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: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <CAFdVvOxH4UQjww4124E2ttuTgknzkHoPxVSFOQgLfoV_dkANwQ@mail.gmail.com>
Date:   Mon, 4 Dec 2023 11:40:55 -0700
From:   Sathya Prakash Veerichetty <sathya.prakash@...adcom.com>
To:     Arnd Bergmann <arnd@...nel.org>
Cc:     Kashyap Desai <kashyap.desai@...adcom.com>,
        Sumit Saxena <sumit.saxena@...adcom.com>,
        Sreekanth Reddy <sreekanth.reddy@...adcom.com>,
        Arnd Bergmann <arnd@...db.de>,
        "James E.J. Bottomley" <jejb@...ux.ibm.com>,
        "Martin K. Petersen" <martin.petersen@...cle.com>,
        Ranjan Kumar <ranjan.kumar@...adcom.com>,
        Tomas Henzl <thenzl@...hat.com>,
        Harshit Mogalapalli <harshit.m.mogalapalli@...cle.com>,
        mpi3mr-linuxdrv.pdl@...adcom.com, linux-scsi@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH] scsi: mpi3mr: reduce stack usage in mpi3mr_refresh_sas_ports()

On Wed, Nov 29, 2023 at 5:06 AM Arnd Bergmann <arnd@...nel.org> wrote:
>
> From: Arnd Bergmann <arnd@...db.de>
>
> Toubling the number of PHYs also doubled the stack usage of this function,
> exceeding the 32-bit limit of 1024 bytes:
>
> drivers/scsi/mpi3mr/mpi3mr_transport.c: In function 'mpi3mr_refresh_sas_ports':
> drivers/scsi/mpi3mr/mpi3mr_transport.c:1818:1: error: the frame size of 1636 bytes is larger than 1024 bytes [-Werror=frame-larger-than=]
>
> Since the sas_io_unit_pg0 structure is already allocated dynamically, use
> the same method here. The size of the allocation can be smaller based on the
> actual number of phys now, so use this as an upper bound.
>
> Fixes: cb5b60894602 ("scsi: mpi3mr: Increase maximum number of PHYs to 64 from 32")
> Signed-off-by: Arnd Bergmann <arnd@...db.de>
> ---
>  drivers/scsi/mpi3mr/mpi3mr_transport.c | 8 +++++++-
>  1 file changed, 7 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/scsi/mpi3mr/mpi3mr_transport.c b/drivers/scsi/mpi3mr/mpi3mr_transport.c
> index c0c8ab586957..ab04596dbdf5 100644
> --- a/drivers/scsi/mpi3mr/mpi3mr_transport.c
> +++ b/drivers/scsi/mpi3mr/mpi3mr_transport.c
> @@ -1671,7 +1671,7 @@ mpi3mr_update_mr_sas_port(struct mpi3mr_ioc *mrioc, struct host_port *h_port,
>  void
>  mpi3mr_refresh_sas_ports(struct mpi3mr_ioc *mrioc)
>  {
> -       struct host_port h_port[64];
> +       struct host_port *h_port = NULL;
>         int i, j, found, host_port_count = 0, port_idx;
>         u16 sz, attached_handle, ioc_status;
>         struct mpi3_sas_io_unit_page0 *sas_io_unit_pg0 = NULL;
> @@ -1685,6 +1685,11 @@ mpi3mr_refresh_sas_ports(struct mpi3mr_ioc *mrioc)
>         sas_io_unit_pg0 = kzalloc(sz, GFP_KERNEL);
>         if (!sas_io_unit_pg0)
>                 return;
> +       h_port = kcalloc(mrioc->sas_hba.num_phys, sizeof(struct host_port),
> +                        GFP_KERNEL);
>> We need more than sas_hba.num_phys, so can you please use 64 instead of sas_hba.num_phys. Otherwise looks good.
> +       if (!h_port)
> +               goto out;
> +
>         if (mpi3mr_cfg_get_sas_io_unit_pg0(mrioc, sas_io_unit_pg0, sz)) {
>                 ioc_err(mrioc, "failure at %s:%d/%s()!\n",
>                     __FILE__, __LINE__, __func__);
> @@ -1814,6 +1819,7 @@ mpi3mr_refresh_sas_ports(struct mpi3mr_ioc *mrioc)
>                 }
>         }
>  out:
> +       kfree(h_port);
>         kfree(sas_io_unit_pg0);
>  }
>
> --
> 2.39.2
>

Download attachment "smime.p7s" of type "application/pkcs7-signature" (4227 bytes)

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ