drm/ssd130x: Fix an oops when attempting to update a disabled plane
Subsystems (3)
Javier Martinez Canillas
maintainerPATCH v3 1/1 drm/ssd130x: Fix an oops when attempting to update a disabled plane
Geert reports that the following NULL pointer dereference happens for him after commit 49d7d581ceaf ("drm/ssd130x: Don't allocate buffers on each plane update"):
[drm] Initialized ssd130x 1.0.0 20220131 for 0-003c on minor 0
ssd130x-i2c 0-003c: [drm] surface width(128), height(32), bpp(1)
and format(R1 little-endian (0x20203152))
Unable to handle kernel NULL pointer dereference at virtual address 00000000
Oops [#1]
CPU: 0 PID: 1 Comm: swapper Not tainted
6.5.0-rc1-orangecrab-02219-g0a529a1e4bf4 #565
epc : ssd130x_update_rect.isra.0+0x13c/0x340
ra : ssd130x_update_rect.isra.0+0x2bc/0x340
...
status: 00000120 badaddr: 00000000 cause: 0000000f
[<c0303d90>] ssd130x_update_rect.isra.0+0x13c/0x340
[<c0304200>] ssd130x_primary_plane_helper_atomic_update+0x26c/0x284
[<c02f8d54>] drm_atomic_helper_commit_planes+0xfc/0x27c
[<c02f9314>] drm_atomic_helper_commit_tail+0x5c/0xb4
[<c02f94fc>] commit_tail+0x190/0x1b8
[<c02f99fc>] drm_atomic_helper_commit+0x194/0x1c0
[<c02c5d00>] drm_atomic_commit+0xa4/0xe4
[<c02cce40>] drm_client_modeset_commit_atomic+0x244/0x278
[<c02ccef0>] drm_client_modeset_commit_locked+0x7c/0x1bc
[<c02cd064>] drm_client_modeset_commit+0x34/0x64
[<c0301a78>] __drm_fb_helper_restore_fbdev_mode_unlocked+0xc4/0xe8
[<c0303424>] drm_fb_helper_set_par+0x38/0x58
[<c027c410>] fbcon_init+0x294/0x534
...The problem is that fbcon calls fbcon_init() which triggers a DRM modeset and this leads to drm_atomic_helper_commit_planes() attempting to commit the atomic state for all planes, even the ones whose CRTC is not enabled.
Since the primary plane buffer is allocated in the encoder .atomic_enable callback, this happens after that initial modeset commit and leads to the mentioned NULL pointer dereference.
Fix this by allocating the buffers in the struct drm_plane_helper_funcs .begin_fb_access callback and free them in to .end_fb_access callback, instead of doing it when the encoder is enabled.
---
Changes in v3:
- Move the buffers allocation to the plane helper funcs .begin_fb_access and the free to .end_fb_access callbacks.
- Always allocate intermediate buffer because is use in ssd130x_clear_screen().
- Don't allocate the buffers as device managed resources.
Changes in v2:
- Move the buffers allocation to .fb_create instead of preventing atomic updates for disable planes.
- Don't allocate the intermediate buffer when using the native R1 format.
- Allocate buffers as device managed resources.
drivers/gpu/drm/solomon/ssd130x.c | 33 ++++++++++++++++++++++++------- 1 file changed, 26 insertions(+), 7 deletions(-)
drivers/gpu/drm/solomon/ssd130x.c +26 -7
@@ -612,6 +612,30 @@ static int ssd130x_fb_blit_rect(struct drm_framebuffer *fb, const struct iosys_mreturn ret; } +static int ssd130x_primary_plane_helper_begin_fb_access (struct drm_plane *plane, +struct drm_plane_state *state) +{ +struct drm_device *drm = plane->dev; +struct ssd130x_device *ssd130x =drm_to_ssd130x (drm); +int ret =ssd130x_buf_alloc (ssd130x); + +if (ret) +return ret; + +return drm_gem_begin_shadow_fb_access (plane, state); +} + +static void ssd130x_primary_plane_helper_end_fb_access (struct drm_plane *plane, +struct drm_plane_state *state) +{ +struct drm_device *drm = plane->dev; +struct ssd130x_device *ssd130x =drm_to_ssd130x (drm); + +ssd130x_buf_free (ssd130x); + +return drm_gem_end_shadow_fb_access (plane, state); +} +static void ssd130x_primary_plane_helper_atomic_update (struct drm_plane *plane,struct drm_atomic_state *state) {
@@ -656,7 +680,8 @@ static void ssd130x_primary_plane_helper_atomic_disable(struct drm_plane *plane, }static const struct drm_plane_helper_funcs ssd130x_primary_plane_helper_funcs = { -DRM_GEM_SHADOW_PLANE_HELPER_FUNCS , + .begin_fb_access = ssd130x_primary_plane_helper_begin_fb_access, + .end_fb_access = ssd130x_primary_plane_helper_end_fb_access, .atomic_check = drm_plane_helper_atomic_check, .atomic_update = ssd130x_primary_plane_helper_atomic_update, .atomic_disable = ssd130x_primary_plane_helper_atomic_disable,
@@ -719,10 +744,6 @@ static void ssd130x_encoder_helper_atomic_enable(struct drm_encoder *encoder,if (ret) goto power_off; - ret =ssd130x_buf_alloc (ssd130x); -if (ret) - goto power_off; -ssd130x_write_cmd (ssd130x,1 ,SSD130X_DISPLAY_ON );backlight_enable (ssd130x->bl_dev);
@@ -744,8 +765,6 @@ static void ssd130x_encoder_helper_atomic_disable(struct drm_encoder *encoder,ssd130x_write_cmd (ssd130x,1 ,SSD130X_DISPLAY_OFF ); -ssd130x_buf_free (ssd130x); -ssd130x_power_off (ssd130x); }
Signature · 1 line
2.41.0
Hi,
Quoted 30 lines · Javier Martinez Canillas
Geert reports that the following NULL pointer dereference happens for him after commit
49d7d581ceaf("drm/ssd130x: Don't allocate buffers on each plane update"):[drm] Initialized ssd130x 1.0.0 20220131 for 0-003c on minor 0 ssd130x-i2c 0-003c: [drm] surface width(128), height(32), bpp(1) and format(R1 little-endian (0x20203152)) Unable to handle kernel NULL pointer dereference at virtual address 00000000 Oops [#1] CPU: 0 PID: 1 Comm: swapper Not tainted 6.5.0-rc1-orangecrab-02219-g0a529a1e4bf4 #565 epc : ssd130x_update_rect.isra.0+0x13c/0x340 ra : ssd130x_update_rect.isra.0+0x2bc/0x340 ... status: 00000120 badaddr: 00000000 cause: 0000000f [<c0303d90>] ssd130x_update_rect.isra.0+0x13c/0x340 [<c0304200>] ssd130x_primary_plane_helper_atomic_update+0x26c/0x284 [<c02f8d54>] drm_atomic_helper_commit_planes+0xfc/0x27c [<c02f9314>] drm_atomic_helper_commit_tail+0x5c/0xb4 [<c02f94fc>] commit_tail+0x190/0x1b8 [<c02f99fc>] drm_atomic_helper_commit+0x194/0x1c0 [<c02c5d00>] drm_atomic_commit+0xa4/0xe4 [<c02cce40>] drm_client_modeset_commit_atomic+0x244/0x278 [<c02ccef0>] drm_client_modeset_commit_locked+0x7c/0x1bc [<c02cd064>] drm_client_modeset_commit+0x34/0x64 [<c0301a78>] __drm_fb_helper_restore_fbdev_mode_unlocked+0xc4/0xe8 [<c0303424>] drm_fb_helper_set_par+0x38/0x58 [<c027c410>] fbcon_init+0x294/0x534 ...
Quoted 7 lines
The problem is that fbcon calls
fbcon_init()which triggers a DRM modeset and this leads todrm_atomic_helper_commit_planes()attempting to commit the atomic state for all planes, even the ones whose CRTC is not enabled.Since the primary plane buffer is allocated in the encoder .atomic_enable callback, this happens after that initial modeset commit and leads to the mentioned NULL pointer dereference.
I think that's where the problem lies: you must not allocate a buffer in atomic_enable.
After drm_atomic_helper_swap_state(), the new commit is being applied and you aren't allowed to fail, and an allocation can fail.
Everything needs to be prepared by the time _swap_state is called, and it's one of the point of atomic_check.
So you need to allocate your buffer there, and use it in whatever atomic_commit related hook you need it in.
The typical way of doing this would be to create a custom state structure that embeds the global one, create your own reset, atomic_duplicate_state and atomic_destroy_state implementations, and store the buffer pointer there.
Maxime
Javier Martinez Canillas
maintainerMaxime Ripard <[email protected]> writes:
Hello Maxime,
Thanks a lot for your feedback.
Quoted 50 lines
Hi,
Quoted 30 lines · Javier Martinez Canillas
Geert reports that the following NULL pointer dereference happens for him after commit
49d7d581ceaf("drm/ssd130x: Don't allocate buffers on each plane update"):[drm] Initialized ssd130x 1.0.0 20220131 for 0-003c on minor 0 ssd130x-i2c 0-003c: [drm] surface width(128), height(32), bpp(1) and format(R1 little-endian (0x20203152)) Unable to handle kernel NULL pointer dereference at virtual address 00000000 Oops [#1] CPU: 0 PID: 1 Comm: swapper Not tainted 6.5.0-rc1-orangecrab-02219-g0a529a1e4bf4 #565 epc : ssd130x_update_rect.isra.0+0x13c/0x340 ra : ssd130x_update_rect.isra.0+0x2bc/0x340 ... status: 00000120 badaddr: 00000000 cause: 0000000f [<c0303d90>] ssd130x_update_rect.isra.0+0x13c/0x340 [<c0304200>] ssd130x_primary_plane_helper_atomic_update+0x26c/0x284 [<c02f8d54>] drm_atomic_helper_commit_planes+0xfc/0x27c [<c02f9314>] drm_atomic_helper_commit_tail+0x5c/0xb4 [<c02f94fc>] commit_tail+0x190/0x1b8 [<c02f99fc>] drm_atomic_helper_commit+0x194/0x1c0 [<c02c5d00>] drm_atomic_commit+0xa4/0xe4 [<c02cce40>] drm_client_modeset_commit_atomic+0x244/0x278 [<c02ccef0>] drm_client_modeset_commit_locked+0x7c/0x1bc [<c02cd064>] drm_client_modeset_commit+0x34/0x64 [<c0301a78>] __drm_fb_helper_restore_fbdev_mode_unlocked+0xc4/0xe8 [<c0303424>] drm_fb_helper_set_par+0x38/0x58 [<c027c410>] fbcon_init+0x294/0x534 ...Quoted 7 lines
The problem is that fbcon calls
fbcon_init()which triggers a DRM modeset and this leads todrm_atomic_helper_commit_planes()attempting to commit the atomic state for all planes, even the ones whose CRTC is not enabled.Since the primary plane buffer is allocated in the encoder .atomic_enable callback, this happens after that initial modeset commit and leads to the mentioned NULL pointer dereference.
I think that's where the problem lies: you must not allocate a buffer in
atomic_enable.After
drm_atomic_helper_swap_state(), the new commit is being applied and you aren't allowed to fail, and an allocation can fail.Everything needs to be prepared by the time
_swap_stateis called, and it's one of the point ofatomic_check.
Quoted 8 lines
So you need to allocate your buffer there, and use it in whatever
atomic_commitrelated hook you need it in.The typical way of doing this would be to create a custom state structure that embeds the global one, create your own reset,
atomic_duplicate_stateandatomic_destroy_stateimplementations, and store the buffer pointer there.
I see. That makes totally sense to me. I'll do that in v4 then!
Quoted 1 line
Maxime
Signature · 5 lines
Best regards, Javier Martinez Canillas Core Platforms Red Hat