[PATCH 07/24] mmap: Add grub_mmap_get_lowest() and grub_mmap_get_highest()

[Sergii Dmytruk]

From: Daniel Kiper <daniel...@oracle.com>

The functions calculate lowest and highest available RAM
addresses respectively.

Both functions are needed to calculate PMR boundaries for
Intel TXT secure launcher introduced by subsequent patches.

Signed-off-by: Daniel Kiper <daniel...@oracle.com>
---
grub-core/mmap/mmap.c | 83 +++++++++++++++++++++++++++++++++++++++++++
include/grub/memory.h | 3 ++
2 files changed, 86 insertions(+)

diff --git a/grub-core/mmap/mmap.c b/grub-core/mmap/mmap.c
index c8c8312c5..80d6c60b8 100644
--- a/grub-core/mmap/mmap.c
+++ b/grub-core/mmap/mmap.c
@@ -26,6 +26,7 @@
#include <grub/command.h>
#include <grub/dl.h>
#include <grub/i18n.h>
+#include <grub/safemath.h>

GRUB_MOD_LICENSE ("GPLv3+");

@@ -343,6 +344,88 @@ grub_mmap_unregister (int handle)

#endif /* ! GRUB_MMAP_REGISTER_BY_FIRMWARE */

+typedef struct
+{
+ grub_uint64_t addr;
+ grub_uint64_t limit;
+} addr_limit_t;
+
+/* Helper for grub_mmap_get_lowest(). */
+static int
+lowest_hook (grub_uint64_t addr, grub_uint64_t size, grub_memory_type_t type,
+ void *data)
+{
+ addr_limit_t *al = data;
+ grub_uint64_t end;
+
+ if (type != GRUB_MEMORY_AVAILABLE)
+ return 0;
+
+ if (grub_add (addr, size, &end))
+ return 0;
+
+ if (addr >= al->limit)
+ al->addr = grub_min (al->addr, addr);
+
+ if ((addr < al->limit) && (end > al->limit))
+ al->addr = al->limit;
+
+ return 0;
+}
+
+/*
+ * This function calculates lowest available RAM address that is at or above
+ * the passed limit. If no RAM exists above the limit, ~0 is returned.
+ */
+grub_uint64_t
+grub_mmap_get_lowest (grub_uint64_t limit)
+{
+ addr_limit_t al = {~0, limit};
+
+ grub_mmap_iterate (lowest_hook, &al);
+
+ return al.addr;
+}
+
+/* Helper for grub_mmap_get_highest(). */
+static int
+highest_hook (grub_uint64_t addr, grub_uint64_t size, grub_memory_type_t type,
+ void *data)
+{
+ addr_limit_t *al = data;
+ grub_uint64_t end;
+
+ if (type != GRUB_MEMORY_AVAILABLE)
+ return 0;
+
+ if (grub_add (addr, size, &end))
+ return 0;
+
+ if (end < al->limit)
+ al->addr = grub_max (al->addr, end);
+
+ if ((addr < al->limit) && (end >= al->limit))
+ al->addr = al->limit;
+
+ return 0;
+}
+
+/*
+ * This function calculates highest available RAM address that is below the
+ * passed limit. Returned address is either one byte after last byte of RAM or
+ * equal to limit, whichever is lower. If no RAM exists below limit, 0 is
+ * returned.
+ */
+grub_uint64_t
+grub_mmap_get_highest (grub_uint64_t limit)
+{
+ addr_limit_t al = {0, limit};
+
+ grub_mmap_iterate (highest_hook, &al);
+
+ return al.addr;
+}
+
#define CHUNK_SIZE 0x400

struct badram_entry {
diff --git a/include/grub/memory.h b/include/grub/memory.h
index 6da114a1b..8f22f7525 100644
--- a/include/grub/memory.h
+++ b/include/grub/memory.h
@@ -69,6 +69,9 @@ void *grub_mmap_malign_and_register (grub_uint64_t align, grub_uint64_t size,

void grub_mmap_free_and_unregister (int handle);

+extern grub_uint64_t grub_mmap_get_lowest (grub_uint64_t limit);
+extern grub_uint64_t grub_mmap_get_highest (grub_uint64_t limit);
+
#ifndef GRUB_MMAP_REGISTER_BY_FIRMWARE

struct grub_mmap_region
--
2.46.0

[ross.ph...@oracle.com]

On 8/26/24 5:44 AM, Sergii Dmytruk wrote:
> From: Daniel Kiper <daniel...@oracle.com>
>
> The functions calculate lowest and highest available RAM
> addresses respectively.
>
> Both functions are needed to calculate PMR boundaries for
> Intel TXT secure launcher introduced by subsequent patches.

After discussing this we think the best course of action is to submit
the first 7 patches standalone (also see my response to patch 11 that is
coming). For this patch I would just remove the last sentence about TXT
so it is not tied to the DRTM work.

Then you can submit this set (with the couple of the changes I suggested
earlier) and we can more easily get this part in.

Thanks
Ross

[Sergii Dmytruk]

Sent 0-7 and part of 11 separately taking all the comments into account:

https://lists.gnu.org/archive/html/grub-devel/2024-09/msg00194.html

Regards,
Sergii

[ross.ph...@oracle.com]

On 9/19/24 3:23 PM, Sergii Dmytruk wrote:
> Sent 0-7 and part of 11 separately taking all the comments into account:
>
> https://lists.gnu.org/archive/html/grub-devel/2024-09/msg00194.html

Great thanks. I will look them over, thanks for doing that. This should
make them easy to get in.

Ross