Remove all mentions of mips-idt. (was: Re: [05/10] watchpoint.exp: Clean up "mips-idt-*" specific re-initialization.)
Pedro Alves
palves@redhat.com
Thu Feb 2 15:25:00 GMT 2012
On 02/02/2012 02:05 PM, Pedro Alves wrote:
> On 01/30/2012 08:14 PM, Stan Shebs wrote:
>> On 1/26/12 6:19 AM, Pedro Alves wrote:
>>> I don't know if there's still anyone who cares about mips-idt. This cleans that
>>> target's specific re-initialization hacks, by doing them close to where a test
>>> starts running the program from scratch.
>>>
>>> If mips-idt is completely dead, we could instead get rid of all this stuff (it was
>>> introduced in 1993), which exists in several tests.
>>>
>>
>> IDT still claims to be selling the MIPS chips, but it's not so clear about the boards that the mips-idt config targets. Their site has board manuals dated as recently as 2006.
>
> Thanks.
>
>> Perhaps this would be a good candidate for official deprecation in 7.5 - I certainly wouldn't miss the funkiness that is remote-mips.c ... :-)
>
> Indeed, though I think remote-mips.c still has users (but for mips-elf only). Kevin brought it back from the dead not so long ago.
>
> This made me go do some archaeology. I noticed that:
>
> - mips-idt really means mips-idt-ecoff.
> - By gdb 4.12, mips-idt-ecoff selected the "idt" target -> config/mips/idt.mt
> - By gdb 5.0, mips-idt-ecoff selected the "embed" target -> config/mips/embed.mt (which reads: # Target: Big-endian mips board, typically an IDT.)
>
> So the idt target turned into a somewhat more generic embed target.
>
> - In http://sourceware.org/ml/gdb-patches/2004-11/msg00300.html
>
> 2004-11-13 Andrew Cagney <cagney@gnu.org>
>
> * regcache.h (deprecated_register_valid): Delete.
> (deprecated_registers): Delete.
> * regcache.c (deprecated_register_valid): Delete.
> (deprecated_registers): Delete.
> (deprecated_read_register_byte, _initialize_regcache)
> (deprecated_write_register_bytes, build_regcache): Update.
> * config/powerpc/ppc-sim.mt (TDEPFILES): Remove ppc-bdm.o and
> remote-sds.o.
> * config/powerpc/ppc-eabi.mt (TDEPFILES): Ditto.
> * config/mips/embed.mt (TDEPFILES): Remove remote-mips.o.
>
> Andrew removes remote-mips.o from embed.mt. This means that the boards in question,
> which talk the MIPS remote protocol, have been undebuggeable/untesteable since this
> commit, more than 9 years ago. We currently only support, in configure.tgt:
>
> mips*-*-elf)
> # Target: MIPS ELF
> gdb_target_obs="mips-tdep.o remote-mips.o"
> gdb_sim=../sim/mips/libsim.a
> ;;
> mips*-*-*)
> # Target: MIPS
> gdb_target_obs="mips-tdep.o"
> gdb_sim=../sim/mips/libsim.a
> ;;
>
>
> So I think we're pretty safe to zap any left over mips-idt bits immediately
> without going through a deprecation phase.
>
Like so. Tested on x86_64 Fedora 16, just in case.
gdb/doc/
2012-02-02 Pedro Alves <palves@redhat.com>
* gdb.texinfo (MIPS boards): Refer to mips-elf instead of
mips-idt-ecoff.
gdb/testsuite/
2012-02-02 Pedro Alves <palves@redhat.com>
* config/mips-idt.exp: Delete.
* gdb.base/a2-run.exp: Remove mips-idt xfails.
* gdb.base/bitfields.exp: Remove mips-idt restarts.
* gdb.base/break.exp: Remove mips-idt references.
* gdb.base/chng-syms.exp: Ditto.
* gdb.base/default.exp
* gdb.base/funcargs.exp (funcargs_reload): Delete.
(top level): Don't call it.
* gdb.base/opaque.exp: Remove mips-idt restarts.
* gdb.base/ptype.exp: Remove mips-idt xfails.
* gdb.base/scope.exp: Remove mips-idt restarts.
* gdb.base/sepdebug.exp: Remove mips-idt references.
* gdb.base/watchpoint.exp (maybe_clean_restart)
(maybe_reinitialize): Delete.
(test_disabling_watchpoints, test_disabling_watchpoints)
(test_watchpoint_triggered_in_syscall)
(test_watchpoint_and_breakpoint, test_inaccessible_watchpoint):
Don't call them.
---
gdb/doc/gdb.texinfo | 4 ++-
gdb/testsuite/config/mips-idt.exp | 16 ------------
gdb/testsuite/gdb.base/a2-run.exp | 4 +--
gdb/testsuite/gdb.base/bitfields.exp | 22 -----------------
gdb/testsuite/gdb.base/break.exp | 3 --
gdb/testsuite/gdb.base/chng-syms.exp | 7 -----
gdb/testsuite/gdb.base/default.exp | 10 ++------
gdb/testsuite/gdb.base/funcargs.exp | 30 -----------------------
gdb/testsuite/gdb.base/opaque.exp | 10 --------
gdb/testsuite/gdb.base/ptype.exp | 6 ++---
gdb/testsuite/gdb.base/scope.exp | 28 ---------------------
gdb/testsuite/gdb.base/sepdebug.exp | 3 --
gdb/testsuite/gdb.base/watchpoint.exp | 44 ---------------------------------
13 files changed, 8 insertions(+), 179 deletions(-)
delete mode 100644 gdb/testsuite/config/mips-idt.exp
diff --git a/gdb/doc/gdb.texinfo b/gdb/doc/gdb.texinfo
index 5738d14..083f30e 100644
--- a/gdb/doc/gdb.texinfo
+++ b/gdb/doc/gdb.texinfo
@@ -19190,7 +19190,7 @@ Show MicroBlaze-specific debugging level.
@cindex MIPS boards
@value{GDBN} can use the MIPS remote debugging protocol to talk to a
MIPS board attached to a serial line. This is available when
-you configure @value{GDBN} with @samp{--target=mips-idt-ecoff}.
+you configure @value{GDBN} with @samp{--target=mips-elf}.
@need 1000
Use these @value{GDBN} commands to specify the connection to your target board:
@@ -19295,7 +19295,7 @@ waiting for an acknowledgment of a packet with the @code{set
retransmit-timeout @var{seconds}} command. The default is 3 seconds.
You can inspect both values with @code{show timeout} and @code{show
retransmit-timeout}. (These commands are @emph{only} available when
-@value{GDBN} is configured for @samp{--target=mips-idt-ecoff}.)
+@value{GDBN} is configured for @samp{--target=mips-elf}.)
The timeout set by @code{set timeout} does not apply when @value{GDBN}
is waiting for your program to stop. In that case, @value{GDBN} waits
diff --git a/gdb/testsuite/config/mips-idt.exp b/gdb/testsuite/config/mips-idt.exp
deleted file mode 100644
index d3aa44e..0000000
--- a/gdb/testsuite/config/mips-idt.exp
+++ /dev/null
@@ -1,16 +0,0 @@
-# Copyright (C) 1993, 1997, 2007-2012 Free Software Foundation, Inc.
-
-# This program is free software; you can redistribute it and/or modify
-# it under the terms of the GNU General Public License as published by
-# the Free Software Foundation; either version 3 of the License, or
-# (at your option) any later version.
-#
-# This program is distributed in the hope that it will be useful,
-# but WITHOUT ANY WARRANTY; without even the implied warranty of
-# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
-# GNU General Public License for more details.
-#
-# You should have received a copy of the GNU General Public License
-# along with this program. If not, see <http://www.gnu.org/licenses/>.
-
-load_lib ../config/monitor.exp
diff --git a/gdb/testsuite/gdb.base/a2-run.exp b/gdb/testsuite/gdb.base/a2-run.exp
index 4e3b54b..e2119d4 100644
--- a/gdb/testsuite/gdb.base/a2-run.exp
+++ b/gdb/testsuite/gdb.base/a2-run.exp
@@ -107,7 +107,7 @@ if [istarget "*-*-vxworks*"] then {
verbose "Timeout is now $timeout seconds" 2
gdb_expect -re "$gdb_prompt $" {}
} else {
- setup_xfail "mips-idt-*" "arm-*-coff"
+ setup_xfail "arm-*-coff"
gdb_run_cmd 5
gdb_expect {
-re ".*120.*$gdb_prompt $"\
@@ -118,7 +118,6 @@ if [istarget "*-*-vxworks*"] then {
}
# Run again with same arguments.
-setup_xfail "mips-idt-*"
gdb_run_cmd
if [istarget "*-*-vxworks*"] then {
@@ -185,7 +184,6 @@ if [istarget "*-*-vxworks*"] then {
}
# Use "set args" command to specify an argument and run again.
-setup_xfail "mips-idt-*"
if [istarget "*-*-vxworks*"] then {
gdb_test_no_output "set args vxmain \"6\""
} else {
diff --git a/gdb/testsuite/gdb.base/bitfields.exp b/gdb/testsuite/gdb.base/bitfields.exp
index 3098ffd..f5777d6 100644
--- a/gdb/testsuite/gdb.base/bitfields.exp
+++ b/gdb/testsuite/gdb.base/bitfields.exp
@@ -253,29 +253,7 @@ gdb_load ${binfile}
gdb_test_no_output "set print sevenbit-strings"
bitfield_uniqueness
-if [istarget "mips-idt-*"] then {
- # Restart because IDT/SIM runs out of file descriptors.
- gdb_exit
- gdb_start
- gdb_reinitialize_dir $srcdir/$subdir
- gdb_load ${binfile}
-}
bitfield_containment
-if [istarget "mips-idt-*"] then {
- # Restart because IDT/SIM runs out of file descriptors.
- gdb_exit
- gdb_start
- gdb_reinitialize_dir $srcdir/$subdir
- gdb_load ${binfile}
-}
bitfield_unsignedness
-if [istarget "mips-idt-*"] then {
- # Restart because IDT/SIM runs out of file descriptors.
- gdb_exit
- gdb_start
- gdb_reinitialize_dir $srcdir/$subdir
- gdb_load ${binfile}
-}
bitfield_signedness
-
bitfield_at_offset
diff --git a/gdb/testsuite/gdb.base/break.exp b/gdb/testsuite/gdb.base/break.exp
index 9e63e2a..54cbae1 100644
--- a/gdb/testsuite/gdb.base/break.exp
+++ b/gdb/testsuite/gdb.base/break.exp
@@ -355,9 +355,6 @@ gdb_test "disable \$1foo" \
# handle arguments.
# Huh? There doesn't *appear* to be anything that passes arguments
# below.
-if [istarget "mips-idt-*"] then {
- return
-}
#
# run until the breakpoint at main is hit. For non-stubs-using targets.
diff --git a/gdb/testsuite/gdb.base/chng-syms.exp b/gdb/testsuite/gdb.base/chng-syms.exp
index ec5b4c8..9c20a44 100644
--- a/gdb/testsuite/gdb.base/chng-syms.exp
+++ b/gdb/testsuite/gdb.base/chng-syms.exp
@@ -18,13 +18,6 @@
# Test that GDB cleans up properly after errors that result when a
# breakpoint is reset.
-
-# IDT/SIM apparently doesn't have enough file descriptors to allow the
-# problem checked by this test to occur.
-if [istarget "mips-idt-*"] {
- return 0;
-}
-
set testfile "chng-syms"
set srcfile ${testfile}.c
set binfile ${objdir}/${subdir}/${testfile}
diff --git a/gdb/testsuite/gdb.base/default.exp b/gdb/testsuite/gdb.base/default.exp
index d0a8295..6a6e24e 100644
--- a/gdb/testsuite/gdb.base/default.exp
+++ b/gdb/testsuite/gdb.base/default.exp
@@ -39,7 +39,6 @@ gdb_test "append binary memory" "Missing filename\."
gdb_test "append binary value" "Missing filename\."
if !$use_gdb_stub {
- setup_xfail "mips-idt-*"
gdb_test_multiple "attach" "attach" {
-re "Argument required .(process-id|program) to attach.*$gdb_prompt $"\
{ pass "attach" }
@@ -61,7 +60,6 @@ if !$use_gdb_stub {
}
- setup_xfail "mips-idt-*"
gdb_test "backtrace" "No stack."
# ba and bac are no longer unique command prefixes. So these tests
@@ -69,18 +67,14 @@ if !$use_gdb_stub {
# GDB needs to be fixed to map unique alias here for ba bac.
#
foreach i "bt ba bac" {
- setup_xfail "mips-idt-*"
gdb_test $i "No stack." "backtrace \"$i\" abbreviation"
}
} else {
warning "Skipping backtrace and break tests because of GDB stub."
}
-# This works on the MIPS IDT board, but confuses future tests.
-if ![istarget "mips-idt-*"] then {
- gdb_test "continue" "The program is not being run." "continue"
- gdb_test "c" "The program is not being run." "continue \"c\" abbreviation"
-}
+gdb_test "continue" "The program is not being run." "continue"
+gdb_test "c" "The program is not being run." "continue \"c\" abbreviation"
#test call
gdb_test "call" "The history is empty..*" "call"
diff --git a/gdb/testsuite/gdb.base/funcargs.exp b/gdb/testsuite/gdb.base/funcargs.exp
index 0b5534a..92161d7 100644
--- a/gdb/testsuite/gdb.base/funcargs.exp
+++ b/gdb/testsuite/gdb.base/funcargs.exp
@@ -965,21 +965,6 @@ proc recursive_structs_by_value {} {
gdb_stop_suppressing_tests;
}
-proc funcargs_reload { } {
- global objdir
- global subdir
- global binfile
- global srcdir
-
- if [istarget "mips-idt-*"] {
- # Restart because IDT/SIM runs out of file descriptors.
- gdb_exit
- gdb_start
- gdb_reinitialize_dir $srcdir/$subdir
- gdb_load ${binfile}
- }
-}
-
#
# Test for accessing local stack variables in functions which call alloca
#
@@ -1234,46 +1219,31 @@ gdb_test_multiple "show endian" "getting target endian" {
gdb_test_no_output "set print frame-arguments all"
integral_args
-funcargs_reload
unsigned_integral_args
-funcargs_reload
if {![target_info exists gdb,skip_float_tests]} {
float_and_integral_args
}
# Test _Complex type here if supported.
if [support_complex_tests] {
- funcargs_reload
complex_args
- funcargs_reload
complex_integral_args
if {![target_info exists gdb,skip_float_tests]} {
- funcargs_reload
complex_float_integral_args
}
}
-funcargs_reload
pointer_args
-funcargs_reload
structs_by_reference
-funcargs_reload
structs_by_value
-funcargs_reload
discard_and_shuffle
-funcargs_reload
shuffle_round_robin
-funcargs_reload
recursive_structs_by_value
-funcargs_reload
localvars_after_alloca
-funcargs_reload
call_after_alloca
-funcargs_reload
localvars_in_indirect_call
-funcargs_reload
test_stepping_over_trampolines
set timeout $prev_timeout
diff --git a/gdb/testsuite/gdb.base/opaque.exp b/gdb/testsuite/gdb.base/opaque.exp
index b7ad265..f5fef76 100644
--- a/gdb/testsuite/gdb.base/opaque.exp
+++ b/gdb/testsuite/gdb.base/opaque.exp
@@ -134,11 +134,6 @@ gdb_test "ptype struct foo" \
# have learned during the static tests.
#
-if [istarget "mips-idt-*"] then {
- # Restart because IDT/SIM runs out of file descriptors.
- gdb_exit
- gdb_start
-}
gdb_reinitialize_dir $srcdir/$subdir
gdb_load ${binfile}
@@ -189,11 +184,6 @@ gdb_test "ptype struct foo" \
# Now reload the symbols again so we forget about anything we might
# have learned reading the symbols during the previous tests.
-if [istarget "mips-idt-*"] then {
- # Restart because IDT/SIM runs out of file descriptors.
- gdb_exit
- gdb_start
-}
gdb_reinitialize_dir $srcdir/$subdir
gdb_load ${binfile}
diff --git a/gdb/testsuite/gdb.base/ptype.exp b/gdb/testsuite/gdb.base/ptype.exp
index adb4cd4..51a3014 100644
--- a/gdb/testsuite/gdb.base/ptype.exp
+++ b/gdb/testsuite/gdb.base/ptype.exp
@@ -169,7 +169,7 @@ gdb_test "ptype red" "type = enum \{red, green, blue\}.*" "ptype unnamed enumera
# acts like whatis for basic types. If it is thought to be necessary to
# test both whatis and ptype for all the types, the tests should be
# merged into whatis.exp, or else maintenance will be a royal pain -kingdon
-#setup_xfail "mips-idt-*" "mips-sgi-*"
+#setup_xfail "mips-sgi-*"
#send "ptype v_char\n"
#gdb_expect {
# -re "type = char.*$gdb_prompt $" { pass "ptype char" }
@@ -273,7 +273,7 @@ gdb_test "ptype v_int" "type = int.*" "ptype int"
#
# test ptype command with arrays
#
-#setup_xfail "mips-idt-*" "mips-sgi-*"
+#setup_xfail "mips-sgi-*"
#send "ptype v_char_array\n"
#gdb_expect {
# -re "type = char .2..*$gdb_prompt $" { pass "ptype char array" }
@@ -374,7 +374,7 @@ gdb_test "ptype t_char_array" "type = (|unsigned )char \\\[0?\\\]"
##
## test ptype command with pointers
##
-#setup_xfail "mips-idt-*" "mips-sgi-*"
+#setup_xfail "mips-sgi-*"
#send "ptype v_char_pointer\n"
#gdb_expect {
# -re "type = char \*.*$gdb_prompt $" { pass "ptype char pointer" }
diff --git a/gdb/testsuite/gdb.base/scope.exp b/gdb/testsuite/gdb.base/scope.exp
index a86e02d..8db7096 100644
--- a/gdb/testsuite/gdb.base/scope.exp
+++ b/gdb/testsuite/gdb.base/scope.exp
@@ -552,37 +552,9 @@ gdb_test "print 'scope0.c'::filelocal" "= 1" \
"print 'scope0.c'::filelocal before run"
if [runto_main] then { test_at_main }
-if [istarget "mips-idt-*"] then {
- # Restart because IDT/SIM runs out of file descriptors.
- gdb_exit
- gdb_start
- gdb_reinitialize_dir $srcdir/$subdir
- gdb_load ${binfile}
-}
if [runto foo] then { test_at_foo }
-if [istarget "mips-idt-*"] then {
- # Restart because IDT/SIM runs out of file descriptors.
- gdb_exit
- gdb_start
- gdb_reinitialize_dir $srcdir/$subdir
- gdb_load ${binfile}
-}
if [runto bar] then { test_at_bar }
-if [istarget "mips-idt-*"] then {
- # Restart because IDT/SIM runs out of file descriptors.
- gdb_exit
- gdb_start
- gdb_reinitialize_dir $srcdir/$subdir
- gdb_load ${binfile}
-}
if [runto localscopes] then { test_at_localscopes }
-if [istarget "mips-idt-*"] then {
- # Restart because IDT/SIM runs out of file descriptors.
- gdb_exit
- gdb_start
- gdb_reinitialize_dir $srcdir/$subdir
- gdb_load ${binfile}
-}
if [runto autovars] then { test_at_autovars }
if [istarget "*-*-vxworks*"] {
diff --git a/gdb/testsuite/gdb.base/sepdebug.exp b/gdb/testsuite/gdb.base/sepdebug.exp
index ba10478..4010d33 100644
--- a/gdb/testsuite/gdb.base/sepdebug.exp
+++ b/gdb/testsuite/gdb.base/sepdebug.exp
@@ -175,9 +175,6 @@ gdb_test "info break" \
# handle arguments.
# Huh? There doesn't *appear* to be anything that passes arguments
# below.
-if [istarget "mips-idt-*"] then {
- return
-}
#
# run until the breakpoint at main is hit. For non-stubs-using targets.
diff --git a/gdb/testsuite/gdb.base/watchpoint.exp b/gdb/testsuite/gdb.base/watchpoint.exp
index 5070914..381c26e 100644
--- a/gdb/testsuite/gdb.base/watchpoint.exp
+++ b/gdb/testsuite/gdb.base/watchpoint.exp
@@ -92,40 +92,6 @@ proc initialize {} {
return 1
}
-proc maybe_clean_restart { } {
- # The IDT/sim monitor only has 8 (!) open files, of which it uses
- # 4 (!). So we have to make sure one program exits before
- # starting another one.
- if [istarget "mips-idt-*"] then {
- global testfile
- global no_hw
-
- clean_restart $testfile
-
- if {$no_hw} {
- gdb_test_no_output "set can-use-hw-watchpoints 0" ""
- }
- }
-}
-
-proc maybe_reinitialize { } {
- # The IDT/sim monitor only has 8 (!) open files, of which it uses
- # 4 (!). So we have to make sure one program exits before
- # starting another one.
- if [istarget "mips-idt-*"] then {
- global testfile
- global no_hw
-
- clean_restart $testfile
-
- if {$no_hw} {
- gdb_test_no_output "set can-use-hw-watchpoints 0" ""
- }
-
- initialize
- }
-}
-
#
# Test simple watchpoint.
#
@@ -246,8 +212,6 @@ proc test_disabling_watchpoints {} {
global decimal
global hex
- maybe_reinitialize
-
gdb_test "info watch" "\[0-9]+\[ \]*.*watchpoint.*ival3.*" "watchpoints found in watchpoint/breakpoint table"
# Ensure that the watchpoint is disabled when we startup.
@@ -312,8 +276,6 @@ proc test_disabling_watchpoints {} {
proc test_stepping {} {
global gdb_prompt
- maybe_clean_restart
-
if [runto marker1] then {
gdb_test "watch ival2" ".*\[Ww\]atchpoint \[0-9\]*: ival2"
@@ -473,8 +435,6 @@ proc test_watchpoint_triggered_in_syscall {} {
proc test_complex_watchpoint {} {
global gdb_prompt
- maybe_clean_restart
-
if [runto marker4] then {
gdb_test "watch ptr1->val" ".*\[Ww\]atchpoint \[0-9\]*: ptr1->val"
gdb_test "break marker5" ".*Breakpoint.*"
@@ -577,8 +537,6 @@ proc test_complex_watchpoint {} {
proc test_watchpoint_and_breakpoint {} {
global gdb_prompt
- maybe_clean_restart
-
# This is a test for PR gdb/38, which involves setting a
# watchpoint right after you've reached a breakpoint.
@@ -708,8 +666,6 @@ proc test_wide_location_2 {} {
proc test_inaccessible_watchpoint {} {
global gdb_prompt
- maybe_clean_restart
-
# This is a test for watchpoints on currently inaccessible (but later
# valid) memory.
--
Pedro Alves
More information about the Gdb-patches
mailing list