| 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
| ||
|
Message-ID: <4b001fde-6442-525e-4d8e-0ba7a85dccc5@loongson.cn>
Date: Tue, 6 Sep 2022 14:06:10 +0800
From: Jinyang He <hejinyang@...ngson.cn>
To: Huacai Chen <chenhuacai@...nel.org>
Cc: WANG Xuerui <kernel@...0n.name>, Oleg Nesterov <oleg@...hat.com>,
loongarch@...ts.linux.dev, LKML <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] LoongArch: Mark kernel stack padding as
TOP_OF_KERNEL_STACK_PADDING
Hi, Huacai,
On 2022/9/5 14:21, Huacai Chen wrote:
> Hi, Jinyang,
>
> On Mon, Sep 5, 2022 at 1:39 PM Jinyang He <hejinyang@...ngson.cn> wrote:
>> On 2022/9/4 20:19, Huacai Chen wrote:
>>
>>> Hi, Jinyang,
>>>
>>> I don't know what scenarios this padding is used and how to use.
>> To be frankly, I don't know its scenarios on LoongArch. It looks like
>> the MIPS legacy. This patch only keep this padding, which can be
>> easily modified if needed in the future. Now we can completely
>> set it to 0 or remove it. Actually, it can be thought of as another
>> buffer for thread_info, while thread_info usually limit to one cache line.
> We should make it clear how MIPS use it at least, then we can make a
> decision whether to remove it or not, otherwise it may break
> something.
I think it is related to shadow parameters area of MIPS o32 Calling
Convention.
This happened a long time ago, starting is 24, then 4*SZREG, and now 32.
But there are few commit-logs I can find. See links. I guess these changes
are due to the ABI, the macro SZREG in [2] is referenced to check stack
alignment. And pt_regs and sigframe also have padding. If we can't find
what will be broken, then we should cancel it rather than keeping it.
We wouldn't have these paddings if we weren't inspired by MIPS.
[1] https://elixir.bootlin.com/linux/1.1.82/source/arch/mips/kernel/head.S
[2] https://elixir.bootlin.com/linux/2.1.44/source/arch/mips/kernel/head.S
[3] https://elixir.bootlin.com/linux/2.1.73/source/arch/mips/kernel/head.S
[4]
https://git.kernel.org/pub/scm/linux/kernel/git/history/history.git/commit?id=484889fc85a223ef56edc31828b86751b2296b7c
>>> And
>>> in x86 there is only TOP_OF_KERNEL_STACK_PADDING but we need
>>> TOP_OF_KERNEL_STACK_PADDING and _TOP_OF_KERNEL_STACK_PADDING. Two
>>> macros are hard to maintain.
>> How about put it into loongarch.h if we need this macro?
> That is good because we only need one macro, but you can research
> whether there is a better place.
As asm and C code need it, how about put it in output_thread_info_defines?
Thanks,
Jinyang
>>> On Fri, Sep 2, 2022 at 11:01 PM Jinyang He <hejinyang@...ngson.cn> wrote:
>>>> Mark the top reserved bytes of the kernel stack by a macro. It collects
>>>> that info so that we can modify or discarded (let it be 0) this padding
>>>> if needed. Its name is referenced from x86_64. It seems that x86_64
>>>> think the padding as extension on top of the kernel stack in other mode
>>>> (normally the top is only pt_regs). Although LoongArch not produce
>>>> such an extension at present, collect them because of the uncertainty
>>>> of the future. Note that this patch does not modify macro size. It
>>>> requires another strict evaluation.
>>>>
>>>> Signed-off-by: Jinyang He <hejinyang@...ngson.cn>
>>>> ---
>>>> arch/loongarch/include/asm/processor.h | 3 ++-
>>>> arch/loongarch/include/asm/ptrace.h | 3 ++-
>>>> arch/loongarch/include/asm/stackframe.h | 1 +
>>>> arch/loongarch/kernel/head.S | 2 +-
>>>> arch/loongarch/kernel/process.c | 4 ++--
>>>> arch/loongarch/kernel/switch.S | 2 +-
>>>> 6 files changed, 9 insertions(+), 6 deletions(-)
>>>>
>>>> diff --git a/arch/loongarch/include/asm/processor.h b/arch/loongarch/include/asm/processor.h
>>>> index 1c4b43083..6077fd8e3 100644
>>>> --- a/arch/loongarch/include/asm/processor.h
>>>> +++ b/arch/loongarch/include/asm/processor.h
>>>> @@ -193,8 +193,9 @@ static inline void flush_thread(void)
>>>>
>>>> unsigned long __get_wchan(struct task_struct *p);
>>>>
>>>> +#define TOP_OF_KERNEL_STACK_PADDING 32
>>>> #define __KSTK_TOS(tsk) ((unsigned long)task_stack_page(tsk) + \
>>>> - THREAD_SIZE - 32 - sizeof(struct pt_regs))
>>>> + THREAD_SIZE - TOP_OF_KERNEL_STACK_PADDING - sizeof(struct pt_regs))
>>>> #define task_pt_regs(tsk) ((struct pt_regs *)__KSTK_TOS(tsk))
>>>> #define KSTK_EIP(tsk) (task_pt_regs(tsk)->csr_era)
>>>> #define KSTK_ESP(tsk) (task_pt_regs(tsk)->regs[3])
>>>> diff --git a/arch/loongarch/include/asm/ptrace.h b/arch/loongarch/include/asm/ptrace.h
>>>> index 17838c6b7..cdceeaa60 100644
>>>> --- a/arch/loongarch/include/asm/ptrace.h
>>>> +++ b/arch/loongarch/include/asm/ptrace.h
>>>> @@ -133,7 +133,8 @@ static inline void die_if_kernel(const char *str, struct pt_regs *regs)
>>>> #define current_pt_regs() \
>>>> ({ \
>>>> unsigned long sp = (unsigned long)__builtin_frame_address(0); \
>>>> - (struct pt_regs *)((sp | (THREAD_SIZE - 1)) + 1 - 32) - 1; \
>>>> + (struct pt_regs *)((sp | (THREAD_SIZE - 1)) + 1 - \
>>>> + TOP_OF_KERNEL_STACK_PADDING) - 1; \
>>>> })
>>>>
>>>> /* Helpers for working with the user stack pointer */
>>>> diff --git a/arch/loongarch/include/asm/stackframe.h b/arch/loongarch/include/asm/stackframe.h
>>>> index 4ca953062..e9f849b29 100644
>>>> --- a/arch/loongarch/include/asm/stackframe.h
>>>> +++ b/arch/loongarch/include/asm/stackframe.h
>>>> @@ -216,4 +216,5 @@
>>>> RESTORE_SP_AND_RET \docfi
>>>> .endm
>>>>
>>>> +#define _TOP_OF_KERNEL_STACK_PADDING 32
>>>> #endif /* _ASM_STACKFRAME_H */
>>>> diff --git a/arch/loongarch/kernel/head.S b/arch/loongarch/kernel/head.S
>>>> index eb3f641d5..290c53847 100644
>>>> --- a/arch/loongarch/kernel/head.S
>>>> +++ b/arch/loongarch/kernel/head.S
>>>> @@ -75,7 +75,7 @@ SYM_CODE_START(kernel_entry) # kernel entry point
>>>>
>>>> la.pcrel tp, init_thread_union
>>>> /* Set the SP after an empty pt_regs. */
>>>> - PTR_LI sp, (_THREAD_SIZE - 32 - PT_SIZE)
>>>> + PTR_LI sp, (_THREAD_SIZE - _TOP_OF_KERNEL_STACK_PADDING - PT_SIZE)
>>>> PTR_ADD sp, sp, tp
>>>> set_saved_sp sp, t0, t1
>>>> PTR_ADDI sp, sp, -4 * SZREG # init stack pointer
>>>> diff --git a/arch/loongarch/kernel/process.c b/arch/loongarch/kernel/process.c
>>>> index 660492f06..b51cf7975 100644
>>>> --- a/arch/loongarch/kernel/process.c
>>>> +++ b/arch/loongarch/kernel/process.c
>>>> @@ -129,7 +129,7 @@ int copy_thread(struct task_struct *p, const struct kernel_clone_args *args)
>>>> unsigned long clone_flags = args->flags;
>>>> struct pt_regs *childregs, *regs = current_pt_regs();
>>>>
>>>> - childksp = (unsigned long)task_stack_page(p) + THREAD_SIZE - 32;
>>>> + childksp = (unsigned long)task_stack_page(p) + THREAD_SIZE - TOP_OF_KERNEL_STACK_PADDING;
>>>>
>>>> /* set up new TSS. */
>>>> childregs = (struct pt_regs *) childksp - 1;
>>>> @@ -236,7 +236,7 @@ bool in_task_stack(unsigned long stack, struct task_struct *task,
>>>> struct stack_info *info)
>>>> {
>>>> unsigned long begin = (unsigned long)task_stack_page(task);
>>>> - unsigned long end = begin + THREAD_SIZE - 32;
>>>> + unsigned long end = begin + THREAD_SIZE - TOP_OF_KERNEL_STACK_PADDING;
>>>>
>>>> if (stack < begin || stack >= end)
>>>> return false;
>>>> diff --git a/arch/loongarch/kernel/switch.S b/arch/loongarch/kernel/switch.S
>>>> index 43ebbc399..1797aceae 100644
>>>> --- a/arch/loongarch/kernel/switch.S
>>>> +++ b/arch/loongarch/kernel/switch.S
>>>> @@ -26,7 +26,7 @@ SYM_FUNC_START(__switch_to)
>>>> move tp, a2
>>>> cpu_restore_nonscratch a1
>>>>
>>>> - li.w t0, _THREAD_SIZE - 32
>>>> + li.w t0, _THREAD_SIZE - _TOP_OF_KERNEL_STACK_PADDING
>>>> PTR_ADD t0, t0, tp
>>>> set_saved_sp t0, t1, t2
>>>>
>>>> --
>>>> 2.31.1
>>>>
Powered by blists - more mailing lists