2019-09-11 05:00:31

by Austin Kim

[permalink] [raw]
Subject: [PATCH] ARM: module: Drop 'rel->r_offset < 0' always false statement

Since rel->r_offset is declared as Elf32_Addr,
this value is always non-negative.
typedef struct elf32_rel {
Elf32_Addr r_offset;
Elf32_Word r_info;
} Elf32_Rel;

typedef __u32 Elf32_Addr;
typedef unsigned int __u32;

Drop 'rel->r_offset < 0' statement which is always false.

Signed-off-by: Austin Kim <[email protected]>
---
arch/arm/kernel/module.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/arm/kernel/module.c b/arch/arm/kernel/module.c
index deef17f..0921ce7 100644
--- a/arch/arm/kernel/module.c
+++ b/arch/arm/kernel/module.c
@@ -92,7 +92,7 @@ apply_relocate(Elf32_Shdr *sechdrs, const char *strtab, unsigned int symindex,
sym = ((Elf32_Sym *)symsec->sh_addr) + offset;
symname = strtab + sym->st_name;

- if (rel->r_offset < 0 || rel->r_offset > dstsec->sh_size - sizeof(u32)) {
+ if (rel->r_offset > dstsec->sh_size - sizeof(u32)) {
pr_err("%s: section %u reloc %u sym '%s': out of bounds relocation, offset %d size %u\n",
module->name, relindex, i, symname,
rel->r_offset, dstsec->sh_size);
--
2.6.2


2019-09-11 09:43:03

by Russell King (Oracle)

[permalink] [raw]
Subject: Re: [PATCH] ARM: module: Drop 'rel->r_offset < 0' always false statement

On Wed, Sep 11, 2019 at 01:54:08PM +0900, Austin Kim wrote:
> Since rel->r_offset is declared as Elf32_Addr,
> this value is always non-negative.
> typedef struct elf32_rel {
> Elf32_Addr r_offset;
> Elf32_Word r_info;
> } Elf32_Rel;
>
> typedef __u32 Elf32_Addr;
> typedef unsigned int __u32;
>
> Drop 'rel->r_offset < 0' statement which is always false.
>
> Signed-off-by: Austin Kim <[email protected]>
> ---
> arch/arm/kernel/module.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/arch/arm/kernel/module.c b/arch/arm/kernel/module.c
> index deef17f..0921ce7 100644
> --- a/arch/arm/kernel/module.c
> +++ b/arch/arm/kernel/module.c
> @@ -92,7 +92,7 @@ apply_relocate(Elf32_Shdr *sechdrs, const char *strtab, unsigned int symindex,
> sym = ((Elf32_Sym *)symsec->sh_addr) + offset;
> symname = strtab + sym->st_name;
>
> - if (rel->r_offset < 0 || rel->r_offset > dstsec->sh_size - sizeof(u32)) {
> + if (rel->r_offset > dstsec->sh_size - sizeof(u32)) {
> pr_err("%s: section %u reloc %u sym '%s': out of bounds relocation, offset %d size %u\n",

Also change %d to %u here.

> module->name, relindex, i, symname,
> rel->r_offset, dstsec->sh_size);
> --
> 2.6.2
>
>

--
RMK's Patch system: https://www.armlinux.org.uk/developer/patches/
FTTC broadband for 0.8mile line in suburbia: sync at 12.1Mbps down 622kbps up
According to speedtest.net: 11.9Mbps down 500kbps up

2019-09-11 11:45:10

by Austin Kim

[permalink] [raw]
Subject: Re: [PATCH] ARM: module: Drop 'rel->r_offset < 0' always false statement

2019년 9월 11일 (수) 오후 6:40, Russell King - ARM Linux admin
<[email protected]>님이 작성:
>
> On Wed, Sep 11, 2019 at 01:54:08PM +0900, Austin Kim wrote:
> > Since rel->r_offset is declared as Elf32_Addr,
> > this value is always non-negative.
> > typedef struct elf32_rel {
> > Elf32_Addr r_offset;
> > Elf32_Word r_info;
> > } Elf32_Rel;
> >
> > typedef __u32 Elf32_Addr;
> > typedef unsigned int __u32;
> >
> > Drop 'rel->r_offset < 0' statement which is always false.
> >
> > Signed-off-by: Austin Kim <[email protected]>
> > ---
> > arch/arm/kernel/module.c | 2 +-
> > 1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/arch/arm/kernel/module.c b/arch/arm/kernel/module.c
> > index deef17f..0921ce7 100644
> > --- a/arch/arm/kernel/module.c
> > +++ b/arch/arm/kernel/module.c
> > @@ -92,7 +92,7 @@ apply_relocate(Elf32_Shdr *sechdrs, const char *strtab, unsigned int symindex,
> > sym = ((Elf32_Sym *)symsec->sh_addr) + offset;
> > symname = strtab + sym->st_name;
> >
> > - if (rel->r_offset < 0 || rel->r_offset > dstsec->sh_size - sizeof(u32)) {
> > + if (rel->r_offset > dstsec->sh_size - sizeof(u32)) {
> > pr_err("%s: section %u reloc %u sym '%s': out of bounds relocation, offset %d size %u\n",
>
> Also change %d to %u here.

Let me resend the patch with the change(%d to %u).

Thanks,
Austin Kim

>
> > module->name, relindex, i, symname,
> > rel->r_offset, dstsec->sh_size);
> > --
> > 2.6.2
> >
> >
>
> --
> RMK's Patch system: https://www.armlinux.org.uk/developer/patches/
> FTTC broadband for 0.8mile line in suburbia: sync at 12.1Mbps down 622kbps up
> According to speedtest.net: 11.9Mbps down 500kbps up