Received-SPF: pass (google.com: domain of linux-kernel+bounces-100894-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:40f1:3f00::1 as permitted sender) client-ip=2604:1380:40f1:3f00::1;
From: Rick Edgecombe <rick.p.edgecombe@intel.com>
To: Liam.Howlett@oracle.com,
	akpm@linux-foundation.org,
	bp@alien8.de,
	broonie@kernel.org,
	dave.hansen@linux.intel.com,
	debug@rivosinc.com,
	hpa@zytor.com,
	keescook@chromium.org,
	kirill.shutemov@linux.intel.com,
	luto@kernel.org,
	mingo@redhat.com,
	peterz@infradead.org,
	tglx@linutronix.de,
	x86@kernel.org,
	christophe.leroy@csgroup.eu
Cc: linux-kernel@vger.kernel.org,
	linux-mm@kvack.org,
	rick.p.edgecombe@intel.com
Subject: [PATCH v3 09/12] mm: Take placement mappings gap into account
Date: Tue, 12 Mar 2024 15:28:40 -0700
Message-Id: <20240312222843.2505560-10-rick.p.edgecombe@intel.com>
In-Reply-To: <20240312222843.2505560-1-rick.p.edgecombe@intel.com>
References: <20240312222843.2505560-1-rick.p.edgecombe@intel.com>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

When memory is being placed, mmap() will take care to respect the guard
gaps of certain types of memory (VM_SHADOWSTACK, VM_GROWSUP and
VM_GROWSDOWN). In order to ensure guard gaps between mappings, mmap()
needs to consider two things:
 1. That the new mapping isn’t placed in an any existing mappings guard
    gaps.
 2. That the new mapping isn’t placed such that any existing mappings
    are not in *its* guard gaps.

The long standing behavior of mmap() is to ensure 1, but not take any care
around 2. So for example, if there is a PAGE_SIZE free area, and a
mmap() with a PAGE_SIZE size, and a type that has a guard gap is being
placed, mmap() may place the shadow stack in the PAGE_SIZE free area. Then
the mapping that is supposed to have a guard gap will not have a gap to
the adjacent VMA.

For MAP_GROWSDOWN/VM_GROWSDOWN and MAP_GROWSUP/VM_GROWSUP this has not
been a problem in practice because applications place these kinds of
mappings very early, when there is not many mappings to find a space
between. But for shadow stacks, they may be placed throughout the lifetime
of the application.

Use the start_gap field to find a space that includes the guard gap for
the new mapping. Take care to not interfere with the alignment.

Signed-off-by: Rick Edgecombe <rick.p.edgecombe@intel.com>
---
v3:
 - Spelling fix in comment

v2:
 - Remove VM_UNMAPPED_START_GAP_SET and have struct vm_unmapped_area_info
   initialized with zeros (in another patch). (Kirill)
 - Drop unrelated space change (Kirill)
 - Add comment around interactions of alignment and start gap step
   (Kirill)
---
 include/linux/mm.h |  1 +
 mm/mmap.c          | 12 +++++++++---
 2 files changed, 10 insertions(+), 3 deletions(-)

diff --git a/include/linux/mm.h b/include/linux/mm.h
index d91cde79aaee..deade7be00d0 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -3418,6 +3418,7 @@ struct vm_unmapped_area_info {
 	unsigned long high_limit;
 	unsigned long align_mask;
 	unsigned long align_offset;
+	unsigned long start_gap;
 };
 
 extern unsigned long vm_unmapped_area(struct vm_unmapped_area_info *info);
diff --git a/mm/mmap.c b/mm/mmap.c
index b889c79d11bd..634e706fd97e 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -1582,7 +1582,7 @@ static unsigned long unmapped_area(struct vm_unmapped_area_info *info)
 	MA_STATE(mas, &current->mm->mm_mt, 0, 0);
 
 	/* Adjust search length to account for worst case alignment overhead */
-	length = info->length + info->align_mask;
+	length = info->length + info->align_mask + info->start_gap;
 	if (length < info->length)
 		return -ENOMEM;
 
@@ -1594,7 +1594,13 @@ static unsigned long unmapped_area(struct vm_unmapped_area_info *info)
 	if (mas_empty_area(&mas, low_limit, high_limit - 1, length))
 		return -ENOMEM;
 
-	gap = mas.index;
+	/*
+	 * Adjust for the gap first so it doesn't interfere with the
+	 * later alignment. The first step is the minimum needed to
+	 * fulill the start gap, the next steps is the minimum to align
+	 * that. It is the minimum needed to fulill both.
+	 */
+	gap = mas.index + info->start_gap;
 	gap += (info->align_offset - gap) & info->align_mask;
 	tmp = mas_next(&mas, ULONG_MAX);
 	if (tmp && (tmp->vm_flags & VM_STARTGAP_FLAGS)) { /* Avoid prev check if possible */
@@ -1633,7 +1639,7 @@ static unsigned long unmapped_area_topdown(struct vm_unmapped_area_info *info)
 
 	MA_STATE(mas, &current->mm->mm_mt, 0, 0);
 	/* Adjust search length to account for worst case alignment overhead */
-	length = info->length + info->align_mask;
+	length = info->length + info->align_mask + info->start_gap;
 	if (length < info->length)
 		return -ENOMEM;
 
-- 
2.34.1