On Thu, May 02, 2024 at 11:56:03PM +0000, Allen Pais wrote:
> Introduce the capability to dynamically configure the maximum file
> note size for ELF core dumps via sysctl. This enhancement removes
> the previous static limit of 4MB, allowing system administrators to
> adjust the size based on system-specific requirements or constraints.
>
> - Remove hardcoded `MAX_FILE_NOTE_SIZE` from `fs/binfmt_elf.c`.
> - Define `max_file_note_size` in `fs/coredump.c` with an initial value
> set to 4MB.
> - Declare `max_file_note_size` as an external variable in
> `include/linux/coredump.h`.
> - Add a new sysctl entry in `kernel/sysctl.c` to manage this setting
> at runtime.
The above bullet points should be clear from the patch itself. The
commit is really more about rationale and examples (which you have
below). I'd remove the bullets.
>
> $ sysctl -a | grep core_file_note_size_max
> kernel.core_file_note_size_max = 4194304
>
> $ sysctl -n kernel.core_file_note_size_max
> 4194304
>
> $echo 519304 > /proc/sys/kernel/core_file_note_size_max
>
> $sysctl -n kernel.core_file_note_size_max
> 519304
>
> Attempting to write beyond the ceiling value of 16MB
> $echo 17194304 > /proc/sys/kernel/core_file_note_size_max
> bash: echo: write error: Invalid argument
>
> Why is this being done?
> We have observed that during a crash when there are more than 65k mmaps
> in memory, the existing fixed limit on the size of the ELF notes section
> becomes a bottleneck. The notes section quickly reaches its capacity,
> leading to incomplete memory segment information in the resulting coredump.
> This truncation compromises the utility of the coredumps, as crucial
> information about the memory state at the time of the crash might be
> omitted.
I'd make this the first paragraph of the commit log. "We have this
problem" goes first, then "Here's what we did to deal with it", then you
examples. :)
>
> Signed-off-by: Vijay Nag <nagvijay@xxxxxxxxxxxxx>
> Signed-off-by: Allen Pais <apais@xxxxxxxxxxxxxxxxxxx>
>
> ---
> Chagnes in v3:
> - Fix commit message to reflect the correct sysctl knob [Kees]
> - Add a ceiling for maximum pssible note size(16M) [Allen]
> - Add a pr_warn_once() [Kees]
> Changes in v2:
> - Move new sysctl to fs/coredump.c [Luis & Kees]
> - rename max_file_note_size to core_file_note_size_max [kees]
> - Capture "why this is being done?" int he commit message [Luis & Kees]
> ---
> fs/binfmt_elf.c | 8 ++++++--
> fs/coredump.c | 15 +++++++++++++++
> include/linux/coredump.h | 1 +
> 3 files changed, 22 insertions(+), 2 deletions(-)
>
> diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c
> index 5397b552fbeb..5294f8f3a9a8 100644
> --- a/fs/binfmt_elf.c
> +++ b/fs/binfmt_elf.c
> @@ -1564,7 +1564,6 @@ static void fill_siginfo_note(struct memelfnote *note, user_siginfo_t *csigdata,
> fill_note(note, "CORE", NT_SIGINFO, sizeof(*csigdata), csigdata);
> }
>
> -#define MAX_FILE_NOTE_SIZE (4*1024*1024)
> /*
> * Format of NT_FILE note:
> *
> @@ -1592,8 +1591,13 @@ static int fill_files_note(struct memelfnote *note, struct coredump_params *cprm
>
> names_ofs = (2 + 3 * count) * sizeof(data[0]);
> alloc:
> - if (size >= MAX_FILE_NOTE_SIZE) /* paranoia check */
> + /* paranoia check */
> + if (size >= core_file_note_size_max) {
> + pr_warn_once("coredump Note size too large: %u "
> + "(does kernel.core_file_note_size_max sysctl need adjustment?)\n",
The string can be on a single line (I think scripts/check_patch.pl will
warn about this, as well as the indentation of "size" below...
> + size);
> return -EINVAL;
> + }
> size = round_up(size, PAGE_SIZE);
> /*
> * "size" can be 0 here legitimately.
> diff --git a/fs/coredump.c b/fs/coredump.c
> index be6403b4b14b..ffaed8c1b3b0 100644
> --- a/fs/coredump.c
> +++ b/fs/coredump.c
> @@ -56,10 +56,16 @@
> static bool dump_vma_snapshot(struct coredump_params *cprm);
> static void free_vma_snapshot(struct coredump_params *cprm);
>
> +#define MAX_FILE_NOTE_SIZE (4*1024*1024)
> +/* Define a reasonable max cap */
> +#define MAX_ALLOWED_NOTE_SIZE (16*1024*1024)
Let's call this CORE_FILE_NOTE_SIZE_DEFAULT and
CORE_FILE_NOTE_SIZE_MAX to match the sysctl.
> +
> static int core_uses_pid;
> static unsigned int core_pipe_limit;
> static char core_pattern[CORENAME_MAX_SIZE] = "core";
> static int core_name_size = CORENAME_MAX_SIZE;
> +unsigned int core_file_note_size_max = MAX_FILE_NOTE_SIZE;
> +unsigned int core_file_note_size_allowed = MAX_ALLOWED_NOTE_SIZE;
The latter can be static and const.
For the note below, perhaps add:
static const unsigned int core_file_note_size_min = CORE_FILE_NOTE_SIZE_DEFAULT;
>
> struct core_name {
> char *corename;
> @@ -1020,6 +1026,15 @@ static struct ctl_table coredump_sysctls[] = {
> .mode = 0644,
> .proc_handler = proc_dointvec,
> },
> + {
> + .procname = "core_file_note_size_max",
> + .data = &core_file_note_size_max,
> + .maxlen = sizeof(unsigned int),
> + .mode = 0644,
> + .proc_handler = proc_douintvec_minmax,
> + .extra1 = &core_file_note_size_max,
This means you can never shrink it if you raise it from the default.
Let's use the core_file_note_size_min above.
> + .extra2 = &core_file_note_size_allowed,
> + },
> };
>
> static int __init init_fs_coredump_sysctls(void)
> diff --git a/include/linux/coredump.h b/include/linux/coredump.h
> index d3eba4360150..14c057643e7f 100644
> --- a/include/linux/coredump.h
> +++ b/include/linux/coredump.h
> @@ -46,6 +46,7 @@ static inline void do_coredump(const kernel_siginfo_t *siginfo) {}
> #endif
>
> #if defined(CONFIG_COREDUMP) && defined(CONFIG_SYSCTL)
> +extern unsigned int core_file_note_size_max;
> extern void validate_coredump_safety(void);
> #else
> static inline void validate_coredump_safety(void) {}
> --
> 2.17.1
>
I think v4 will be all good to go, assuming no one else pops up. :)
Thanks for the changes!
-Kees
--
Kees Cook
