linux-kernel - Re: [PATCH v2 7/7] ARM: mm: Fix PXN process with LPAE feature

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <c9c4a8ec-da51-4de9-4404-b5bf7f017441@huawei.com>
Date:   Thu, 3 Jun 2021 17:38:51 +0800
From:   Kefeng Wang <wangkefeng.wang@...wei.com>
To:     "Russell King (Oracle)" <linux@...linux.org.uk>
CC:     <linux-arm-kernel@...ts.infradead.org>,
        Catalin Marinas <catalin.marinas@....com>,
        <linux-kernel@...r.kernel.org>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Jungseung Lee <js07.lee@...il.com>
Subject: Re: [PATCH v2 7/7] ARM: mm: Fix PXN process with LPAE feature


On 2021/6/2 23:58, Russell King (Oracle) wrote:
> On Wed, Jun 02, 2021 at 11:13:14PM +0800, Kefeng Wang wrote:
>>    IFSR format when using the Short-descriptor translation table format
>>
>>      Domain fault      01001            First level   01011     Second level
>>
>>      Permission fault 01101            First level   01111     Second level
>>
>>    IFSR format when using the Long-descriptor translation table format
>>
>>     0011LL Permission fault. LL bits indicate levelb.
>>
>> After check the ARM spec, I think for the permission fault,  we should panic
>> with or without LPAE, will change to
> As I explained in one of the previous patches, the page tables that get
> used for mapping kernel space are the _tasks_ own page tables. Any new
> kernel mappings are lazily copied to the task page tables - such as
> when a module is loaded.
>
> The first time we touch a page, we could end up with a page translation
> fault. This will call do_page_fault(), and so with your proposal,
> loading a module will potentially cause a kernel panic in this case,
> probably leading to systems that panic early during userspace boot.

Could we add some FSR_FS check， only panic when the permission fault， eg，

+static inline bool is_permission_fault(unsigned int fsr)
+{
+       int fs = fsr_fs(fsr);
+#ifdef CONFIG_ARM_LPAE
+       if ((fs & FS_PERM_NOLL_MASK) == FS_PERM_NOLL)
+               return true;
+#else
+       if (fs == FS_L1_PERM || fs == )
+               return true;
+#endif
+       return false;
+}
+
  static int __kprobes
  do_page_fault(unsigned long addr, unsigned int fsr, struct pt_regs *regs)
  {
@@ -255,8 +268,7 @@ do_page_fault(unsigned long addr, unsigned int fsr, 
struct pt_regs *regs)

         if (fsr & FSR_LNX_PF) {
                 vm_flags = VM_EXEC;
-
-               if (!user_mode(regs))
+               if (is_permission_fault && !user_mode(regs))
                         die_kernel_fault("execution of memory",
                                          mm, addr, fsr, regs);
         }

diff --git a/arch/arm/mm/fault.h b/arch/arm/mm/fault.h
index 9ecc2097a87a..187954b4acca 100644
--- a/arch/arm/mm/fault.h
+++ b/arch/arm/mm/fault.h
@@ -14,6 +14,8 @@

  #ifdef CONFIG_ARM_LPAE
  #define FSR_FS_AEA             17
+#define FS_PERM_NOLL           0xC
+#define FS_PERM_NOLL_MASK      0x3C

  static inline int fsr_fs(unsigned int fsr)
  {
@@ -21,6 +23,8 @@ static inline int fsr_fs(unsigned int fsr)
  }
  #else
  #define FSR_FS_AEA             22
+#define FS_L1_PERM             0xD
+#define FS_L2_PERM             0xF

and suggestion or proper solution to solve the issue？

>