diff options
author | Mark Johnston <markj@FreeBSD.org> | 2021-08-10 20:52:36 +0000 |
---|---|---|
committer | Mark Johnston <markj@FreeBSD.org> | 2021-08-11 01:27:53 +0000 |
commit | 8978608832c28572bbf5adadb9cfb077e8f15255 (patch) | |
tree | 3f085afcbbee03e9f73ae749eba50c36bdbbfbf3 /sys/vm | |
parent | 5dda15adbcf7b650fb69b5259090b16c66d1cf1a (diff) | |
download | src-8978608832c28572bbf5adadb9cfb077e8f15255.tar.gz src-8978608832c28572bbf5adadb9cfb077e8f15255.zip |
amd64: Populate the KMSAN shadow maps and integrate with the VM
- During boot, allocate PDP pages for the shadow maps. The region above
KERNBASE is currently not shadowed.
- Create a dummy shadow for the vm page array. For now, this array is
not protected by the shadow map to help reduce kernel memory usage.
- Grow shadows when growing the kernel map.
- Increase the default kernel stack size when KMSAN is enabled. As with
KASAN, sanitizer instrumentation appears to create stack frames large
enough that the default value is not sufficient.
- Disable UMA's use of the direct map when KMSAN is configured. KMSAN
cannot validate the direct map.
- Disable unmapped I/O when KMSAN configured.
- Lower the limit on paging buffers when KMSAN is configured. Each
buffer has a static MAXPHYS-sized allocation of KVA, which in turn
eats 2*MAXPHYS of space in the shadow map.
Reviewed by: alc, kib
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D31295
Diffstat (limited to 'sys/vm')
-rw-r--r-- | sys/vm/vm_pager.c | 9 |
1 files changed, 9 insertions, 0 deletions
diff --git a/sys/vm/vm_pager.c b/sys/vm/vm_pager.c index 640e3d977e99..69f0a2dc2bbb 100644 --- a/sys/vm/vm_pager.c +++ b/sys/vm/vm_pager.c @@ -217,6 +217,15 @@ pbuf_zsecond_create(const char *name, int max) zone = uma_zsecond_create(name, pbuf_ctor, pbuf_dtor, NULL, NULL, pbuf_zone); + +#ifdef KMSAN + /* + * Shrink the size of the pbuf pools if KMSAN is enabled, otherwise the + * shadows of the large KVA allocations eat up too much memory. + */ + max /= 3; +#endif + /* * uma_prealloc() rounds up to items per slab. If we would prealloc * immediately on every pbuf_zsecond_create(), we may accumulate too |