diff mbox series

[09/11] sparc32: use PUD rather than PGD to get PMD in srmmu_nocache_init()

Message ID 20200523052309.4caVN81-C%akpm@linux-foundation.org (mailing list archive)
State New, archived
Headers show
Series [01/11] device-dax: don't leak kernel memory to user space after unloading kmem | expand

Commit Message

Andrew Morton May 23, 2020, 5:23 a.m. UTC 
From: Mike Rapoport <rppt@linux.ibm.com>
Subject: sparc32: use PUD rather than PGD to get PMD in srmmu_nocache_init()

The kbuild test robot reported the following warning:

arch/sparc/mm/srmmu.c: In function 'srmmu_nocache_init':
>> arch/sparc/mm/srmmu.c:300:9: error: variable 'pud' set but not used
>> [-Werror=unused-but-set-variable]
300 |  pud_t *pud;

This warning is caused by misprint in the page table traversal in
srmmu_nocache_init() function which accessed a PMD entry using PGD rather
than PUD.

Since sparc32 has only 3 page table levels, the PGD and PUD are
essentially the same and usage of __nocache_fix() removed the type
checking.

Use PUD for the consistency and to silence the compiler warning.

Link: http://lkml.kernel.org/r/20200520132005.GM1059226@linux.ibm.com
Fixes: 7235db268a2777bc38 ("sparc32: use pgtable-nopud instead of 4level-fixup")
Signed-off-by: Mike Rapoport <rppt@linux.ibm.com>
Reported-by: kbuild test robot <lkp@intel.com>
Cc: David S. Miller <davem@davemloft.net>
Cc: Anatoly Pugachev <matorola@gmail.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---

 arch/sparc/mm/srmmu.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

Comments

Mike Rapoport May 23, 2020, 7:01 p.m. UTC | #1 
On Fri, May 22, 2020 at 10:23:09PM -0700, Andrew Morton wrote:
> From: Mike Rapoport <rppt@linux.ibm.com>
> Subject: sparc32: use PUD rather than PGD to get PMD in srmmu_nocache_init()
> 
> The kbuild test robot reported the following warning:
> 
> arch/sparc/mm/srmmu.c: In function 'srmmu_nocache_init':
> >> arch/sparc/mm/srmmu.c:300:9: error: variable 'pud' set but not used
> >> [-Werror=unused-but-set-variable]
> 300 |  pud_t *pud;
> 
> This warning is caused by misprint in the page table traversal in
> srmmu_nocache_init() function which accessed a PMD entry using PGD rather
> than PUD.
> 
> Since sparc32 has only 3 page table levels, the PGD and PUD are
> essentially the same and usage of __nocache_fix() removed the type
> checking.
> 
> Use PUD for the consistency and to silence the compiler warning.

Unfortunately, this fixes a compile warning but breaks the boot :(

Since pgd, p4d and pud are eesentially the same thing and pgd_offset()
and p4d_offset() are no-ops, the __nocache_fix() should be done only at
pud level.

The correcteted patch is below, boot tested with qemu-systems-sparc.

> Link: http://lkml.kernel.org/r/20200520132005.GM1059226@linux.ibm.com
> Fixes: 7235db268a2777bc38 ("sparc32: use pgtable-nopud instead of 4level-fixup")
> Signed-off-by: Mike Rapoport <rppt@linux.ibm.com>
> Reported-by: kbuild test robot <lkp@intel.com>
> Cc: David S. Miller <davem@davemloft.net>
> Cc: Anatoly Pugachev <matorola@gmail.com>
> Cc: <stable@vger.kernel.org>
> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
> ---

From 9b8b3214f7f079864dca0c6a7b684ab442eeb437 Mon Sep 17 00:00:00 2001
From: Mike Rapoport <rppt@linux.ibm.com>
Date: Wed, 20 May 2020 15:39:22 +0300
Subject: [PATCH] sparc32: use PUD rather than PGD to get PMD in  srmmu_nocache_init()

The kbuild test robot reported the following warning:

arch/sparc/mm/srmmu.c: In function 'srmmu_nocache_init':
>> arch/sparc/mm/srmmu.c:300:9: error: variable 'pud' set but not used
>> [-Werror=unused-but-set-variable]
300 |  pud_t *pud;

This warning is caused by misprint in the page table traversal in
srmmu_nocache_init() function which accessed a PMD entry using PGD rather
than PUD.

Since pgd, p4d and pud are eesentially the same thing and pgd_offset()
and p4d_offset() are no-ops, the __nocache_fix() should be done only at
pud level.

Remove __nocache_fix() for p4d_offset() and pud_offset() and use it only
for PUD and lower levels.

Link: http://lkml.kernel.org/r/20200520132005.GM1059226@linux.ibm.com
Fixes: 7235db268a2777bc38 ("sparc32: use pgtable-nopud instead of 4level-fixup")
Signed-off-by: Mike Rapoport <rppt@linux.ibm.com>
Reported-by: kbuild test robot <lkp@intel.com>
Cc: David S. Miller <davem@davemloft.net>
Cc: Anatoly Pugachev <matorola@gmail.com>
Cc: <stable@vger.kernel.org>
---
 arch/sparc/mm/srmmu.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/arch/sparc/mm/srmmu.c b/arch/sparc/mm/srmmu.c
index 6cb1ea2d2b5c..3d9690d940a3 100644
--- a/arch/sparc/mm/srmmu.c
+++ b/arch/sparc/mm/srmmu.c
@@ -302,9 +302,9 @@ static void __init srmmu_nocache_init(void)
 
 	while (vaddr < srmmu_nocache_end) {
 		pgd = pgd_offset_k(vaddr);
-		p4d = p4d_offset(__nocache_fix(pgd), vaddr);
-		pud = pud_offset(__nocache_fix(p4d), vaddr);
-		pmd = pmd_offset(__nocache_fix(pgd), vaddr);
+		p4d = p4d_offset(pgd, vaddr);
+		pud = pud_offset(p4d, vaddr);
+		pmd = pmd_offset(__nocache_fix(pud), vaddr);
 		pte = pte_offset_kernel(__nocache_fix(pmd), vaddr);
 
 		pteval = ((paddr >> 4) | SRMMU_ET_PTE | SRMMU_PRIV);
Linus Torvalds May 23, 2020, 7:10 p.m. UTC | #2 
On Sat, May 23, 2020 at 12:02 PM Mike Rapoport <rppt@linux.ibm.com> wrote:
>
> Unfortunately, this fixes a compile warning but breaks the boot :(

Argh. I delayed applying/merging this overnight to see if there were
any reports, but this came in after I'd already merged Andrew's
patches and pushed them out.

So it's in my tree now as commit c2bc26f7ca1f ("sparc32: use PUD
rather than PGD to get PMD in srmmu_nocache_init()")

> The correcteted patch is below, boot tested with qemu-systems-sparc.

Mind sending a patch relative to the previous one that already got merged?

Also, would it perhaps be worth it to just make __nocache_fix() not
throw the type away? IOW, make it do something like

  #define __nocache_fix(VADDR) \
        ((__typeof__(VADDR))__va(__nocache_pa(VADDR)))

or whatever? Wouldn't that show when those pgd/p4d/pud pointers get
mis-used because they don't end up dropping the type info..

              Linus
Mike Rapoport May 23, 2020, 7:57 p.m. UTC | #3 
On Sat, May 23, 2020 at 12:10:27PM -0700, Linus Torvalds wrote:
> On Sat, May 23, 2020 at 12:02 PM Mike Rapoport <rppt@linux.ibm.com> wrote:
> >
> > Unfortunately, this fixes a compile warning but breaks the boot :(
> 
> Argh. I delayed applying/merging this overnight to see if there were
> any reports, but this came in after I'd already merged Andrew's
> patches and pushed them out.

Actually, it's really by chance I noticed it tonight, although still it
was too late :)

> So it's in my tree now as commit c2bc26f7ca1f ("sparc32: use PUD
> rather than PGD to get PMD in srmmu_nocache_init()")
>
> > The correcteted patch is below, boot tested with qemu-systems-sparc.
> 
> Mind sending a patch relative to the previous one that already got merged?
 
Sure.

> Also, would it perhaps be worth it to just make __nocache_fix() not
> throw the type away? IOW, make it do something like
> 
>   #define __nocache_fix(VADDR) \
>         ((__typeof__(VADDR))__va(__nocache_pa(VADDR)))
> 
> or whatever? Wouldn't that show when those pgd/p4d/pud pointers get
> mis-used because they don't end up dropping the type info..

Yes, I'll look into it.

>               Linus
diff mbox series

Patch

--- a/arch/sparc/mm/srmmu.c~sparc32-use-pud-rather-than-pgd-to-get-pmd-in-srmmu_nocache_init
+++ a/arch/sparc/mm/srmmu.c
@@ -333,7 +333,7 @@  static void __init srmmu_nocache_init(vo
 		pgd = pgd_offset_k(vaddr);
 		p4d = p4d_offset(__nocache_fix(pgd), vaddr);
 		pud = pud_offset(__nocache_fix(p4d), vaddr);
-		pmd = pmd_offset(__nocache_fix(pgd), vaddr);
+		pmd = pmd_offset(__nocache_fix(pud), vaddr);
 		pte = pte_offset_kernel(__nocache_fix(pmd), vaddr);
 
 		pteval = ((paddr >> 4) | SRMMU_ET_PTE | SRMMU_PRIV);