2020-06-15 21:44:15

by Gustavo A. R. Silva

[permalink] [raw]
Subject: [PATCH] scsi: megaraid_sas: Use array_size() helper

The get_order() function has no 2-factor argument form, so multiplication
factors need to be wrapped in array_size().

This issue was found with the help of Coccinelle and, audited and fixed
manually.

Addresses-KSPP-ID: https://github.com/KSPP/linux/issues/83
Signed-off-by: Gustavo A. R. Silva <[email protected]>
---
drivers/scsi/megaraid/megaraid_sas_fusion.c | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/scsi/megaraid/megaraid_sas_fusion.c b/drivers/scsi/megaraid/megaraid_sas_fusion.c
index 319f241da4b6..6de44ed4cde7 100644
--- a/drivers/scsi/megaraid/megaraid_sas_fusion.c
+++ b/drivers/scsi/megaraid/megaraid_sas_fusion.c
@@ -5180,8 +5180,8 @@ megasas_alloc_fusion_context(struct megasas_instance *instance)

fusion = instance->ctrl_context;

- fusion->log_to_span_pages = get_order(MAX_LOGICAL_DRIVES_EXT *
- sizeof(LD_SPAN_INFO));
+ fusion->log_to_span_pages = get_order(array_size(MAX_LOGICAL_DRIVES_EXT,
+ sizeof(LD_SPAN_INFO)));
fusion->log_to_span =
(PLD_SPAN_INFO)__get_free_pages(GFP_KERNEL | __GFP_ZERO,
fusion->log_to_span_pages);
@@ -5196,8 +5196,8 @@ megasas_alloc_fusion_context(struct megasas_instance *instance)
}
}

- fusion->load_balance_info_pages = get_order(MAX_LOGICAL_DRIVES_EXT *
- sizeof(struct LD_LOAD_BALANCE_INFO));
+ fusion->load_balance_info_pages = get_order(array_size(MAX_LOGICAL_DRIVES_EXT,
+ sizeof(struct LD_LOAD_BALANCE_INFO)));
fusion->load_balance_info =
(struct LD_LOAD_BALANCE_INFO *)__get_free_pages(GFP_KERNEL | __GFP_ZERO,
fusion->load_balance_info_pages);
--
2.27.0


2020-06-15 23:31:26

by James Bottomley

[permalink] [raw]
Subject: Re: [PATCH] scsi: megaraid_sas: Use array_size() helper

On Mon, 2020-06-15 at 16:47 -0500, Gustavo A. R. Silva wrote:
> The get_order() function has no 2-factor argument form, so
> multiplication
> factors need to be wrapped in array_size().
>
> This issue was found with the help of Coccinelle and, audited and
> fixed
> manually.
>
> Addresses-KSPP-ID: https://github.com/KSPP/linux/issues/83
> Signed-off-by: Gustavo A. R. Silva <[email protected]>
> ---
> drivers/scsi/megaraid/megaraid_sas_fusion.c | 8 ++++----
> 1 file changed, 4 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/scsi/megaraid/megaraid_sas_fusion.c
> b/drivers/scsi/megaraid/megaraid_sas_fusion.c
> index 319f241da4b6..6de44ed4cde7 100644
> --- a/drivers/scsi/megaraid/megaraid_sas_fusion.c
> +++ b/drivers/scsi/megaraid/megaraid_sas_fusion.c
> @@ -5180,8 +5180,8 @@ megasas_alloc_fusion_context(struct
> megasas_instance *instance)
>
> fusion = instance->ctrl_context;
>
> - fusion->log_to_span_pages = get_order(MAX_LOGICAL_DRIVES_EXT
> *
> - sizeof(LD_SPAN_INFO));
> + fusion->log_to_span_pages =
> get_order(array_size(MAX_LOGICAL_DRIVES_EXT,
> + sizeof(LD_SPAN_INFO)))
> ;

What's the point of this? You're replacing a constant multiplication
the compiler can compute with one it can't on the theory there might be
an overflow, which is pretty far fetched given MAX_LOGICAL_DRIVES_EXT
is 256 and sizeof(LD_SPAN_INFO) is around 82.

I thought the whole point of overflow detection was to use it for
instances where we could be tricked into triggering one by userspace
which may result in a buffer under or overflow ... this is two
constants, how could this ever be a source of an exploit?

James