netdev - RE: [RDMA RFC v6 00/16] Intel RDMA Driver Updates 2020-05-19

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]

Date:   Tue, 2 Jun 2020 22:59:46 +0000
From:   "Saleem, Shiraz" <shiraz.saleem@...el.com>
To:     Jason Gunthorpe <jgg@...lanox.com>
CC:     Leon Romanovsky <leon@...nel.org>,
        "Kirsher, Jeffrey T" <jeffrey.t.kirsher@...el.com>,
        "dledford@...hat.com" <dledford@...hat.com>,
        "davem@...emloft.net" <davem@...emloft.net>,
        "gregkh@...uxfoundation.org" <gregkh@...uxfoundation.org>,
        "linux-rdma@...r.kernel.org" <linux-rdma@...r.kernel.org>,
        "netdev@...r.kernel.org" <netdev@...r.kernel.org>,
        "nhorman@...hat.com" <nhorman@...hat.com>,
        "sassmann@...hat.com" <sassmann@...hat.com>,
        "poswald@...e.com" <poswald@...e.com>
Subject: RE: [RDMA RFC v6 00/16] Intel RDMA Driver Updates 2020-05-19

> Subject: Re: [RDMA RFC v6 00/16] Intel RDMA Driver Updates 2020-05-19
> 
> On Fri, May 29, 2020 at 03:21:05PM +0000, Saleem, Shiraz wrote:
> > > Subject: Re: [RDMA RFC v6 00/16] Intel RDMA Driver Updates
> > > 2020-05-19
> > >
> >
> > [......]
> >
> > >
> > > I'm looking on it and see static assignments, to by dynamic you will need "to
> play"
> > > with hw_shifts/hw_masks later, but you don't. What am I missing?
> > >
> > > +	for (i = 0; i < IRDMA_MAX_SHIFTS; ++i)
> > > +		dev->hw_shifts[i] = i40iw_shifts[i];
> > > +
> > > +	for (i = 0; i < IRDMA_MAX_MASKS; ++i)
> > > +		dev->hw_masks[i] = i40iw_masks[i];
> > >
> > > >
> > > > we still need to use the custom macro FLD_LS_64 without FIELD_PREP
> > > > in this case as FIELD_PREP expects compile time constants.
> > > > +#define FLD_LS_64(dev, val, field)	\
> > > > +	(((u64)(val) << (dev)->hw_shifts[field ## _S]) &
> > > > +(dev)->hw_masks[field ## _M])
> > > > And the shifts are still required for these fields which causes a
> > > > bit of inconsistency
> > > >
> >
> >
> > The device hw_masks/hw_shifts array store masks/shifts of those
> > descriptor fields that have same name across HW generations but differ
> > in some attribute such as field width. Yes they are statically
> > assigned, initialized with values from i40iw_masks and icrdma_masks,
> > depending on the HW generation. We can even use GENMASK for the values
> > in i40iw_masks[] , icrdma_masks[] but FIELD_PREP cant be used on
> > dev->hw_masks[]
> 
> So compute the shift and mask when building i40iw_shifts array using the compile
> time constant?
> 

i40iw_shifts[] and i40iw_mask[] are setup as compile constants
and used to initialize dev->hw_masks[], dev->hw_shifts[] if the device is gen1.
I still don't see how FIELD_PREP can be used on a value and
dev->hw_masks[i].