Re: [PATCH v3 RESEND] media: vimc: fla: Add virtual flash subdevice
From: Dafna Hirschfeld
Date: Thu Dec 05 2019 - 13:01:46 EST
Hi
On 12/5/19 12:12 AM, Lucas MagalhÃes wrote:
Hi Dafna,
Thanks for the review.
On Wed, Dec 4, 2019 at 7:03 AM Dafna Hirschfeld
<dafna.hirschfeld@xxxxxxxxxxxxx> wrote:
Hi,
Thanks for the patch
On 12/4/19 12:01 AM, Lucas A. M. MagalhÃes wrote:
From: Lucas A. M. Magalhaes <lucmaga@xxxxxxxxx>
Add a virtual subdevice to simulate the flash control API.
Those are the supported controls:
v4l2-ctl -d /dev/v4l-subdev6 -L
Flash Controls
led_mode 0x009c0901 (menu) : min=0 max=2 default=1 value=1
0: Off
1: Flash
2: Torch
strobe_source 0x009c0902 (menu) : min=0 max=1 default=0 value=0
0: Software
1: External
strobe 0x009c0903 (button) : flags=write-only, execute-on-write
stop_strobe 0x009c0904 (button) : flags=write-only, execute-on-write
strobe_status 0x009c0905 (bool) : default=0 value=0 flags=read-only
strobe_timeout 0x009c0906 (int) : min=50 max=400 step=50 default=50 value=400
intensity_flash_mode 0x009c0907 (int) : min=23040 max=1499600 step=11718 default=23040 value=23040
intensity_torch_mode 0x009c0908 (int) : min=2530 max=187100 step=1460 default=2530 value=2530
intensity_indicator 0x009c0909 (int) : min=0 max=255 step=1 default=0 value=0
faults 0x009c090a (bitmask): max=0x00000002 default=0x00000000 value=0x00000000
Co-authored-by: Eduardo Barretto <edusbarretto@xxxxxxxxx>
Signed-off-by: Eduardo Barretto <edusbarretto@xxxxxxxxx>
Signed-off-by: Lucas A. M. MagalhÃes <lucmaga@xxxxxxxxx>
---
Hi,
I've copied some values from another driver (lm3646) to make it more
realistic, as suggested by Hans. All values except for
V4L2_CID_FLASH_INDICATOR_INTENSITY, which I couldn't find any
implementation.
The v4l-compliance is failing. From the documentation
V4L2_CID_FLASH_STROBE should just work if the
V4L2_CID_FLASH_STROBE_SOURCE is "Software" and the
V4L2_CID_FLASH_LED_MODE is "Flash", otherwise it should fail. With the
standard values configured for the V4L2_CID_FLASH_STROBE will not fail.
But during the tests v4l-compliance sets V4L2_CID_FLASH_LED_MODE to
"Torch" and V4L2_CID_FLASH_STROBE_SOURCE to "External" which makes
V4L2_CID_FLASH_STROBE to fail. How do I proceed? Should the
v4l-compliance be changed?
Changes in v3:
- Fix style errors
- Use more realistic numbers for the controllers
- Change from kthread to workqueue
- Change commit message for the new controllers values
Changes in v2:
- Fix v4l2-complience errors
- Add V4L2_CID_FLASH_STROBE_STATUS behavior
- Add V4L2_CID_FLASH_STROBE restrictions
- Remove vimc_fla_g_volatile_ctrl
- Remove unnecessarie V4L2_CID_FLASH_CLASS
- Change varables names
drivers/media/platform/vimc/Makefile | 2 +-
drivers/media/platform/vimc/vimc-common.c | 2 +
drivers/media/platform/vimc/vimc-common.h | 4 +
drivers/media/platform/vimc/vimc-core.c | 5 +
drivers/media/platform/vimc/vimc-flash.c | 248 ++++++++++++++++++++++
5 files changed, 260 insertions(+), 1 deletion(-)
create mode 100644 drivers/media/platform/vimc/vimc-flash.c
diff --git a/drivers/media/platform/vimc/Makefile b/drivers/media/platform/vimc/Makefile
index a53b2b532e9f..e759bbb04b14 100644
--- a/drivers/media/platform/vimc/Makefile
+++ b/drivers/media/platform/vimc/Makefile
@@ -1,6 +1,6 @@
# SPDX-License-Identifier: GPL-2.0
vimc-y := vimc-core.o vimc-common.o vimc-streamer.o vimc-capture.o \
- vimc-debayer.o vimc-scaler.o vimc-sensor.o
+ vimc-debayer.o vimc-scaler.o vimc-sensor.o vimc-flash.o
obj-$(CONFIG_VIDEO_VIMC) += vimc.o
diff --git a/drivers/media/platform/vimc/vimc-common.c b/drivers/media/platform/vimc/vimc-common.c
index a3120f4f7a90..cb786de75573 100644
--- a/drivers/media/platform/vimc/vimc-common.c
+++ b/drivers/media/platform/vimc/vimc-common.c
@@ -203,6 +203,8 @@ struct media_pad *vimc_pads_init(u16 num_pads, const unsigned long *pads_flag)
struct media_pad *pads;
unsigned int i;
+ if (!num_pads)
+ return NULL;
Please rebase on top of latest master,
this function was removed in patch 'media: vimc: embed the pads of entities in the entities' structs'
Ok.
/* Allocate memory for the pads */
pads = kcalloc(num_pads, sizeof(*pads), GFP_KERNEL);
if (!pads)
diff --git a/drivers/media/platform/vimc/vimc-common.h b/drivers/media/platform/vimc/vimc-common.h
index 698db7c07645..19815f0f4d40 100644
--- a/drivers/media/platform/vimc/vimc-common.h
+++ b/drivers/media/platform/vimc/vimc-common.h
@@ -169,6 +169,10 @@ struct vimc_ent_device *vimc_sen_add(struct vimc_device *vimc,
const char *vcfg_name);
void vimc_sen_rm(struct vimc_device *vimc, struct vimc_ent_device *ved);
+struct vimc_ent_device *vimc_fla_add(struct vimc_device *vimc,
+ const char *vcfg_name);
This should be lined with the opening '('
That's strange. I'm sure it's not like this here. Maybe my smtp is alterating
my code for some reason. I will look this.
Hi, apparently it is my email client that does not show the alignment correctly
sorry for the false comment
+void vimc_fla_rm(struct vimc_device *vimc, struct vimc_ent_device *ved);
+
/**
* vimc_pads_init - initialize pads
*
diff --git a/drivers/media/platform/vimc/vimc-core.c b/drivers/media/platform/vimc/vimc-core.c
index 6e3e5c91ae39..5f6c750d3d8d 100644
--- a/drivers/media/platform/vimc/vimc-core.c
+++ b/drivers/media/platform/vimc/vimc-core.c
@@ -91,6 +91,11 @@ static struct vimc_ent_config ent_config[] = {
.add = vimc_cap_add,
.rm = vimc_cap_rm,
},
+ {
+ .name = "Flash Controller",
+ .add = vimc_fla_add,
+ .rm = vimc_fla_rm,
+ }
};
static const struct vimc_ent_link ent_links[] = {
diff --git a/drivers/media/platform/vimc/vimc-flash.c b/drivers/media/platform/vimc/vimc-flash.c
new file mode 100644
index 000000000000..3918beecec57
--- /dev/null
+++ b/drivers/media/platform/vimc/vimc-flash.c
@@ -0,0 +1,248 @@
+// SPDX-License-Identifier: GPL-2.0+
+/*
+ * vimc-flash.c Virtual Media Controller Driver
+ *
+ * Copyright (C) 2019
+ * Contributors: Lucas A. M. MagalhÃes <lamm@xxxxxxxxxxx>
+ * Eduardo Barretto <edusbarretto@xxxxxxxxx>
+ *
+ */
+
+#include <linux/delay.h>
+#include <linux/workqueue.h>
+#include <linux/sched.h>
+#include <linux/vmalloc.h>
+#include <media/v4l2-ctrls.h>
+#include <media/v4l2-event.h>
+#include <media/v4l2-subdev.h>
+
+#include "vimc-common.h"
+
+/*
+ * Flash timeout in ms
+ */
+#define VIMC_FLASH_TIMEOUT_MS_MIN 50
+#define VIMC_FLASH_TIMEOUT_MS_MAX 400
+#define VIMC_FLASH_TIMEOUT_MS_STEP 50
+
+/*
+ * Torch intencity in uA
+ */
+#define VIMC_FLASH_TORCH_UA_MIN 2530
+#define VIMC_FLASH_TORCH_UA_MAX 187100
+#define VIMC_FLASH_TORCH_UA_STEP 1460
+
+/*
+ * Flash intencity in uA
+ */
+#define VIMC_FLASH_FLASH_UA_MIN 23040
+#define VIMC_FLASH_FLASH_UA_MAX 1499600
+#define VIMC_FLASH_FLASH_UA_STEP 11718
+
+struct vimc_fla_device {
+ struct vimc_ent_device ved;
+ struct v4l2_subdev sd;
+ struct v4l2_ctrl_handler hdl;
+ int strobe_source;
+ bool is_strobe;
+ int led_mode;
+ int indicator_intensity;
+ int torch_intensity;
+ int flash_intensity;
+ u64 timeout;
+ u64 last_strobe;
+ struct workqueue_struct *wq;
+ struct work_struct work;
+ struct v4l2_ctrl *strobe_status_ctl;
+};
+
+static void vimc_fla_strobe_work(struct work_struct *work)
+{
+ struct vimc_fla_device *vfla =
+ container_of(work, struct vimc_fla_device, work);
+ v4l2_ctrl_s_ctrl(vfla->strobe_status_ctl, true);
+ vfla->last_strobe = ktime_get_ns();
+ while (vfla->is_strobe &&
+ vfla->last_strobe + vfla->timeout > ktime_get_ns()) {
+ msleep_interruptible(VIMC_FLASH_TIMEOUT_MS_STEP);
+ }
+ v4l2_ctrl_s_ctrl(vfla->strobe_status_ctl, false);
+}
+
+static int vimc_fla_s_ctrl(struct v4l2_ctrl *c)
+{
+ struct vimc_fla_device *vfla =
+ container_of(c->handler, struct vimc_fla_device, hdl);
+
+ switch (c->id) {
+ case V4L2_CID_FLASH_LED_MODE:
+ vfla->led_mode = c->val;
+ return 0;
+ case V4L2_CID_FLASH_STROBE_SOURCE:
+ vfla->strobe_source = c->val;
+ return 0;
+ case V4L2_CID_FLASH_STROBE:
+ if (vfla->led_mode != V4L2_FLASH_LED_MODE_FLASH ||
+ vfla->strobe_source != V4L2_FLASH_STROBE_SOURCE_SOFTWARE){
you can add error/warning debug here,
if you choose not to, then the parentheses are redundant.
Additionally, the opening '{' should come after a space
You can run srctipts/checkpatch.pl before submitting to catch such issues
+ return -EINVAL;
+ }
+ queue_work(vfla->wq, &vfla->work);
+ return 0;
+ case V4L2_CID_FLASH_STROBE_STATUS:
+ vfla->is_strobe = c->val;
+ return 0;
+ case V4L2_CID_FLASH_STROBE_STOP:
+ vfla->is_strobe = false;
+ return 0;
+ case V4L2_CID_FLASH_TIMEOUT:
+ vfla->timeout = c->val * 1000000; /* MS to NS */
+ return 0;
+ case V4L2_CID_FLASH_INTENSITY:
+ vfla->flash_intensity = c->val;
+ return 0;
+ case V4L2_CID_FLASH_TORCH_INTENSITY:
+ vfla->torch_intensity = c->val;
+ return 0;
+ case V4L2_CID_FLASH_INDICATOR_INTENSITY:
+ vfla->indicator_intensity = c->val;
+ return 0;
+ default:
+ return -EINVAL;
+ }
+ return 0;
+}
+
+static const struct v4l2_ctrl_ops vimc_fla_ctrl_ops = {
+ .s_ctrl = vimc_fla_s_ctrl,
+};
+
+static const struct v4l2_subdev_core_ops vimc_fla_core_ops = {
+ .subscribe_event = v4l2_ctrl_subdev_subscribe_event,
+ .unsubscribe_event = v4l2_event_subdev_unsubscribe,
+};
+
+static const struct v4l2_subdev_ops vimc_fla_ops = {
+ .core = &vimc_fla_core_ops,
+};
+
+static void vimc_fla_release(struct v4l2_subdev *sd)
+{
+ struct vimc_fla_device *vfla =
+ container_of(sd, struct vimc_fla_device, sd);
one tab is enough
+
+ v4l2_ctrl_handler_free(&vfla->hdl);
+ kfree(vfla);
+}
+
+static const struct v4l2_subdev_internal_ops vimc_fla_int_ops = {
+ .release = vimc_fla_release,
+};
+
+/* initialize device */
+struct vimc_ent_device *vimc_fla_add(struct vimc_device *vimc,
+ const char *vcfg_name)
+{
+ struct v4l2_device *v4l2_dev = &vimc->v4l2_dev;
+ struct vimc_fla_device *vfla;
+ int ret;
+
+ /* Allocate the vfla struct */
+ vfla = kzalloc(sizeof(*vfla), GFP_KERNEL);
+ if (!vfla)
+ return NULL;
I think it is better to change the return values of the .add calbbacks
to return ERR_PTR upon error and not just NULL so that different types of
errors can be distinguished.
(This is not related specifically to this patch),
If you want you can send a patch fixing that.
Sure. I will change it.
cool, thanks
+
+ v4l2_ctrl_handler_init(&vfla->hdl, 4);
+ v4l2_ctrl_new_std_menu(&vfla->hdl, &vimc_fla_ctrl_ops,
+ V4L2_CID_FLASH_LED_MODE,
+ V4L2_FLASH_LED_MODE_TORCH, ~0x7,
+ V4L2_FLASH_LED_MODE_FLASH);
+ v4l2_ctrl_new_std_menu(&vfla->hdl, &vimc_fla_ctrl_ops,
+ V4L2_CID_FLASH_STROBE_SOURCE, 0x1, ~0x3,
+ V4L2_FLASH_STROBE_SOURCE_SOFTWARE);
+ v4l2_ctrl_new_std(&vfla->hdl, &vimc_fla_ctrl_ops,
+ V4L2_CID_FLASH_STROBE, 0, 0, 0, 0);
+ v4l2_ctrl_new_std(&vfla->hdl, &vimc_fla_ctrl_ops,
+ V4L2_CID_FLASH_STROBE_STOP, 0, 0, 0, 0);
+ v4l2_ctrl_new_std(&vfla->hdl, &vimc_fla_ctrl_ops,
+ V4L2_CID_FLASH_TIMEOUT, VIMC_FLASH_TIMEOUT_MS_MIN,
+ VIMC_FLASH_TIMEOUT_MS_MAX,
+ VIMC_FLASH_TIMEOUT_MS_STEP,
+ VIMC_FLASH_TIMEOUT_MS_MIN);
+ v4l2_ctrl_new_std(&vfla->hdl, &vimc_fla_ctrl_ops,
+ V4L2_CID_FLASH_TORCH_INTENSITY,
+ VIMC_FLASH_TORCH_UA_MIN,
+ VIMC_FLASH_TORCH_UA_MAX,
+ VIMC_FLASH_TORCH_UA_STEP,
+ VIMC_FLASH_TORCH_UA_MIN);
+ v4l2_ctrl_new_std(&vfla->hdl, &vimc_fla_ctrl_ops,
+ V4L2_CID_FLASH_INTENSITY,
+ VIMC_FLASH_FLASH_UA_MIN,
+ VIMC_FLASH_FLASH_UA_MAX,
+ VIMC_FLASH_FLASH_UA_STEP,
+ VIMC_FLASH_FLASH_UA_MIN);
+ v4l2_ctrl_new_std(&vfla->hdl, &vimc_fla_ctrl_ops,
+ V4L2_CID_FLASH_INDICATOR_INTENSITY,
+ 0,
+ 255,
+ 1,
+ 0);
why not having one line with "0,255,1,0);" ?
+ v4l2_ctrl_new_std(&vfla->hdl, &vimc_fla_ctrl_ops,
+ V4L2_CID_FLASH_STROBE_STATUS, 0, 1, 1, 0);
+ v4l2_ctrl_new_std(&vfla->hdl, &vimc_fla_ctrl_ops,
+ V4L2_CID_FLASH_FAULT, 0,
+ V4L2_FLASH_FAULT_TIMEOUT, 0, 0);
+ vfla->sd.ctrl_handler = &vfla->hdl;
+ if (vfla->hdl.error) {
+ ret = vfla->hdl.error;
+ goto err_free_vfla;
+ }
+ vfla->strobe_status_ctl = v4l2_ctrl_find(&vfla->hdl,
+ V4L2_CID_FLASH_STROBE_STATUS);
+
+ /* Initialize ved and sd */
+ ret = vimc_ent_sd_register(&vfla->ved, &vfla->sd, v4l2_dev,
+ vcfg_name,
+ MEDIA_ENT_F_FLASH, 0, NULL,
+ &vimc_fla_int_ops, &vimc_fla_ops);
+ if (ret)
+ goto err_free_hdl;
+
+ /* Create processing workqueue */
+ vfla->wq = alloc_workqueue("%s", 0, 0, "vimc-flash thread");
+ if (!vfla->wq)
+ goto err_unregister;
+
+ INIT_WORK(&vfla->work, vimc_fla_strobe_work);
+ /* Initialize standard values */
+ vfla->indicator_intensity = 0;
+ vfla->torch_intensity = 0;
+ vfla->flash_intensity = 0;
+ vfla->is_strobe = false;
+ vfla->timeout = 0;
+ vfla->last_strobe = 0;
+ vfla->strobe_source = V4L2_FLASH_STROBE_SOURCE_SOFTWARE;
+ vfla->led_mode = V4L2_FLASH_LED_MODE_FLASH;
+
+ return &vfla->ved;
+
+err_unregister:
+ vimc_ent_sd_unregister(&vfla->ved, &vfla->sd);
+err_free_hdl:
+ v4l2_ctrl_handler_free(&vfla->hdl);
+err_free_vfla:
+ kfree(vfla);
+
+ return NULL;
+}
+
+void vimc_fla_rm(struct vimc_device *vimc, struct vimc_ent_device *ved)
+{
+ struct vimc_fla_device *vfla;
+
+ if (!ved)
+ return;
+
+ vfla = container_of(ved, struct vimc_fla_device, ved);
+ destroy_workqueue(vfla->wq);
+ vimc_ent_sd_unregister(ved, &vfla->sd);
+}
Not sure but I think there are indentation issues in this patch, a tab should be 8 spaces
This is strange. I already had style issues on the v2 so I runned the
checkpatch.pl on the
patch before sent and it didn't point this identation issues.
Here is the checkpatch.pl output. It could be the case that my smtp is
alterating the
it before transmission.
Sorry, as I wrote, this is my thunderbird client
Dafna
$ scripts/checkpatch.pl
patchset/0001-media-vimc-fla-Add-virtual-flash-subdevice.patch
WARNING: Possible unwrapped commit description (prefer a maximum 75
chars per line)
#14:
led_mode 0x009c0901 (menu) : min=0 max=2
default=1 value=1
WARNING: Non-standard signature: Co-authored-by:
#30:
Co-authored-by: Eduardo Barretto <edusbarretto@xxxxxxxxx>
WARNING: added, moved or deleted file(s), does MAINTAINERS need updating?
#131:
new file mode 100644
total: 0 errors, 3 warnings, 284 lines checked
NOTE: For some of the reported defects, checkpatch may be able to
mechanically convert to the typical style using --fix or --fix-inplace.
patchset/0001-media-vimc-fla-Add-virtual-flash-subdevice.patch has
style problems, please review.
NOTE: If any of the errors are false positives, please report
them to the maintainer, see CHECKPATCH in MAINTAINERS.
Thanks,
Lucas