On Monday 25 September 2006 20:34, Jeff Dike wrote: > The UML/x86_64 headers were missing ptrace support for some segment > registers. The underlying problem was that the x86_64 kernel uses > user_regs_struct rather than the ptrace register definitions in ptrace. > This patch switches UML/x86_64 to using user_regs_struct for its > definitions of the host's registers.
> Signed-off-by: Jeff Dike <[EMAIL PROTECTED]> > Cc: Paolo 'Blaisorblade' Giarrusso <[EMAIL PROTECTED]> > Signed-off-by: Andrew Morton <[EMAIL PROTECTED]> > --- > > arch/um/include/sysdep-x86_64/ptrace.h | 43 +++++++++++++++++++++-- > arch/um/include/sysdep-x86_64/sc.h | 2 - > include/asm-um/ptrace-x86_64.h | 3 + > 3 files changed, 44 insertions(+), 4 deletions(-) > > diff -puN > arch/um/include/sysdep-x86_64/ptrace.h~uml-fix-missing-x86_64-register-defi >nitions arch/um/include/sysdep-x86_64/ptrace.h --- > a/arch/um/include/sysdep-x86_64/ptrace.h~uml-fix-missing-x86_64-register-de >finitions +++ a/arch/um/include/sysdep-x86_64/ptrace.h > @@ -50,6 +50,21 @@ > #define HOST_FS 25 > #define HOST_GS 26 > > +/* Also defined in asm/ptrace-x86_64.h, but not in libc headers. So, > these + * are already defined for kernel code, but not for userspace code. > + */ > +#ifndef FS_BASE > +/* These aren't defined in ptrace.h, but exist in struct user_regs_struct, > + * which is what x86_64 ptrace actually uses. > + */ > +#define FS_BASE (HOST_FS_BASE * sizeof(long)) > +#define GS_BASE (HOST_GS_BASE * sizeof(long)) > +#define DS (HOST_DS * sizeof(long)) > +#define ES (HOST_ES * sizeof(long)) > +#define FS (HOST_FS * sizeof(long)) > +#define GS (HOST_GS * sizeof(long)) > +#endif > + > #define REGS_FS_BASE(r) ((r)[HOST_FS_BASE]) > #define REGS_GS_BASE(r) ((r)[HOST_GS_BASE]) > #define REGS_DS(r) ((r)[HOST_DS]) > diff -puN > include/asm-um/ptrace-x86_64.h~uml-fix-missing-x86_64-register-definitions > include/asm-um/ptrace-x86_64.h --- > a/include/asm-um/ptrace-x86_64.h~uml-fix-missing-x86_64-register-definition >s +++ a/include/asm-um/ptrace-x86_64.h > @@ -16,12 +16,15 @@ > > #define HOST_AUDIT_ARCH AUDIT_ARCH_X86_64 > > +/* Also defined in sysdep/ptrace.h, so may already be defined. */ > +#ifndef FS_BASE > #define FS_BASE (21 * sizeof(unsigned long)) > #define GS_BASE (22 * sizeof(unsigned long)) > #define DS (23 * sizeof(unsigned long)) > #define ES (24 * sizeof(unsigned long)) > #define FS (25 * sizeof(unsigned long)) > #define GS (26 * sizeof(unsigned long)) > +#endif > > #define PT_REGS_RBX(r) UPT_RBX(&(r)->regs) > #define PT_REGS_RCX(r) UPT_RCX(&(r)->regs) The patch is ok for me, but frankly, this hunk could be further cleaned up - there is an awful hardcoded duplication of code which could be removed (the definition could be split away from <sysdep/ptrace.h> if inclusion order hell starts). -- Inform me of my mistakes, so I can keep imitating Homer Simpson's "Doh!". Paolo Giarrusso, aka Blaisorblade http://www.user-mode-linux.org/~blaisorblade Chiacchiera con i tuoi amici in tempo reale! http://it.yahoo.com/mail_it/foot/*http://it.messenger.yahoo.com ------------------------------------------------------------------------- Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV _______________________________________________ User-mode-linux-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel
