Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Tue, 7 Aug 2018 10:07:28 -0400 (EDT)
From:   Mikulas Patocka <mpatocka@redhat.com>
To:     David Laight <David.Laight@ACULAB.COM>
cc:     "'Ard Biesheuvel'" <ard.biesheuvel@linaro.org>,
        Ramana Radhakrishnan <ramana.gcc@googlemail.com>,
        Florian Weimer <fweimer@redhat.com>,
        Thomas Petazzoni <thomas.petazzoni@free-electrons.com>,
        GNU C Library <libc-alpha@sourceware.org>,
        Andrew Pinski <pinskia@gmail.com>,
        Catalin Marinas <catalin.marinas@arm.com>,
        Will Deacon <will.deacon@arm.com>,
        Russell King <linux@armlinux.org.uk>,
        LKML <linux-kernel@vger.kernel.org>,
        linux-arm-kernel <linux-arm-kernel@lists.infradead.org>
Subject: RE: framebuffer corruption due to overlapping stp instructions on
 arm64
In-Reply-To: <51a6c4e102ad4193b3f42498f0ff11a4@AcuMS.aculab.com>
Message-ID: <alpine.LRH.2.02.1808070939320.6020@file01.intranet.prod.int.rdu2.redhat.com>
References: <alpine.LRH.2.02.1808021242320.31834@file01.intranet.prod.int.rdu2.redhat.com> <CA+=Sn1mWkjuwVnjw6OWWUM=UcP76bdFa680FebCseewHfx3NpA@mail.gmail.com> <9acdacdb-3bd5-b71a-3003-e48132ee1371@redhat.com> <CAJA7tRZbmnZq7RfvQeYEy_a1ZObWqpFpVdvgsXgsioQ3RyPMuA@mail.gmail.com>
 <CAKv+Gu97QvwoLLK_zueiA_gjg_4Q5cqU4YVUyHUVFFfffdyJaw@mail.gmail.com> <f696ebe8605840e3bb04bb78b60a6cfa@AcuMS.aculab.com> <alpine.LRH.2.02.1808030759480.12341@file01.intranet.prod.int.rdu2.redhat.com> <a1564e8d091648bcad9b5ec58ab6cc95@AcuMS.aculab.com>
 <alpine.LRH.2.02.1808051018360.23136@file01.intranet.prod.int.rdu2.redhat.com> <51a6c4e102ad4193b3f42498f0ff11a4@AcuMS.aculab.com>
User-Agent: Alpine 2.02 (LRH 1266 2009-07-14)
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk


On Mon, 6 Aug 2018, David Laight wrote:

> From: Mikulas Patocka
> > Sent: 05 August 2018 15:36
> > To: David Laight
> ...
> > There's an instruction movntdqa (and vmovntdqa) that can actually do
> > prefetch on write-combining memory type. It's the only instruction that
> > can do it.
> > 
> > It this instruction is used on non-write-combining memory type, it behaves
> > like movdqa.
> > 
> ...
> > I benchmarked it on a processor with ERMS - for writes to the framebuffer,
> > there's no difference between memcpy, 8-byte writes, rep stosb, rep stosq,
> > mmx, sse, avx - all this method achieve 16-17 GB/s
> 
> The combination of write-combining, posted writes and a fast PCIe slave
> are probably why there is little difference.
> 
> > For reading from the framebuffer:
> >  323 MB/s - memcpy (using avx2)
> >   91 MB/s - explicit 8-byte reads
> >  249 MB/s - rep movsq
> >  307 MB/s - rep movsb
> 
> You must be getting the ERMS hardware optimised 'rep movsb'.
> 
> >   90 MB/s - mmx
> >  176 MB/s - sse
> > 4750 MB/s - sse movntdqa
> >  330 MB/s - avx
> 
> avx512 is probably faster still.
> 
> > 5369 MB/s - avx vmovntdqa
> > 
> > So - it may make sense to introduce a function memcpy_from_framebuffer()
> > that uses movntdqa or vmovntdqa on CPUs that support it.
> 
> For kernel space it ought to be just memcpy_fromio().

I meant for userspace. Unaccelerated scrolling is still painfully slow 
even on modern computers because of slow framebuffer read. If glibc 
provided a function memcpy_from_framebuffer() that used movntdqa and the 
fbdev Xorg driver used it, it would help the users who use unaccelerated 
drivers for some reason.

> Can you easily repeat the tests using a non-write-combining map of the
> same PCIe slave?

I mapped the framebuffer as uncached and these are the results:

reading from the framebuffer:
318 MB/s - memcpy
 74 MB/s - explicit 8-byte reads
 73 MB/s - rep movsq
 11 MB/s - rep movsb
 87 MB/s - mmx
173 MB/s - sse
173 MB/s - sse movntdqa
323 MB/s - avx
284 MB/s - avx vmovntdqa

zeroing the framebuffer:
 19 MB/s - memset
154 MB/s - explicit 8-byte writes
152 MB/s - rep stosq
 19 MB/s - rep stosb
152 MB/s - mmx
306 MB/s - sse
621 MB/s - avx

copying data to the framebuffer:
618 MB/s - memcpy (using avx2)
152 MB/s - explicit 8-byte writes
139 MB/s - rep movsq
 17 MB/s - rep movsb
154 MB/s - mmx
305 MB/s - sse
306 MB/s - sse movntdqa
619 MB/s - avx
619 MB/s - avx movntdqa

> I can probably run the same measurements against our rather leisurely
> FPGA based PCIe slave.
> IIRC PCIe reads happen every 128 clocks of the cards 62.5MHz clock,
> increasing the size of the registers makes a significant different.
> I've not tried mapping write-combining and using (v)movntdaq.
> I'm not sure what effect write-combining would have if the whole BAR
> were mapped that way - so I'll either have to map the physical addresses
> twice or add in another BAR.
> 
> 	David

Mikulas