Commit Graph

143 Commits

Author SHA1 Message Date
David E. O'Brien
9ceffc28f3 Fix GNU coding violations. 2002-09-18 16:20:49 +00:00
David E. O'Brien
71791a1d60 Stop using 3 distinct and far varied coding styles in a single file.
Pretend to actually adhere to the GNU coding standards.
2002-09-18 16:14:36 +00:00
Peter Wemm
0f258b75d3 Recognize frames for lcall_syscall and int0x80_syscall.
Reminded by:	bde
2002-07-28 18:46:49 +00:00
David E. O'Brien
e2a86a90af Make our i386 hacks in a less vendor file disturbing manner. 2002-07-10 07:48:45 +00:00
David E. O'Brien
2b1f5a4271 Stop trying to reimplement libkvm on i386. Instead use it directly and
consistently on all platforms.

Submitted by:	gallatin
2002-07-10 06:58:25 +00:00
David E. O'Brien
bad0e574ca Target specific FreeBSD kgdb hacks. 2002-07-10 06:40:03 +00:00
David E. O'Brien
68464ed689 Dike out bits specific to i386. This pretty much means no kgdb for
non-i386 platforms.

I would however like to see a shared file here.  If a function or two cannot
be shared we should create ${TARGET_ARCH}/kvm-fbsd-${TARGET_ARCH}.c.
2002-06-30 04:47:43 +00:00
David E. O'Brien
a9b6be3e9e Bmake bits for GDB 5.2. 2002-06-30 03:24:05 +00:00
David E. O'Brien
c4819db060 I don't see any reason to conditionalize this. 2002-06-29 17:55:17 +00:00
David E. O'Brien
cc5dcb202c Bmake bits for GDB 5.2. 2002-06-29 16:52:15 +00:00
David E. O'Brien
b73f3861c3 Best guess at configurations for our other arches. (totally untested) 2002-06-29 03:18:28 +00:00
David E. O'Brien
2587e03b72 Allow to compile and at least startup on sparc64. 2002-06-29 03:16:40 +00:00
David E. O'Brien
7b726fea14 Bmake bits for GDB 5.2. 2002-06-29 01:16:01 +00:00
David E. O'Brien
b1ae3e66e6 NO! We DON'T wany any libintl action here. 2002-06-28 01:33:53 +00:00
David E. O'Brien
99c8489d5b Things Gdb 5.2 wants to know about us. 2002-06-28 00:34:05 +00:00
David E. O'Brien
22cb25a945 When linking staticaly libtermcap is a postrequisite of libreadline.
Otherwise `tgoto' (only used by libreadline) isn't resolved.

Submitted by:	bde
2002-06-22 17:07:45 +00:00
David E. O'Brien
457170e99c We need to link with libiberty.a before libreadline.so. Both supply
xmalloc() and xrealloc() and the mixed usage of xmalloc in some .c's from
libiberty.a and other .c's from libreadline.so produces an unusable binary
on the Alpha.

While I am here, preventatively move other libs in the link order.

Submitted by:	gallatin
2002-06-21 19:05:20 +00:00
Bruce Evans
6c37e4212e phsyical -> physical
Submitted by:	Andre Guibert de Bruet <andy@siliconlandmark.com>
2002-05-27 13:08:08 +00:00
David E. O'Brien
ecdd3d1cab This is only WARNS=0 clean now. 2002-05-19 18:25:41 +00:00
David E. O'Brien
4728018172 This is enough to get thru a cross-build. 2002-05-18 18:27:09 +00:00
David E. O'Brien
3ce5798474 Segregate MD weirdness a little bit more. Improve cross buildability. 2002-05-18 04:59:57 +00:00
Mark Peek
117af193f6 Make our version tag unique due to our local changes.
PR:		19733
Submitted by:	sheldonh
MFC after:	3 days
2002-05-14 18:17:59 +00:00
David E. O'Brien
596d4501ed Use PTRACE_ARG3_TYPE of caddr_t.
Obtained from:	gdb 5.2
2002-04-11 20:42:13 +00:00
David E. O'Brien
6fc7b42212 Enhance GDB's abillity WRT stack traces where signal handlers are involved.
Correct backtrace was made more complex when the new signal trampoline
was introduced to support more than 32 signals, while keeping a modified
version of the old signal trampoline.

The 'where' command will now show:

	#2  <signal handler called>

where appropiate.

Submitted by:	Tor.Egge@fast.no
2002-03-27 04:55:43 +00:00
David E. O'Brien
194788fafc GDB 4.18 does not use the mmalloc library by default. So GC a little. 2002-03-23 17:15:49 +00:00
David E. O'Brien
8c157f6041 Fix setting of CLEANFILES. 2002-03-23 04:52:02 +00:00
David E. O'Brien
d27504cd28 Fix unaligned access in the GDB binary on Alpha.
Submitted by:	gallatin
2002-03-16 19:33:23 +00:00
David E. O'Brien
2cadc0a98e Turn up WARNS as high as possible.
Tested by:	gcc31, gcc32, /usr/bin/cc
2002-03-13 05:14:31 +00:00
Daniel Eischen
68ace64ebe Sync this up to recent changes to our (userland) thread structure. 2002-02-23 19:28:01 +00:00
Andrew Gallatin
c48944beea Fix kgdb in the face of the last round of KSE commits.
The alpha world may actually build now..
2002-02-18 14:13:59 +00:00
Daniel Eischen
a06523b4f6 Don't rely on <sys/signal.h> to include <sys/ucontext.h>. 2002-02-17 17:19:14 +00:00
Peter Wemm
489b85b027 Add missing & 2002-02-08 04:17:33 +00:00
Peter Wemm
656c73dbdf Make this part compile and try and use the p_threads tailq. It might work
but I am not sure, I haven't been able to compile the rest of gdb yet.
2002-02-08 03:19:56 +00:00
John Baldwin
b0492e310b Catch up to the globaldata -> pcpu changes. 2001-12-12 21:15:30 +00:00
Peter Wemm
8dc7644269 *** empty log message *** 2001-12-11 07:04:48 +00:00
Bruce Evans
43ea907a42 Backed out "Compensate for header dethreading [mistakes]" mistakes in
alpha files too.
2001-10-13 04:38:46 +00:00
Bruce Evans
6eabd84580 Compensate for "Compensate for header dethreading" by backing it out. 2001-10-10 17:48:44 +00:00
Ian Dowse
3c7bcedd06 Remove the Xresume* labels from the i386 interrupt handlers; the
code in ipl.s and icu_ipl.s that used them was removed when the
interrupt thread system was committed. Debuggers also knew about
Xresume* because these labels hide the real names of the interrupt
handlers (Xintr*), and debuggers need to special-case interrupt
handlers to get the interrupt frame.

Both gdb and ddb will now use the Xintr* and Xfastintr* symbols to
detect interrupt frames. Fast interrupt frames were never identified
correctly before, so this fixes the problem of the running stack
frame getting lost in a ddb or gdb trace generated from a fast
interrupt - e.g. when debugging a simple infinite loop in the kernel
using a serial console, the frame containing the loop would never
appear in a gdb or ddb trace.

Reviewed by:	jhb, bde
2001-10-09 19:54:52 +00:00
Ian Dowse
e5cef9b61f Catch up with the SMPng reduced interrupt frame size. The corresponding
change was made to DDB months ago (i386/i386/db_trace.c revision 1.37).

Reviewed by:	bde
2001-10-08 12:46:01 +00:00
Peter Wemm
d1c276c383 Fix alpha gdb -k on "live" kernels. Use offsetof() instead of some evil
hand-rolled macros to do the same thing.
2001-09-20 06:31:23 +00:00
Bill Fenner
9c52cf5785 Don't try to dereference a kernel pointer in userland; use offsetof()
to get the right address.

This fixes kernel GDB after KSE2.

Reviewed by:	jhb, jake
2001-09-19 18:42:19 +00:00
Peter Wemm
e433a4a84b Update this to compile on the Alpha. 2001-09-14 11:08:44 +00:00
Julian Elischer
b40ce4165d KSE Milestone 2
Note ALL MODULES MUST BE RECOMPILED
make the kernel aware that there are smaller units of scheduling than the
process. (but only allow one thread per process at this time).
This is functionally equivalent to teh previousl -current except
that there is a thread associated with each process.

Sorry john! (your next MFC will be a doosie!)

Reviewed by: peter@freebsd.org, dillon@freebsd.org

X-MFC after:    ha ha ha ha
2001-09-12 08:38:13 +00:00
Peter Wemm
fd131b1483 Dynamically adapt to kernbase changes on crashdumps, falling back to
KERNBASE if the "kernbase" symbol is not present on older kernels.
2001-08-24 09:12:04 +00:00
David E. O'Brien
fcd08e35ec Define _KERNEL as this grubs around where no userland should go. 2001-08-16 20:47:01 +00:00
Peter Wemm
da81118c1b Use the real structure names explicitly. Note that this is mostly for
reading old a.out core files, which are totally 100% non-understandable
to the gdb floating-point reader if you have SSE turned on.

This should be the last of the world build breakers...
2001-07-12 13:01:17 +00:00
Matt Jacob
8dde563eb3 like i386, add inclusion of lock.h 2001-05-03 06:33:33 +00:00
Mark Murray
573fbdc21f Compensate for dethreaded headers.
(I have no religion about how this was done. Follow-up commits welcome)
2001-05-01 09:08:09 +00:00
Jake Burkholder
aa17032fcb - Add #defines for the symbol names of the kernel interrupt, system
call and trap entry points so they're easy to find and change
- Use the cpuhead and allcpu list to locate globaldata for the current
  cpu, rather than SMP_prvspace or __globaldata
- Use offsets into struct globaldata directly to find per-cpu variables,
  rather than symbols in globals.o

Glanced at by:	peter
2001-01-10 18:15:25 +00:00
Peter Wemm
57a0ee63f0 Fix gdb -k after jake's most recent commit. The gd_XXX symbols are now
offsets in all cases, and we have to find the base address (&__globaldata)
ourselves for the UP case as well as SMP.
2001-01-07 05:08:39 +00:00