From: "Maíra Canal" <mcanal@igalia.com>
To: marius.vlad@collabora.com
Cc: mwen@igalia.com, rodrigosiqueiramelo@gmail.com,
"Maíra Canal" <mcanal@igalia.com>,
dri-devel@lists.freedesktop.org, melissa.srw@gmail.com,
igormtorrente@gmail.com
Subject: Re: [PATCH 2/2] vkms: Add support for multiple connectors
Date: Wed, 5 Apr 2023 10:51:46 -0300 [thread overview]
Message-ID: <20230405135145.305293-1-mcanal@igalia.com> (raw)
In-Reply-To: <20221202142051.136651-3-marius.vlad@collabora.com>
Hi Marius,
> This patch adds support for creating multiple virtual connectors, in
> case one would need it. Use module parameters to specify how many,
> defaulting to just one, allocating from the start, a maximum of 4
> possible outputs.
I got a bit confused by this description. The commit message says that you
are creating multiple connectors, but it seems like you are creating multiple
pipes. From what I could see, for each new connector created, you are also
creating a new CRTC and new planes.
Moreover, if your real intention is to create multiple pipes, I believe that
you don't really need to duplicate the planes as well.
I ran the VKMS CI [1] with your patches applied and, although most of the
tests are now passing with multiple pipes, which is really nice, the test
kms_writeback crashes with the following error:
[ 1997.244347] [IGT] kms_writeback: starting subtest writeback-check-output
[ 1997.250673] BUG: kernel NULL pointer dereference, address: 000000000000016c
[ 1997.250691] #PF: supervisor read access in kernel mode
[ 1997.250693] #PF: error_code(0x0000) - not-present page
[ 1997.250694] PGD 800000001a877067 P4D 800000001a877067 PUD 1a872067 PMD 0
[ 1997.250697] Oops: 0000 [#1] PREEMPT SMP PTI
[ 1997.250699] CPU: 2 PID: 3223 Comm: kms_writeback Not tainted 6.3.0-rc4-g8c2c29ba129f-dirty #257
[ 1997.250701] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.16.2-1.fc37 04/01/2014
[ 1997.250703] RIP: 0010:drm_vblank_get+0xa/0xe0
[ 1997.250710] Code: a9 66 66 66 66 66 66 2e 0f 1f 84 00 00 00 00 00 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 55 41 57 41 56 41 55 41 54 53 <8b> 87 6c 01 00 00 41 bc ea ff ff ff 85 c0 74 74 89 f5 48 89 fb 39
[ 1997.250712] RSP: 0018:ffffb84d407a3b08 EFLAGS: 00010202
[ 1997.250714] RAX: 0000000000000000 RBX: ffffa2eb02bf8b70 RCX: 0000000000000000
[ 1997.250718] RDX: ffffa2eb180d2420 RSI: 0000000000000000 RDI: 0000000000000000
[ 1997.250719] RBP: ffffa2eb02bf99e8 R08: 0000000000000036 R09: ffffa2eb01d620c0
[ 1997.250720] R10: ffffe82b84027e40 R11: ffffffffc0042520 R12: ffffa2eb01c01900
[ 1997.250721] R13: ffffa2eb02bf9b60 R14: 0000000000000001 R15: ffffa2ea1ecd6b80
[ 1997.250722] FS: 00007f7d44e89a80(0000) GS:ffffa2eb3bd00000(0000) knlGS:0000000000000000
[ 1997.250723] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 1997.250725] CR2: 000000000000016c CR3: 000000001ec02000 CR4: 00000000000006e0
[ 1997.250728] Call Trace:
[ 1997.250735] <TASK>
[ 1997.250736] vkms_set_composer+0x18/0x60 [vkms]
[ 1997.250745] vkms_wb_atomic_commit+0x93/0x150 [vkms]
[ 1997.250749] drm_atomic_helper_commit_modeset_enables+0x1d9/0x250
[ 1997.250754] vkms_atomic_commit_tail+0x33/0xb0 [vkms]
[ 1997.250758] commit_tail+0x8d/0x170
[ 1997.250761] drm_atomic_helper_commit+0x26b/0x280
[ 1997.250763] drm_atomic_commit+0x9f/0xc0
[ 1997.250766] ? __pfx___drm_printfn_info+0x10/0x10
[ 1997.250769] drm_mode_atomic_ioctl+0x34c/0x480
[ 1997.250771] ? __pfx_drm_mode_atomic_ioctl+0x10/0x10
[ 1997.250773] drm_ioctl_kernel+0xd7/0x150
[ 1997.250780] drm_ioctl+0x31f/0x490
[ 1997.250781] ? __pfx_drm_mode_atomic_ioctl+0x10/0x10
[ 1997.250783] __se_sys_ioctl+0x6d/0xb0
[ 1997.250789] do_syscall_64+0x43/0x90
[ 1997.250795] entry_SYSCALL_64_after_hwframe+0x72/0xdc
[1] https://gitlab.freedesktop.org/drm/igt-gpu-tools/-/tree/master/tests/vkms_ci
Best Regards,
- Maíra Canal
>
> This is of particular importance when testing out the DRM backend in
> compositors, but also to be able to independently handle multiple
> outputs/connectors, like setting one to off/sleep on while the others
> are on, and combinations that arise from that.
>
> Signed-off-by: Marius Vlad <marius.vlad@collabora.com>
> ---
> drivers/gpu/drm/vkms/vkms_crtc.c | 3 +--
> drivers/gpu/drm/vkms/vkms_drv.c | 26 ++++++++++++++++++++++----
> drivers/gpu/drm/vkms/vkms_drv.h | 8 +++++---
> drivers/gpu/drm/vkms/vkms_output.c | 5 ++---
> drivers/gpu/drm/vkms/vkms_writeback.c | 18 ++++++++----------
> 5 files changed, 38 insertions(+), 22 deletions(-)
>
next prev parent reply other threads:[~2023-04-05 13:53 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-12-02 14:20 [PATCH 0/2] vkms: Add support for multiple connectors Marius Vlad
2022-12-02 14:20 ` [PATCH 1/2] vkms: Pass the correct bitmask for possible crtcs Marius Vlad
2022-12-02 14:20 ` [PATCH 2/2] vkms: Add support for multiple connectors Marius Vlad
2023-04-05 13:51 ` Maíra Canal [this message]
2023-04-06 9:17 ` Marius Vlad
2023-04-06 11:29 ` Thomas Zimmermann
2023-04-06 14:04 ` Marius Vlad
2023-04-06 14:18 ` Thomas Zimmermann
2023-04-10 18:17 ` Melissa Wen
2023-04-20 8:44 ` Marius Vlad
2023-04-06 10:32 ` Thomas Zimmermann
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=20230405135145.305293-1-mcanal@igalia.com \
--to=mcanal@igalia.com \
--cc=dri-devel@lists.freedesktop.org \
--cc=igormtorrente@gmail.com \
--cc=marius.vlad@collabora.com \
--cc=melissa.srw@gmail.com \
--cc=mwen@igalia.com \
--cc=rodrigosiqueiramelo@gmail.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;
as well as URLs for NNTP newsgroup(s).