linux-stable.git/drivers/video/fbdev/amba-clcd.c, branch linux-5.4.y

video: fbdev: amba-clcd: Fix refcount leak bugs

2022-08-25T09:18:06+00:00

[ Upstream commit 26c2b7d9fac42eb8317f3ceefa4c1a9a9170ca69 ]

In clcdfb_of_init_display(), we should call of_node_put() for the
references returned by of_graph_get_next_endpoint() and
of_graph_get_remote_port_parent() which have increased the refcount.

Besides, we should call of_node_put() both in fail path or when
the references are not used anymore.

Fixes: d10715be03bd ("video: ARM CLCD: Add DT support")
Signed-off-by: Liang He 
Signed-off-by: Helge Deller 
Signed-off-by: Sasha Levin

video: fbdev: clcdfb: Fix refcount leak in clcdfb_of_vram_setup

2022-06-14T16:11:43+00:00

[ Upstream commit b23789a59fa6f00e98a319291819f91fbba0deb8 ]

of_parse_phandle() returns a node pointer with refcount incremented, we should
use of_node_put() on it when not need anymore.  Add missing of_node_put() to
avoid refcount leak.

Fixes: d10715be03bd ("video: ARM CLCD: Add DT support")
Signed-off-by: Miaoqian Lin 
Signed-off-by: Helge Deller 
Signed-off-by: Sasha Levin

video: amba-clcd: Spout an error if of_get_display_timing() gives an error

2019-07-26T14:32:28+00:00

In the patch ("video: of: display_timing: Don't yell if no timing node
is present") we'll stop spouting an error directly in
of_get_display_timing() if no node is present.  Presumably amba-clcd
should take charge of spouting its own error now.

NOTE: we'll print two errors if the node was present but there were
problems parsing the timing node (one in of_parse_display_timing() and
this new one).  Since this is a fatal error for the driver's probe
(and presumably someone will be debugging), this should be OK.

Signed-off-by: Douglas Anderson 
Reviewed-by: Sam Ravnborg 
Acked-by: Linus Walleij 
Cc: Thierry Reding 
Cc: David Airlie 
Cc: Philipp Zabel 
Cc: Tomi Valkeinen 
Cc: Laurent Pinchart 
Cc: Russell King 
Cc: Daniel Vetter 
Signed-off-by: Bartlomiej Zolnierkiewicz 
Link: https://patchwork.freedesktop.org/patch/msgid/20190722182439.44844-5-dianders@chromium.org

video: amba-clcd: Decomission Versatile and Nomadik

2019-04-11T17:25:12+00:00

These board families are now handled in the DRM subsystem
where we can have reusable panel drivers and some other
stuff. The PL111 there is now the driver used in the
defconfig for Versatile and Nomadik so no need to keep
this code around.

There are a few minor machines in arch/arm/ such as
mach-netx still using the old driver, so we need to keep
the core fbdev driver around for some time.

Signed-off-by: Linus Walleij 
Cc: Russell King 
Signed-off-by: Bartlomiej Zolnierkiewicz

video: ARM CLCD: Improve a size determination in clcdfb_probe()

2018-03-28T14:34:29+00:00

Replace the specification of a data structure by a pointer dereference
as the parameter for the operator "sizeof" to make the corresponding size
determination a bit safer according to the Linux coding style convention.

This issue was detected by using the Coccinelle software.

Signed-off-by: Markus Elfring 
Cc: Russell King 
Signed-off-by: Bartlomiej Zolnierkiewicz

video: ARM CLCD: Delete an error message for a failed memory allocation in clcdfb_probe()

2018-03-28T14:34:29+00:00

Omit an extra message for a memory allocation failure in this function.

This issue was detected by using the Coccinelle software.

Signed-off-by: Markus Elfring 
Cc: Russell King 
Signed-off-by: Bartlomiej Zolnierkiewicz

video: ARM CLCD: constify amba_id

2017-09-04T14:00:49+00:00

amba_id are not supposed to change at runtime. All functions
working with const amba_id. So mark the non-const structs as const.

Signed-off-by: Arvind Yadav 
Cc: Russell King 
Signed-off-by: Bartlomiej Zolnierkiewicz

video: ARM CLCD: fix dma allocation size

2017-04-07T15:03:24+00:00

This patch forces the frambuffer size to be aligned on kernel pages.

During the board startup, the splash screed did appear;
the "ts_test" program or our application were not able to start.

The following error message was reported:
error: failed to map framebuffer device to memory.
LinuxFB: driver cannot connect

The issue was discovered, on the LPC32xx platform, during the migration
of the LCD definition from the board file to the device tree.

Signed-off-by: Liam Beguin 
Signed-off-by: Sylvain Lemieux 
Cc: Vladimir Zapolskiy 
Cc: Russell King 
Signed-off-by: Bartlomiej Zolnierkiewicz

video: ARM CLCD: use panel device node for getting backlight and mode

2017-01-30T16:39:48+00:00

Having obtained panel device node in clcdfb_of_init_display() it allows to
generalize and simplify two more helper functions clcdfb_of_get_backlight()
and clcdfb_of_get_mode().

Cc: Tomi Valkeinen 
Cc: Russell King 
Cc: Linus Walleij 
Cc: Arnd Bergmann 
Signed-off-by: Vladimir Zapolskiy 
Signed-off-by: Bartlomiej Zolnierkiewicz

video: ARM CLCD: use panel device node for panel initialization

2017-01-30T16:39:48+00:00

There is no necessity to pass an endpoint device node to custom panel
initialization functions, because a child panel device node should be
sufficient, note that the existing init_panel() callback declaration from
linux/amba/clcd.h already prompts to use the panel device node here.

Cc: Tomi Valkeinen 
Cc: Russell King 
Cc: Linus Walleij 
Cc: Arnd Bergmann 
Signed-off-by: Vladimir Zapolskiy 
Signed-off-by: Bartlomiej Zolnierkiewicz