linux-kernel - Re: [PATCH 1/4] fpga mgr: Introduce FPGA capabilities

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: <alpine.DEB.2.10.1611141720200.6488@atull-730U3E-740U3E>
Date:   Mon, 14 Nov 2016 17:23:08 -0600
From:   atull <atull@...nsource.altera.com>
To:     Moritz Fischer <moritz.fischer@...us.com>
CC:     Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        <moritz.fischer.private@...il.com>,
        Michal Simek <michal.simek@...inx.com>,
        Sören Brinkmann <soren.brinkmann@...inx.com>,
        linux-arm-kernel <linux-arm-kernel@...ts.infradead.org>,
        Julia Cartwright <julia@...com>
Subject: Re: [PATCH 1/4] fpga mgr: Introduce FPGA capabilities

On Mon, 14 Nov 2016, Moritz Fischer wrote:

Hi Moritz,

> Hi Alan,
> 
> On Mon, Nov 14, 2016 at 6:06 AM, atull <atull@...nsource.altera.com> wrote:
> > On Mon, 7 Nov 2016, Moritz Fischer wrote:
> >
> >> Add FPGA capabilities as a way to express the capabilities
> >> of a given FPGA manager.
> >>
> >> Removes code duplication by comparing the low-level driver's
> >> capabilities at the framework level rather than having each driver
> >> check for supported operations in the write_init() callback.
> >>
> >> This allows for extending with additional capabilities, similar
> >> to the the dmaengine framework's implementation.
> >>
> >> Signed-off-by: Moritz Fischer <moritz.fischer@...us.com>
> >> Cc: Alan Tull <atull@...nsource.altera.com>
> >> Cc: Michal Simek <michal.simek@...inx.com>
> >> Cc: Sören Brinkmann <soren.brinkmann@...inx.com>
> >> Cc: linux-kernel@...r.kernel.org
> >> Cc: linux-arm-kernel@...ts.infradead.org
> >> ---
> >>
> >> Changes from RFC:
> >> * in the RFC the caps weren't actually stored into the struct fpga_mgr
> >>
> >> Note:
> >>
> >> If people disagree on the typedef being a 'false positive' I can fix
> >> that in a future rev of the patchset.
> >>
> >> Thanks,
> >>
> >>     Moritz
> >>
> >> ---
> >>  drivers/fpga/fpga-mgr.c       | 15 ++++++++++++++
> >>  drivers/fpga/socfpga.c        | 10 +++++-----
> >>  drivers/fpga/zynq-fpga.c      |  7 ++++++-
> >>  include/linux/fpga/fpga-mgr.h | 46 ++++++++++++++++++++++++++++++++++++++++++-
> >>  4 files changed, 71 insertions(+), 7 deletions(-)
> >>
> >> diff --git a/drivers/fpga/fpga-mgr.c b/drivers/fpga/fpga-mgr.c
> >> index 953dc91..ed57c17 100644
> >> --- a/drivers/fpga/fpga-mgr.c
> >> +++ b/drivers/fpga/fpga-mgr.c
> >> @@ -49,6 +49,18 @@ int fpga_mgr_buf_load(struct fpga_manager *mgr, u32 flags, const char *buf,
> >>       struct device *dev = &mgr->dev;
> >>       int ret;
> >>
> >> +     if (flags & FPGA_MGR_PARTIAL_RECONFIG &&
> >> +         !fpga_mgr_has_cap(FPGA_MGR_CAP_PARTIAL_RECONF, mgr->caps)) {
> >> +             dev_err(dev, "Partial reconfiguration not supported\n");
> >> +             return -ENOTSUPP;
> >> +     }
> >> +
> >> +     if (flags & FPGA_MGR_FULL_RECONFIG &&
> >> +         !fpga_mgr_has_cap(FPGA_MGR_CAP_FULL_RECONF, mgr->caps)) {
> >> +             dev_err(dev, "Full reconfiguration not supported\n");
> >> +             return -ENOTSUPP;
> >> +     }
> >> +
> >
> > Could you move the checks to their own function like
> > 'fpga_mgr_check_caps()' or something?  I really like it if we can keep
> > the functions short, like a screen or so where it's practicle to do
> > so and I could see the number of caps growing here.
> 
> Absolutely. Great suggestion.
> 
> > The only counter argument I could think of is if a cap affects the sequence
> > in this function.  Hmmm...
> 
> Oh you mean the cap being there affecting the sequence in *this* function?
> I'd suggest we address that when we run into a cap that requires this.

Yes, I agree.

Alan

> 
> Cheers,
> 
> Moritz
>