debugger: reachable from the console, and named on the crash screen

Two things, both about the debugger being available at the moment you
actually need it rather than once a shell exists.

**Ctrl-G on the console runs every view.** No filesystem, no /dev, no
shell -- it is the last thing main() sets up before entering the scheduler,
so it answers from early in boot, which is exactly when boot problems
happen. It sits beside the older Ctrl-P, and the difference between them is
the interesting part: procdump() takes no lock at all, deliberately,
because a lock is a way to hang while reporting a hang. The views cannot
follow that rule -- a free list read torn is worse than not read -- so they
take locks, and a lock must never be taken from an interrupt handler that
may have interrupted its holder. The key therefore *requests* and a kernel
thread *performs*. A scheduler too wedged to run a thread will not answer
Ctrl-G; Ctrl-P still will, and that division is the honest one.

**PSOGM says where the report went.** Its closing line used to be an
addr2line command, which the call chain no longer needs -- it prints names
now. What replaced it is the thing nobody would guess:

    This report is in /proc/kmsg after a restart.
    schdbg, or Ctrl-G on the console, asks more.
    PSOGM  --  Purple Screen of Guru Meditation

"Restart and read it" beats "photograph this screen" by a wide margin, and
until the log outlived the kernel it was not true. Both panic.c and
graphics.md had claimed in writing that the screen was the only artefact of
a crash; neither says so now. The QR code stays, because it is still the
only channel that works when the machine will not come back at all.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VQ7Qx8E1q1M1bbCq8QLHS2

docs/architecture/debugging.md

@@ -200,10 +200,24 @@ fine.
 
 | | Works when |
 |---|---|
-| `sysrq g` | always — every view to the console, no process, no filesystem, no client program |
+| **Ctrl-G** on the console | there is a scheduler — so from early in boot, with nothing mounted and nothing in userspace started |
-| `/dev/dbg` | there is a scheduler and a filesystem |
+| `sysrq g` | `/proc` is mounted |
+| `/dev/dbg` | `/dev` is mounted |
 | `schdbg` | there is a shell |
 
+**Ctrl-G is the one that matters for boot problems**, because it is the only
+one that does not need a filesystem. It sits beside the older Ctrl-P, which
+dumps the process table — and the difference between them is worth knowing.
+`procdump()` takes **no lock at all**, deliberately: a lock is a way to hang
+while reporting a hang. The views cannot follow that rule, because a free list
+or a socket table read torn is worse than not read. So they take locks, and a
+lock must never be taken from an interrupt handler that may have interrupted
+its holder.
+
+The key therefore *requests* and a kernel thread *performs*. The cost is stated
+plainly: a scheduler too wedged to run a thread will not answer Ctrl-G. Ctrl-P
+still will, and that division of labour is the honest one.
+
 A watchpoint's *hits* go to the console rather than to any of these, because
 they happen while you are not looking, which is the point of having them.
 
@@ -342,6 +356,10 @@ something that reports nonsense.
   watchpoint on something `cprintf` itself writes is silently dropped rather
   than looping; the handler guards against re-entering itself per CPU.
 - **A watchpoint takes up to one timer tick** to reach the other CPUs.
+- **Ctrl-G needs a working scheduler**, per the note above. Before
+  `dbg_thread_start()` — which is the last thing `main()` does before entering
+  the scheduler — nothing answers but the log, and the log is enough because it
+  is read on the *next* boot.
 - **No network transport.** The plan puts a TCP bridge in a userspace daemon,
   deliberately outside the kernel; it does not exist, so the debugger is local
   or serial only.

docs/architecture/graphics.md

@@ -179,17 +179,36 @@ you it stopped — which is both truer and much less alarming.
 face that is obviously a cartoon, then the facts. Whoever is reading this is
 already having a bad time.
 
-**It says considerably more than it needs to.** The screen is the only artefact
+**It says considerably more than it needs to.** Everything cheap to collect is
-of a panic — no core dump, no log to go back to — so everything cheap to collect
+collected: stop code, fault address, which CPU, which process, how far into the
-is collected: stop code, fault address, which CPU, which process, how far into
+boot, where the kernel is loaded, and a call chain — and since
-the boot, where the kernel is loaded, and a call chain. Anything left out is
+[debugging.md](debugging.md)'s symbol table, the call chain is *names*:
-gone. All of it goes to the serial line in the same order, which is what
+`panic+0x5f`, `proc_sysrq_cmd+0xb3`, `sys_writev+0x12b`. All of it goes to the
-`bootcheck` reads and what you get when the crash happens before there is a
+serial line in the same order, which is what `bootcheck` reads and what you get
-framebuffer at all.
+when the crash happens before there is a framebuffer at all.
 
 Once painted, `panicked` is set and `consputc` spins rather than writing, so
 nothing can scribble over the report.
 
+### It is no longer the only copy
+
+This screen used to be the only artefact of a panic — this file said so, and so
+did `panic.c`'s own header. That stopped being true when the kernel log started
+outliving the kernel, and the closing lines changed to match:
+
+```
+This report is in /proc/kmsg after a restart.
+schdbg, or Ctrl-G on the console, asks more.
+PSOGM  --  Purple Screen of Guru Meditation
+```
+
+The first line replaced an `addr2line` command, which the call chain no longer
+needs. It is the more useful instruction by a wide margin, and it is the one
+nobody would guess: **restart and read it** beats photographing the screen.
+
+The QR code below is still drawn, and still earns its place — it is the only
+channel that works when the machine will not come back at all.
+
 ### The QR code
 
 `kernel/core/qr.c`. **A panic report is worth nothing if it cannot leave a

include/defs.h

@@ -373,6 +373,8 @@ int             watch_trap(struct trapframe *tf);
 
 // dbg.c -- /dev/dbg, the introspection registry. See include/dbg.h.
 int             dbg_read(struct inode*, uint64, char*, int);
+void            dbg_request_dump(void);
+void            dbg_thread_start(void);
 int             dbg_write(struct inode*, uint64, char*, int);
 
 // klog.c -- the kernel log that outlives the kernel. See include/klog.h.

kernel/core/console.c

@@ -2,6 +2,7 @@
 #include "types.h"
 #include "defs.h"
 #include "klog.h"
+#include "dbg.h"
 #include "vt.h"
 #include "kbd.h"   /* KEY_VT_SWITCH -- the terminal-switch chords */
 #include "gfx.h"
@@ -462,6 +463,19 @@ consoleintr(int (*getc)(void))
       continue;
     }
 
+    // Ctrl-G: every debug view, on the console.
+    //
+    // The same thing `sysrq g` and `schdbg all` produce, reachable with no
+    // filesystem mounted and no shell running -- which is the point. It only
+    // asks; a kernel thread does the work, because the views take locks and
+    // this is an interrupt handler. See dbg_request_dump().
+    if (c == C('G')) {
+      release(&input_lock);
+      dbg_request_dump();
+      acquire(&input_lock);
+      continue;
+    }
+
     if (v->tio.c_lflag & ICANON) {
       if (c == v->tio.c_cc[VERASE] || c == C('H')) {
         erase_one(v);

kernel/core/main.c

@@ -217,6 +217,12 @@ kmain(uint32 fb_addr)
   // until mpmain() below enters the scheduler.
   netinit();
   if (boot_verbose) cprintf("verbose boot: netinit done\n");
+  // The debugger's own thread, for the same reason netinit's comes after
+  // sinit: kthread_create needs a usable process table. From here on Ctrl-G
+  // on the console answers, with no filesystem mounted and nothing in
+  // userspace started -- which is the whole point of putting it this early.
+  dbg_thread_start();
+  if (boot_verbose) cprintf("verbose boot: dbg thread started\n");
   mpmain();
 }
 

kernel/core/panic.c

@@ -497,16 +497,25 @@ paint(char *reason, addr_t *pcs, int npcs)
     trace_line(line, sizeof(line), i, pcs[i]);
     gfx_text_at(vx - 3 * cw, y, line, PSOGM_FG);
     y += ch;
-    if (y > sh - ch * 5)
+    if (y > sh - ch * 6)
       break;
   }
 
-  // Two closing lines, bottom left: what to do with the numbers above, and
+  // Three closing lines, bottom left. The first used to be an addr2line
-  // what this screen is called, so it can be searched for by name. Shortened
+  // command, which the call chain above no longer needs -- it prints names.
-  // from the full command line, which at 62 characters ran under the QR card.
+  //
-  gfx_text_at(left, sh - ch * 3,
+  // What replaces it is the thing nobody would guess: **this screen is no
-              "Resolve:  addr2line -e kernel <address>", PSOGM_DIM);
+  // longer the only copy.** The log survives into the next boot (see
-  gfx_text_at(left, sh - ch * 3 + ch + 4,
+  // include/klog.h), so the right instruction on a crash screen is now
+  // "restart and read it" rather than "photograph this".
+  //
+  // Kept under about 45 characters each: the QR card is on the right, and the
+  // 62-character command line this replaces used to run underneath it.
+  gfx_text_at(left, sh - ch * 4,
+              "This report is in /proc/kmsg after a restart.", PSOGM_DIM);
+  gfx_text_at(left, sh - ch * 4 + ch + 4,
+              "schdbg, or Ctrl-G on the console, asks more.", PSOGM_DIM);
+  gfx_text_at(left, sh - ch * 4 + (ch + 4) * 2,
               "PSOGM  --  Purple Screen of Guru Meditation", PSOGM_DIM);
 }
 
@@ -542,6 +551,14 @@ report_serial(char *reason, addr_t *pcs, int npcs)
     cprintf("\n    Resolve with: addr2line -e build/kernel/kernel <address>\n");
   else
     cprintf("\n    For a line number: addr2line -e build/kernel/kernel <address>\n");
+
+  // Where to go next. Worth printing even though the machine is about to stop,
+  // because this text is also what ends up in the log -- and the log is the
+  // thing being pointed at.
+  cprintf("    This report is in /proc/kmsg after a restart.\n");
+  cprintf("    Then: schdbg whose <addr> -- who owns a page\n");
+  cprintf("          schdbg watch <addr> -- who writes to one\n");
+  cprintf("          Ctrl-G on the console -- every view, no shell needed\n");
   cprintf("    Reason: %s\n", reason);
   cprintf("    System halted.\n");
 }

kernel/dev/dbg.c

@@ -328,6 +328,56 @@ fill(const char *cmd)
   return sb.len;
 }
 
+// --- the console key --------------------------------------------------------
+//
+// Ctrl-G runs the debugger with no filesystem, no shell and no client program,
+// which is what makes it available *during boot* rather than once userspace is
+// up. docs/plan/debugger.md calls this level 0, and it is the level that
+// matters when the thing you are debugging is the boot.
+//
+// It cannot do the work in the interrupt handler, though, and the reason is
+// worth stating because the neighbouring Ctrl-P does exactly that. procdump()
+// reads the process table with no lock held, deliberately: a lock is a way to
+// hang while reporting a hang. The views cannot follow that rule -- they read
+// free lists and socket tables that are meaningless torn -- so they take locks,
+// and taking a lock from an interrupt handler deadlocks against whatever was
+// interrupted while holding it.
+//
+// So the key *requests* and a kernel thread *performs*. The cost is that a
+// scheduler too wedged to run a thread will not answer; Ctrl-P still does, and
+// that division is the honest one.
+static int dbg_wanted;
+
+void
+dbg_request_dump(void)
+{
+  dbg_wanted = 1;
+  wakeup(&dbg_wanted);
+}
+
+static void
+dbg_thread(void)
+{
+  for (;;) {
+    acquire(&dbglock);
+    while (!dbg_wanted)
+      sleep(&dbg_wanted, &dbglock);
+    dbg_wanted = 0;
+    release(&dbglock);
+
+    cprintf("\n--- debug views, requested from the console ---\n");
+    dbg_dump_all();
+  }
+}
+
+void
+dbg_thread_start(void)
+{
+  if (!dbginit) { initlock(&dbglock, "dbg"); dbginit = 1; }
+  if (kthread_create(dbg_thread, "dbgd") < 0)
+    cprintf("dbg: no kernel thread; Ctrl-G will not work\n");
+}
+
 void
 dbg_dump_all(void)
 {