From: Alexandre Ghiti <alex@ghiti.fr>
To: Paul Walmsley <paul.walmsley@sifive.com>,
Palmer Dabbelt <palmer@dabbelt.com>, Zong Li <zong.li@sifive.com>,
Anup Patel <anup@brainfault.org>, Christoph Hellwig <hch@lst.de>,
Ard Biesheuvel <ardb@kernel.org>, Arnd Bergmann <arnd@arndb.de>,
linux-efi@vger.kernel.org, linux-arch@vger.kernel.org,
linux-riscv@lists.infradead.org, linux-kernel@vger.kernel.org
Cc: Alexandre Ghiti <alex@ghiti.fr>
Subject: [RFC PATCH 02/12] riscv: Protect the kernel linear mapping
Date: Mon, 4 Jan 2021 14:58:30 -0500 [thread overview]
Message-ID: <20210104195840.1593-3-alex@ghiti.fr> (raw)
In-Reply-To: <20210104195840.1593-1-alex@ghiti.fr>
The kernel is now mapped at the end of the address space and it should
be accessed through this mapping only: so map the whole kernel in the
linear mapping as read only.
Signed-off-by: Alexandre Ghiti <alex@ghiti.fr>
---
arch/riscv/include/asm/page.h | 9 ++++++++-
arch/riscv/mm/init.c | 29 +++++++++++++++++++++--------
2 files changed, 29 insertions(+), 9 deletions(-)
diff --git a/arch/riscv/include/asm/page.h b/arch/riscv/include/asm/page.h
index 98188e315e8d..a93e35aaa717 100644
--- a/arch/riscv/include/asm/page.h
+++ b/arch/riscv/include/asm/page.h
@@ -102,8 +102,15 @@ extern unsigned long pfn_base;
extern unsigned long max_low_pfn;
extern unsigned long min_low_pfn;
extern unsigned long kernel_virt_addr;
+extern uintptr_t load_pa, load_sz;
+
+#define linear_mapping_pa_to_va(x) ((void *)((unsigned long)(x) + va_pa_offset))
+#define kernel_mapping_pa_to_va(x) \
+ ((void *)((unsigned long) (x) + va_kernel_pa_offset))
+#define __pa_to_va_nodebug(x) \
+ ((x >= load_pa && x < load_pa + load_sz) ? \
+ kernel_mapping_pa_to_va(x): linear_mapping_pa_to_va(x))
-#define __pa_to_va_nodebug(x) ((void *)((unsigned long) (x) + va_pa_offset))
#define linear_mapping_va_to_pa(x) ((unsigned long)(x) - va_pa_offset)
#define kernel_mapping_va_to_pa(x) \
((unsigned long)(x) - va_kernel_pa_offset)
diff --git a/arch/riscv/mm/init.c b/arch/riscv/mm/init.c
index 9d06ff0e015a..7b87c14f1d24 100644
--- a/arch/riscv/mm/init.c
+++ b/arch/riscv/mm/init.c
@@ -159,8 +159,6 @@ void __init setup_bootmem(void)
{
phys_addr_t mem_start = 0;
phys_addr_t start, end = 0;
- phys_addr_t vmlinux_end = __pa_symbol(&_end);
- phys_addr_t vmlinux_start = __pa_symbol(&_start);
u64 i;
/* Find the memory region containing the kernel */
@@ -168,7 +166,7 @@ void __init setup_bootmem(void)
phys_addr_t size = end - start;
if (!mem_start)
mem_start = start;
- if (start <= vmlinux_start && vmlinux_end <= end)
+ if (start <= load_pa && (load_pa + load_sz) <= end)
BUG_ON(size == 0);
}
@@ -179,8 +177,13 @@ void __init setup_bootmem(void)
*/
memblock_enforce_memory_limit(mem_start - PAGE_OFFSET);
- /* Reserve from the start of the kernel to the end of the kernel */
- memblock_reserve(vmlinux_start, vmlinux_end - vmlinux_start);
+ /*
+ * Reserve from the start of the kernel to the end of the kernel
+ * and make sure we align the reservation on PMD_SIZE since we will
+ * map the kernel in the linear mapping as read-only: we do not want
+ * any allocation to happen between _end and the next pmd aligned page.
+ */
+ memblock_reserve(load_pa, (load_sz + PMD_SIZE - 1) & ~(PMD_SIZE - 1));
max_pfn = PFN_DOWN(memblock_end_of_DRAM());
max_low_pfn = max_pfn;
@@ -438,7 +441,9 @@ static uintptr_t __init best_map_size(phys_addr_t base, phys_addr_t size)
#error "setup_vm() is called from head.S before relocate so it should not use absolute addressing."
#endif
-static uintptr_t load_pa, load_sz;
+uintptr_t load_pa, load_sz;
+EXPORT_SYMBOL(load_pa);
+EXPORT_SYMBOL(load_sz);
static void __init create_kernel_page_table(pgd_t *pgdir, uintptr_t map_size)
{
@@ -596,9 +601,17 @@ static void __init setup_vm_final(void)
map_size = best_map_size(start, end - start);
for (pa = start; pa < end; pa += map_size) {
- va = (uintptr_t)__va(pa);
+ pgprot_t prot = PAGE_KERNEL;
+
+ /* Protect the kernel mapping that lies in the linear mapping */
+ if (pa >= __pa(_start) && pa < __pa(_end))
+ prot = PAGE_KERNEL_READ;
+
+ /* Make sure we get virtual addresses in the linear mapping */
+ va = (uintptr_t)linear_mapping_pa_to_va(pa);
+
create_pgd_mapping(swapper_pg_dir, va, pa,
- map_size, PAGE_KERNEL);
+ map_size, prot);
}
}
--
2.20.1
next prev parent reply other threads:[~2021-01-04 20:14 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-01-04 19:58 [RFC PATCH 00/12] Introduce sv48 support without relocable kernel Alexandre Ghiti
2021-01-04 19:58 ` [RFC PATCH 01/12] riscv: Move kernel mapping outside of linear mapping Alexandre Ghiti
2021-01-05 11:40 ` Anup Patel
2021-01-06 6:36 ` Alex Ghiti
2021-01-06 6:44 ` Anup Patel
2021-01-06 8:16 ` Alex Ghiti
2021-01-04 19:58 ` Alexandre Ghiti [this message]
2021-01-05 11:50 ` [RFC PATCH 02/12] riscv: Protect the kernel " Anup Patel
2021-01-04 19:58 ` [RFC PATCH 03/12] riscv: Get rid of compile time logic with MAX_EARLY_MAPPING_SIZE Alexandre Ghiti
2021-01-04 19:58 ` [RFC PATCH 04/12] riscv: Allow to dynamically define VA_BITS Alexandre Ghiti
2021-01-05 12:06 ` Anup Patel
2021-01-06 6:38 ` Alex Ghiti
2021-01-04 19:58 ` [RFC PATCH 05/12] riscv: Simplify MAXPHYSMEM config Alexandre Ghiti
2021-01-04 19:58 ` [RFC PATCH 06/12] riscv: Prepare ptdump for vm layout dynamic addresses Alexandre Ghiti
2021-01-04 19:58 ` [RFC PATCH 07/12] asm-generic: Prepare for riscv use of pud_alloc_one and pud_free Alexandre Ghiti
2021-01-04 19:58 ` [RFC PATCH 08/12] riscv: Implement sv48 support Alexandre Ghiti
2021-01-04 19:58 ` [RFC PATCH 09/12] riscv: Allow user to downgrade to sv39 when hw supports sv48 Alexandre Ghiti
2021-01-04 19:58 ` [RFC PATCH 10/12] riscv: Use pgtable_l4_enabled to output mmu type in cpuinfo Alexandre Ghiti
2021-01-04 19:58 ` [RFC PATCH 11/12] riscv: Explicit comment about user virtual address space size Alexandre Ghiti
2021-01-04 19:58 ` [RFC PATCH 12/12] riscv: Improve virtual kernel memory layout dump Alexandre Ghiti
2021-01-30 9:33 ` [RFC PATCH 00/12] Introduce sv48 support without relocable kernel Alex Ghiti
2021-02-03 3:04 ` Palmer Dabbelt
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20210104195840.1593-3-alex@ghiti.fr \
--to=alex@ghiti.fr \
--cc=anup@brainfault.org \
--cc=ardb@kernel.org \
--cc=arnd@arndb.de \
--cc=hch@lst.de \
--cc=linux-arch@vger.kernel.org \
--cc=linux-efi@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-riscv@lists.infradead.org \
--cc=palmer@dabbelt.com \
--cc=paul.walmsley@sifive.com \
--cc=zong.li@sifive.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).