All of lore.kernel.org
 help / color / mirror / Atom feed
From: zourongrong@huawei.com (Rongrong Zou)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH v6 7/9] drm/hisilicon/hibmc: Add connector for VDAC
Date: Mon, 14 Nov 2016 22:07:30 +0800	[thread overview]
Message-ID: <5829C522.1040908@huawei.com> (raw)
In-Reply-To: <CAOw6vbJA3Oe3GmRZBZjF24wtu_5hyTdVL2HK-1p3BLbiY1G7Tw@mail.gmail.com>

? 2016/11/11 9:45, Sean Paul ??:
> On Fri, Oct 28, 2016 at 3:28 AM, Rongrong Zou <zourongrong@gmail.com> wrote:
>> Add connector funcs and helper funcs for VDAC.
>>
>> Signed-off-by: Rongrong Zou <zourongrong@gmail.com>
>> ---
>>   drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c  |  8 +++
>>   drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.h  |  2 +
>>   drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c | 76 ++++++++++++++++++++++++
>>   3 files changed, 86 insertions(+)
>>
>> diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c
>> index ba191e1..4253603 100644
>> --- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c
>> +++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c
>> @@ -131,6 +131,14 @@ static int hibmc_kms_init(struct hibmc_drm_device *hidev)
>>                  return ret;
>>          }
>>
>> +       ret = hibmc_connector_init(hidev);
>> +       if (ret) {
>> +               DRM_ERROR("failed to init connector\n");
>> +               return ret;
>> +       }
>> +
>> +       drm_mode_connector_attach_encoder(&hidev->connector,
>> +                                         &hidev->encoder);
>
> The connector should be initialized in the vdac driver with the
> encoder, not in the drv file (same as plane/crtc)

applied, thanks.

>
>>          return 0;
>>   }
>>
>> diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.h b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.h
>> index 401cea4..450247d 100644
>> --- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.h
>> +++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.h
>> @@ -48,6 +48,7 @@ struct hibmc_drm_device {
>>          struct drm_plane plane;
>>          struct drm_crtc crtc;
>>          struct drm_encoder encoder;
>> +       struct drm_connector connector;
>
> No need to keep track here

will allocate with devm_kzalloc(), thanks.

>
>>          bool mode_config_initialized;
>>
>>          /* ttm */
>> @@ -89,6 +90,7 @@ static inline struct hibmc_bo *gem_to_hibmc_bo(struct drm_gem_object *gem)
>>   int hibmc_plane_init(struct hibmc_drm_device *hidev);
>>   int hibmc_crtc_init(struct hibmc_drm_device *hidev);
>>   int hibmc_encoder_init(struct hibmc_drm_device *hidev);
>> +int hibmc_connector_init(struct hibmc_drm_device *hidev);
>>   int hibmc_fbdev_init(struct hibmc_drm_device *hidev);
>>   void hibmc_fbdev_fini(struct hibmc_drm_device *hidev);
>>
>> diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c
>> index 953f659..ebefcd1 100644
>> --- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c
>> +++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c
>> @@ -87,3 +87,79 @@ int hibmc_encoder_init(struct hibmc_drm_device *hidev)
>>          drm_encoder_helper_add(encoder, &hibmc_encoder_helper_funcs);
>>          return 0;
>>   }
>> +
>> +static int hibmc_connector_get_modes(struct drm_connector *connector)
>> +{
>> +       int count;
>> +
>> +       count = drm_add_modes_noedid(connector, 800, 600);
>> +       drm_set_preferred_mode(connector, defx, defy);
>
> So you have defx/defy as module parameters, but then hardcode the
> 800x600 mode. If defx/defy is anything other than 800/600, this won't
> work. I think you should just remove the defx/defy module params and
> rely on userspace adding modes as appropriate.

will remove them, thanks.

>
>> +       return count;
>> +}
>> +
>> +static int hibmc_connector_mode_valid(struct drm_connector *connector,
>> +                                     struct drm_display_mode *mode)
>> +{
>> +       struct hibmc_drm_device *hiprivate =
>> +        container_of(connector, struct hibmc_drm_device, connector);
>> +       unsigned long size = mode->hdisplay * mode->vdisplay * 4;
>
> Why * 4 here and why * 2 below? You support formats less than 32 bpp,
> so the * 4 isn't necessarily correct for all formats. Is the * 2 to
> account for front & back buffer?
>

it is from bochs, the original comments below:
	/*
	 * Make sure we can fit two framebuffers into video memory.
	 * This allows up to 1600x1200 with 16 MB (default size).
	 * If you want more try this:
	 *     'qemu -vga std -global VGA.vgamem_mb=32 $otherargs'
	 */
i think it is not necessary for hibmc, so will remove it and return
MODE_OK, thanks.

>
>> +
>> +       if (size * 2 > hiprivate->fb_size)
>> +               return MODE_BAD;
>> +
>> +       return MODE_OK;
>> +}
>> +
>> +static struct drm_encoder *
>> +hibmc_connector_best_encoder(struct drm_connector *connector)
>> +{
>> +       int enc_id = connector->encoder_ids[0];
>> +
>> +       /* pick the encoder ids */
>> +       if (enc_id)
>> +               return drm_encoder_find(connector->dev, enc_id);
>
> Can't you just do return drm_encoder_find(connector->dev,
> connector->encoder_ids[0]); ?
>
> ie: won't drm_encoder_find do the right thing if you pass in id == 0?

hibmc_connector_best_encoder(struct drm_connector *connector)
{
	return drm_encoder_find(connector->dev, connector->encoder_ids[0]);
}
looks more simple, i like it, thanks:)

>
>> +
>> +       return NULL;
>> +}
>> +
>> +static enum drm_connector_status hibmc_connector_detect(struct drm_connector
>> +                                                *connector, bool force)
>> +{
>> +       return connector_status_connected;
>
> Perhaps this should be connector_status_unknown, since you don't
> necessarily know it's connected.

it is always connected in hibmc's scenario, usually it is used as a
management chip on server, we use KVM(keyboard, mouse, video) rather than
a directly connected monitor. I will modify if a phisical monitor
is connected later.

>
>> +}
>> +
>> +static const struct drm_connector_helper_funcs
>> +       hibmc_connector_connector_helper_funcs = {
>> +       .get_modes = hibmc_connector_get_modes,
>> +       .mode_valid = hibmc_connector_mode_valid,
>> +       .best_encoder = hibmc_connector_best_encoder,
>> +};
>> +
>> +static const struct drm_connector_funcs hibmc_connector_connector_funcs = {
>> +       .dpms = drm_atomic_helper_connector_dpms,
>> +       .detect = hibmc_connector_detect,
>> +       .fill_modes = drm_helper_probe_single_connector_modes,
>> +       .destroy = drm_connector_cleanup,
>> +       .reset = drm_atomic_helper_connector_reset,
>> +       .atomic_duplicate_state = drm_atomic_helper_connector_duplicate_state,
>> +       .atomic_destroy_state = drm_atomic_helper_connector_destroy_state,
>> +};
>> +
>> +int hibmc_connector_init(struct hibmc_drm_device *hidev)
>> +{
>> +       struct drm_device *dev = hidev->dev;
>> +       struct drm_connector *connector = &hidev->connector;
>> +       int ret;
>> +
>> +       ret = drm_connector_init(dev, connector,
>> +                                &hibmc_connector_connector_funcs,
>> +                                DRM_MODE_CONNECTOR_VGA);
>> +       if (ret) {
>> +               DRM_ERROR("failed to init connector\n");
>> +               return ret;
>> +       }
>> +       drm_connector_helper_add(connector,
>> +                                &hibmc_connector_connector_helper_funcs);
>> +
>> +       return 0;
>> +}
>> --
>> 1.9.1
>>
>>
>> _______________________________________________
>> linux-arm-kernel mailing list
>> linux-arm-kernel at lists.infradead.org
>> http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
> _______________________________________________
> linuxarm mailing list
> linuxarm at huawei.com
> http://rnd-openeuler.huawei.com/mailman/listinfo/linuxarm
>
> .
>


-- 
Regards, Rongrong

WARNING: multiple messages have this Message-ID (diff)
From: Rongrong Zou <zourongrong@huawei.com>
To: Sean Paul <seanpaul@chromium.org>, Rongrong Zou <zourongrong@gmail.com>
Cc: Mark Rutland <mark.rutland@arm.com>,
	Archit <architt@codeaurora.org>,
	shenhui@huawei.com, Tomeu Vizoso <tomeu.vizoso@collabora.com>,
	Jonathan Corbet <corbet@lwn.net>, Dave Airlie <airlied@linux.ie>,
	catalin.marinas@arm.com, Emil Velikov <emil.l.velikov@gmail.com>,
	linuxarm@huawei.com, dri-devel <dri-devel@lists.freedesktop.org>,
	Xinliang Liu <xinliang.liu@linaro.org>,
	james.xiong@huawei.com, Daniel Stone <daniel@fooishbar.org>,
	Daniel Vetter <daniel@ffwll.ch>,
	Will Deacon <will.deacon@arm.com>,
	lijianhua@huawei.com,
	Linux ARM Kernel <linux-arm-kernel@lists.infradead.org>,
	Benjamin Gaignard <benjamin.gaignard@linaro.org>
Subject: Re: [PATCH v6 7/9] drm/hisilicon/hibmc: Add connector for VDAC
Date: Mon, 14 Nov 2016 22:07:30 +0800	[thread overview]
Message-ID: <5829C522.1040908@huawei.com> (raw)
In-Reply-To: <CAOw6vbJA3Oe3GmRZBZjF24wtu_5hyTdVL2HK-1p3BLbiY1G7Tw@mail.gmail.com>

在 2016/11/11 9:45, Sean Paul 写道:
> On Fri, Oct 28, 2016 at 3:28 AM, Rongrong Zou <zourongrong@gmail.com> wrote:
>> Add connector funcs and helper funcs for VDAC.
>>
>> Signed-off-by: Rongrong Zou <zourongrong@gmail.com>
>> ---
>>   drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c  |  8 +++
>>   drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.h  |  2 +
>>   drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c | 76 ++++++++++++++++++++++++
>>   3 files changed, 86 insertions(+)
>>
>> diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c
>> index ba191e1..4253603 100644
>> --- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c
>> +++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c
>> @@ -131,6 +131,14 @@ static int hibmc_kms_init(struct hibmc_drm_device *hidev)
>>                  return ret;
>>          }
>>
>> +       ret = hibmc_connector_init(hidev);
>> +       if (ret) {
>> +               DRM_ERROR("failed to init connector\n");
>> +               return ret;
>> +       }
>> +
>> +       drm_mode_connector_attach_encoder(&hidev->connector,
>> +                                         &hidev->encoder);
>
> The connector should be initialized in the vdac driver with the
> encoder, not in the drv file (same as plane/crtc)

applied, thanks.

>
>>          return 0;
>>   }
>>
>> diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.h b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.h
>> index 401cea4..450247d 100644
>> --- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.h
>> +++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.h
>> @@ -48,6 +48,7 @@ struct hibmc_drm_device {
>>          struct drm_plane plane;
>>          struct drm_crtc crtc;
>>          struct drm_encoder encoder;
>> +       struct drm_connector connector;
>
> No need to keep track here

will allocate with devm_kzalloc(), thanks.

>
>>          bool mode_config_initialized;
>>
>>          /* ttm */
>> @@ -89,6 +90,7 @@ static inline struct hibmc_bo *gem_to_hibmc_bo(struct drm_gem_object *gem)
>>   int hibmc_plane_init(struct hibmc_drm_device *hidev);
>>   int hibmc_crtc_init(struct hibmc_drm_device *hidev);
>>   int hibmc_encoder_init(struct hibmc_drm_device *hidev);
>> +int hibmc_connector_init(struct hibmc_drm_device *hidev);
>>   int hibmc_fbdev_init(struct hibmc_drm_device *hidev);
>>   void hibmc_fbdev_fini(struct hibmc_drm_device *hidev);
>>
>> diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c
>> index 953f659..ebefcd1 100644
>> --- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c
>> +++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c
>> @@ -87,3 +87,79 @@ int hibmc_encoder_init(struct hibmc_drm_device *hidev)
>>          drm_encoder_helper_add(encoder, &hibmc_encoder_helper_funcs);
>>          return 0;
>>   }
>> +
>> +static int hibmc_connector_get_modes(struct drm_connector *connector)
>> +{
>> +       int count;
>> +
>> +       count = drm_add_modes_noedid(connector, 800, 600);
>> +       drm_set_preferred_mode(connector, defx, defy);
>
> So you have defx/defy as module parameters, but then hardcode the
> 800x600 mode. If defx/defy is anything other than 800/600, this won't
> work. I think you should just remove the defx/defy module params and
> rely on userspace adding modes as appropriate.

will remove them, thanks.

>
>> +       return count;
>> +}
>> +
>> +static int hibmc_connector_mode_valid(struct drm_connector *connector,
>> +                                     struct drm_display_mode *mode)
>> +{
>> +       struct hibmc_drm_device *hiprivate =
>> +        container_of(connector, struct hibmc_drm_device, connector);
>> +       unsigned long size = mode->hdisplay * mode->vdisplay * 4;
>
> Why * 4 here and why * 2 below? You support formats less than 32 bpp,
> so the * 4 isn't necessarily correct for all formats. Is the * 2 to
> account for front & back buffer?
>

it is from bochs, the original comments below:
	/*
	 * Make sure we can fit two framebuffers into video memory.
	 * This allows up to 1600x1200 with 16 MB (default size).
	 * If you want more try this:
	 *     'qemu -vga std -global VGA.vgamem_mb=32 $otherargs'
	 */
i think it is not necessary for hibmc, so will remove it and return
MODE_OK, thanks.

>
>> +
>> +       if (size * 2 > hiprivate->fb_size)
>> +               return MODE_BAD;
>> +
>> +       return MODE_OK;
>> +}
>> +
>> +static struct drm_encoder *
>> +hibmc_connector_best_encoder(struct drm_connector *connector)
>> +{
>> +       int enc_id = connector->encoder_ids[0];
>> +
>> +       /* pick the encoder ids */
>> +       if (enc_id)
>> +               return drm_encoder_find(connector->dev, enc_id);
>
> Can't you just do return drm_encoder_find(connector->dev,
> connector->encoder_ids[0]); ?
>
> ie: won't drm_encoder_find do the right thing if you pass in id == 0?

hibmc_connector_best_encoder(struct drm_connector *connector)
{
	return drm_encoder_find(connector->dev, connector->encoder_ids[0]);
}
looks more simple, i like it, thanks:)

>
>> +
>> +       return NULL;
>> +}
>> +
>> +static enum drm_connector_status hibmc_connector_detect(struct drm_connector
>> +                                                *connector, bool force)
>> +{
>> +       return connector_status_connected;
>
> Perhaps this should be connector_status_unknown, since you don't
> necessarily know it's connected.

it is always connected in hibmc's scenario, usually it is used as a
management chip on server, we use KVM(keyboard, mouse, video) rather than
a directly connected monitor. I will modify if a phisical monitor
is connected later.

>
>> +}
>> +
>> +static const struct drm_connector_helper_funcs
>> +       hibmc_connector_connector_helper_funcs = {
>> +       .get_modes = hibmc_connector_get_modes,
>> +       .mode_valid = hibmc_connector_mode_valid,
>> +       .best_encoder = hibmc_connector_best_encoder,
>> +};
>> +
>> +static const struct drm_connector_funcs hibmc_connector_connector_funcs = {
>> +       .dpms = drm_atomic_helper_connector_dpms,
>> +       .detect = hibmc_connector_detect,
>> +       .fill_modes = drm_helper_probe_single_connector_modes,
>> +       .destroy = drm_connector_cleanup,
>> +       .reset = drm_atomic_helper_connector_reset,
>> +       .atomic_duplicate_state = drm_atomic_helper_connector_duplicate_state,
>> +       .atomic_destroy_state = drm_atomic_helper_connector_destroy_state,
>> +};
>> +
>> +int hibmc_connector_init(struct hibmc_drm_device *hidev)
>> +{
>> +       struct drm_device *dev = hidev->dev;
>> +       struct drm_connector *connector = &hidev->connector;
>> +       int ret;
>> +
>> +       ret = drm_connector_init(dev, connector,
>> +                                &hibmc_connector_connector_funcs,
>> +                                DRM_MODE_CONNECTOR_VGA);
>> +       if (ret) {
>> +               DRM_ERROR("failed to init connector\n");
>> +               return ret;
>> +       }
>> +       drm_connector_helper_add(connector,
>> +                                &hibmc_connector_connector_helper_funcs);
>> +
>> +       return 0;
>> +}
>> --
>> 1.9.1
>>
>>
>> _______________________________________________
>> linux-arm-kernel mailing list
>> linux-arm-kernel@lists.infradead.org
>> http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
> _______________________________________________
> linuxarm mailing list
> linuxarm@huawei.com
> http://rnd-openeuler.huawei.com/mailman/listinfo/linuxarm
>
> .
>


-- 
Regards, Rongrong

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

  reply	other threads:[~2016-11-14 14:07 UTC|newest]

Thread overview: 64+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-10-28  7:27 [PATCH v6 0/9] Add DRM driver for Hisilicon Hibmc Rongrong Zou
2016-10-28  7:27 ` Rongrong Zou
2016-10-28  7:27 ` [PATCH v6 1/9] drm/hisilicon/hibmc: Add hisilicon hibmc drm master driver Rongrong Zou
2016-10-28  7:27   ` Rongrong Zou
2016-11-10 17:35   ` Sean Paul
2016-11-10 17:35     ` Sean Paul
2016-11-11  3:10     ` Rongrong Zou
2016-11-11  3:10       ` Rongrong Zou
2016-10-28  7:27 ` [PATCH v6 2/9] drm/hisilicon/hibmc: Add video memory management Rongrong Zou
2016-10-28  7:27   ` Rongrong Zou
2016-11-10 17:35   ` Sean Paul
2016-11-10 17:35     ` Sean Paul
2016-11-11 11:16     ` Rongrong Zou
2016-11-11 11:16       ` Rongrong Zou
2016-11-11 13:25       ` Sean Paul
2016-11-11 13:25         ` Sean Paul
2016-11-11 13:57         ` Rongrong Zou
2016-11-11 13:57           ` Rongrong Zou
2016-10-28  7:27 ` [PATCH v6 3/9] drm/hisilicon/hibmc: Add support for frame buffer Rongrong Zou
2016-10-28  7:27   ` Rongrong Zou
2016-11-10 18:30   ` Sean Paul
2016-11-10 18:30     ` Sean Paul
2016-11-11 13:16     ` Rongrong Zou
2016-11-11 13:16       ` Rongrong Zou
2016-11-11 13:27       ` Sean Paul
2016-11-11 13:27         ` Sean Paul
2016-10-28  7:27 ` [PATCH v6 4/9] drm/hisilicon/hibmc: Add plane for DE Rongrong Zou
2016-10-28  7:27   ` Rongrong Zou
2016-11-10 21:53   ` Sean Paul
2016-11-10 21:53     ` Sean Paul
2016-11-12  5:11     ` Rongrong Zou
2016-11-12  5:11       ` Rongrong Zou
2016-11-14 17:08       ` Sean Paul
2016-11-14 17:08         ` Sean Paul
2016-10-28  7:27 ` [PATCH v6 5/9] drm/hisilicon/hibmc: Add crtc " Rongrong Zou
2016-10-28  7:27   ` Rongrong Zou
2016-11-10 22:14   ` Sean Paul
2016-11-10 22:14     ` Sean Paul
2016-11-12 10:19     ` Rongrong Zou
2016-11-12 10:19       ` Rongrong Zou
2016-11-14 17:10       ` Sean Paul
2016-11-14 17:10         ` Sean Paul
2016-10-28  7:27 ` [PATCH v6 6/9] drm/hisilicon/hibmc: Add encoder for VDAC Rongrong Zou
2016-10-28  7:27   ` Rongrong Zou
2016-11-10 22:20   ` Sean Paul
2016-11-10 22:20     ` Sean Paul
2016-11-12 10:36     ` Rongrong Zou
2016-11-12 10:36       ` Rongrong Zou
2016-10-28  7:28 ` [PATCH v6 7/9] drm/hisilicon/hibmc: Add connector " Rongrong Zou
2016-10-28  7:28   ` Rongrong Zou
2016-11-11  1:45   ` Sean Paul
2016-11-11  1:45     ` Sean Paul
2016-11-14 14:07     ` Rongrong Zou [this message]
2016-11-14 14:07       ` Rongrong Zou
2016-10-28  7:28 ` [PATCH v6 8/9] drm/hisilicon/hibmc: Add vblank interruput Rongrong Zou
2016-10-28  7:28   ` Rongrong Zou
2016-11-11  1:49   ` Sean Paul
2016-11-11  1:49     ` Sean Paul
2016-11-14 14:10     ` Rongrong Zou
2016-11-14 14:10       ` Rongrong Zou
2016-10-28  7:28 ` [PATCH v6 9/9] MAINTAINERS: Update HISILICON DRM entries Rongrong Zou
2016-10-28  7:28   ` Rongrong Zou
2016-11-11  1:50   ` Sean Paul
2016-11-11  1:50     ` Sean Paul

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=5829C522.1040908@huawei.com \
    --to=zourongrong@huawei.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.