linux-kernel - Re: [PATCH 1/2] nvme: Update type from size_t to u16 for opts->queue

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Wed, 29 Nov 2023 15:15:04 +0300
From:   Dan Carpenter <dan.carpenter@...aro.org>
To:     Nitesh Shetty <nj.shetty@...sung.com>
Cc:     Keith Busch <kbusch@...nel.org>, Jens Axboe <axboe@...nel.dk>,
        Christoph Hellwig <hch@....de>,
        Sagi Grimberg <sagi@...mberg.me>,
        James Smart <james.smart@...adcom.com>,
        Chaitanya Kulkarni <kch@...dia.com>, error27@...il.com,
        gost.dev@...sung.com, nitheshshetty@...il.com,
        linux-nvme@...ts.infradead.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 1/2] nvme: Update type from size_t to u16 for
 opts->queue_size

On Wed, Nov 29, 2023 at 12:26:24PM +0300, Dan Carpenter wrote:
> The ctrl->ops assignment happens in nvme_init_ctrl() and it should have
> been easy to track.  I am not sure what went wrong there.  I'll take a
> look at that and fix it.

I suspect on your system you don't have the cross function DB built so
it's not aware what happens in nvme_init_ctrl() or the other function I
mentioned.

On my system there is too much data so it isn't parsed.
1) When there is too much data it should just mark ctrl->ops as unknown
2) I should try to consolidate the data.  I should just mark all of
   ctrl->ctrl_device as unknown instead of recording any of the struct
   members individually.  There might also be stuff that isn't used like
   ctrl->namespaces_rwsem internals.
3) On my system, I have 2187 bogus entries that say we removed an item
   from a list.

Probably any of these fixes would silence the false positive but it
would be better to do all three.

If you don't have the cross function database though, you're probably
out of luck.  There are always going to be more false positives if you
don't have the cross function information.

regards,
dan carpenter