mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Scott Wood <swood@redhat.com>
To: Alan Tull <atull@kernel.org>
Cc: Wu Hao <hao.wu@intel.com>, Moritz Fischer <mdf@kernel.org>,
	linux-fpga@vger.kernel.org,
	linux-kernel <linux-kernel@vger.kernel.org>,
	linux-api@vger.kernel.org,
	Ananda Ravuri <ananda.ravuri@intel.com>,
	Xu Yilun <yilun.xu@intel.com>
Subject: Re: [PATCH 03/17] fpga: dfl: fme: support 512bit data width PR
Date: Tue, 26 Mar 2019 16:22:34 -0500	[thread overview]
Message-ID: <580c6c604d3915c105f076fc7f22ab5da98598fc.camel@redhat.com> (raw)
In-Reply-To: <CANk1AXSBHAudsLfNkSK1OuE6R9bogo6jW9bYmmZdvweTt3eCiA@mail.gmail.com>

On Tue, 2019-03-26 at 14:33 -0500, Alan Tull wrote:
> On Mon, Mar 25, 2019 at 5:58 PM Scott Wood <swood@redhat.com> wrote:
> 
> Hi Scott,
> 
> > On Mon, 2019-03-25 at 17:53 -0500, Scott Wood wrote:
> > > On Mon, 2019-03-25 at 11:07 +0800, Wu Hao wrote:
> > > > +#if defined(CONFIG_X86) && defined(CONFIG_AS_AVX512)
> > > > +
> > > > +#include <asm/fpu/api.h>
> > > > +
> > > > +static inline void copy512(void *src, void __iomem *dst)
> > > > +{
> > > > +   kernel_fpu_begin();
> > > > +
> > > > +   asm volatile("vmovdqu64 (%0), %%zmm0;"
> > > > +                "vmovntdq %%zmm0, (%1);"
> > > > +                :
> > > > +                : "r"(src), "r"(dst));
> > > > +
> > > > +   kernel_fpu_end();
> > > > +}
> > > 
> > > Shouldn't there be some sort of check that AVX512 is actually
> > > supported
> > > on the running system?
> > > 
> > > Also, src should be const, and the asm statement should have a memory
> > > clobber.
> > > 
> > > > +#else
> > > > +static inline void copy512(void *src, void __iomem *dst)
> > > > +{
> > > > +   WARN_ON_ONCE(1);
> > > > +}
> > > > +#endif
> > > 
> > > Likewise, this will be called if a revision 2 device is used on non-
> > > x86
> > > (or on x86 with an old binutils).  The driver should fall back to 32-
> > > bit
> > > in such cases.
> > 
> > Sorry, I missed the comment about revision 2 only being on integrated
> > devices -- but will that always be the case?  Seems worthwhile to check
> > for
> > AVX512 support anyway.  And there's still the possibility of being built
> > with an old binutils such that CONFIG_AS_AVX512 is not set, or running
> > on a
> > kernel where avx512 was disabled via a boot option.
> 
> The code checks for CONFIG_AS_AVX512 above.

That just indicates that binutils supports it.  Plus, the code does not
check for CONFIG_AS_AVX512 when deciding whether to set pr_datawidth to 64
(and thus call copy512), so you'll get a WARN_ON rather than falling back to
32-bit.

> What boot option are you referring to?

clearcpuid=304

-Scott



  reply	other threads:[~2019-03-26 21:22 UTC|newest]

Thread overview: 62+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-03-25  3:07 [PATCH 00/17] add new features for FPGA DFL drivers Wu Hao
2019-03-25  3:07 ` [PATCH 01/17] fpga: dfl-fme-mgr: fix FME_PR_INTFC_ID register address Wu Hao
2019-03-25 17:28   ` Alan Tull
2019-04-01 19:54   ` Moritz Fischer
2019-04-02  4:38     ` Wu Hao
2019-04-02 13:33       ` Moritz Fischer
2019-03-25  3:07 ` [PATCH 02/17] fpga: dfl: fme: align PR buffer size per PR datawidth Wu Hao
2019-03-25 17:50   ` Alan Tull
2019-03-26  0:28     ` Wu Hao
2019-03-28 18:50       ` Alan Tull
2019-03-25  3:07 ` [PATCH 03/17] fpga: dfl: fme: support 512bit data width PR Wu Hao
2019-03-25 18:48   ` Alan Tull
2019-03-25 22:53   ` Scott Wood
2019-03-25 22:58     ` Scott Wood
2019-03-26 19:33       ` Alan Tull
2019-03-26 21:22         ` Scott Wood [this message]
2019-03-27  4:37           ` Wu Hao
2019-03-27  6:10             ` Scott Wood
2019-03-27  6:03               ` Wu Hao
2019-03-27  5:10       ` Wu Hao
2019-03-27  6:19         ` Scott Wood
2019-03-27  7:10           ` Wu Hao
2019-03-27  5:46     ` Wu Hao
2019-03-25  3:07 ` [PATCH 04/17] Documentation: fpga: dfl: add descriptions for virtualization and new interfaces Wu Hao
2019-03-25  3:07 ` [PATCH 05/17] fpga: dfl: fme: add DFL_FPGA_FME_PORT_RELEASE/ASSIGN ioctl support Wu Hao
2019-03-28 22:03   ` Alan Tull
2019-03-25  3:07 ` [PATCH 06/17] fpga: dfl: pci: enable SRIOV support Wu Hao
2019-03-28 22:03   ` Alan Tull
2019-03-25  3:07 ` [PATCH 07/17] fpga: dfl: afu: add AFU state related sysfs interfaces Wu Hao
2019-03-28 17:13   ` Alan Tull
2019-03-25  3:07 ` [PATCH 08/17] fpga: dfl: afu: add userclock " Wu Hao
2019-04-01 21:41   ` Alan Tull
2019-03-25  3:07 ` [PATCH 09/17] fpga: dfl: add id_table for dfl private feature driver Wu Hao
2019-04-02 15:09   ` Moritz Fischer
2019-04-11 20:55     ` Alan Tull
2019-03-25  3:07 ` [PATCH 10/17] fpga: dfl: afu: export __port_enable/disable function Wu Hao
2019-04-02 15:50   ` Moritz Fischer
2019-04-11 20:45     ` Alan Tull
2019-03-25  3:07 ` [PATCH 11/17] fpga: dfl: afu: add error reporting support Wu Hao
2019-04-09 20:57   ` Alan Tull
2019-04-10  1:43     ` Wu Hao
2019-03-25  3:07 ` [PATCH 12/17] fpga: dfl: afu: add STP (SignalTap) support Wu Hao
2019-04-02 15:07   ` Moritz Fischer
2019-04-11 20:41     ` Alan Tull
2019-03-25  3:07 ` [PATCH 13/17] fpga: dfl: fme: add capability sysfs interfaces Wu Hao
2019-04-09 21:05   ` Alan Tull
2019-03-25  3:07 ` [PATCH 14/17] fpga: dfl: fme: add thermal management support Wu Hao
2019-04-02 14:59   ` Moritz Fischer
2019-04-03 16:31     ` Wu Hao
2019-04-03 18:09       ` Moritz Fischer
2019-04-03 23:43         ` Wu Hao
2019-03-25  3:07 ` [PATCH 15/17] fpga: dfl: fme: add power " Wu Hao
2019-04-11 20:07   ` Alan Tull
2019-04-12  2:50     ` Wu Hao
2019-04-15 21:17       ` Alan Tull
2019-04-17  7:36         ` Wu Hao
2019-04-12 21:05     ` Moritz Fischer
2019-04-17  7:31       ` Wu Hao
2019-03-25  3:07 ` [PATCH 16/17] fpga: dfl: fme: add global error reporting support Wu Hao
2019-04-09 21:35   ` Alan Tull
2019-04-10  1:34     ` Wu Hao
2019-03-25  3:07 ` [PATCH 17/17] fpga: dfl: fme: add performance " Wu Hao

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=580c6c604d3915c105f076fc7f22ab5da98598fc.camel@redhat.com \
    --to=swood@redhat.com \
    --cc=ananda.ravuri@intel.com \
    --cc=atull@kernel.org \
    --cc=hao.wu@intel.com \
    --cc=linux-api@vger.kernel.org \
    --cc=linux-fpga@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mdf@kernel.org \
    --cc=yilun.xu@intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®