re-entered it fast enough, we wouldn't leave the loop and thus cause a
deadlock, since we wouldn't process the ICI message telling us to halt. We do
now call smp_intercpu_int_handler() in the loop and guard the function from
being re-entered. This also has the advantage that we can execute code on all
CPUs in the kernel debugger, if we have to.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@30208 a95241bf-73f2-0310-859d-f6bbb57e9c96
Check gKernelStartup to avoid sending ICI when entering the kernel debugger
(after ICI is enabled). Moreover it's more obvious than to use that instead
of relying on ICIs not being enabled.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@29907 a95241bf-73f2-0310-859d-f6bbb57e9c96
* Added debugger commands to resolve usb_ids to pipes.
* Adjusted the physical memory allocator to be usable in a slimmed down mode
when running inside the kernel debugger.
* Implemented USB keyboard support for KDL through a kernel debugger add-on.
* Added kgetc() and made use of it where previously individual methods were used
to ensure that reading characters always goes through the kernel debugger
add-ons and the other methods.
This has some preconditions to meet though:
1) The keyboard must be in the boot protocol (currently the case but needs to
be revisited once we have a full usb_hid).
2) The keyboard must be attached to a UHCI root port (i.e. not use EHCI or OHCI,
also not through hubs unless those are USB 1.1).
3) the usb_hid driver has to be opened for this to work. This means that for the
time between initializing USB and when usb_hid is opened by the input_server
there is no keyboard support.
Also note that this has no way of detecting hot-plug, meaning that you can't
re-attach your USB keyboard from the hub to the root port once in KDL.
On the bright side of things, since this is a non-destructive mechanism it is
possible to enter and leave KDL without loosing the USB state.
Tested OK in QEMU, not tested on real hardware yet, will see in a few minutes.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@29291 a95241bf-73f2-0310-859d-f6bbb57e9c96
#2919 (OMG :-)).
* While I was at it, I also cleaned up some style inconsistencies.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28842 a95241bf-73f2-0310-859d-f6bbb57e9c96
and colors for the on-screen debug output when leaving KDL. Thanks, Adrian!
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28841 a95241bf-73f2-0310-859d-f6bbb57e9c96
demangle module (yet). This could cause a read fault in x86's
print_demangled_call() (as it assumed there must be a ':' when this is true).
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28375 a95241bf-73f2-0310-859d-f6bbb57e9c96
that dprintf_no_syslog() can avoid sending something to the syslog.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28286 a95241bf-73f2-0310-859d-f6bbb57e9c96
the <DROP> marker at the place data was dropped, not somewhere else. See
ticket #156; this might already fix this bug.
* Minor cleanup.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28278 a95241bf-73f2-0310-859d-f6bbb57e9c96
* Introduced flag "invoke_scheduler" in the per CPU structure. It is
evaluated in hardware_interrupt() (x86 only ATM).
* Introduced SMP_MSG_RESCHEDULE_IF_IDLE message, which enters the
scheduler when the CPU currently runs an idle thread.
* Don't do dprintf() "CPU x halted!" when handling a SMP_MSG_CPU_HALT
ICI message. It uses nested spinlocks and could thus potentially
deadlock itself (acquire_spinlock() processes ICI messages, so it
could already hold one of the locks). This is a pretty likely scenario
on machines with more than two CPUs, but is also possible when the
panic()ing thread holds the threads spinlock. Probably fixes#2572.
* Reworked the way the kernel debugger is entered and added a "cpu"
command that allows switching the CPU once in KDL. It is thus possible
to get a stack trace of the thread not on the panic()ing CPU.
* When a thread is added to the run queue, we do now check, if another
CPU is idle and ask it to reschedule, if it is. Before this change, the
CPU was continuing to idle until the quantum of the idle thread
expired. Speeds up the libbe.so build about 8% on my machine (haven't
tested the full Haiku image build yet).
* When spinlock debugging is enabled (DEBUG_SPINLOCKS) we also record
the spinlock acquirer on non-smp machines. Added "spinlock" debugger
command to get the info.
* Added debugger commands "ici" and "ici_message", printing info on
pending ICI message respectively on a given one.
* Process not only a single ICI message in acquire_spinlock() and other
places, but all pending ones.
* Also process ICI messages when waiting for a free one -- avoids a
potential deadlock.
* Mask out non-existing CPUs in send_multicast_ici(). panic() instead of
just returning when there's no target CPU left.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28223 a95241bf-73f2-0310-859d-f6bbb57e9c96
debugger add-on set a demangle hook, all modules under debugger/demangle/ are
now considered demangle modules.
* Added another function to the demangle module interface that gives you access
to the arguments.
* Implemented a demangling module for GCC2.
* The older demangling module is now called "gcc3+", but doesn't support
getting the arguments yet.
* The "call" KDL command is now using demangling to automatically show you
the arguments of a call from a stack crawl.
* Minor cleanup.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28018 a95241bf-73f2-0310-859d-f6bbb57e9c96
various system information.
* Implemented retrieving some VM stats via this call.
* The VM now maintains a page fault counter, and sets system_info::page_faults
accordingly.
* Added a (pretty simple) "vmstat" command line app.
* Minor cleanup.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@27597 a95241bf-73f2-0310-859d-f6bbb57e9c96
debugger and added respective getter/setter methods
debug_{get,set}_debugged_thread(). By default the currently debugged
thread is the thread that dropped into the kernel debugger, but commands
like "in_context" can change it.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@27164 a95241bf-73f2-0310-859d-f6bbb57e9c96
(fault address, pc, read/write) when a page fault occurs, and print them
in case this caused the termination of a debugger command.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@26949 a95241bf-73f2-0310-859d-f6bbb57e9c96
empty thread structure. This causes the thread::team pointer to be NULL
during the early boot process, but our kernel debugger didn't like that.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@26735 a95241bf-73f2-0310-859d-f6bbb57e9c96
* Added "syslog" command that dumps the contents of the syslog ring buffer into
KDL. Use the '-n' option to only show what hasn't been sent to the syslog
daemon yet.
* When entering the kernel debugger, the current thread ID and name are also
printed (not only the current CPU).
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@26684 a95241bf-73f2-0310-859d-f6bbb57e9c96
It does its job by using part of libsupc++, with fake malloc and friends to make sure it won't double fault.
Works here, but cp-demangle must be extracted by hand from the lib, can't get the rule to work as I want, Ingo ?
Maybe using it directly without malloc hack would be ok with 16KB buffer, but I'm not sure of that.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@26581 a95241bf-73f2-0310-859d-f6bbb57e9c96
* Fixed most of the warnings resulting from that by removing actually not used
variables or moving declaration into the #IF. Left unused functions there
though, as I wouldn't know if they are supposed to be used again.
* Fixed two statements with no effect (clamping to MAX_ANCILLARY_DATA_LEN in
socket.cpp and unsetting fCounterSem in MessagingService.cpp).
* Some style cleanups.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@26253 a95241bf-73f2-0310-859d-f6bbb57e9c96
got a debugger_getchar hook, it just wasn't used yet. Not that there would be
any debugger add-on implementing that hook currently...
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@25809 a95241bf-73f2-0310-859d-f6bbb57e9c96
syslog so syslogs always contain the revision number too (as does serial
output).
This is only necessary because the early part of the serial output (that
contains the revision number) does not make it into the syslog. So fixing that
might be the better way.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@25804 a95241bf-73f2-0310-859d-f6bbb57e9c96
B_DEBUG_SPINLOCK_CONTENTION is defined to 1. It typedefs spinlock to a
structure (thus breaking BeOS binary compatibility), containing a
counter which is incremented whenever a thread has to wait for the
spinlock.
* Added macros for spinlock initialization and access and changed
code using spinlocks accordingly. This breaks compilation for BeOS --
the macros should be defined in the respective compatibility wrappers.
* Added generic syscall to get the spinlock counters for the thread and
the team spinlocks.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@25752 a95241bf-73f2-0310-859d-f6bbb57e9c96
pipes in the shell, though the semantics is a little different: The
second command is invoked whenever the first command has written a
complete line. The line is passed as last argument to the second
command. The new command flag B_KDEBUG_PIPE_FINAL_RERUN causes the
second command to be invoked again (with NULL argument) after the
first command is done.
* Added kprintf_unfiltered() and kputs_unfiltered() which bypass the
pipe mechanism and directly print to the bluescreen/serial output.
* Moved most commands from debug.cpp to the new
debug_builtin_commands.cpp.
* B_KDEBUG_DONT_PARSE_ARGUMENTS commands don't get an argument anymore,
if it would consist of white space only.
* Added new debugger command return value B_KDEBUG_ERROR, which
indicates that executing the command failed. This return code will
abort a complete pipe.
* Since debugger commands can nest (i.e. one command can invoke another
one) the setjmp()/longjmp() mechanism to restore the stack after a
page fault in a command needs more than one jump buffer.
* Added abort_debugger_command(), which longjmp()s out of the currently
executed command. This will also abort the current pipe.
* When pagination is enabled pressing "a" will abort the running command
(as opposed to "q" which only disables the blue screen output, but
lets the command continue).
* Added debugger commands:
- "grep" which can be used to filter output by pattern. Removed the
"filter" command and the underlying mechanism that did that before.
- "head" which prints only the first lines of output of another
command.
- "wc" counts lines, words, and characters of another command's
output.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@25744 a95241bf-73f2-0310-859d-f6bbb57e9c96
* The boot splash code now checks wether debug screen output is enabled or not
using the above function.
* The boot splash code no longer maps it's own copy of the frame buffer, instead,
it will use the boot item feature as the VESA driver does. Also removed the
lock, as that's not needed at all.
* Renamed splash.cpp to boot_splash.cpp, and boot/splash.h to boot_splash.h
(it's not part of the boot loader, but the kernel).
* Removed dead code from boot_splash.cpp, added license. Replaced license
header in boot_splash.h to a style guide conforming one.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@24489 a95241bf-73f2-0310-859d-f6bbb57e9c96
When specified it desigantes that the interrupt handler should not lock the
vector with a spinlock when executing the installed interrupt handlers. This
is necessary to allow the same interrupt vector to be handled in parallel on
different CPUs. And it is required for the CPU halt to work synchronously when
there is more than one AP CPU. Though the acquire_spinlock() should cause IPIs
to be processed, only this fixed the SMP_MSG_FLAG_SYNC problem for me.
Not locking is safe as long as it is guaranteed that no interrupt handler is
registered or removed while the interrupt handler is running. We can guarantee
this for the SMP interrupt handlers we install in arch_smp_init() as they are
never uninstalled. Probably this flag should be made private though.
Restored the SMP_MSG_FLAG_SYNC when entering the kernel debugger.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@23838 a95241bf-73f2-0310-859d-f6bbb57e9c96
* The kernel now opens up to 8 debugger modules (and puts them into an array;
maybe we'll want to switch to a doubly linked list when there is the need).
* Implemented an example debugger module that prints a stack trace of the
current thread when the kernel debugger is entered (not included in the
image).
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@23794 a95241bf-73f2-0310-859d-f6bbb57e9c96
was a race condition when more than one CPU would enter the debugger at the
same time (or rather before one CPU could stop all the others). We now use the
inDebugger variable to tell if someone is already in the debugger and then
only process inter CPU messages and retry entering the debugger.
Since sending the synchronous broadcast most of the time hung over here with
SMP enabled I removed the synchronous flag and added a simple spin to give the
other CPUs a chance to process the halt request. Added comments that explain
the reasons and a ToDo to revert to synchronous delivery once we fixed the
problem. The kernel debugger is now usable on my quad core.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@23751 a95241bf-73f2-0310-859d-f6bbb57e9c96
For commands with this flag set, the parser won't parse the
arguments provided (it will only check for matching parentheses and
brackets), but will pass the unparsed argument list string to the
command instead.
* Set the new flag for the "expr" command, so one doesn't have to quote
the expression to evaluate anymore (or put it in parentheses).
* Fixed tokenizing of quoted and unquoted strings in expression mode.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@23697 a95241bf-73f2-0310-859d-f6bbb57e9c96
command usage texts, which can be too long for kprintf().
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@23677 a95241bf-73f2-0310-859d-f6bbb57e9c96
invoke_debugger_command() is now automatically preceded by
"usage: <command name>", so the string passed to
add_debugger_command_etc() shouldn't contain it anymore.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@23584 a95241bf-73f2-0310-859d-f6bbb57e9c96
add_debugger_command(), but additionally takes parameters "usage"
and "flags".
* Added add_debugger_command_alias() which creates another name for an
existing command.
* Added print_debugger_command_usage() to print a command's usage.
* invoke_debugger_command() intercepts invocations with "--help" and
prints the command's usage text, if it is known. If unknown, the
command will be called normally.
* Made use of the new functions in debug.cpp.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@23573 a95241bf-73f2-0310-859d-f6bbb57e9c96
* Fixed the usage text of the "help" KDL command.
* Made the "ambiguous command" warning shorter, so that it will still fit in
the exception buffer.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@23565 a95241bf-73f2-0310-859d-f6bbb57e9c96
* The kernel debugger does now define some temporary variables when
entered (current thread and team, etc.).
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@23549 a95241bf-73f2-0310-859d-f6bbb57e9c96
that is a little more powerful than BeOS'. It features:
- Persistent and temporary uint64 variables. The former kind is set
only by the user. The latter (those prefixed "_") can be set
automatically by commands, thus e.g. making it easier to access
members of a dumped structure. They are unset when the next command
is invoked. The special temporary variable "_" is defined as a
command's return value.
- Expressions can contain nested command invocations using brackets
("[ ... ]").
- Command lines are parsed by the expression parser, too. They can
contain command invocations (in brackets) and expressions (in
parentheses).
* Added debugger commands:
- expr: Evaluates the given expression and prints the result.
- unset: Undefines a variable.
- vars: Prints the values of all defined variables.
* Moved debugger command code into its own source file.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@23546 a95241bf-73f2-0310-859d-f6bbb57e9c96
* Don't execute a command anymore, if the given prefix is ambiguous.
* Added tab completion for commands.
* Added on-the-fly help while typing a command line. It is triggered
by pressing tab at a position after the space following the command.
It is implemented by calling the command with argument "--help", which
doesn't work yet with most commands, but some already print their
usage in this case.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@23521 a95241bf-73f2-0310-859d-f6bbb57e9c96
Ingo Weinhold