linux-kernel - Re: [PATCH v3 11/22] firmware: arm_scmi: add support for polling based SCMI transfers

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

Message-ID: <CAK8P3a2wRFif2zYKKmdcNm5wgbGdhqwLdA-YV7AX_0LmM+OiQQ@mail.gmail.com>
Date:   Wed, 4 Oct 2017 13:13:54 +0200
From:   Arnd Bergmann <arnd@...db.de>
To:     Sudeep Holla <sudeep.holla@....com>
Cc:     ALKML <linux-arm-kernel@...ts.infradead.org>,
        LKML <linux-kernel@...r.kernel.org>,
        DTML <devicetree@...r.kernel.org>,
        Roy Franz <roy.franz@...ium.com>,
        Harb Abdulhamid <harba@...eaurora.org>,
        Nishanth Menon <nm@...com>, Loc Ho <lho@....com>,
        Alexey Klimov <alexey.klimov@....com>,
        Ryan Harkin <Ryan.Harkin@....com>,
        Jassi Brar <jassisinghbrar@...il.com>
Subject: Re: [PATCH v3 11/22] firmware: arm_scmi: add support for polling
 based SCMI transfers

On Thu, Sep 28, 2017 at 3:11 PM, Sudeep Holla <sudeep.holla@....com> wrote:
> It would be useful to have options to perform some SCMI transfers
> atomically by polling for the completion flag instead of interrupt
> driven. The SCMI specification has option to disable the interrupt and
> poll for the completion flag in the shared memory.
>
> This patch adds support for polling based SCMI transfers using that
> option. This might be used for uninterrupted/atomic DVFS operations
> from the scheduler context.

multi-millisecond timeouts from inside the scheduler sound like a
really bad idea. Could this maybe get changed to an asynchronous
operation?

> +       if (xfer->hdr.poll_completion) {
> +               timeout = info->desc->max_rx_timeout_ms * 100;
> +               while (!scmi_xfer_poll_done(info, xfer) && timeout--)
> +                       udelay(10);

The timeout calculation is bad as well, since both the
scmi_xfer_poll_done() call and udelay() can take much longer
than the 10 microsecond delay that you use for the calculation.

If you want to do a timeout check like this, it should generally
be done using ktime_get()/ktime_add()/ktime_before().

       Arnd