qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Yongbok Kim <yongbok.kim@imgtec.com>
To: James Hogan <james.hogan@imgtec.com>, qemu-devel@nongnu.org
Cc: Aurelien Jarno <aurelien@aurel32.net>
Subject: Re: [Qemu-devel] [PATCH 4/9] target-mips: Check memory permissions with mem_idx
Date: Fri, 7 Oct 2016 16:48:31 +0100	[thread overview]
Message-ID: <6785d7db-8866-9f50-b182-24c7ed69e34d@imgtec.com> (raw)
In-Reply-To: <e713111f058685914efa8ce0be14197b151cfa9a.1473159543.git-series.james.hogan@imgtec.com>



On 06/09/2016 12:03, James Hogan wrote:
> When performing virtual to physical address translation, check the
> required privilege level based on the mem_idx rather than the mode in
> the hflags. This will allow EVA loads & stores to operate safely only on
> user memory from kernel mode.
> 
> For the cases where the mmu_idx doesn't need to be overridden
> (mips_cpu_get_phys_page_debug() and cpu_mips_translate_address()), we
> calculate the required mmu_idx using cpu_mmu_index(). Note that this
> only tests the MIPS_HFLAG_KSU bits rather than MIPS_HFLAG_MODE, so we
> don't test the debug mode hflag MIPS_HFLAG_DM any longer. This should be
> fine as get_physical_address() only compares against MIPS_HFLAG_UM and
> MIPS_HFLAG_SM, neither of which should get set by compute_hflags() when
> MIPS_HFLAG_DM is set.
> 
> Signed-off-by: James Hogan <james.hogan@imgtec.com>
> Cc: Leon Alrae <leon.alrae@imgtec.com>
> Cc: Aurelien Jarno <aurelien@aurel32.net>
> ---
>  target-mips/helper.c | 17 +++++++++--------
>  1 file changed, 9 insertions(+), 8 deletions(-)
> 
> diff --git a/target-mips/helper.c b/target-mips/helper.c
> index 29ebf391cb94..2065fc3ec119 100644
> --- a/target-mips/helper.c
> +++ b/target-mips/helper.c
> @@ -109,11 +109,11 @@ int r4k_map_address (CPUMIPSState *env, hwaddr *physical, int *prot,
>  
>  static int get_physical_address (CPUMIPSState *env, hwaddr *physical,
>                                  int *prot, target_ulong real_address,
> -                                int rw, int access_type)
> +                                int rw, int access_type, int mmu_idx)
>  {
>      /* User mode can only access useg/xuseg */
> -    int user_mode = (env->hflags & MIPS_HFLAG_MODE) == MIPS_HFLAG_UM;
> -    int supervisor_mode = (env->hflags & MIPS_HFLAG_MODE) == MIPS_HFLAG_SM;
> +    int user_mode = mmu_idx == MIPS_HFLAG_UM;
> +    int supervisor_mode = mmu_idx == MIPS_HFLAG_SM;
>      int kernel_mode = !user_mode && !supervisor_mode;
>  #if defined(TARGET_MIPS64)
>      int UX = (env->CP0_Status & (1 << CP0St_UX)) != 0;
> @@ -413,11 +413,12 @@ static void raise_mmu_exception(CPUMIPSState *env, target_ulong address,
>  hwaddr mips_cpu_get_phys_page_debug(CPUState *cs, vaddr addr)
>  {
>      MIPSCPU *cpu = MIPS_CPU(cs);
> +    CPUMIPSState *env = &cpu->env;

Not really useful change as it is used only once but it is ok.

>      hwaddr phys_addr;
>      int prot;
>  
> -    if (get_physical_address(&cpu->env, &phys_addr, &prot, addr, 0,
> -                             ACCESS_INT) != 0) {
> +    if (get_physical_address(env, &phys_addr, &prot, addr, 0, ACCESS_INT,
> +                             cpu_mmu_index(env, false)) != 0) {
>          return -1;
>      }
>      return phys_addr;
> @@ -449,7 +450,7 @@ int mips_cpu_handle_mmu_fault(CPUState *cs, vaddr address, int rw,
>         correctly */
>      access_type = ACCESS_INT;
>      ret = get_physical_address(env, &physical, &prot,
> -                               address, rw, access_type);
> +                               address, rw, access_type, mmu_idx);
>      qemu_log_mask(CPU_LOG_MMU,
>               "%s address=%" VADDR_PRIx " ret %d physical " TARGET_FMT_plx
>               " prot %d\n",
> @@ -479,8 +480,8 @@ hwaddr cpu_mips_translate_address(CPUMIPSState *env, target_ulong address, int r
>  
>      /* data access */
>      access_type = ACCESS_INT;
> -    ret = get_physical_address(env, &physical, &prot,
> -                               address, rw, access_type);
> +    ret = get_physical_address(env, &physical, &prot, address, rw, access_type,
> +                               cpu_mmu_index(env, false));
>      if (ret != TLBRET_MATCH) {
>          raise_mmu_exception(env, address, rw, ret);
>          return -1LL;
> 

Otherwise,

Reviewed-by: Yongbok Kim <yongbok.kim@imgtec.com>


Regards,
Yongbok

  reply	other threads:[~2016-10-07 15:48 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-09-06 11:03 [Qemu-devel] [PATCH 0/9] target-mips: Add Enhanced Virtual Addressing (EVA) support James Hogan
2016-09-06 11:03 ` [Qemu-devel] [PATCH 1/9] target-mips: Add CP0_Ebase.WG (write gate) support James Hogan
2016-10-07 13:42   ` Yongbok Kim
2016-10-07 15:37     ` James Hogan
2016-09-06 11:03 ` [Qemu-devel] [PATCH 2/9] target-mips: Prepare loads/stores for EVA James Hogan
2016-10-07 15:32   ` Yongbok Kim
2016-09-06 11:03 ` [Qemu-devel] [PATCH 3/9] target-mips: Decode EVA load & store instructions James Hogan
2016-10-07 15:34   ` Yongbok Kim
2016-10-07 15:48     ` James Hogan
2016-10-07 16:05       ` Yongbok Kim
2016-10-07 16:16         ` James Hogan
2016-09-06 11:03 ` [Qemu-devel] [PATCH 4/9] target-mips: Check memory permissions with mem_idx James Hogan
2016-10-07 15:48   ` Yongbok Kim [this message]
2017-07-06 20:50     ` James Hogan
2016-09-06 11:03 ` [Qemu-devel] [PATCH 5/9] target-mips: Abstract mmu_idx from hflags James Hogan
2016-10-07 16:08   ` Yongbok Kim
2017-07-06 20:55     ` James Hogan
2016-09-06 11:03 ` [Qemu-devel] [PATCH 6/9] target-mips: Add an MMU mode for ERL James Hogan
2016-10-13 13:18   ` Yongbok Kim
2017-07-06 20:58     ` James Hogan
2016-09-06 11:03 ` [Qemu-devel] [PATCH 7/9] target-mips: Add segmentation control registers James Hogan
2016-10-10 14:57   ` Yongbok Kim
2017-07-06 22:05     ` James Hogan
2016-09-06 11:03 ` [Qemu-devel] [PATCH 8/9] target-mips: Implement segmentation control James Hogan
2016-10-13 13:06   ` Yongbok Kim
2017-07-06 23:15     ` James Hogan
2016-09-06 11:03 ` [Qemu-devel] [PATCH 9/9] target-mips: Add EVA support to P5600 James Hogan
2016-10-13 13:15   ` Yongbok Kim
2016-09-06 11:18 ` [Qemu-devel] [PATCH 0/9] target-mips: Add Enhanced Virtual Addressing (EVA) support no-reply
2016-09-06 12:09   ` James Hogan

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=6785d7db-8866-9f50-b182-24c7ed69e34d@imgtec.com \
    --to=yongbok.kim@imgtec.com \
    --cc=aurelien@aurel32.net \
    --cc=james.hogan@imgtec.com \
    --cc=qemu-devel@nongnu.org \
    /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).