Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
From:   sxwjean@me.com
To:     linux-mm@kvack.org, x86@kernel.org
Cc:     Xiongwei Song <sxwjean@gmail.com>,
        Thomas Gleixner <tglx@linutronix.de>,
        Ingo Molnar <mingo@redhat.com>, Borislav Petkov <bp@alien8.de>,
        Kees Cook <keescook@chromium.org>,
        "H. Peter Anvin" <hpa@zytor.com>,
        Dave Hansen <dave.hansen@linux.intel.com>,
        Andy Lutomirski <luto@kernel.org>,
        Peter Zijlstra <peterz@infradead.org>,
        Andrew Morton <akpm@linux-foundation.org>,
        Arnd Bergmann <arnd@arndb.de>,
        Al Viro <viro@zeniv.linux.org.uk>,
        Gabriel Krisman Bertazi <krisman@collabora.com>,
        Lai Jiangshan <laijs@linux.alibaba.com>,
        Huang Rui <ray.huang@amd.com>,
        Yazen Ghannam <yazen.ghannam@amd.com>,
        Kim Phillips <kim.phillips@amd.com>,
        Oleg Nesterov <oleg@redhat.com>,
        Balbir Singh <sblbir@amazon.com>,
        "David S. Miller" <davem@davemloft.net>, sxwjean@me.com,
        linux-kernel@vger.kernel.org
Subject: [PATCH v2 2/6] mm/util: Allow to pass a specific task size when getting mmapping base
Date:   Mon, 11 Oct 2021 22:31:46 +0800
Message-Id: <20211011143150.318239-3-sxwjean@me.com>
In-Reply-To: <20211011143150.318239-1-sxwjean@me.com>
References: <20211011143150.318239-1-sxwjean@me.com>
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Precedence: bulk

From: Xiongwei Song <sxwjean@gmail.com>

In x86, a 64bit task may invoke a 32 bit syscall, which is in compat
syscall. Then we have to provide 32bit mapping base.

Signed-off-by: Xiongwei Song <sxwjean@gmail.com>
---
 mm/util.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/mm/util.c b/mm/util.c
index 40b1a8837c0b..38326ef21a3b 100644
--- a/mm/util.c
+++ b/mm/util.c
@@ -385,14 +385,15 @@ static int mmap_is_legacy(struct rlimit *rlim_stack)
 #define MIN_GAP		(SZ_128M)
 #define MAX_GAP		(STACK_TOP / 6 * 5)
 
-static unsigned long mmap_base(unsigned long rnd, struct rlimit *rlim_stack)
+static unsigned long mmap_base(unsigned long rnd, unsigned long task_size,
+			       struct rlimit *rlim_stack)
 {
 	unsigned long gap = rlim_stack->rlim_cur;
 	unsigned long pad = stack_guard_gap;
 
 	/* Account for stack randomization if necessary */
 	if (current->flags & PF_RANDOMIZE)
-		pad += (STACK_RND_MASK << PAGE_SHIFT);
+		pad += (in_compat_syscall() ? 0x7ff : STACK_RND_MASK) << PAGE_SHIFT;
 
 	/* Values close to RLIM_INFINITY can overflow. */
 	if (gap + pad > gap)
@@ -403,7 +404,7 @@ static unsigned long mmap_base(unsigned long rnd, struct rlimit *rlim_stack)
 	else if (gap > MAX_GAP)
 		gap = MAX_GAP;
 
-	return PAGE_ALIGN(STACK_TOP - gap - rnd);
+	return PAGE_ALIGN(task_size - gap - rnd);
 }
 
 void arch_pick_mmap_layout(struct mm_struct *mm, struct rlimit *rlim_stack)
@@ -418,7 +419,7 @@ void arch_pick_mmap_layout(struct mm_struct *mm, struct rlimit *rlim_stack)
 		mm->mmap_base = mm->mmap_legacy_base;
 		mm->get_unmapped_area = arch_get_unmapped_area;
 	} else {
-		mm->mmap_base = mmap_base(random_factor, rlim_stack);
+		mm->mmap_base = mmap_base(random_factor, STACK_TOP, rlim_stack);
 		mm->get_unmapped_area = arch_get_unmapped_area_topdown;
 	}
 }
-- 
2.30.2