https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=288236
--- Comment #2 from Tomoaki AOKI <[email protected]> ---
(In reply to Konstantin Belousov from comment #1)
I prefer using KV4ADDR(DMPML4I, 0, 0, 0) and KV4ADDR(DMPML4I + NDMPML4E, 0, 0,
0), as
*kva_layout.dmap_low and kva_layout.dmap_high are version dependent,
*DMAP_MIN_ADDRESS is #define'ed with KV4ADDR(DMPML4I, 0, 0, 0) and
DMAP_MAX_ADDRESS is #define'ed with KV4ADDR(DMPML4I + NDMPML4E, 0, 0, 0)
on all supported versions of base.
And more, kva_layout.dmap_low and kva_layout.dmap_high are initialized with
KV4ADDR(DMPML4I, 0, 0, 0) and KV4ADDR(DMPML4I + NDMPML4E, 0, 0, 0)
respectively.
So, this way, I'll need to add REINPLACE_CMD without conditionals.
Anyway, I'm currently on full poudriere rebuilds caused by bump on stable/14,
so I need to wait at least nvidia-related ports finishes to be rebuilt.
(I'm sharing ports tree between bare-metal and poudriere.)
But considering possibility that other kmod ports are using DMAP_MIN_ADDRESS
and DMAP_MAX_ADDRESS, something like below would be considerable, too.
How do you think?
--- a/sys/amd64/include/vmparam.h
+++ b/sys/amd64/include/vmparam.h
@@ -184,6 +184,9 @@
#define VM_MIN_KERNEL_ADDRESS_LA48 KV4ADDR(KPML4BASE, 0, 0, 0)
#define VM_MIN_KERNEL_ADDRESS kva_layout.km_low
#define VM_MAX_KERNEL_ADDRESS kva_layout.km_high
+
+#define DMAP_MIN_ADDRESS kva_layout.dmap_low
+#define DMAP_MAX_ADDRESS kva_layout.dmap_high
#define KASAN_MIN_ADDRESS KV4ADDR(KASANPML4I, 0, 0, 0)
#define KASAN_MAX_ADDRESS KV4ADDR(KASANPML4I + NKASANPML4E, 0, 0,
0)
--
You are receiving this mail because:
You are the assignee for the bug.
lmpx.com only provides a reader for public news (NNTP) servers. It is not
affiliated with the servers or forums shown here and is not responsible for
the content of articles, which is written by their respective authors.