On Mon, Mar 25, 2019 at 05:24:49PM -0700, Andy Lutomirski wrote:
> On Mon, Mar 25, 2019 at 5:12 PM Christian Brauner
> wrote:
> >
> > On Mon, Mar 25, 2019 at 05:00:17PM -0700, Andy Lutomirski wrote:
> > > On Mon, Mar 25, 2019 at 4:45 PM Christian Brauner
> > > wrote:
> > > >
> > > > On Mon, Mar
Moved code to configure sync to where check enable_sync option before.
There is no need to check enable_sync twice. Configuring sync should be
executed immediately after enabling sync.
Signed-off-by: Sidong Yang
---
v2: remove obvious comment. reordered size/value/enable sync functions.
drivers
On Wed, 2019-03-27 at 23:49 +0530, Anirudh Rayabharam wrote:
> Checkpatch.pl complains that these lines are over 80 characters. Use the
> "psecuritypriv" pointer for consistency, remove unnecessary parantheses
> and fix the alignment.
>
> This patch just cleans up a condition, it doesn't affect ru
This patch changes the access flags of the channel attributes to
read-only. This is needed, because configuration is done via configfs.
Signed-off-by: Christian Gromm
---
v2:
drivers/staging/most/core.c | 122 +++-
1 file changed, 7 insertions(+), 115 del
This patch adds the param argument to the function parameter of
the call-back probe_channel. This parameter is needed to configure
the channels of an attached device.
Signed-off-by: Christian Gromm
---
v2:
drivers/staging/most/cdev/cdev.c | 2 +-
drivers/staging/most/core.c| 6 --
This patch updates the driver documentation files to reflect the
latest changes regarding configfs.
Signed-off-by: Christian Gromm
v2:
- changed kernel version to 5.2
---
.../most/Documentation/ABI/configfs-most.txt | 204 +
.../staging/most/Documentation/drive
This patch enables the configfs functionality of the driver by
registering the configfs subsystems and compiling the configfs
part of the sources.
Signed-off-by: Christian Gromm
---
v2:
- remove call to configfs_exit function
drivers/staging/most/Makefile | 1 +
drivers/staging/mo
This patch adds code that cleans up established links whenever the destroy
attribute is set or if the config_item (directory) is being removed.
Signed-off-by: Christian Gromm
---
v2:
- follow-up adaptions due to changes introduced w/ [PATCH v2 01/14]
drivers/staging/most/configfs.c | 38
This patch introduces attribute names that are more self explaining.
Signed-off-by: Christian Gromm
---
v2:
- follow-up adaptions due to changes introduced w/ [PATCH v2 01/14]
drivers/staging/most/configfs.c | 97 +
1 file changed, 49 insertions(+
This patch adapts the sound card management to the configfs changes.
Signed-off-by: Christian Gromm
---
v2:
drivers/staging/most/sound/sound.c | 41 +-
1 file changed, 23 insertions(+), 18 deletions(-)
diff --git a/drivers/staging/most/sound/sound.c
b/drive
This patch adds the core's interface to configfs file.
Signed-off-by: Christian Gromm
---
v2:
- remove configfs_exit prototype
drivers/staging/most/core.c | 142
drivers/staging/most/core.h | 16 -
2 files changed, 157 insertions(+), 1 d
This patch cuts off the usb_device prefix of the description string.
It is not needed, as the interface type is already available with the
interface attribute of a channel.
Signed-off-by: Christian Gromm
---
v2:
drivers/staging/most/usb/usb.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-
This patch uses the device description to clearly identity a device
attached to the bus. It is needed as the currently useed mdevX
notation is not sufficiant in case more than one network
interface controller is being used at the same time.
Signed-off-by: Christian Gromm
---
v2:
drivers/staging
Reading the create attribute that triggers the creation of a link to
a certain channel is not necessary. Hence, it is being removed.
Signed-off-by: Christian Gromm
---
v2:
drivers/staging/most/configfs.c | 14 ++
1 file changed, 2 insertions(+), 12 deletions(-)
diff --git a/drivers
This patch set makes the driver provide its configuration interface via
configfs. The configuration interface is being switched to simplify the
process of setting up the driver and to introduce the new feature of
speculative configuration.
v2:
Christian Gromm (14):
staging: most: add new file c
This patch adds the file configfs.c to the driver directory. The file
registers the necessary subsystems with configfs in order to move the
driver configuration from sysfs to configfs.
Signed-off-by: Christian Gromm
---
v2:
Reported-by: Dan Carpenter
- remove unnecessary variable init
This patch removes the driver attribute add_link. It is not needed, because
the link management is now done via configfs.
Signed-off-by: Christian Gromm
---
v2:
drivers/staging/most/core.c | 61 -
1 file changed, 61 deletions(-)
diff --git a/drivers/
This patch makes the driver accept a link confiiguration eventhough no
device is attached to the bus. Instead the configuration is being applied
as soon as a device is being registered with the core.
Signed-off-by: Christian Gromm
---
v2:
- follow-up adaptions due to changes introduced w/
On Thu, Mar 28, 2019 at 02:17:29PM +0100, Christian Gromm wrote:
> +static ssize_t mdev_link_direction_store(struct config_item *item,
> + const char *page, size_t count)
> +{
> + struct mdev_link *mdev_link = to_mdev_link(item);
> +
> + if (sysfs_streq(
On Thu, Mar 28, 2019 at 02:17:31PM +0100, Christian Gromm wrote:
> +int most_set_cfg_direction(char *mdev, char *mdev_ch, char *buf)
> +{
> + struct most_channel *c = get_channel(mdev, mdev_ch);
> +
> + if (!c)
> + return -ENODEV;
> + if (!strcmp(buf, "dir_rx\n")) {
> +
On Thu, Mar 28, 2019 at 02:17:32PM +0100, Christian Gromm wrote:
> +static int audio_create_sound_card(void)
> +{
> + int ret;
> + struct sound_adapter *adpt;
> +
> + list_for_each_entry(adpt, &adpt_list, list) {
> + if (!adpt->registered)
> + goto adpt_a
On Thu, Mar 28, 2019 at 02:17:38PM +0100, Christian Gromm wrote:
> This patch makes the driver accept a link confiiguration eventhough no
> device is attached to the bus. Instead the configuration is being applied
> as soon as a device is being registered with the core.
>
> Signed-off-by: Christia
On Wed, Mar 27, 2019 at 11:43:20PM +, David Howells wrote:
> +static void binderfs_free_fc(struct fs_context *fc)
> +{
> + struct binderfs_info *info = fc->s_fs_info;
> +
> + if (info) {
> + struct ipc_namespace *ipc_ns = fc->s_fs_info;
> + put_ipc_ns(ipc_ns);
I
On Thu, Mar 14, 2019 at 02:22:10PM +0100, Sergio Paracuellos wrote:
> Add bindings to describe Mediatek MT7621 PCIe PHY.
Binding should come before the driver.
>
> Signed-off-by: Sergio Paracuellos
> ---
> .../bindings/phy/mediatek,mt7621-pci-phy.txt | 54 +++
> 1 file changed
Hi Rob,
Thanks for the review.
On Thu, Mar 28, 2019 at 4:42 PM Rob Herring wrote:
>
> On Thu, Mar 14, 2019 at 02:22:10PM +0100, Sergio Paracuellos wrote:
> > Add bindings to describe Mediatek MT7621 PCIe PHY.
>
> Binding should come before the driver.
Do you mean this should be PATCH 1 of the s
Hello,
syzbot found the following crash on:
HEAD commit:1a9df9e2 Merge git://git.kernel.org/pub/scm/linux/kernel/g..
git tree: upstream
console output: https://syzkaller.appspot.com/x/log.txt?x=11fed11f20
kernel config: https://syzkaller.appspot.com/x/.config?x=8dcdce25ea72bedf
da
The fix for this is in Greg KH's char-misc-linus branch. Pointing
syzbot at it to verify:
#syz test: git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc.git
char-misc-linus
On Thu, Mar 28, 2019 at 9:14 AM syzbot
wrote:
>
> Hello,
>
> syzbot found the following crash on:
>
> HEAD comm
This series change both bindings and driver to don't use child nodes in the
device tree.
This changes are inspired after Rob's bindings review which can be found here:
http://driverdev.linuxdriverproject.org/pipermail/driverdev-devel/2019-March/133123.html
Hope this helps.
Best regards,
Se
If each phy port doesn't have its own resources, then we don't need
child nodes. Handle it using #phy-cells to 1 or 0 conveniently.
Signed-off-by: Sergio Paracuellos
---
drivers/staging/mt7621-dts/mt7621.dtsi | 23 +++
1 file changed, 3 insertions(+), 20 deletions(-)
diff --
Device tree has been simplified to don't use child nodes and use
the #phy-cells property instead. Change the driver accordly implementing
custom 'xlate' function to return the correct phy for each port.
Signed-off-by: Sergio Paracuellos
---
.../staging/mt7621-pci-phy/pci-mt7621-phy.c | 26
Device tree has been simplified to use phy-cells instead of
using child nodes. Update documentation accordly.
Signed-off-by: Sergio Paracuellos
---
.../mediatek,mt7621-pci-phy.txt | 44 ---
1 file changed, 9 insertions(+), 35 deletions(-)
diff --git a/drivers/stagi
Hello,
syzbot tried to test the proposed patch but build/boot failed:
patch is already applied
Tested on:
commit: 0532a1b0 virt: vbox: Implement passing requestor info to t..
git tree:
git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc.git
char-misc-linus
compil
From: Gao Xiang
After commit 419d6efc50e9, kernel cannot be crashed in the namei
path. However, corrupted nameoff can do harm in the process of
readdir for scenerios without dm-verity as well. Fix it now.
Fixes: 3aa8ec716e52 ("staging: erofs: add directory operations")
Cc: # 4.19+
Signed-off-by
syzbot has bisected this bug to:
commit 1a7c3d9bb7a926e88d5f57643e75ad1abfc55013
Author: Todd Kjos
Date: Fri Feb 8 18:35:14 2019 +
binder: create userspace-to-binder-buffer copy function
bisection log: https://syzkaller.appspot.com/x/bisect.txt?x=15b9383f20
start commit: 1a9df
Signed-off-by: Mikhail Gusarov
---
drivers/staging/wlan-ng/cfg80211.c | 48 +++---
drivers/staging/wlan-ng/p80211metadef.h | 278 +---
drivers/staging/wlan-ng/p80211netdev.c | 8 +-
drivers/staging/wlan-ng/p80211req.c | 28 ++--
drivers/staging/wlan-ng/p8
From: Colin Ian King
Currently the lower 8 bits of ret are being masked and left
shifted by 8 bits always leaving a result of zero. The mask
appears to be incorrect and should probably be 0xff00 instead
of 0xff. Fix this. (Note: not tested).
Fixes: 16feab644fd1 ("staging: rtl8192u: check retur
Hello,
syzbot tried to test the proposed patch but build/boot failed:
timed out
Tested on:
commit: 0532a1b0 virt: vbox: Implement passing requestor info to t..
git tree:
git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc.git
char-misc-linus
kernel config: h
Sigh. One more try... master with the same patch as an attachment.
(the patch is already queued up in gregkh's 'char-misc-linus' branch).
#syz test: git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable.git
master
On Thu, Mar 28, 2019 at 5:20 PM syzbot
wrote:
>
> Hello,
>
> syzbot tr
On Thu, Mar 28 2019, Sergio Paracuellos wrote:
> If each phy port doesn't have its own resources, then we don't need
> child nodes. Handle it using #phy-cells to 1 or 0 conveniently.
>
> Signed-off-by: Sergio Paracuellos
> ---
> drivers/staging/mt7621-dts/mt7621.dtsi | 23 +++
On Thu, Mar 28 2019, Sergio Paracuellos wrote:
> Device tree has been simplified to don't use child nodes and use
> the #phy-cells property instead. Change the driver accordly implementing
> custom 'xlate' function to return the correct phy for each port.
>
> Signed-off-by: Sergio Paracuellos
> -
Device tree has been simplified to use phy-cells instead of
using child nodes. Update documentation accordly.
Signed-off-by: Sergio Paracuellos
---
.../mediatek,mt7621-pci-phy.txt | 44 ---
1 file changed, 9 insertions(+), 35 deletions(-)
diff --git a/drivers/stagi
Device tree has been simplified to don't use child nodes and use
the #phy-cells property instead. Change the driver accordly implementing
custom 'xlate' function to return the correct phy for each port.
Signed-off-by: Sergio Paracuellos
---
.../staging/mt7621-pci-phy/pci-mt7621-phy.c | 26
If each phy port doesn't have its own resources, then we don't need
child nodes. Handle it using #phy-cells to 1 for both phy's.
Signed-off-by: Sergio Paracuellos
---
drivers/staging/mt7621-dts/mt7621.dtsi | 23 +++
1 file changed, 3 insertions(+), 20 deletions(-)
diff --git
This series change both bindings and driver to don't use child nodes in the
device tree.
This changes are inspired after Rob's bindings review which can be found here:
http://driverdev.linuxdriverproject.org/pipermail/driverdev-devel/2019-March/133123.html
Changes in v2:
- Use #phy_cells=1
Hi Neil,
Thanks for the feedback.
On Fri, Mar 29, 2019 at 4:01 AM NeilBrown wrote:
>
> On Thu, Mar 28 2019, Sergio Paracuellos wrote:
>
> > If each phy port doesn't have its own resources, then we don't need
> > child nodes. Handle it using #phy-cells to 1 or 0 conveniently.
> >
> > Signed-off-b
On Fri, Mar 29, 2019 at 4:06 AM NeilBrown wrote:
>
> On Thu, Mar 28 2019, Sergio Paracuellos wrote:
>
> > Device tree has been simplified to don't use child nodes and use
> > the #phy-cells property instead. Change the driver accordly implementing
> > custom 'xlate' function to return the correct
On Fri, Mar 29 2019, Sergio Paracuellos wrote:
> This series change both bindings and driver to don't use child nodes in the
> device tree.
>
> This changes are inspired after Rob's bindings review which can be found here:
>
> http://driverdev.linuxdriverproject.org/pipermail/driverdev-devel/2019
On Thu, Mar 28, 2019 at 11:53:29PM +, Mikhail Gusarov wrote:
> Signed-off-by: Mikhail Gusarov
> ---
I can not take patches without any changelog text at all, sorry.
___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverpro
drivers/staging/wlan-ng/p80211metadef.h was generated before
submitting to Linux, so it contains long constants and
ugly code.
- Shorten macro names for DIDMSG_*/DIDMIB_* constants generators
- Drop repeating prefixes from DIDMSG_*/DIDMIB_* constants
- Use abbreviations where spelling full words i
On 2019/3/29 4:14, Gao Xiang wrote:
> From: Gao Xiang
>
> After commit 419d6efc50e9, kernel cannot be crashed in the namei
> path. However, corrupted nameoff can do harm in the process of
> readdir for scenerios without dm-verity as well. Fix it now.
>
> Fixes: 3aa8ec716e52 ("staging: erofs: add
On Fri, Mar 29, 2019 at 06:25:09AM +, Mikhail Gusarov wrote:
> drivers/staging/wlan-ng/p80211metadef.h was generated before
> submitting to Linux, so it contains long constants and
> ugly code.
>
> - Shorten macro names for DIDMSG_*/DIDMIB_* constants generators
> - Drop repeating prefixes fro
51 matches
Mail list logo
