2022-07-11 07:55:07

by Jianglei Nie

Subject: [PATCH v3] proc/vmcore: fix potential memory leak in vmcore_init()

elfcorehdr_alloc() allocates a memory chunk for elfcorehdr_addr with
kzalloc(). If is_vmcore_usable() returns false, elfcorehdr_addr is a
predefined value. If parse_crash_elf_headers() occurs some error and
returns a negetive value, the elfcorehdr_addr should be released with
elfcorehdr_free().

We can fix by calling elfcorehdr_free() when parse_crash_elf_headers()
fails.

Signed-off-by: Jianglei Nie <[email protected]>
---
fs/proc/vmcore.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/fs/proc/vmcore.c b/fs/proc/vmcore.c
index 4eaeb645e759..125efe63f281 100644
--- a/fs/proc/vmcore.c
+++ b/fs/proc/vmcore.c
@@ -1569,7 +1569,7 @@ static int __init vmcore_init(void)
rc = parse_crash_elf_headers();
if (rc) {
pr_warn("Kdump: vmcore not initialized\n");
- return rc;
+ goto fail;
}
elfcorehdr_free(elfcorehdr_addr);
elfcorehdr_addr = ELFCORE_ADDR_ERR;
@@ -1578,6 +1578,9 @@ static int __init vmcore_init(void)
if (proc_vmcore)
proc_vmcore->size = vmcore_size;
return 0;
+fail:
+ elfcorehdr_free(elfcorehdr_addr);
+ return rc;
}
fs_initcall(vmcore_init);

--
2.25.1

2022-07-11 08:02:37

by Baoquan He

Subject: Re: [PATCH v3] proc/vmcore: fix potential memory leak in vmcore_init()

On 07/11/22 at 03:34pm, Jianglei Nie wrote:
> elfcorehdr_alloc() allocates a memory chunk for elfcorehdr_addr with
> kzalloc(). If is_vmcore_usable() returns false, elfcorehdr_addr is a
> predefined value. If parse_crash_elf_headers() occurs some error and
> returns a negetive value, the elfcorehdr_addr should be released with
> elfcorehdr_free().
>
> We can fix by calling elfcorehdr_free() when parse_crash_elf_headers()
> fails.
>
> Signed-off-by: Jianglei Nie <[email protected]>
> ---
> fs/proc/vmcore.c | 5 ++++-
> 1 file changed, 4 insertions(+), 1 deletion(-)
>
> diff --git a/fs/proc/vmcore.c b/fs/proc/vmcore.c
> index 4eaeb645e759..125efe63f281 100644
> --- a/fs/proc/vmcore.c
> +++ b/fs/proc/vmcore.c
> @@ -1569,7 +1569,7 @@ static int __init vmcore_init(void)
> rc = parse_crash_elf_headers();
> if (rc) {
> pr_warn("Kdump: vmcore not initialized\n");
> - return rc;
> + goto fail;

Sigh. Why don't you copy my suggested code directly?

> }
> elfcorehdr_free(elfcorehdr_addr);

Remove above line.

> elfcorehdr_addr = ELFCORE_ADDR_ERR;
> @@ -1578,6 +1578,9 @@ static int __init vmcore_init(void)
> if (proc_vmcore)
> proc_vmcore->size = vmcore_size;
> return 0;

Remove above line too.

> +fail:
> + elfcorehdr_free(elfcorehdr_addr);
> + return rc;
> }
> fs_initcall(vmcore_init);
>
> --
> 2.25.1
>

2022-07-11 12:32:32

by Matthew Wilcox

Subject: Re: [PATCH v3] proc/vmcore: fix potential memory leak in vmcore_init()

On Mon, Jul 11, 2022 at 03:49:51PM +0800, Baoquan He wrote:
> On 07/11/22 at 03:34pm, Jianglei Nie wrote:
> > elfcorehdr_alloc() allocates a memory chunk for elfcorehdr_addr with
> > kzalloc(). If is_vmcore_usable() returns false, elfcorehdr_addr is a
> > predefined value. If parse_crash_elf_headers() occurs some error and
> > returns a negetive value, the elfcorehdr_addr should be released with
> > elfcorehdr_free().
> >
> > We can fix by calling elfcorehdr_free() when parse_crash_elf_headers()
> > fails.
> >
> > Signed-off-by: Jianglei Nie <[email protected]>
> > ---
> > fs/proc/vmcore.c | 5 ++++-
> > 1 file changed, 4 insertions(+), 1 deletion(-)
> >
> > diff --git a/fs/proc/vmcore.c b/fs/proc/vmcore.c
> > index 4eaeb645e759..125efe63f281 100644
> > --- a/fs/proc/vmcore.c
> > +++ b/fs/proc/vmcore.c
> > @@ -1569,7 +1569,7 @@ static int __init vmcore_init(void)
> > rc = parse_crash_elf_headers();
> > if (rc) {
> > pr_warn("Kdump: vmcore not initialized\n");
> > - return rc;
> > + goto fail;
>
> Sigh. Why don't you copy my suggested code directly?

I think at this point, you should just submit your own patch
and credit this person with Reported-by:

2022-07-12 01:17:15

by Baoquan He

Subject: Re: [PATCH v3] proc/vmcore: fix potential memory leak in vmcore_init()

On 07/11/22 at 01:18pm, Matthew Wilcox wrote:
> On Mon, Jul 11, 2022 at 03:49:51PM +0800, Baoquan He wrote:
> > On 07/11/22 at 03:34pm, Jianglei Nie wrote:
> > > elfcorehdr_alloc() allocates a memory chunk for elfcorehdr_addr with
> > > kzalloc(). If is_vmcore_usable() returns false, elfcorehdr_addr is a
> > > predefined value. If parse_crash_elf_headers() occurs some error and
> > > returns a negetive value, the elfcorehdr_addr should be released with
> > > elfcorehdr_free().
> > >
> > > We can fix by calling elfcorehdr_free() when parse_crash_elf_headers()
> > > fails.
> > >
> > > Signed-off-by: Jianglei Nie <[email protected]>
> > > ---
> > > fs/proc/vmcore.c | 5 ++++-
> > > 1 file changed, 4 insertions(+), 1 deletion(-)
> > >
> > > diff --git a/fs/proc/vmcore.c b/fs/proc/vmcore.c
> > > index 4eaeb645e759..125efe63f281 100644
> > > --- a/fs/proc/vmcore.c
> > > +++ b/fs/proc/vmcore.c
> > > @@ -1569,7 +1569,7 @@ static int __init vmcore_init(void)
> > > rc = parse_crash_elf_headers();
> > > if (rc) {
> > > pr_warn("Kdump: vmcore not initialized\n");
> > > - return rc;
> > > + goto fail;
> >
> > Sigh. Why don't you copy my suggested code directly?
>
> I think at this point, you should just submit your own patch
> and credit this person with Reported-by:

Thanks for telling. I will consider doing this in the future.