x86/boot/KASLR: Add two new functions for 1GB huge pages handling
authorBaoquan He <bhe@redhat.com>
Mon, 25 Jun 2018 03:16:55 +0000 (11:16 +0800)
committerIngo Molnar <mingo@kernel.org>
Tue, 3 Jul 2018 08:50:12 +0000 (10:50 +0200)
Introduce two new functions: parse_gb_huge_pages() and process_gb_huge_pages(),
which handle a conflict between KASLR and huge pages of 1GB.

These two functions will be used in the next patch:

- parse_gb_huge_pages() is used to parse kernel command-line to get
  how many 1GB huge pages have been specified. A static global
  variable 'max_gb_huge_pages' is added to store the number.

- process_gb_huge_pages() is used to skip as many 1GB huge pages
  as possible from the passed in memory region according to the
  specified number.

Signed-off-by: Baoquan He <bhe@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: douly.fnst@cn.fujitsu.com
Cc: fanc.fnst@cn.fujitsu.com
Cc: indou.takao@jp.fujitsu.com
Cc: keescook@chromium.org
Cc: lcapitulino@redhat.com
Cc: yasu.isimatu@gmail.com
Link: http://lkml.kernel.org/r/20180625031656.12443-2-bhe@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
arch/x86/boot/compressed/kaslr.c

index b87a758..d97647b 100644 (file)
@@ -215,6 +215,35 @@ static void mem_avoid_memmap(char *str)
                memmap_too_large = true;
 }
 
+/* Store the number of 1GB huge pages which users specified: */
+static unsigned long max_gb_huge_pages;
+
+static void parse_gb_huge_pages(char *param, char *val)
+{
+       static bool gbpage_sz;
+       char *p;
+
+       if (!strcmp(param, "hugepagesz")) {
+               p = val;
+               if (memparse(p, &p) != PUD_SIZE) {
+                       gbpage_sz = false;
+                       return;
+               }
+
+               if (gbpage_sz)
+                       warn("Repeatedly set hugeTLB page size of 1G!\n");
+               gbpage_sz = true;
+               return;
+       }
+
+       if (!strcmp(param, "hugepages") && gbpage_sz) {
+               p = val;
+               max_gb_huge_pages = simple_strtoull(p, &p, 0);
+               return;
+       }
+}
+
+
 static int handle_mem_memmap(void)
 {
        char *args = (char *)get_cmd_line_ptr();
@@ -466,6 +495,60 @@ static void store_slot_info(struct mem_vector *region, unsigned long image_size)
        }
 }
 
+/*
+ * Skip as many 1GB huge pages as possible in the passed region
+ * according to the number which users specified:
+ */
+static void
+process_gb_huge_pages(struct mem_vector *region, unsigned long image_size)
+{
+       unsigned long addr, size = 0;
+       struct mem_vector tmp;
+       int i = 0;
+
+       if (!max_gb_huge_pages) {
+               store_slot_info(region, image_size);
+               return;
+       }
+
+       addr = ALIGN(region->start, PUD_SIZE);
+       /* Did we raise the address above the passed in memory entry? */
+       if (addr < region->start + region->size)
+               size = region->size - (addr - region->start);
+
+       /* Check how many 1GB huge pages can be filtered out: */
+       while (size > PUD_SIZE && max_gb_huge_pages) {
+               size -= PUD_SIZE;
+               max_gb_huge_pages--;
+               i++;
+       }
+
+       /* No good 1GB huge pages found: */
+       if (!i) {
+               store_slot_info(region, image_size);
+               return;
+       }
+
+       /*
+        * Skip those 'i'*1GB good huge pages, and continue checking and
+        * processing the remaining head or tail part of the passed region
+        * if available.
+        */
+
+       if (addr >= region->start + image_size) {
+               tmp.start = region->start;
+               tmp.size = addr - region->start;
+               store_slot_info(&tmp, image_size);
+       }
+
+       size  = region->size - (addr - region->start) - i * PUD_SIZE;
+       if (size >= image_size) {
+               tmp.start = addr + i * PUD_SIZE;
+               tmp.size = size;
+               store_slot_info(&tmp, image_size);
+       }
+}
+
 static unsigned long slots_fetch_random(void)
 {
        unsigned long slot;