Re: [PATCH 0/2] dis: Introduce the -f option

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 




----- Original Message -----
> 
> 
> ----- Original Message -----
> > Dave,
> > 
> > Since our last discussion [1], a further change to the "consolidation"
> > patch has been made. The assumption now is that the text address generated
> > by gdb is _always_ in base 16 form. Hence the if clause to check buf2 is
> > prefixed with 0x has been removed as indicated in the hunk below:
> > 
> > -				if (STRNEQ(buf2, "0x"))
> > -					extract_hex(buf2, &curaddr, ':', TRUE);
> > +			extract_hex(buf2, &curaddr, ':', TRUE);
> > 
> > Please let me know if this assumption is valid.
> 
> I don't think it is.
> 
> The reason for the "0x" check is that gdb may split a single line of disassembly
> into two lines, or at least it used to.  As I recall, I think I later added a gdb
> option initialization to prevent that from happening (?, but nonetheless, it could
> still be undone by the user.  So it's best to leave the check in place.

BTW, that's what all the ":" checks are there for, both in cmd_dis() and in the
per-arch dis_filter functions.  When gdb would split the line, it would end the
first line with the colon following the text address and offset.

Dave



> 
> Dave
> 
> > 
> > With optimisation level 2 (default) the following warning is displayed:
> > 
> > $ make warn
> > TARGET: X86_64
> >  CRASH: 7.1.2
> >    GDB: 7.6
> > 
> > cc -c -g -DX86_64  -DGDB_7_6  build_data.c -Wall -O2 -Wstrict-prototypes
> > -Wmissing-prototypes -fstack-protector -Wformat-security
> > cc -c -g -DX86_64  -DGDB_7_6  kernel.c -Wall -O2 -Wstrict-prototypes
> > -Wmissing-prototypes -fstack-protector -Wformat-security
> > kernel.c: In function ‘cmd_dis’:
> > kernel.c:1663:8: warning: ‘target’ may be used uninitialized in this
> > function
> > [-Wmaybe-uninitialized]
> >      if (curaddr != target)
> >         ^
> > Note that this is not seen when optimisation is disabled. To work around
> > this (what I suspect is a false positive), where target is declared it is
> > now initialised to zero even though its value is not considered if forward
> > or reverse is not set.
> > 
> 
> That's not a problem -- there are many place where that's done to work around
> that kind of warning.
> 
> Thanks,
>   Dave
> 
> 
> > [1]:
> > https://www.redhat.com/archives/crash-utility/2015-August/msg00025.html
> > 
> > Aaron Tomlin (2):
> >   dis: Consolidate cmd_dis
> >   dis: Introduce the -f option
> > 
> >  help.c   |   2 +
> >  kernel.c | 287
> >  +++++++++++++++++++++++----------------------------------------
> >  2 files changed, 106 insertions(+), 183 deletions(-)
> > 
> > --
> > 2.4.3
> > 
> > --
> > Crash-utility mailing list
> > Crash-utility@xxxxxxxxxx
> > https://www.redhat.com/mailman/listinfo/crash-utility
> 
> --
> Crash-utility mailing list
> Crash-utility@xxxxxxxxxx
> https://www.redhat.com/mailman/listinfo/crash-utility

--
Crash-utility mailing list
Crash-utility@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/crash-utility




[Index of Archives]     [Fedora Development]     [Fedora Desktop]     [Fedora SELinux]     [Yosemite News]     [KDE Users]     [Fedora Tools]

 

Powered by Linux