From: Liviu Dudau <liviu.dudau@arm.com>
To: Chris Wilson <chris@chris-wilson.co.uk>
Cc: IGT GPU Tools <igt-dev@lists.freedesktop.org>,
Boris Brezillon <boris.brezillon@free-electrons.com>,
Intel GFX ML <intel-gfx@lists.freedesktop.org>,
Brian Starkey <brian.starkey@arm.com>
Subject: Re: [igt-dev] [Intel-gfx] [PATCH i-g-t v5 3/6] lib: Add function to hash a framebuffer
Date: Wed, 16 Jan 2019 11:20:09 +0000 [thread overview]
Message-ID: <20190116112009.GZ20661@e110455-lin.cambridge.arm.com> (raw)
In-Reply-To: <154757805730.30063.1853125668491466900@skylake-alporthouse-com>
On Tue, Jan 15, 2019 at 06:47:47PM +0000, Chris Wilson wrote:
> Quoting Liviu Dudau (2019-01-15 17:47:44)
> > +int igt_fb_get_crc(struct igt_fb *fb, igt_crc_t *crc)
> > +{
> > +#define FNV1a_OFFSET_BIAS 2166136261
> > +#define FNV1a_PRIME 16777619
> > + uint32_t hash;
> > + void *map;
> > + char *ptr, *line = NULL;
> > + int x, y, cpp = igt_drm_format_to_bpp(fb->drm_format) / 8;
> > + uint32_t stride = calc_plane_stride(fb, 0);
> > +
> > + if (fb->is_dumb)
> > + map = kmstest_dumb_map_buffer(fb->fd, fb->gem_handle, fb->size,
> > + PROT_READ);
> > + else
> > + map = gem_mmap__gtt(fb->fd, fb->gem_handle, fb->size,
> > + PROT_READ);
> > + ptr = map;
> > +
> > + /*
> > + * Framebuffers are often uncached, which can make byte-wise accesses
> > + * very slow. We copy each line of the FB into a local buffer to speed
> > + * up the hashing.
> > + */
> > + line = malloc(stride);
> > + if (!line) {
> > + munmap(map, fb->size);
> > + return -ENOMEM;
> > + }
> > +
> > + hash = FNV1a_OFFSET_BIAS;
> > +
> > + for (y = 0; y < fb->height; y++, ptr += stride) {
> > +
> > + memcpy(line, ptr, stride);
>
> igt_memcpy_from_wc() for the reasons cited above.
> -Chris
Hi Chris,
Thanks for pointing out the function, I was not aware of it!
Now, looking at the implementation, and ignoring the fact that it is in a
file called igt_x86.c, it looks to me that it will end up calling memcpy
anyway for Arm drivers. Not being a GCC expert, I am wondering if the
ifunc() wrapper around resolve_memcpy_from_wc() will not actually
prevent the compiler from choosing an optimised version of memcpy for Arm.
I can refresh the patch if you think it is safe to use igt_memcpy_from_wc() for
non-x86 architectures.
Best regards,
Liviu
--
====================
| I would like to |
| fix the world, |
| but they're not |
| giving me the |
\ source code! /
---------------
¯\_(ツ)_/¯
_______________________________________________
igt-dev mailing list
igt-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/igt-dev
next prev parent reply other threads:[~2019-01-16 11:20 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-01-15 17:47 [igt-dev] [PATCH i-g-t v5 0/6] igt: Add support for testing writeback connectors Liviu Dudau
2019-01-15 17:47 ` [igt-dev] [PATCH i-g-t v5 1/6] lib/igt_kms: Add writeback support Liviu Dudau
2019-02-04 13:31 ` [Intel-gfx] " Brian Starkey
2019-03-06 21:30 ` [Intel-gfx] [igt-dev] " Rodrigo Siqueira
2019-03-18 10:41 ` Liviu Dudau
2019-03-18 22:05 ` Rodrigo Siqueira
2019-03-20 14:46 ` Liviu Dudau
2019-01-15 17:47 ` [igt-dev] [PATCH i-g-t v5 2/6] kms_writeback: Add initial writeback tests Liviu Dudau
2019-01-15 17:47 ` [Intel-gfx] [PATCH i-g-t v5 3/6] lib: Add function to hash a framebuffer Liviu Dudau
2019-01-15 18:47 ` [igt-dev] " Chris Wilson
2019-01-16 11:20 ` Liviu Dudau [this message]
2019-01-16 11:50 ` Chris Wilson
2019-01-16 12:21 ` [igt-dev] [PATCH i-g-t v6] " Liviu Dudau
2019-01-15 17:47 ` [igt-dev] [PATCH i-g-t v5 4/6] kms_writeback: Add writeback-check-output Liviu Dudau
2019-01-15 17:47 ` [igt-dev] [PATCH i-g-t v5 5/6] lib/igt_kms: Add igt_output_clone_pipe for cloning Liviu Dudau
2019-02-04 13:27 ` Brian Starkey
2019-01-15 17:47 ` [igt-dev] [PATCH i-g-t v5 6/6] kms_writeback: Add tests using a cloned output Liviu Dudau
2019-01-15 19:08 ` [igt-dev] ✓ Fi.CI.BAT: success for igt: Add support for testing writeback connectors (rev3) Patchwork
2019-01-16 0:13 ` [igt-dev] ✗ Fi.CI.IGT: failure " Patchwork
2019-01-16 12:41 ` [igt-dev] ✗ Fi.CI.BAT: failure for igt: Add support for testing writeback connectors (rev4) Patchwork
2019-02-04 13:36 ` [igt-dev] [PATCH i-g-t v5 0/6] igt: Add support for testing writeback connectors Brian Starkey
2019-03-06 21:42 ` [igt-dev] ✗ Fi.CI.BAT: failure for igt: Add support for testing writeback connectors (rev5) Patchwork
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=20190116112009.GZ20661@e110455-lin.cambridge.arm.com \
--to=liviu.dudau@arm.com \
--cc=boris.brezillon@free-electrons.com \
--cc=brian.starkey@arm.com \
--cc=chris@chris-wilson.co.uk \
--cc=igt-dev@lists.freedesktop.org \
--cc=intel-gfx@lists.freedesktop.org \
/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