From: Tomi Valkeinen <tomi.valkeinen@ti.com>
To: Laurent Pinchart <laurent.pinchart@ideasonboard.com>,
dri-devel@lists.freedesktop.org
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>,
Thierry Reding <treding@nvidia.com>,
Benjamin Gaignard <benjamin.gaignard@linaro.org>
Subject: Re: [PATCH 00/31] Convert omapdrm to the atomic updates API
Date: Mon, 20 Apr 2015 14:22:53 +0300 [thread overview]
Message-ID: <5534E18D.1080902@ti.com> (raw)
In-Reply-To: <1429135788-31979-1-git-send-email-laurent.pinchart@ideasonboard.com>
[-- Attachment #1.1: Type: text/plain, Size: 7244 bytes --]
Hi,
On 16/04/15 01:09, Laurent Pinchart wrote:
> Hello,
>
> This patch set converts the omapdrm to the atomic update API in 31 small(ish)
> and hopefully reviewable steps.
Thanks! I'll start reviewing the patches later this week.
> The series is based on the latest drm/next branch and depends on the pending
> "drm: omapdrm: Store the rotation property in dev->mode_config" patch by
> Daniel Vetter.
>
> The omapdrm loses support for its custom fences in the process. This isn't
> deemed to be an issue, as the custom fences API is used by the SGX driver
> only, which requires out-of-tree patches anyway. Fences support will be
> reimplemented at a later point on top of the atomic updates conversion using
> the mainline fence API.
>
> One known issue is a race condition between asynchronous commits and
> drm_release() which can trigger a warning if a commit is applied between the
> drm_events_release() call and the WARN_ON(!list_empty(&file_priv->event_list)).
> Investigation is ongoing to find out the best way to fix the problem, the
> problem seems not to be limited to omapdrm (but can be more or less difficult
> to reproduce depending on the driver). Ideas would be welcome, I've CC'ed
> maintainers of possibly affected drivers.
>
> There's not much else to be said, details are explained in individual patches.
>
> The patches can be found in my git tree at
>
> git://linuxtv.org/pinchartl/fbdev.git omapdrm/next
I ran a few quick tests with your branch, on Pandaboard.
Something related to vblanks/timing is broken, and I can only get half
the fps. The "db" test app from my omapdrm-tests shows this easily.
I also got a few warnings (below), but that was before I applied your
"drm: omapdrm: Support unlinking page flip events prematurely" patch.
After that I didn't see those anymore.
I didn't see any other issues, but my tests were quite quick and simple.
Tomi
[ 190.434112] ------------[ cut here ]------------
[ 190.439575] WARNING: CPU: 0 PID: 175 at
drivers/gpu/drm/omapdrm/omap_gem.c:1292 omap_gem_free_object+0x260/0x278
[omapdrm]()
[ 190.451385] Modules linked in: omapdrm drm_kms_helper drm cfbfillrect
cfbimgblt cfbcopyarea panel_nec_nl8048hl11 panel_sony_acx565akm
panel_lgphilips_lb035q02 panel_dsi_cm pan
el_dpi connector_analog_tv connector_hdmi connector_dvi
encoder_tpd12s015 encoder_tfp410 omapdss
[ 190.475860] CPU: 0 PID: 175 Comm: producer Not tainted
4.0.0-rc7-00759-g7290d47d0475 #4
[ 190.485198] Hardware name: Generic OMAP4 (Flattened Device Tree)
[ 190.485626] Backtrace:
[ 190.493469] [<c00137dc>] (dump_backtrace) from [<c00139f0>]
(show_stack+0x18/0x1c)
[ 190.502075] r6:c0a07598 r5:c0a07598 r4:00000000 r3:00000000
[ 190.505676] [<c00139d8>] (show_stack) from [<c068bf8c>]
(dump_stack+0xa4/0xd4)
[ 190.515686] [<c068bee8>] (dump_stack) from [<c0045084>]
(warn_slowpath_common+0x84/0xc0)
[ 190.524230] r6:0000050c r5:bf145890 r4:00000000 r3:ed148b80
[ 190.526794] [<c0045000>] (warn_slowpath_common) from [<c0045164>]
(warn_slowpath_null+0x24/0x2c)
[ 190.539459] r8:edac6678 r7:edf9eb98 r6:ee6b7800 r5:ed1b689c r4:ee713b80
[ 190.545288] [<c0045140>] (warn_slowpath_null) from [<bf145890>]
(omap_gem_free_object+0x260/0x278 [omapdrm])
[ 190.556945] [<bf145630>] (omap_gem_free_object [omapdrm]) from
[<bf0ae0bc>] (drm_gem_object_free+0x30/0x38 [drm])
[ 190.567993] r9:00000010 r8:edac6678 r7:edf9eb98 r6:ee2c2b10
r5:ee6b7834 r4:ee713b80
[ 190.575134] [<bf0ae08c>] (drm_gem_object_free [drm]) from
[<bf146154>] (omap_gem_dmabuf_release+0xa8/0xb4 [omapdrm])
[ 190.587432] [<bf1460ac>] (omap_gem_dmabuf_release [omapdrm]) from
[<c04211e0>] (dma_buf_release+0x54/0x17c)
[ 190.597717] r5:00000000 r4:ee60e800
[ 190.601501] [<c042118c>] (dma_buf_release) from [<c016ba6c>]
(__fput+0x9c/0x1ec)
[ 190.609283] r9:00000010 r8:edac6678 r7:edf9eb98 r6:ee2c2b10
r5:edac6678 r4:ee6fce80
[ 190.614654] [<c016b9d0>] (__fput) from [<c016bc24>] (____fput+0x10/0x14)
[ 190.619750] r10:ed148b80 r9:00000001 r8:00000000 r7:ed148b80
r6:c0a53738 r5:ed02ce40
[ 190.632812] r4:ed149084
[ 190.635467] [<c016bc14>] (____fput) from [<c006099c>]
(task_work_run+0xbc/0x100)
[ 190.643432] [<c00608e0>] (task_work_run) from [<c0047b20>]
(do_exit+0x390/0xb00)
[ 190.651214] r8:ebc16658 r7:ebc16600 r6:b6f75750 r5:c0a402c4
r4:ed14909c r3:0000003c
[ 190.654876] [<c0047790>] (do_exit) from [<c0048314>]
(do_group_exit+0x48/0xc4)
[ 190.667144] r7:000000f8
[ 190.670074] [<c00482cc>] (do_group_exit) from [<c00483a8>]
(__wake_up_parent+0x0/0x28)
[ 190.678466] r6:b6f75750 r5:00000000 r4:00000001 r3:ffffffff
[ 190.684753] [<c0048390>] (SyS_exit_group) from [<c000f880>]
(ret_fast_syscall+0x0/0x4c)
[ 190.693237] ---[ end trace 0f3a4a8e4c514be3 ]---
[ 226.385437] ------------[ cut here ]------------
[ 226.385437] WARNING: CPU: 0 PID: 185 at
drivers/gpu/drm/drm_fops.c:459 drm_release+0x418/0x464 [drm]()
[ 226.385437] Modules linked in: omapdrm drm_kms_helper drm cfbfillrect
cfbimgblt cfbcopyarea panel_nec_nl8048hl11 panel_sony_acx565akm
panel_lgphilips_lb035q02 panel_dsi_cm pan
el_dpi connector_analog_tv connector_hdmi connector_dvi
encoder_tpd12s015 encoder_tfp410 omapdss
[ 226.425781] CPU: 0 PID: 185 Comm: consumer Tainted: G W
4.0.0-rc7-00759-g7290d47d0475 #4
[ 226.435668] Hardware name: Generic OMAP4 (Flattened Device Tree)
[ 226.441986] Backtrace:
[ 226.442413] [<c00137dc>] (dump_backtrace) from [<c00139f0>]
(show_stack+0x18/0x1c)
[ 226.452545] r6:c0a07598 r5:c0a07598 r4:00000000 r3:00000000
[ 226.456451] [<c00139d8>] (show_stack) from [<c068bf8c>]
(dump_stack+0xa4/0xd4)
[ 226.466156] [<c068bee8>] (dump_stack) from [<c0045084>]
(warn_slowpath_common+0x84/0xc0)
[ 226.474700] r6:000001cb r5:bf0ad8ac r4:00000000 r3:e380a280
[ 226.475097] [<c0045000>] (warn_slowpath_common) from [<c0045164>]
(warn_slowpath_null+0x24/0x2c)
[ 226.489959] r8:e57c52f4 r7:ee6b7834 r6:ee6b7800 r5:e57c52f8 r4:e57c5200
[ 226.494720] [<c0045140>] (warn_slowpath_null) from [<bf0ad8ac>]
(drm_release+0x418/0x464 [drm])
[ 226.506378] [<bf0ad494>] (drm_release [drm]) from [<c016ba6c>]
(__fput+0x9c/0x1ec)
[ 226.506896] r10:00000000 r9:00000008 r8:ed34d838 r7:edf3c240
r6:ee698910 r5:ed34d838
[ 226.520416] r4:ee762d40
[ 226.525451] [<c016b9d0>] (__fput) from [<c016bc24>] (____fput+0x10/0x14)
[ 226.525573] r10:e380a280 r9:00000001 r8:00000000 r7:e380a280
r6:c0a53738 r5:ee6f4680
[ 226.535339] r4:e380a784
[ 226.543457] [<c016bc14>] (____fput) from [<c006099c>]
(task_work_run+0xbc/0x100)
[ 226.551239] [<c00608e0>] (task_work_run) from [<c0047b20>]
(do_exit+0x390/0xb00)
[ 226.559020] r8:e57c5658 r7:e57c5600 r6:b6f46750 r5:c0a402c4
r4:e380a79c r3:0000003d
[ 226.567230] [<c0047790>] (do_exit) from [<c0048314>]
(do_group_exit+0x48/0xc4)
[ 226.574829] r7:000000f8
[ 226.575164] [<c00482cc>] (do_group_exit) from [<c00483a8>]
(__wake_up_parent+0x0/0x28)
[ 226.585845] r6:b6f46750 r5:00000001 r4:00000001 r3:ffffffff
[ 226.586181] [<c0048390>] (SyS_exit_group) from [<c000f880>]
(ret_fast_syscall+0x0/0x4c)
[ 226.600250] ---[ end trace 0f3a4a8e4c514be4 ]---
[-- Attachment #1.2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 836 bytes --]
[-- Attachment #2: Type: text/plain, Size: 159 bytes --]
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/dri-devel
prev parent reply other threads:[~2015-04-20 11:23 UTC|newest]
Thread overview: 34+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-04-15 22:09 [PATCH 00/31] Convert omapdrm to the atomic updates API Laurent Pinchart
2015-04-15 22:09 ` [PATCH 01/31] drm: omapdrm: Store the rotation property in dev->mode_config Laurent Pinchart
2015-04-15 22:09 ` [PATCH 02/31] drm: omapdrm: Apply settings synchronously Laurent Pinchart
2015-04-15 22:09 ` [PATCH 03/31] drm: omapdrm: Rename omap_crtc_page_flip_locked to omap_crtc_page_flip Laurent Pinchart
2015-04-15 22:09 ` [PATCH 04/31] drm: omapdrm: Rename omap_crtc page flip-related fields Laurent Pinchart
2015-04-15 22:09 ` [PATCH 05/31] drm: omapdrm: Simplify IRQ registration Laurent Pinchart
2015-04-15 22:09 ` [PATCH 06/31] drm: omapdrm: Cancel pending page flips when closing device Laurent Pinchart
2015-04-15 22:09 ` [PATCH 07/31] drm: omapdrm: Rework page flip handling Laurent Pinchart
2015-04-15 22:09 ` [PATCH 08/31] drm: omapdrm: Turn vblank on/off when enabling/disabling CRTC Laurent Pinchart
2015-04-15 22:09 ` [PATCH 09/31] drm: omapdrm: Fix page flip race with CRTC disable Laurent Pinchart
2015-04-15 22:09 ` [PATCH 10/31] drm: omapdrm: Clean up #include's Laurent Pinchart
2015-04-15 22:09 ` [PATCH 11/31] drm: omapdrm: Rename CRTC DSS operations with an omap_crtc_dss_ prefix Laurent Pinchart
2015-04-15 22:09 ` [PATCH 12/31] drm: omapdrm: Rework CRTC enable/disable for atomic updates Laurent Pinchart
2015-04-15 22:09 ` [PATCH 13/31] drm: omapdrm: Implement encoder .disable() and .enable() operations Laurent Pinchart
2015-04-15 22:09 ` [PATCH 14/31] drm: omapdrm: Wire up atomic state object scaffolding Laurent Pinchart
2015-04-15 22:09 ` [PATCH 15/31] drm: omapdrm: Implement planes atomic operations Laurent Pinchart
2015-04-15 22:09 ` [PATCH 16/31] drm: omapdrm: Handle primary plane config through atomic plane ops Laurent Pinchart
2015-04-15 22:09 ` [PATCH 17/31] drm: omapdrm: Switch plane update to atomic helpers Laurent Pinchart
2015-04-15 22:09 ` [PATCH 18/31] drm: omapdrm: Switch mode config " Laurent Pinchart
2015-04-15 22:09 ` [PATCH 19/31] drm: omapdrm: Switch connector DPMS " Laurent Pinchart
2015-04-15 22:09 ` [PATCH 20/31] drm: omapdrm: Replace encoder mode_fixup with atomic_check Laurent Pinchart
2015-04-15 22:09 ` [PATCH 21/31] drm: omapdrm: Implement asynchronous commit support Laurent Pinchart
2015-04-15 22:09 ` [PATCH 22/31] drm: omapdrm: Switch page flip to atomic helpers Laurent Pinchart
2015-04-15 22:09 ` [PATCH 23/31] drm: omapdrm: Drop manual framebuffer pin handling Laurent Pinchart
2015-04-15 22:09 ` [PATCH 24/31] drm: omapdrm: Switch crtc and plane set_property to atomic helpers Laurent Pinchart
2015-04-15 22:09 ` [PATCH 25/31] drm: omapdrm: Move plane info and win out of the plane structure Laurent Pinchart
2015-04-15 22:09 ` [PATCH 26/31] drm: omapdrm: Move crtc info out of the crtc structure Laurent Pinchart
2015-04-15 22:09 ` [PATCH 27/31] drm: omapdrm: Remove omap_crtc enabled field Laurent Pinchart
2015-04-15 22:09 ` [PATCH 28/31] drm: omapdrm: Remove omap_plane " Laurent Pinchart
2015-04-15 22:09 ` [PATCH 29/31] drm: omapdrm: Make the omap_crtc_flush function static Laurent Pinchart
2015-04-15 22:09 ` [PATCH 30/31] drm: omapdrm: Don't get/put dispc in omap_crtc_flush() Laurent Pinchart
2015-04-15 22:09 ` [PATCH 31/31] drm: omapdrm: omap_crtc_flush() isn't called with modeset locked Laurent Pinchart
2015-04-16 9:12 ` [PATCH 00/31] Convert omapdrm to the atomic updates API Daniel Vetter
2015-04-20 11:22 ` Tomi Valkeinen [this message]
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=5534E18D.1080902@ti.com \
--to=tomi.valkeinen@ti.com \
--cc=benjamin.gaignard@linaro.org \
--cc=daniel.vetter@ffwll.ch \
--cc=dri-devel@lists.freedesktop.org \
--cc=laurent.pinchart@ideasonboard.com \
--cc=treding@nvidia.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