/*
* Copyright (c) 2000 Apple Computer, Inc. All rights reserved.
*
- * @APPLE_LICENSE_HEADER_START@
+ * @APPLE_OSREFERENCE_LICENSE_HEADER_START@
*
- * The contents of this file constitute Original Code as defined in and
- * are subject to the Apple Public Source License Version 1.1 (the
- * "License"). You may not use this file except in compliance with the
- * License. Please obtain a copy of the License at
- * http://www.apple.com/publicsource and read it before using this file.
+ * This file contains Original Code and/or Modifications of Original Code
+ * as defined in and that are subject to the Apple Public Source License
+ * Version 2.0 (the 'License'). You may not use this file except in
+ * compliance with the License. The rights granted to you under the License
+ * may not be used to create, or enable the creation or redistribution of,
+ * unlawful or unlicensed copies of an Apple operating system, or to
+ * circumvent, violate, or enable the circumvention or violation of, any
+ * terms of an Apple operating system software license agreement.
*
- * This Original Code and all software distributed under the License are
- * distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY KIND, EITHER
+ * Please obtain a copy of the License at
+ * http://www.opensource.apple.com/apsl/ and read it before using this file.
+ *
+ * The Original Code and all software distributed under the License are
+ * distributed on an 'AS IS' basis, WITHOUT WARRANTY OF ANY KIND, EITHER
* EXPRESS OR IMPLIED, AND APPLE HEREBY DISCLAIMS ALL SUCH WARRANTIES,
* INCLUDING WITHOUT LIMITATION, ANY WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE OR NON-INFRINGEMENT. Please see the
- * License for the specific language governing rights and limitations
- * under the License.
+ * FITNESS FOR A PARTICULAR PURPOSE, QUIET ENJOYMENT OR NON-INFRINGEMENT.
+ * Please see the License for the specific language governing rights and
+ * limitations under the License.
*
- * @APPLE_LICENSE_HEADER_END@
+ * @APPLE_OSREFERENCE_LICENSE_HEADER_END@
*/
#include <mach/mach_types.h>
#include <libsa/types.h>
+#include <string.h> /* bcopy */
+
+#include <kern/processor.h>
+#include <kern/thread.h>
+#include <vm/vm_map.h>
+#include <vm/vm_kern.h>
+
int kdp_vm_read( caddr_t, caddr_t, unsigned int);
int kdp_vm_write( caddr_t, caddr_t, unsigned int);
#endif
static kdp_dispatch_t
- dispatch_table[KDP_REATTACH - KDP_CONNECT +1] =
+ dispatch_table[KDP_HOSTREBOOT - KDP_CONNECT +1] =
{
/* 0 */ kdp_connect,
/* 1 */ kdp_disconnect,
/* F */ kdp_breakpoint_set,
/*10 */ kdp_breakpoint_remove,
/*11 */ kdp_regions,
-/*12 */ kdp_reattach
+/*12 */ kdp_reattach,
+/*13 */ kdp_reboot
};
kdp_glob_t kdp;
static kdp_breakpoint_record_t breakpoint_list[MAX_BREAKPOINTS];
static unsigned int breakpoints_initialized = 0;
+
int reattach_wait = 0;
+int noresume_on_disconnect = 0;
+
+#define MAXCOMLEN 16
+
+struct thread_snapshot {
+ uint32_t snapshot_magic;
+ thread_t thread_id;
+ int32_t state;
+ wait_queue_t wait_queue;
+ event64_t wait_event;
+ vm_offset_t kernel_stack;
+ vm_offset_t reserved_stack;
+ thread_continue_t continuation;
+ uint32_t nkern_frames;
+ char user64_p;
+ uint32_t nuser_frames;
+ int32_t pid;
+ char p_comm[MAXCOMLEN + 1];
+};
+
+typedef struct thread_snapshot *thread_snapshot_t;
+
+extern int
+machine_trace_thread(thread_t thread, uint32_t tracepos, uint32_t tracebound, int nframes, boolean_t user_p);
+extern int
+machine_trace_thread64(thread_t thread, uint32_t tracepos, uint32_t tracebound, int nframes, boolean_t user_p);
+extern int
+proc_pid(void *p);
+extern void
+proc_name(int pid, char *buf, int size);
+extern void
+kdp_snapshot_postflight(void);
+
+static int
+pid_from_task(task_t task);
+
+int
+kdp_stackshot(int pid, uint32_t tracebuf, uint32_t tracebuf_size, unsigned trace_options, uint32_t *pbytesTraced);
+
+extern char version[];
boolean_t
kdp_packet(
}
req = rd->hdr.request;
- if ((req < KDP_CONNECT) || (req > KDP_REATTACH)) {
+ if ((req < KDP_CONNECT) || (req > KDP_HOSTREBOOT)) {
printf("kdp_packet bad request %x len %d seq %x key %x\n",
rd->hdr.request, rd->hdr.len, rd->hdr.seq, rd->hdr.key);
kdp.is_halted = kdp.is_conn = FALSE;
kdp.exception_seq = kdp.conn_seq = 0;
+ if (noresume_on_disconnect == 1) {
+ reattach_wait = 1;
+ noresume_on_disconnect = 0;
+ }
+
rp->hdr.is_reply = 1;
rp->hdr.len = sizeof (*rp);
int plen = *len;
kdp_readmem_reply_t *rp = &pkt->readmem_reply;
int cnt;
-
+#if __i386__
+ void *pversion = &version;
+#endif
if (plen < sizeof (*rq))
return (FALSE);
unsigned int n = rq->nbytes;
dprintf(("kdp_readmem addr %x size %d\n", rq->address, rq->nbytes));
-
+#if __i386__
+ /* XXX This is a hack to facilitate the "showversion" macro
+ * on i386, which is used to obtain the kernel version without
+ * symbols - a pointer to the version string should eventually
+ * be pinned at a fixed address when an equivalent of the
+ * VECTORS segment (loaded at a fixed load address, and contains
+ * a table) is implemented on x86, as with PPC.
+ */
+ if (rq->address == (void *)0x501C)
+ rq->address = &pversion;
+#endif
cnt = kdp_vm_read((caddr_t)rq->address, (caddr_t)rp->data, rq->nbytes);
rp->error = KDPERR_NO_ERROR;
}
return breakpoint_found;
}
+
+
+#define MAX_FRAMES 1000
+
+static int pid_from_task(task_t task)
+{
+ int pid = -1;
+
+ if (task->bsd_info)
+ pid = proc_pid(task->bsd_info);
+
+ return pid;
+}
+
+int
+kdp_stackshot(int pid, uint32_t tracebuf, uint32_t tracebuf_size, unsigned trace_options, uint32_t *pbytesTraced)
+{
+ uint32_t tracepos = (uint32_t) tracebuf;
+ uint32_t tracebound = tracepos + tracebuf_size;
+ uint32_t tracebytes = 0;
+ int error = 0;
+
+ processor_set_t pset = &default_pset;
+ task_t task = TASK_NULL;
+ thread_t thread = THREAD_NULL;
+ int nframes = trace_options;
+ thread_snapshot_t tsnap = NULL;
+ unsigned framesize = 2 * sizeof(vm_offset_t);
+
+ if ((nframes <= 0) || nframes > MAX_FRAMES)
+ nframes = MAX_FRAMES;
+
+ queue_iterate(&pset->tasks, task, task_t, pset_tasks) {
+ /* Trace everything, unless a process was specified */
+ if ((pid == -1) || (pid == pid_from_task(task)))
+ queue_iterate(&task->threads, thread, thread_t, task_threads){
+ if ((tracepos + 4 * sizeof(struct thread_snapshot)) > tracebound) {
+ error = -1;
+ goto error_exit;
+ }
+/* Populate the thread snapshot header */
+ tsnap = (thread_snapshot_t) tracepos;
+ tsnap->thread_id = thread;
+ tsnap->state = thread->state;
+ tsnap->wait_queue = thread->wait_queue;
+ tsnap->wait_event = thread->wait_event;
+ tsnap->kernel_stack = thread->kernel_stack;
+ tsnap->reserved_stack = thread->reserved_stack;
+ tsnap->continuation = thread->continuation;
+/* Add the BSD process identifiers */
+ if ((tsnap->pid = pid_from_task(task)) != -1)
+ proc_name(tsnap->pid, tsnap->p_comm, MAXCOMLEN + 1);
+ else
+ tsnap->p_comm[0] = '\0';
+
+ tsnap->snapshot_magic = 0xfeedface;
+ tracepos += sizeof(struct thread_snapshot);
+
+/* Call through to the machine specific trace routines
+ * Frames are added past the snapshot header.
+ */
+ if (tsnap->kernel_stack != 0)
+ tracebytes = machine_trace_thread(thread, tracepos, tracebound, nframes, FALSE);
+ tsnap->nkern_frames = tracebytes/(2 * sizeof(vm_offset_t));
+ tracepos += tracebytes;
+ tracebytes = 0;
+ tsnap->user64_p = 0;
+/* Trace user stack, if any */
+ if (thread->task->map != kernel_map) {
+/* 64-bit task? */
+ if (task_has_64BitAddr(thread->task)) {
+ tracebytes = machine_trace_thread64(thread, tracepos, tracebound, nframes, TRUE);
+ tsnap->user64_p = 1;
+ framesize = 2 * sizeof(addr64_t);
+ }
+ else {
+ tracebytes = machine_trace_thread(thread, tracepos, tracebound, nframes, TRUE);
+ framesize = 2 * sizeof(vm_offset_t);
+ }
+ }
+ tsnap->nuser_frames = tracebytes/framesize;
+ tracepos += tracebytes;
+ tracebytes = 0;
+ }
+ }
+
+error_exit:
+ /* Release stack snapshot wait indicator */
+ kdp_snapshot_postflight();
+
+ *pbytesTraced = tracepos - tracebuf;
+
+ return error;
+}