linux-kernel - Re: [PATCH 6.10 000/809] 6.10.3-rc3 review

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: <314821a5-43e1-4c96-ad8d-188bca8ae0a1@roeck-us.net>
Date: Thu, 8 Aug 2024 17:50:32 -0700
From: Guenter Roeck <linux@...ck-us.net>
To: John David Anglin <dave.anglin@...l.net>,
 Linus Torvalds <torvalds@...ux-foundation.org>,
 Thomas Gleixner <tglx@...utronix.de>
Cc: Vlastimil Babka <vbabka@...e.cz>, linux-kernel@...r.kernel.org,
 Linux-MM <linux-mm@...ck.org>, Helge Deller <deller@....de>,
 linux-parisc@...r.kernel.org,
 Richard Henderson <richard.henderson@...aro.org>
Subject: Re: [PATCH 6.10 000/809] 6.10.3-rc3 review

On 8/8/24 14:50, John David Anglin wrote:
> On 2024-08-08 4:52 p.m., Guenter Roeck wrote:
>> On 8/8/24 11:19, Linus Torvalds wrote:
>>> On Thu, 8 Aug 2024 at 10:48, Thomas Gleixner <tglx@...utronix.de> wrote:
>>>>
>>>> Here is the disassembly from my latest crashing debug kernel which
>>>> shifts it up a couple of pages. Add 0x10 or sub 0x20 to make it work.
>>>
>>> Looks like I was off by an instruction, it's the 28th divide-step (not
>>> 29) that does the page crosser:
>>>
>>>>      4121dffc:   0b 21 04 41     ds r1,r25,r1
>>>>      4121e000:   0b bd 07 1d     add,c ret1,ret1,ret1
> I think this macro might clobber the C/B bits on a ITLB missing:
> 
>          /* This is for ILP32 PA2.0 only.  The TLB insertion needs
>           * to extend into I/O space if the address is 0xfXXXXXXX
>           * so we extend the f's into the top word of the pte in
>           * this case */
>          .macro          f_extend        pte,tmp
>          extrd,s         \pte,42,4,\tmp
>          addi,<>         1,\tmp,%r0
>          extrd,s         \pte,63,25,\pte
>          .endm
> 
> The addi instruction affects the C/B bits.  However, it is only used for 32-bit PA 2.0 kernels.
> A second tmp register would be needed to change the addi to an add logical.
> 
> The mode likely problem is the shladd instruction in the following macro in entry.S:
> 
>          .macro          L2_ptep pmd,pte,index,va,fault
> #if CONFIG_PGTABLE_LEVELS == 3
>          extru_safe      \va,31-ASM_PMD_SHIFT,ASM_BITS_PER_PMD,\index
> #else
>          extru_safe \va,31-ASM_PGDIR_SHIFT,ASM_BITS_PER_PGD,\index
> #endif
>          dep             %r0,31,PAGE_SHIFT,\pmd  /* clear offset */
> #if CONFIG_PGTABLE_LEVELS < 3
>          copy            %r0,\pte
> #endif
>          ldw,s           \index(\pmd),\pmd
>          bb,>=,n         \pmd,_PxD_PRESENT_BIT,\fault
>          dep             %r0,31,PxD_FLAG_SHIFT,\pmd /* clear flags */
>          SHLREG          \pmd,PxD_VALUE_SHIFT,\pmd
>          extru_safe      \va,31-PAGE_SHIFT,ASM_BITS_PER_PTE,\index
>          dep             %r0,31,PAGE_SHIFT,\pmd  /* clear offset */
>          shladd          \index,BITS_PER_PTE_ENTRY,\pmd,\pmd /* pmd is now pte */
>          .endm
> 
> I believe the shladd instruction should be changed to shladd,l (shift left and add logical).
> 

That doesn't help, at least not in qemu.

Guenter