| Message ID | 1291491689-11982-1-git-send-email-weil@mail.berlios.de |
|---|---|
| State | Superseded |
| Headers | show |
Am 04.12.2010 20:41, schrieb Stefan Weil: > QEMU always uses POSIX format specifiers, even with mingw32. > > Therefore the old definitions of the PRI*64 macros were wrong. > They should be removed, but as long as the mingw32 system > include inttypes.h provides wrong definitions, too, > we correct them here. > > Cc: Blue Swirl <blauwirbel@gmail.com> > Signed-off-by: Stefan Weil <weil@mail.berlios.de> > --- > qemu-common.h | 16 ++++++++++++---- > 1 files changed, 12 insertions(+), 4 deletions(-) > > diff --git a/qemu-common.h b/qemu-common.h > index de82c2e..c739f45 100644 > --- a/qemu-common.h > +++ b/qemu-common.h > @@ -96,10 +96,18 @@ static inline char *realpath(const char *path, > char *resolved_path) > return resolved_path; > } > > -#define PRId64 "I64d" > -#define PRIx64 "I64x" > -#define PRIu64 "I64u" > -#define PRIo64 "I64o" > +/* inttypes.h (mingw32) provides wrong definitions, so fix them here. */ > +/* TODO: remove this workaround as soon as mingw32 is fixed. */ > + > +#undef PRId64 > +#undef PRIx64 > +#undef PRIu64 > +#undef PRIo64 > + > +#define PRId64 "lld" > +#define PRIx64 "llx" > +#define PRIu64 "llu" > +#define PRIo64 "llo" > #endif > > /* FIXME: Remove NEED_CPU_H. */ What about this patch? It is still missing in QEMU git master. Kind regards, Stefan
On Sun, Jan 30, 2011 at 9:27 PM, Stefan Weil <weil@mail.berlios.de> wrote: > Am 04.12.2010 20:41, schrieb Stefan Weil: >> >> QEMU always uses POSIX format specifiers, even with mingw32. >> >> Therefore the old definitions of the PRI*64 macros were wrong. >> They should be removed, but as long as the mingw32 system >> include inttypes.h provides wrong definitions, too, >> we correct them here. >> >> Cc: Blue Swirl <blauwirbel@gmail.com> >> Signed-off-by: Stefan Weil <weil@mail.berlios.de> >> --- >> qemu-common.h | 16 ++++++++++++---- >> 1 files changed, 12 insertions(+), 4 deletions(-) >> >> diff --git a/qemu-common.h b/qemu-common.h >> index de82c2e..c739f45 100644 >> --- a/qemu-common.h >> +++ b/qemu-common.h >> @@ -96,10 +96,18 @@ static inline char *realpath(const char *path, char >> *resolved_path) >> return resolved_path; >> } >> >> -#define PRId64 "I64d" >> -#define PRIx64 "I64x" >> -#define PRIu64 "I64u" >> -#define PRIo64 "I64o" >> +/* inttypes.h (mingw32) provides wrong definitions, so fix them here. */ >> +/* TODO: remove this workaround as soon as mingw32 is fixed. */ >> + >> +#undef PRId64 >> +#undef PRIx64 >> +#undef PRIu64 >> +#undef PRIo64 >> + >> +#define PRId64 "lld" >> +#define PRIx64 "llx" >> +#define PRIu64 "llu" >> +#define PRIo64 "llo" >> #endif >> >> /* FIXME: Remove NEED_CPU_H. */ > > What about this patch? It is still missing in QEMU git master. It would appear to suppress quite a few warnings about formats. But on my version of inttypes.h there is the following comment: /* 7.8.1 Macros for format specifiers * * MS runtime does not yet understand C9x standard "ll" * length specifier. It appears to treat "ll" as "l". * The non-standard I64 length specifier causes warning in GCC, * but understood by MS runtime functions. */ So is this change OK after all?
Am 30.01.2011 22:39, schrieb Blue Swirl: > On Sun, Jan 30, 2011 at 9:27 PM, Stefan Weil <weil@mail.berlios.de> wrote: >> Am 04.12.2010 20:41, schrieb Stefan Weil: >>> >>> QEMU always uses POSIX format specifiers, even with mingw32. >>> >>> Therefore the old definitions of the PRI*64 macros were wrong. >>> They should be removed, but as long as the mingw32 system >>> include inttypes.h provides wrong definitions, too, >>> we correct them here. >>> >>> Cc: Blue Swirl <blauwirbel@gmail.com> >>> Signed-off-by: Stefan Weil <weil@mail.berlios.de> >>> --- >>> qemu-common.h | 16 ++++++++++++---- >>> 1 files changed, 12 insertions(+), 4 deletions(-) >>> >>> diff --git a/qemu-common.h b/qemu-common.h >>> index de82c2e..c739f45 100644 >>> --- a/qemu-common.h >>> +++ b/qemu-common.h >>> @@ -96,10 +96,18 @@ static inline char *realpath(const char *path, char >>> *resolved_path) >>> return resolved_path; >>> } >>> >>> -#define PRId64 "I64d" >>> -#define PRIx64 "I64x" >>> -#define PRIu64 "I64u" >>> -#define PRIo64 "I64o" >>> +/* inttypes.h (mingw32) provides wrong definitions, so fix them >>> here. */ >>> +/* TODO: remove this workaround as soon as mingw32 is fixed. */ >>> + >>> +#undef PRId64 >>> +#undef PRIx64 >>> +#undef PRIu64 >>> +#undef PRIo64 >>> + >>> +#define PRId64 "lld" >>> +#define PRIx64 "llx" >>> +#define PRIu64 "llu" >>> +#define PRIo64 "llo" >>> #endif >>> >>> /* FIXME: Remove NEED_CPU_H. */ >> >> What about this patch? It is still missing in QEMU git master. > > It would appear to suppress quite a few warnings about formats. But on > my version of inttypes.h there is the following comment: > /* 7.8.1 Macros for format specifiers > * > * MS runtime does not yet understand C9x standard "ll" > * length specifier. It appears to treat "ll" as "l". > * The non-standard I64 length specifier causes warning in GCC, > * but understood by MS runtime functions. > */ > So is this change OK after all? Yes, it is. MS runtime indeed does not understand "%lld" and similar format specifiers. Mingw does, because it replaces the printf family functions by inline functions which call __mingw_vfprintf as soon as __USE_MINGW_ANSI_STDIO is defined. If your MinGW stdio.h does not use __USE_MINGW_ANSI_STDIO, it is too old. QEMU defines __USE_MINGW_ANSI_STDIO, and __mingw_vfprintf understands C9x standard length specifiers. Stefan
On Sun, Jan 30, 2011 at 9:50 PM, Stefan Weil <weil@mail.berlios.de> wrote: > Am 30.01.2011 22:39, schrieb Blue Swirl: >> >> On Sun, Jan 30, 2011 at 9:27 PM, Stefan Weil <weil@mail.berlios.de> wrote: >>> >>> Am 04.12.2010 20:41, schrieb Stefan Weil: >>>> >>>> QEMU always uses POSIX format specifiers, even with mingw32. >>>> >>>> Therefore the old definitions of the PRI*64 macros were wrong. >>>> They should be removed, but as long as the mingw32 system >>>> include inttypes.h provides wrong definitions, too, >>>> we correct them here. >>>> >>>> Cc: Blue Swirl <blauwirbel@gmail.com> >>>> Signed-off-by: Stefan Weil <weil@mail.berlios.de> >>>> --- >>>> qemu-common.h | 16 ++++++++++++---- >>>> 1 files changed, 12 insertions(+), 4 deletions(-) >>>> >>>> diff --git a/qemu-common.h b/qemu-common.h >>>> index de82c2e..c739f45 100644 >>>> --- a/qemu-common.h >>>> +++ b/qemu-common.h >>>> @@ -96,10 +96,18 @@ static inline char *realpath(const char *path, char >>>> *resolved_path) >>>> return resolved_path; >>>> } >>>> >>>> -#define PRId64 "I64d" >>>> -#define PRIx64 "I64x" >>>> -#define PRIu64 "I64u" >>>> -#define PRIo64 "I64o" >>>> +/* inttypes.h (mingw32) provides wrong definitions, so fix them here. >>>> */ >>>> +/* TODO: remove this workaround as soon as mingw32 is fixed. */ >>>> + >>>> +#undef PRId64 >>>> +#undef PRIx64 >>>> +#undef PRIu64 >>>> +#undef PRIo64 >>>> + >>>> +#define PRId64 "lld" >>>> +#define PRIx64 "llx" >>>> +#define PRIu64 "llu" >>>> +#define PRIo64 "llo" >>>> #endif >>>> >>>> /* FIXME: Remove NEED_CPU_H. */ >>> >>> What about this patch? It is still missing in QEMU git master. >> >> It would appear to suppress quite a few warnings about formats. But on >> my version of inttypes.h there is the following comment: >> /* 7.8.1 Macros for format specifiers >> * >> * MS runtime does not yet understand C9x standard "ll" >> * length specifier. It appears to treat "ll" as "l". >> * The non-standard I64 length specifier causes warning in GCC, >> * but understood by MS runtime functions. >> */ >> So is this change OK after all? > > Yes, it is. MS runtime indeed does not understand "%lld" > and similar format specifiers. > > Mingw does, because it replaces the printf family functions > by inline functions which call __mingw_vfprintf as soon > as __USE_MINGW_ANSI_STDIO is defined. If your MinGW stdio.h > does not use __USE_MINGW_ANSI_STDIO, it is too old. Well, mine doesn't. The change was introduced in mingw-runtime 3.15, which was released in September 2008, but Debian still hasn't updated from 3.13. Maybe other distros are not so lagging and someone who wishes to build QEMU on Windows is not pampered with distro support for MinGW anyway. Perhaps a configure time check should be added? > QEMU defines __USE_MINGW_ANSI_STDIO, and __mingw_vfprintf > understands C9x standard length specifiers. BTW, MinGW FAQ page http://www.mingw.org/wiki/FAQ still mentions that %ll formats are not supported.
On 30/01/11 22:14, Blue Swirl wrote: > Well, mine doesn't. The change was introduced in mingw-runtime 3.15, > which was released in September 2008, but Debian still hasn't updated > from 3.13. Maybe other distros are not so lagging and someone who > wishes to build QEMU on Windows is not pampered with distro support > for MinGW anyway. Perhaps a configure time check should be added? FWIW anyone using MingW now should be using the new GCC 4.5 based builds with the associated runtime rather than anything that old. I suspect most people using MingW for anything serious will be doing this already. I think adding the configure check would be the best solution here. >> QEMU defines __USE_MINGW_ANSI_STDIO, and __mingw_vfprintf >> understands C9x standard length specifiers. > BTW, MinGW FAQ page http://www.mingw.org/wiki/FAQ still mentions that > %ll formats are not supported. Since MingW uses the MSVCRT then by default it won't accept %ll, but my understanding is that Stefan is correct, i.e. defining __USE_MINGW_ANSI_STDIO causes MingW to use an internal standards-compliant implementation instead of the in-built MSVCRT version that *does* understand these extras. HTH, Mark.
Am 30.01.2011 23:14, schrieb Blue Swirl: > On Sun, Jan 30, 2011 at 9:50 PM, Stefan Weil <weil@mail.berlios.de> wrote: >> Am 30.01.2011 22:39, schrieb Blue Swirl: >>> It would appear to suppress quite a few warnings about formats. But on >>> my version of inttypes.h there is the following comment: >>> /* 7.8.1 Macros for format specifiers >>> * >>> * MS runtime does not yet understand C9x standard "ll" >>> * length specifier. It appears to treat "ll" as "l". >>> * The non-standard I64 length specifier causes warning in GCC, >>> * but understood by MS runtime functions. >>> */ >>> So is this change OK after all? >> >> Yes, it is. MS runtime indeed does not understand "%lld" >> and similar format specifiers. >> >> Mingw does, because it replaces the printf family functions >> by inline functions which call __mingw_vfprintf as soon >> as __USE_MINGW_ANSI_STDIO is defined. If your MinGW stdio.h >> does not use __USE_MINGW_ANSI_STDIO, it is too old. > > Well, mine doesn't. The change was introduced in mingw-runtime 3.15, > which was released in September 2008, but Debian still hasn't updated > from 3.13. Maybe other distros are not so lagging and someone who > wishes to build QEMU on Windows is not pampered with distro support > for MinGW anyway. Perhaps a configure time check should be added? >> QEMU defines __USE_MINGW_ANSI_STDIO, and __mingw_vfprintf >> understands C9x standard length specifiers. > > BTW, MinGW FAQ page http://www.mingw.org/wiki/FAQ still mentions that > %ll formats are not supported. The FAQ says "You should use %I64 instead of %ll when using msvcrt". => If you don't use msvcrt for printf, you no longer have to use %I64. Even with latest versions, MinGW's support for ANSI stdio is still incomplete - inttypes.h is one example, others are missing GCC atributes and missing ANSI scanf support. I'll add a page to QEMU's wiki which lists all changes needed. Regards, Stefan
diff --git a/qemu-common.h b/qemu-common.h index de82c2e..c739f45 100644 --- a/qemu-common.h +++ b/qemu-common.h @@ -96,10 +96,18 @@ static inline char *realpath(const char *path, char *resolved_path) return resolved_path; } -#define PRId64 "I64d" -#define PRIx64 "I64x" -#define PRIu64 "I64u" -#define PRIo64 "I64o" +/* inttypes.h (mingw32) provides wrong definitions, so fix them here. */ +/* TODO: remove this workaround as soon as mingw32 is fixed. */ + +#undef PRId64 +#undef PRIx64 +#undef PRIu64 +#undef PRIo64 + +#define PRId64 "lld" +#define PRIx64 "llx" +#define PRIu64 "llu" +#define PRIo64 "llo" #endif /* FIXME: Remove NEED_CPU_H. */
QEMU always uses POSIX format specifiers, even with mingw32. Therefore the old definitions of the PRI*64 macros were wrong. They should be removed, but as long as the mingw32 system include inttypes.h provides wrong definitions, too, we correct them here. Cc: Blue Swirl <blauwirbel@gmail.com> Signed-off-by: Stefan Weil <weil@mail.berlios.de> --- qemu-common.h | 16 ++++++++++++---- 1 files changed, 12 insertions(+), 4 deletions(-)