Subject: [PATCH v2] kprobes: remove duplicated strong free_insn_page in x86 and s390

free_insn_page() in x86 and s390 are same with the common weak function
in kernel/kprobes.c.
Plus, the comment "Recover page to RW mode before releasing it" in x86
seems insensible to be there since resetting mapping is done by common
code in vfree() of module_memfree().
So drop these two duplicated strong functions and related comment, then
mark the common one in kernel/kprobes.c strong.

Cc: Christoph Hellwig <[email protected]>
Signed-off-by: Barry Song <[email protected]>
---
-v2:
remove free_insn_page in s390 as well and remove the __weak in common
code according to Christoph's comment;

arch/s390/kernel/kprobes.c | 5 -----
arch/x86/kernel/kprobes/core.c | 6 ------
kernel/kprobes.c | 2 +-
3 files changed, 1 insertion(+), 12 deletions(-)

diff --git a/arch/s390/kernel/kprobes.c b/arch/s390/kernel/kprobes.c
index aae24dc75df6..60cfbd24229b 100644
--- a/arch/s390/kernel/kprobes.c
+++ b/arch/s390/kernel/kprobes.c
@@ -44,11 +44,6 @@ void *alloc_insn_page(void)
return page;
}

-void free_insn_page(void *page)
-{
- module_memfree(page);
-}
-
static void *alloc_s390_insn_page(void)
{
if (xchg(&insn_page_in_use, 1) == 1)
diff --git a/arch/x86/kernel/kprobes/core.c b/arch/x86/kernel/kprobes/core.c
index d3d65545cb8b..3bce67d3a03c 100644
--- a/arch/x86/kernel/kprobes/core.c
+++ b/arch/x86/kernel/kprobes/core.c
@@ -422,12 +422,6 @@ void *alloc_insn_page(void)
return page;
}

-/* Recover page to RW mode before releasing it */
-void free_insn_page(void *page)
-{
- module_memfree(page);
-}
-
/* Kprobe x86 instruction emulation - only regs->ip or IF flag modifiers */

static void kprobe_emulate_ifmodifiers(struct kprobe *p, struct pt_regs *regs)
diff --git a/kernel/kprobes.c b/kernel/kprobes.c
index 745f08fdd7a6..ddb643f3879f 100644
--- a/kernel/kprobes.c
+++ b/kernel/kprobes.c
@@ -106,7 +106,7 @@ void __weak *alloc_insn_page(void)
return module_alloc(PAGE_SIZE);
}

-void __weak free_insn_page(void *page)
+void free_insn_page(void *page)
{
module_memfree(page);
}
--
2.25.1


2021-06-07 09:35:44

by Christoph Hellwig

[permalink] [raw]
Subject: Re: [PATCH v2] kprobes: remove duplicated strong free_insn_page in x86 and s390

Please also mark free_insn_page static now.

2021-06-07 11:00:44

by Heiko Carstens

[permalink] [raw]
Subject: Re: [PATCH v2] kprobes: remove duplicated strong free_insn_page in x86 and s390

On Mon, Jun 07, 2021 at 09:18:54PM +1200, Barry Song wrote:
> free_insn_page() in x86 and s390 are same with the common weak function
> in kernel/kprobes.c.
> Plus, the comment "Recover page to RW mode before releasing it" in x86
> seems insensible to be there since resetting mapping is done by common
> code in vfree() of module_memfree().
> So drop these two duplicated strong functions and related comment, then
> mark the common one in kernel/kprobes.c strong.
>
> Cc: Christoph Hellwig <[email protected]>
> Signed-off-by: Barry Song <[email protected]>
> ---
> -v2:
> remove free_insn_page in s390 as well and remove the __weak in common
> code according to Christoph's comment;
>
> arch/s390/kernel/kprobes.c | 5 -----
> arch/x86/kernel/kprobes/core.c | 6 ------
> kernel/kprobes.c | 2 +-
> 3 files changed, 1 insertion(+), 12 deletions(-)

For the s390 part:
Acked-by: Heiko Carstens <[email protected]>

2021-06-07 12:33:46

by Masami Hiramatsu

[permalink] [raw]
Subject: Re: [PATCH v2] kprobes: remove duplicated strong free_insn_page in x86 and s390

On Mon, 7 Jun 2021 21:18:54 +1200
Barry Song <[email protected]> wrote:

> free_insn_page() in x86 and s390 are same with the common weak function
> in kernel/kprobes.c.
> Plus, the comment "Recover page to RW mode before releasing it" in x86
> seems insensible to be there since resetting mapping is done by common
> code in vfree() of module_memfree().
> So drop these two duplicated strong functions and related comment, then
> mark the common one in kernel/kprobes.c strong.

Hm, OK. Actually riscv and arm64 uses __vmalloc() but anyway
since module_memfree() calls vfree(). So it has no problem.

Acked-by: Masami Hiramatsu <[email protected]>

Thank you,

>
> Cc: Christoph Hellwig <[email protected]>
> Signed-off-by: Barry Song <[email protected]>
> ---
> -v2:
> remove free_insn_page in s390 as well and remove the __weak in common
> code according to Christoph's comment;
>
> arch/s390/kernel/kprobes.c | 5 -----
> arch/x86/kernel/kprobes/core.c | 6 ------
> kernel/kprobes.c | 2 +-
> 3 files changed, 1 insertion(+), 12 deletions(-)
>
> diff --git a/arch/s390/kernel/kprobes.c b/arch/s390/kernel/kprobes.c
> index aae24dc75df6..60cfbd24229b 100644
> --- a/arch/s390/kernel/kprobes.c
> +++ b/arch/s390/kernel/kprobes.c
> @@ -44,11 +44,6 @@ void *alloc_insn_page(void)
> return page;
> }
>
> -void free_insn_page(void *page)
> -{
> - module_memfree(page);
> -}
> -
> static void *alloc_s390_insn_page(void)
> {
> if (xchg(&insn_page_in_use, 1) == 1)
> diff --git a/arch/x86/kernel/kprobes/core.c b/arch/x86/kernel/kprobes/core.c
> index d3d65545cb8b..3bce67d3a03c 100644
> --- a/arch/x86/kernel/kprobes/core.c
> +++ b/arch/x86/kernel/kprobes/core.c
> @@ -422,12 +422,6 @@ void *alloc_insn_page(void)
> return page;
> }
>
> -/* Recover page to RW mode before releasing it */
> -void free_insn_page(void *page)
> -{
> - module_memfree(page);
> -}
> -
> /* Kprobe x86 instruction emulation - only regs->ip or IF flag modifiers */
>
> static void kprobe_emulate_ifmodifiers(struct kprobe *p, struct pt_regs *regs)
> diff --git a/kernel/kprobes.c b/kernel/kprobes.c
> index 745f08fdd7a6..ddb643f3879f 100644
> --- a/kernel/kprobes.c
> +++ b/kernel/kprobes.c
> @@ -106,7 +106,7 @@ void __weak *alloc_insn_page(void)
> return module_alloc(PAGE_SIZE);
> }
>
> -void __weak free_insn_page(void *page)
> +void free_insn_page(void *page)
> {
> module_memfree(page);
> }
> --
> 2.25.1
>


--
Masami Hiramatsu <[email protected]>