You're missing "ath10k: " prefix in the patch subject.
On 20 October 2014 18:38, Yanbo Li wrote:
> The debugfs interface reg_addr®_val used to read and write the target
> register.
> The interface mem_addr&mem_val used to dump the targer memory and also
> can be used to assign value to target me
Kalle Valo writes:
> All warning and error messages should end with '\n', fix that.
>
> Signed-off-by: Kalle Valo
Joe pointed out that I had missed a few, will send v2.
--
Kalle Valo
--
To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
the body of a message to major
On Tue, 2014-10-21 at 09:26 +0530, Sujith Manoharan wrote:
> Johannes Berg wrote:
> > Maybe we need something like this:
> >
> > diff --git a/net/mac80211/scan.c b/net/mac80211/scan.c
> > index af0d094b2f2f..45b74ab1c59d 100644
> > --- a/net/mac80211/scan.c
> > +++ b/net/mac80211/scan.c
> > @@ -98
On Tue, Oct 21, 2014 at 12:06 PM, Krishna Chaitanya
wrote:
> On Tue, Oct 21, 2014 at 11:01 AM, Emmanuel Grumbach
> wrote:
>> On Tue, Oct 21, 2014 at 8:02 AM, Emmanuel Grumbach
>> wrote:
>>> On Mon, Oct 20, 2014 at 10:29 PM, Laurențiu Nicola wrote:
No, nothing in the DD-WRT syslog. The IP
On Tue, Oct 21, 2014 at 11:01 AM, Emmanuel Grumbach wrote:
> On Tue, Oct 21, 2014 at 8:02 AM, Emmanuel Grumbach
> wrote:
>> On Mon, Oct 20, 2014 at 10:29 PM, Laurențiu Nicola wrote:
>>> No, nothing in the DD-WRT syslog. The IP seems to be the one cached by
>>> my DHCP client. I also tried vario
On Tue, Oct 21, 2014 at 8:02 AM, Emmanuel Grumbach wrote:
> On Mon, Oct 20, 2014 at 10:29 PM, Laurențiu Nicola wrote:
>> No, nothing in the DD-WRT syslog. The IP seems to be the one cached by
>> my DHCP client. I also tried various suggestions found on the internet,
>> but nothing helped.
>>
>
>
On Mon, Oct 20, 2014 at 10:29 PM, Laurențiu Nicola wrote:
> No, nothing in the DD-WRT syslog. The IP seems to be the one cached by
> my DHCP client. I also tried various suggestions found on the internet,
> but nothing helped.
>
Can you please try to have another device in monitor mode to sniff t
Johannes Berg wrote:
> Maybe we need something like this:
>
> diff --git a/net/mac80211/scan.c b/net/mac80211/scan.c
> index af0d094b2f2f..45b74ab1c59d 100644
> --- a/net/mac80211/scan.c
> +++ b/net/mac80211/scan.c
> @@ -985,7 +985,6 @@ void ieee80211_scan_cancel(struct ieee80211_local *local)
>
From: Ben Greear
This is useful when creating virtual-stations, and probably
APs and other devices as well. Keeps udev from mucking with
things it shouldn't, since the default MAC is never seen
by udev when specified on the cmd-line during creation.
Signed-off-by: Ben Greear
---
include/net/c
From: Ben Greear
This will be helpful when using the mac80211_hwsim
wiphys and automated testing. Let user create the
wlan devs as needed, and named as expected.
Signed-off-by: Ben Greear
---
include/net/mac80211.h | 7 ++-
net/mac80211/main.c| 3 ++-
2 files changed, 8 insertions(+),
From: Ben Greear
Support creating wiphy devices with an optional name.
This will be used by hwsim to have better automated control
over virtual radio creation/deletion.
Signed-off-by: Ben Greear
---
include/net/mac80211.h | 25 -
net/mac80211/main.c| 9 +
2
From: Ben Greear
Good for automated testing, where user can create wlan
interfaces with specified names.
Signed-off-by: Ben Greear
---
drivers/net/wireless/mac80211_hwsim.c | 13 ++---
drivers/net/wireless/mac80211_hwsim.h | 2 ++
2 files changed, 12 insertions(+), 3 deletions(-)
dif
From: Ben Greear
Otherwise, it can be very difficult to know which is which
if you are trying to do detailed testing.
Signed-off-by: Ben Greear
---
drivers/net/wireless/mac80211_hwsim.c | 13 +
1 file changed, 9 insertions(+), 4 deletions(-)
diff --git a/drivers/net/wireless/mac80
From: Ben Greear
Kernel will attempt to use the name if it is supplied,
but if name cannot be used for some reason, the default
phyX name will be used instead.
Signed-off-by: Ben Greear
---
include/net/cfg80211.h | 22 +-
net/wireless/core.c| 45
Use "%s" in the workqueue allocation to make sure the rtl_hal_cfg name
can never accidentally leak information via a format string.
Signed-off-by: Kees Cook
---
drivers/net/wireless/rtlwifi/base.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/wireless/rtlwifi/ba
On 10/20/2014 05:58 AM, Michal Kazior wrote:
> Latest firmware revision introduced a new WMI ABI
> called wmi-tlv. It is not a tlv strictly speaking
> but something that resembles it because it is
> ordered and may have duplicate id entries.
>
> This will be used by new hardware eventually.
>
>
On Mon, 2014-10-20 at 21:33 +0200, Johannes Berg wrote:
> On Mon, 2014-10-20 at 17:34 +0200, Karl Beldan wrote:
> > > Indeed, will resending only 4/4 do ?
>
> Sure.
I merged patches 1-3 now.
johannes
--
To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
the body of a m
On Mon, 2014-10-20 at 17:34 +0200, Karl Beldan wrote:
> > Indeed, will resending only 4/4 do ?
Sure.
> > > CHECK
> > > /home/johannes/sys/wireless/net/mac80211/rc80211_minstrel_ht_debugfs.c
> > > /home/johannes/sys/wireless/net/mac80211/rc80211_minstrel_ht_debugfs.c:47
> > > minstrel_ht_sta
On Mon, 2014-10-20 at 21:09 +0200, Oleksij Rempel wrote:
> Hi Johannes,
>
> since this function was added by you, may be you can help here (commit
> f09603a259). ath9k_htc support max 8 STAs in AP or ADHOC mode. If more
> then 8 stations connected, driver will return error, but
> sta_info_insert_
No, nothing in the DD-WRT syslog. The IP seems to be the one cached by
my DHCP client. I also tried various suggestions found on the internet,
but nothing helped.
Laurentiu
On Mon, Oct 20, 2014, at 11:23, Emmanuel Grumbach wrote:
> On Mon, Oct 20, 2014 at 10:22 AM, Laurențiu Nicola
> wrote:
> >
Hi Johannes,
since this function was added by you, may be you can help here (commit
f09603a259). ath9k_htc support max 8 STAs in AP or ADHOC mode. If more
then 8 stations connected, driver will return error, but
sta_info_insert_drv_state() setting it to zero.
What was initial purpose of this?
net
The debugfs interface reg_addr®_val used to read and write the target
register.
The interface mem_addr&mem_val used to dump the targer memory and also
can be used to assign value to target memory
The basic usage explain as below:
Register read/write:
reg_addr (the register address)
On Mon, Oct 20, 2014 at 05:13:02PM +0200, Karl Beldan wrote:
> On Mon, Oct 20, 2014 at 04:49:06PM +0200, Johannes Berg wrote:
> > On Mon, 2014-10-20 at 15:45 +0200, Karl Beldan wrote:
> > > From: Karl Beldan
> > >
> > > Hi,
> > >
> > > Varka Bhadram reported checkpatch is noisy on this series.
>
On Mon, Oct 20, 2014 at 04:49:06PM +0200, Johannes Berg wrote:
> On Mon, 2014-10-20 at 15:45 +0200, Karl Beldan wrote:
> > From: Karl Beldan
> >
> > Hi,
> >
> > Varka Bhadram reported checkpatch is noisy on this series.
> > I made another pass which affects [2/4](trivially) and [4/4].
> > This c
On Mon, 2014-10-20 at 15:45 +0200, Karl Beldan wrote:
> From: Karl Beldan
>
> Hi,
>
> Varka Bhadram reported checkpatch is noisy on this series.
> I made another pass which affects [2/4](trivially) and [4/4].
> This could have been a patch on top of what Felix acked, I hope you
> won't bother to
On Mon, 2014-10-20 at 13:20 +0300, Jouni Malinen wrote:
> The optional NL80211_ATTR_MGMT_SUBTYPE and NL80211_ATTR_REASON_CODE
> attributes can now be included in NL80211_CMD_DEL_STATION to indicate to
> the driver which frame (Deauthentication/Disassociation) and reason code
> in that frame should
On Mon, 2014-10-20 at 10:54 +0200, Karl Beldan wrote:
> From: Karl Beldan
>
> ATM an HT rc_stats line is 106 chars.
> Times 8(MCS_GROUP_RATES)*3(SS)*2(GI)*2(BW) + CCK(4), i.e. x100, this is
> well above the current 8192 - sizeof(*ms) currently allocated.
Applied.
johannes
--
To unsubscribe fro
The netbuf pop can return NULL. Make sure to check
for that. It shouldn't happen but better safe than
sorry.
Signed-off-by: Michal Kazior
---
drivers/net/wireless/ath/ath10k/htt_rx.c | 9 +
1 file changed, 9 insertions(+)
diff --git a/drivers/net/wireless/ath/ath10k/htt_rx.c
b/drivers/
The rx descriptor variable was no longer used in
the rx handler.
Signed-off-by: Michal Kazior
---
drivers/net/wireless/ath/ath10k/htt_rx.c | 5 -
1 file changed, 5 deletions(-)
diff --git a/drivers/net/wireless/ath/ath10k/htt_rx.c
b/drivers/net/wireless/ath/ath10k/htt_rx.c
index 63c69d8..6
In theory it was possible to drain entire HTT Rx
ring via fragmented Rx leading to Rx lockup.
In practice non-data traffic would always trigger
replenishment via the regular Rx handler.
For correctness sake make sure to replenish the
ring on fragmented Rx.
Signed-off-by: Michal Kazior
---
driv
There's little point in dropping, e.g. frames with
FCS error early in ath10k.
This simplifies amsdu_allowed() and gets rid of
htt_rx_mpdu_status usage finally.
Signed-off-by: Michal Kazior
---
drivers/net/wireless/ath/ath10k/htt_rx.c | 15 ---
1 file changed, 15 deletions(-)
diff -
Once driver entered the rx_confused state it would
refuse to rx even after firmware is restarted.
Make sure to clear it so that rx works after, e.g.
hw restart or after all interfaces are stopped.
Signed-off-by: Michal Kazior
---
drivers/net/wireless/ath/ath10k/htt_rx.c | 2 ++
1 file changed, 2
Tracing function was called before buffers were
unmapped from DMA.
Signed-off-by: Michal Kazior
---
drivers/net/wireless/ath/ath10k/htt_rx.c | 7 ---
1 file changed, 4 insertions(+), 3 deletions(-)
diff --git a/drivers/net/wireless/ath/ath10k/htt_rx.c
b/drivers/net/wireless/ath/ath10k/htt_
Use the globally defined ieee80211 values instead
of re-defining them in the driver again.
Signed-off-by: Michal Kazior
---
drivers/net/wireless/ath/ath10k/htt_rx.c | 34 +++-
1 file changed, 20 insertions(+), 14 deletions(-)
diff --git a/drivers/net/wireless/ath/ath
Treat non-chained and chained popping the same
way. Also this makes netbuf pop fully symmetrical
to (re)filling.
Signed-off-by: Michal Kazior
---
drivers/net/wireless/ath/ath10k/htt_rx.c | 30 +-
1 file changed, 9 insertions(+), 21 deletions(-)
diff --git a/drivers/n
Hi,
I've found a couple of issues while reworking rx
path.
Michal Kazior (9):
ath10k: don't drop control and null func Rx
ath10k: remove unused variable
ath10k: use ieee80211 defines for crypto param lengths
ath10k: fix rx buffer tracing
ath10k: deduplicate htt rx dma unmapping
ath10
HTT_RX_IND_MPDU_STATUS_MGMT_CTRL was pretty greedy
and because of that ath10k ended up dropping
Control Frames as well as Null Func frames.
Reported-by: Okhwan Lee
Signed-off-by: Michal Kazior
---
drivers/net/wireless/ath/ath10k/htt_rx.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
From: Karl Beldan
When the new CONFIG_MAC80211_RC_MINSTREL_VHT is not set (default 'N'),
there is no behavioral change including in sampling and MCS_GROUP_RATES
remains 8.
Otherwise MCS_GROUP_RATES is 10, and a module parameter *vht_only*
(default 'true'), restricts the rates selection to VHT whe
From: Karl Beldan
No functional change.
Signed-off-by: Karl Beldan
Cc: Felix Fietkau
---
net/mac80211/rc80211_minstrel_ht.c | 51 +++---
net/mac80211/rc80211_minstrel_ht.h | 15 +++--
net/mac80211/rc80211_minstrel_ht_debugfs.c | 10 +++---
3 files c
From: Karl Beldan
ATM, we grep cck rates idx with idx / MCS_GROUP_RATES ==
MINSTREL_CCK_GROUP.
Matching neither-cck-non-ht rates could be done by replacing '==' with
'>', however it would be less versatile or explicit.
This will allow to match VHT rates with IEEE80211_TX_RC_VHT_MCS.
Signed-off-b
From: Karl Beldan
Hi,
Varka Bhadram reported checkpatch is noisy on this series.
I made another pass which affects [2/4](trivially) and [4/4].
This could have been a patch on top of what Felix acked, I hope you
won't bother too much.
Thanks for reviewing,
Karl Beldan (4):
mac80211: minstrel_h
From: Karl Beldan
Since 5935839ad735 ("mac80211: improve minstrel_ht rate sorting by
throughput & probability"), the rate indexes are manipulated via u8's
and hence allow for a maximum of 256 mcs_group entries in
minstrel_mcs_groups.
ATM, minstrel_ht advertizes support up to 3HTSS@40MHz, consumi
On 20 October 2014 14:58, Michal Kazior wrote:
> Latest firmware revision introduced a new WMI ABI
> called wmi-tlv. It is not a tlv strictly speaking
> but something that resembles it because it is
> ordered and may have duplicate id entries.
>
> This will be used by new hardware eventually.
Ok,
Latest firmware revision introduced a new WMI ABI
called wmi-tlv. It is not a tlv strictly speaking
but something that resembles it because it is
ordered and may have duplicate id entries.
This will be used by new hardware eventually.
Signed-off-by: Michal Kazior
---
drivers/net/wireless/ath/at
Since the 10.x fw branch support was introduced it
became apparent ath10k will need to be able to
deal with different fw ABIs eventually.
The patch creates an abstraction for dealing with
command and event structures across different ABIs
and mostly gets rid of the
ATH10K_FW_FEATURE_WMI_10X flag u
This splits the actual event parsing into
intermediary structures to facilitate future
support of vastly different ABI WMI backends.
Signed-off-by: Michal Kazior
---
drivers/net/wireless/ath/ath10k/wmi.c | 383 +-
drivers/net/wireless/ath/ath10k/wmi.h | 60 +-
Some functions can be shared across different WMI
ABIs. Make them public so different WMI backends
can use them from different source files in the
future.
Signed-off-by: Michal Kazior
---
drivers/net/wireless/ath/ath10k/wmi.c | 136 ++
drivers/net/wireless/ath/ath
The dump print uses sizeof() but since service_map
was a mere pointer the dump was too short.
Signed-off-by: Michal Kazior
---
drivers/net/wireless/ath/ath10k/wmi.c | 4 +++-
drivers/net/wireless/ath/ath10k/wmi.h | 1 +
2 files changed, 4 insertions(+), 1 deletion(-)
diff --git a/drivers/net/wi
Hi,
First 2 patches are simple fixes. Patch 3 isn't
strictly related to wmi-tlv but I decided to
include it anyway. Other patches gradually
implement wmi-tlv.
There's a bunch of false-positive checkpatch
warnings regarding `CHECK: No space is necessary
after a cast`.
There's also a bunch of `WAR
The service bitmap can be used to determine
firmware capabilities.
Signed-off-by: Michal Kazior
---
drivers/net/wireless/ath/ath10k/core.h | 2 +-
drivers/net/wireless/ath/ath10k/debug.c | 11 +++
drivers/net/wireless/ath/ath10k/debug.h | 9 -
drivers/net/wireless/ath/ath10k/w
This shouldn't really happen but take into account
the original service bitmap length when mapping
service ids.
Signed-off-by: Michal Kazior
---
drivers/net/wireless/ath/ath10k/wmi.c | 6 +-
drivers/net/wireless/ath/ath10k/wmi.h | 121 +-
2 files changed, 66 ins
All the callers of ieee80211_mgd_probe_ap_send return right
after they call the flush() callback. This means that calling
flush() is uneeded since its meaning is to wait until the
queues of the device are empty.
Devices that know how to report status on Tx will do so using
the regular path (ieee80
On Mon, 2014-10-20 at 14:02 +0200, Johannes Berg wrote:
> On Mon, 2014-10-20 at 16:26 +0530, Sujith Manoharan wrote:
>
> > Here is a trace log showing the issue - cfg80211_scan_done()
> > is called after NETDEV_DOWN is received in cfg80211_netdev_notifier_call().
> >
> > http://pastebin.com/raw.p
Until now it was possible to simulate soft and
hard fw crashes but it wasn't possible to trigger
an immediately hw restart itself (without the fw
crash).
This can be useful when stress testing hw
restarting stability, e.g. during heavy tx/rx
traffic.
Signed-off-by: Michal Kazior
---
Notes:
Patch #1 fixes a bug I've found while testing
patches #2 and #3. It's safe to cherry-pick it in
case patches #2 and/or #3 aren't accepted or need
a re-spin.
I've mainly used patch #2 to test reset and
recovery. It's pretty handy for (stress-)testing.
Patch #3 should improve recovery speed in some
While testing other things I've found that CE
items aren't cleared properly. This could lead to
null dereferences in BMI.
To prevent that make sure CE revoking clears the
nbytes value (which is used as a buffer completion
indication) and memset the entire CE rings when
(re)initializing.
Also make
In some cases hw recovery was taking an absurdly
long time due to ath10k waiting for things that
would never really complete.
Instead of waiting for inevitable timeouts poke
all completions and wakequeues and check if it's
still worth waiting.
Reading/writing ar->state requires conf_mutex.
Since
Currently hif_power_up performs effectively a
reset and hif_stop resets the chip as well so
there's no point in resetting here.
Signed-off-by: Michal Kazior
---
Notes:
v2:
* this patch replaces `replace power up/down with reset callback` [Kalle]
drivers/net/wireless/ath/ath10k/pci.c |
One of the problems with warm reset I've found is
that it must be guaranteed that copy engine
registers are not being accessed while being
reset. Otherwise in worst case scenario the host
may lock up.
Instead of using sleeps and hoping the device is
operational in some arbitrary timeframes use
fir
This was the final missing bit to making sure the
device doesn't assert interrupts to host.
This should fix possible race when target crashes
during driver teardown.
This also removes an early warm reset workaround
during pci probing.
Signed-off-by: Michal Kazior
---
Notes:
v2:
* remo
Calling init to reinit ce pipe state would also
re-set all static structure links and setting
(which don't change over driver lifecycle).
Make it so alloc links structures and initializes
static data and init part to setup state
variables and clear stuff.
Signed-off-by: Michal Kazior
---
Notes:
The power up procedure was overly complex due to
warm/cold reset workarounds and issues.
Signed-off-by: Michal Kazior
---
Notes:
v2:
* fix indentation [Kalle]
drivers/net/wireless/ath/ath10k/pci.c | 151 ++
1 file changed, 80 insertions(+), 71 deletions
Hi,
First 2 patches address some irq issues.
Other 4 fix and cleanup device resetting.
Hopefully this will make ath10k a little more
robust when it comes to warm reset.
As a side effect of the changes ath10k now boots a
little faster (slightly over a second instead of
two).
v2:
* don't remove
If MSI isn't configured device ROM program expects
legacy interrupts to be enabled before it can
fully boot. Don't forget to disable legacy
interrupts after that.
While at it re-use the legacy irq enabling helper
instead of calling ath10k_pci_write32().
Signed-off-by: Michal Kazior
---
drivers/
On Mon, 2014-10-20 at 16:26 +0530, Sujith Manoharan wrote:
> Here is a trace log showing the issue - cfg80211_scan_done()
> is called after NETDEV_DOWN is received in cfg80211_netdev_notifier_call().
>
> http://pastebin.com/raw.php?i=AJqFLtZR
>
> I've removed entries to show the issue clearly, i
Johannes Berg wrote:
> I'm not sure that's true - the WARN_ON you're modifying keys off
> something that happens in cfg80211_scan_done(), not
> __cfg80211_scan_done(), so as long as the driver called
> cfg80211_scan_done() properly things should be OK?
Here is a trace log showing the issue - cfg80
On Mon, 2014-10-20 at 12:46 +0200, Marcel Holtmann wrote:
> Maybe relaxing the check and allow ip link to move a wireless netdev
> into a namespace (and having the wiphy follow) could be allowed if it
> is the only netdev or the original wlan0 that each wiphy creates. I
> really do not know if thi
Hi Johannes,
>> The reason for this was to make possible to change netns for wireless
>> dev via 'ip link' too like for 'iw' util. I just think that changing
>> namespace for netdev should have the generic way. May be you can suggest
>> a better way
>
> That's a respectable goal, but I think you'
>
> On Thu, 2014-10-16 at 23:25 +0300, Emmanuel Grumbach wrote:
> > All the callers of ieee80211_mgd_probe_ap_send return right after they
> > call the flush() callback. This means that calling
> > flush() is uneeded since its meaning is to wait until the queues of
> > the device are empty. It is
On Thu, 2014-10-16 at 23:25 +0300, Emmanuel Grumbach wrote:
> All the callers of ieee80211_mgd_probe_ap_send return right
> after they call the flush() callback. This means that calling
> flush() is uneeded since its meaning is to wait until the
> queues of the device are empty. It is pointless to
On Mon, 2014-10-20 at 15:09 +0530, Sujith Manoharan wrote:
> Johannes Berg wrote:
> > If it's a mac80211 race, wouldn't it make sense to fix it in mac80211?
>
> Yes, definitely. :-)
>
> But, we also ran into issues where starting a BA session
> when HW scan was in progress (with MCC enabled) caus
On Mon, 2014-10-20 at 15:57 +0530, Sujith Manoharan wrote:
> Johannes Berg wrote:
> > But the driver can/should just do it when the interface is removed from
> > it? Otherwise it has no valid way to run/continue the scan anyway?
>
> cancel_hw_scan() is called before remove_interface() and if the
>
Johannes Berg wrote:
> But the driver can/should just do it when the interface is removed from
> it? Otherwise it has no valid way to run/continue the scan anyway?
cancel_hw_scan() is called before remove_interface() and if the
interface is removed in cfg80211 before __cfg80211_scan_done() gets
a
The optional NL80211_ATTR_MGMT_SUBTYPE and NL80211_ATTR_REASON_CODE
attributes can now be included in NL80211_CMD_DEL_STATION to indicate to
the driver which frame (Deauthentication/Disassociation) and reason code
in that frame should be used to indicate removal to the specific
station. This is use
On Mon, 2014-10-20 at 15:12 +0530, Sujith Manoharan wrote:
> Johannes Berg wrote:
> > Those commits say they *fix* the race though, so in case I was wrong you
> > should report the sequence of events so we can look at fixing it
> > properly?
>
> The commits fix both the issues (use-after-free, new
On Tue, 2014-10-14 at 10:01 +, Malinen, Jouni wrote:
> I think I would be fine with cfg80211 default being Deauthentication frame
> with reason code 2, so if that is what you would like to see here (rather
> than leaving this to the drivers), I can update the patch to behave in
> that way.
As
On Tue, 2014-10-14 at 15:16 +0300, vadi...@gmail.com wrote:
> The reason for this was to make possible to change netns for wireless
> dev via 'ip link' too like for 'iw' util. I just think that changing
> namespace for netdev should have the generic way. May be you can suggest
> a better way
That
On Thu, 2014-10-16 at 19:20 +0200, Rostislav Lisovy wrote:
> Now I realized it is not that easy (and I confused "interface running"
> and being "connected to the network"). There seems not to be a solid
> indication that we are no longer "connected" to the network.
Oops, ignore my previous reply.
On Thu, 2014-10-16 at 18:33 +0200, Rostislav Lisovy wrote:
> You are right. I hope the following is a reasonable solution (in form of
> a patch to my previous patch; comment stolen from some prehistoric
> version of mesh.c):
>
> @@ -127,6 +127,9 @@ void ieee80211_ocb_work(struct ieee80211_sub_if_
Johannes Berg wrote:
> Those commits say they *fix* the race though, so in case I was wrong you
> should report the sequence of events so we can look at fixing it
> properly?
The commits fix both the issues (use-after-free, new scan) correctly,
so I don't see a race ? If the driver is slow in noti
Hi Varka Bhadram,
On Mon, Oct 20, 2014 at 02:31:31PM +0530, Varka Bhadram wrote:
> Hi Karl Beldan,
>
> On 10/20/2014 02:25 PM, Karl Beldan wrote:
> >From: Karl Beldan
> >
> >Hi,
> >
> >This is just a respin rebased on v2 of "mac80211: minstrels: fix buffer
> >overflow in HT debugfs" which only a
Johannes Berg wrote:
> If it's a mac80211 race, wouldn't it make sense to fix it in mac80211?
Yes, definitely. :-)
But, we also ran into issues where starting a BA session
when HW scan was in progress (with MCC enabled) caused connectivity
problems, hence this fix in ath9k.
Regarding the BA race
On Fri, 2014-10-17 at 07:02 +0530, Sujith Manoharan wrote:
> Yes, the scan cancellation is initiated on interface removal,
> but it's async. These commits explain the race:
>
> https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/net/wireless/core.c?id=a617302c531eaf497ccd02a61d
On Fri, 2014-10-17 at 07:40 +0530, Sujith Manoharan wrote:
> From: Sujith Manoharan
>
> mac80211 currently has a race which can be hit
> with this sequence:
If it's a mac80211 race, wouldn't it make sense to fix it in mac80211?
johannes
--
To unsubscribe from this list: send the line "unsubscr
Hi Karl Beldan,
On 10/20/2014 02:25 PM, Karl Beldan wrote:
From: Karl Beldan
Hi,
This is just a respin rebased on v2 of "mac80211: minstrels: fix buffer
overflow in HT debugfs" which only affects 4/4.
Karl Beldan (4):
mac80211: minstrel_ht: Increase the range of handled rate indexes
ma
Felix, I added your Acks in the patches but git-send-email did not send
them to the Acked-by addresses.
Karl
On Mon, Oct 20, 2014 at 10:55:38AM +0200, Karl Beldan wrote:
> From: Karl Beldan
>
> Hi,
>
> This is just a respin rebased on v2 of "mac80211: minstrels: fix buffer
> overflow in HT de
On 2014-10-20 10:15, Sujith Manoharan wrote:
> Felix Fietkau wrote:
>> The "goto chip_reset" is a bit misleading, because it does not actually
>> issue a chip reset. Instead it is bypassing processing of other
>> interrupts and assumes that the tasklet will issue a chip reset.
>
> Well, we kill in
From: Karl Beldan
Since 5935839ad735 ("mac80211: improve minstrel_ht rate sorting by
throughput & probability"), the rate indexes are manipulated via u8's
and hence allow for a maximum of 256 mcs_group entries in
minstrel_mcs_groups.
ATM, minstrel_ht advertizes support up to 3HTSS@40MHz, consumi
From: Karl Beldan
ATM, we grep cck rates idx with idx / MCS_GROUP_RATES ==
MINSTREL_CCK_GROUP.
Matching neither-cck-non-ht rates could be done by replacing '==' with
'>', however it would be less versatile or explicit.
This will allow to match VHT rates with IEEE80211_TX_RC_VHT_MCS.
Signed-off-b
From: Karl Beldan
Hi,
This is just a respin rebased on v2 of "mac80211: minstrels: fix buffer
overflow in HT debugfs" which only affects 4/4.
Karl Beldan (4):
mac80211: minstrel_ht: Increase the range of handled rate indexes
mac80211: minstrel_ht: macros adjustments for future VHT_GROUPs
From: Karl Beldan
No functional change.
Signed-off-by: Karl Beldan
Acked-by: Felix Fietkau
---
net/mac80211/rc80211_minstrel_ht.c | 51 +++---
net/mac80211/rc80211_minstrel_ht.h | 14 ++--
net/mac80211/rc80211_minstrel_ht_debugfs.c | 10 +++---
3 fi
From: Karl Beldan
When the new CONFIG_MAC80211_RC_MINSTREL_VHT is not set (default 'N'),
there is no behavioral change including in sampling and MCS_GROUP_RATES
remains 8.
Otherwise MCS_GROUP_RATES is 10, and a module parameter *vht_only*
(default 'true'), restricts the rates selection to VHT whe
On 2014-10-20 09:58, Sujith Manoharan wrote:
> Felix Fietkau wrote:
>> -else if (AR_SREV_9340(ah))
>> -INIT_INI_ARRAY(&ah->iniModesTxGain,
>> -ar9340Modes_ub124_tx_gain_table_1p0);
>
> Shouldn't the array in ar9340_initvals.h also be removed ?
We can do that lat
From: Karl Beldan
ATM an HT rc_stats line is 106 chars.
Times 8(MCS_GROUP_RATES)*3(SS)*2(GI)*2(BW) + CCK(4), i.e. x100, this is
well above the current 8192 - sizeof(*ms) currently allocated.
Fix this by squeezing the output as follows (not that we're short on
memory but this also improves readab
On 2014-10-20 09:57, Sujith Manoharan wrote:
> Felix Fietkau wrote:
>> diff --git a/drivers/net/wireless/ath/ath9k/hw.h
>> b/drivers/net/wireless/ath/ath9k/hw.h
>> index 975074f..7a81f5b 100644
>> --- a/drivers/net/wireless/ath/ath9k/hw.h
>> +++ b/drivers/net/wireless/ath/ath9k/hw.h
>> @@ -244,13
Felix Fietkau wrote:
> The "goto chip_reset" is a bit misleading, because it does not actually
> issue a chip reset. Instead it is bypassing processing of other
> interrupts and assumes that the tasklet will issue a chip reset.
Well, we kill interrupts and the tasklet sets ATH_OP_HW_RESET,
so no m
Felix Fietkau wrote:
> + if (!ar9002_hw_per_calibration(ah, chan, rxchainmask, currCal))
> + return 0;
> +
> + ah->cal_list_curr = currCal = currCal->calNext;
> + if (currCal->calState == CAL_WAITING) {
> + ath9k_hw_reset_c
Felix Fietkau wrote:
> diff --git a/drivers/net/wireless/ath/ath9k/link.c
> b/drivers/net/wireless/ath/ath9k/link.c
> index 2343f56..713b040 100644
> --- a/drivers/net/wireless/ath/ath9k/link.c
> +++ b/drivers/net/wireless/ath/ath9k/link.c
> @@ -371,9 +371,14 @@ void ath_ani_calibrate(unsigned lon
Felix Fietkau wrote:
> ar9002_hw_pa_cal(ah, true);
> + ath9k_hw_start_nfcal(ah, true);
>
> if (ah->caldata)
> set_bit(NFCAL_PENDING, &ah->caldata->cal_flags);
> diff --git a/drivers/net/wireless/ath/ath9k/hw.c
> b/drivers/net/wireless/ath/ath9k/hw.c
> index 8be4b14.
Felix Fietkau wrote:
> - else if (AR_SREV_9340(ah))
> - INIT_INI_ARRAY(&ah->iniModesTxGain,
> - ar9340Modes_ub124_tx_gain_table_1p0);
Shouldn't the array in ar9340_initvals.h also be removed ?
Sujith
--
To unsubscribe from this list: send the line "unsubscribe
1 - 100 of 106 matches
Mail list logo
