mm: the loader jumped out from under itself, and 3 GB was never the limit

Three things, and the first one is the freeze.

**The loader jumped out from under itself.** A machine with a lot of RAM
stopped at "Starting the kernel" with the splash still up: no panic, no reset,
nothing. The firmware's own exception handler had the answer the whole time,
once a machine was found that printed it:

    X64 Exception Type - 0E(#PF - Page-Fault)
    ExceptionData - 0x10   I:1 P:0
    RIP - 0000000140004DEE      <- the loader's own code, at 5.4 GB
    RAX - 0000000000001000      <- the page tables it had just installed
    RCX - 00000000001000F0      <- the kernel it was about to jump to

I:1 P:0 is an instruction fetch from a page that is not present. The loader
built page tables covering four gigabytes, loaded cr3 -- and the firmware had
placed the loader image *above* four gigabytes, so the next instruction fetch
was unmapped. It reproduces exactly on `-m 8G` and never on `-m 1G`, which is
why every test here passed.

The PDPT is one page whether four entries are filled or all 512, so all 512
are: 512 GB of identity map, replaced by kvmalloc() a few hundred instructions
later.

**And 3 GB was never the limit, it was a circle.** PHYSTOP_MAX was 0xC0000000
because kvmalloc() mapped exactly four 1 GB pages, and kvmalloc mapped four
because that was the cap -- each holding the other up, throwing away thirteen
gigabytes of a sixteen gigabyte machine. kvmalloc now maps as many gigabytes as
the machine has RAM and tells ioremap where the direct map stopped, instead of
ioremap assuming it stopped at four.

Freeing memory becomes two passes, and the ordering is forced rather than
chosen: kinit1() runs *before* kvmalloc because kvmalloc calls kalloc for its
own page tables, and freeing a page means writing into it -- so kinit1 can only
touch what the loader mapped. kinit_high() adds the rest afterwards.
`phystop` and the new `lowtop` are now two different numbers: the DMA arena
stays below 4 GB because a device is not obliged to reach the top of memory,
and an RTL8139's buffer address register is thirty-two bits wide.

**And the 0 MB.** The System Information screen reported "0 MB usable" on a
machine with sixteen gigabytes in it, because memory_mb() had its own 16 KB
guess at how big a memory map gets, and returned zero when GetMemoryMap said
EFI_BUFFER_TOO_SMALL. That is the same bug as the previous commit's, in a
second copy, and this one failed silently. There is one buffer and one
map_read() now.

Verified at 1 GB, 8 GB and 16 GB: all three boot, no traps, no panics, socktest
and memfdtest green in each. 8 GB and 16 GB did not boot at all before -- and
16 GB now reports 17408 MB with "mem: 13312 MB above 4 GB added to the
allocator", where the old kernel would have used three. Both display
configurations still clean.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

kernel/core/kalloc.c

@@ -163,6 +163,41 @@ mem_whose(struct sbuf *out, struct whose *w)
   }
 }
 
+// The memory the loader's page tables could not reach.
+//
+// kinit1() runs before kvmalloc(), on page tables that map four gigabytes, and
+// freeing a page means writing a free-list pointer *into* it -- so anything
+// higher would fault. kvmalloc() has now mapped as much of physical memory as
+// the machine has, so the rest can be handed over.
+//
+// Two passes rather than one because of that ordering, and the ordering is
+// forced: kvmalloc() calls kalloc() for its own page tables, so the allocator
+// has to be working before the map it needs exists.
+void
+kinit_high(void)
+{
+  uint64 added = 0;
+
+  for (int i = 0; i < nmem_usable; i++) {
+    uint64 lo = mem_usable[i].base;
+    uint64 hi = mem_usable[i].end;
+
+    if (lo < DIRECTMAP_BOOT)
+      lo = DIRECTMAP_BOOT;
+    if (hi > phystop)
+      hi = phystop;
+    if (lo >= hi)
+      continue;
+
+    freerange(P2V(lo), P2V(hi));
+    added += hi - lo;
+  }
+
+  if (added)
+    kinfo("mem: %d MB above 4 GB added to the allocator\n",
+          (int)(added / (1024 * 1024)));
+}
+
 void
 kinit2()
 {

kernel/core/main.c

@@ -188,13 +188,22 @@ kmain(uint32 fb_addr)
   // is a parameter, so moving it up by KLOG_BYTES is the whole of it.
   klog_base = (void *)PGROUNDUP((addr_t)end);
   klog_init(klog_base, KLOG_BYTES);
-  kinit1((char *)klog_base + KLOG_BYTES, P2V(PHYSTOP - DMA_ARENA_BYTES));
+  // `lowtop`, not PHYSTOP: this runs on the loader's page tables, which map
+  // four gigabytes, and freeing a page writes into it. Everything above that
+  // is added by kinit_high() once kvmalloc has mapped it.
+  kinit1((char *)klog_base + KLOG_BYTES, P2V(lowtop - DMA_ARENA_BYTES));
   if (boot_verbose) cprintf("verbose boot: kinit1 done\n");
-  dma_init(P2V(PHYSTOP - DMA_ARENA_BYTES), P2V(PHYSTOP));
+  // The arena stays low for the same reason it is contiguous: a device is not
+  // obliged to be able to reach the top of memory. An RTL8139's buffer address
+  // register is thirty-two bits wide, and it is not the only one.
+  dma_init(P2V(lowtop - DMA_ARENA_BYTES), P2V(lowtop));
   if (boot_verbose) cprintf("verbose boot: dma_init done\n");
   kvmalloc();
   if (boot_verbose) cprintf("verbose boot: kvmalloc done\n");
 
+  // Now that the direct map reaches it, the rest of RAM can be allocated from.
+  kinit_high();
+
   // The other kind of framebuffer: one the direct map does not reach.
   //
   // Any machine with a discrete GPU is entitled to put its aperture above