nasty, though. We reply on unix tools like cp, rm, etc, but they are often bundled together with sh. Maybe add the above note to "* For the impatient" section as well would be a good idea? -- William http://xwl.appspot.com ^ permalink raw reply [flat|nested] 6+ messages in thread
* bug#8564: 24.0.50; bootstrap on windows 7: `gl-stamp' target failed 2011-04-28 6:19 ` William Xu @ 2011-04-28 17:26 ` Eli Zaretskii 0 siblings, 0 replies; 6+ messages in thread From: Eli Zaretskii @ 2011-04-28 17:26 UTC (permalink / raw) To: William Xu; +Cc: 8564-done > From: William Xu <william.xwl@gmail.com> > Date: Thu, 28 Apr 2011 14:19:20 +0800 > > OK. That is the problem! It is a MSYS bash, included in msys-git. > Thanks. Thanks for helping me unlock this mystery. > It seems kind of nasty, though. We reply on unix tools like cp, rm, > etc, but they are often bundled together with sh. Only the Cygwin and MSYS versions of these tools come with sh. The GnuWin32 ports don't. > Maybe add the above note to "* For the impatient" section as well > would be a good idea? Done. I'm closing this bug. ^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2011-04-28 17:26 UTC | newest] Thread overview: 6+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2011-04-27 5:20 bug#8564: 24.0.50; bootstrap on windows 7: `gl-stamp' target failed William Xu 2011-04-27 10:29 ` Eli Zaretskii 2011-04-28 2:24 ` William Xu 2011-04-28 4:53 ` Eli Zaretskii 2011-04-28 6:19 ` William Xu 2011-04-28 17:26 ` Eli Zaretskii
Code repositories for project(s) associated with this public inbox https://git.savannah.gnu.org/cgit/emacs.git This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).