* Re: ath10k: check power save support in STA mode through FW IE
From: Kalle Valo @ 2017-10-13 11:38 UTC (permalink / raw)
To: Kalle Valo; +Cc: ath10k, linux-wireless
In-Reply-To: <150670357817.24239.7065449335785410650.stgit@potku.adurom.net>
Kalle Valo <kvalo@qca.qualcomm.com> wrote:
> Currently ath10k host enables power save support in station mode by
> default for all firmwares but Power save for station mode still not supported
> in some of the firmware versions. Which results in firmware crash while
> issueing multiple scan commands.
>
> Fix this problem by introducing new FW feature flag to check power save
> support in firmware and then the firmware image can tell to ath10k that power
> save mode is not supported in station mode.
>
> Signed-off-by: Venkateswara Naralasetty <vnaralas@qti.qualcomm.com>
> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
Patch applied to ath-next branch of ath.git, thanks.
36d9cdb6fb4a ath10k: check power save support in STA mode through FW IE
--
https://patchwork.kernel.org/patch/9978455/
https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches
^ permalink raw reply
* Re: [v2] ath10k: add new cipher suite support
From: Kalle Valo @ 2017-10-13 11:40 UTC (permalink / raw)
To: Rajkumar Manoharan
Cc: ath10k, linux-wireless, rmanohar, Rajkumar Manoharan,
Jouni Malinen
In-Reply-To: <1506378581-13598-1-git-send-email-rmanohar@qti.qualcomm.com>
Rajkumar Manoharan <rmanohar@qti.qualcomm.com> wrote:
> QCA99x0 and QCA4019 family chips support CCMP-256, GCMP-128, and
> GCMP-256 ciphers in hardware, so advertise support for these. As
> firmware does not support group management frame ciphers (BIP),
> handle them in software (mac80211).
>
> Reviewed-by: Sebastian Gottschall <s.gottschall@dd-wrt.com>
> Cc: Jouni Malinen <jouni@qca.qualcomm.com>
> Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
Patch applied to ath-next branch of ath.git, thanks.
2ea9f12cefe4 ath10k: add new cipher suite support
--
https://patchwork.kernel.org/patch/9970613/
https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches
^ permalink raw reply
* Re: ath9k: fix tx99 potential info leak
From: Kalle Valo @ 2017-10-13 11:41 UTC (permalink / raw)
To: miaoqing pan; +Cc: linux-wireless, ath9k-devel, sssa, Miaoqing Pan
In-Reply-To: <1506474814-18118-1-git-send-email-miaoqing@codeaurora.org>
miaoqing pan <miaoqing@codeaurora.org> wrote:
> When the user sets count to zero the string buffer would remain
> completely uninitialized which causes the kernel to parse its
> own stack data, potentially leading to an info leak. In addition
> to that, the string might be not terminated properly when the
> user data does not contain a 0-terminator.
>
> Signed-off-by: Miaoqing Pan <miaoqing@codeaurora.org>
> Reviewed-by: Christoph Böhmwalder <christoph@boehmwalder.at>
> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
Patch applied to ath-next branch of ath.git, thanks.
ee0a47186e2f ath9k: fix tx99 potential info leak
--
https://patchwork.kernel.org/patch/9972889/
https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches
^ permalink raw reply
* pull-request: iwlwifi firmwares update 2017-10-13
From: Luca Coelho @ 2017-10-13 11:50 UTC (permalink / raw)
To: Kyle McMartin
Cc: linux-wireless, linuxwifi, linux-firmware, Grumbach, Emmanuel
[-- Attachment #1: Type: text/plain, Size: 3848 bytes --]
Hi Kyle,
I have updated the firmwares for the 3160, 3168, 7260, 7265D, 8000C and
8265 devices.
Please pull or let me know if there are any issues.
Cheers,
Luca.
The following changes since commit bf04291309d3169c0ad3b8db52564235bbd08e30:
WHENCE: Add new qed firmware (2017-10-09 18:03:26 +0100)
are available in the git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/iwlwifi/linux-firmware.git tags/iwlwifi-fw-2017-10-13
for you to fetch changes up to 97a3e95ecd6439790d2cfe14a32af7a21714800d:
iwlwifi: update firmwares for 3160, 3168, 7265D, 8000C and 8265 (2017-10-13 14:22:26 +0300)
----------------------------------------------------------------
Update firmwares for 3160, 3168, 7260, 7265D, 8000C and 8265
* update -17 for 3160 and 7260;
* update -29 for 3168 and 7265D;
* update -31 for 8000C and 8265.
----------------------------------------------------------------
Luca Coelho (1):
iwlwifi: update firmwares for 3160, 3168, 7265D, 8000C and 8265
WHENCE | 12 ++++++------
iwlwifi-3160-17.ucode | Bin 918268 -> 918268 bytes
iwlwifi-3168-29.ucode | Bin 1036372 -> 1036372 bytes
iwlwifi-7260-17.ucode | Bin 1049340 -> 1049340 bytes
iwlwifi-7265D-29.ucode | Bin 1036528 -> 1036528 bytes
iwlwifi-8000C-31.ucode | Bin 2309768 -> 2310116 bytes
iwlwifi-8265-31.ucode | Bin 2303024 -> 2307104 bytes
7 files changed, 6 insertions(+), 6 deletions(-)
diff --git a/WHENCE b/WHENCE
index b2dc6a10abb2..c8d1d4b4b15c 100644
--- a/WHENCE
+++ b/WHENCE
@@ -861,7 +861,7 @@ File: iwlwifi-7260-16.ucode
Version 16.242414.0
File: iwlwifi-7260-17.ucode
-Version 17.459231.0
+Version 17.589903.0
File: iwlwifi-3160-7.ucode
Version: 22.1.7.0
@@ -885,7 +885,7 @@ File: iwlwifi-3160-16.ucode
Version 16.242414.0
File: iwlwifi-3160-17.ucode
-Version 17.459231.0
+Version 17.589903.0
File: iwlwifi-7265-8.ucode
Version: 22.24.8.0
@@ -933,7 +933,7 @@ File: iwlwifi-7265D-27.ucode
Version 27.541033.0
File: iwlwifi-7265D-29.ucode
-Version 29.541020.0
+Version 29.588277.0
File: iwlwifi-3168-21.ucode
Version 21.302800.0
@@ -945,7 +945,7 @@ File: iwlwifi-3168-27.ucode
Version 27.541033.0
File: iwlwifi-3168-29.ucode
-Version 29.541020.0
+Version 29.588277.0
File: iwlwifi-8000C-13.ucode
Version: 25.30.13.0
@@ -963,7 +963,7 @@ File: iwlwifi-8000C-27.ucode
Version 27.541033.0
File: iwlwifi-8000C-31.ucode
-Version 31.532993.0
+Version 31.560484.0
File: iwlwifi-8265-21.ucode
Version 21.302800.0
@@ -975,7 +975,7 @@ File: iwlwifi-8265-27.ucode
Version 27.541033.0
File: iwlwifi-8265-31.ucode
-Version 31.532993.0
+Version 31.560484.0
Licence: Redistributable. See LICENCE.iwlwifi_firmware for details
diff --git a/iwlwifi-3160-17.ucode b/iwlwifi-3160-17.ucode
index 90c8d6205d31..fb22aa9dc436 100644
Binary files a/iwlwifi-3160-17.ucode and b/iwlwifi-3160-17.ucode differ
diff --git a/iwlwifi-3168-29.ucode b/iwlwifi-3168-29.ucode
index c9ba02305754..3e9a166803bb 100644
Binary files a/iwlwifi-3168-29.ucode and b/iwlwifi-3168-29.ucode differ
diff --git a/iwlwifi-7260-17.ucode b/iwlwifi-7260-17.ucode
index db7b4c20bf67..4379a8effdca 100644
Binary files a/iwlwifi-7260-17.ucode and b/iwlwifi-7260-17.ucode differ
diff --git a/iwlwifi-7265D-29.ucode b/iwlwifi-7265D-29.ucode
index 711c170f6089..bcb6eafedbe6 100644
Binary files a/iwlwifi-7265D-29.ucode and b/iwlwifi-7265D-29.ucode differ
diff --git a/iwlwifi-8000C-31.ucode b/iwlwifi-8000C-31.ucode
index 00812b5c31e0..9b41e50b61ed 100644
Binary files a/iwlwifi-8000C-31.ucode and b/iwlwifi-8000C-31.ucode differ
diff --git a/iwlwifi-8265-31.ucode b/iwlwifi-8265-31.ucode
index f6cace11135c..608ec749ffe8 100644
Binary files a/iwlwifi-8265-31.ucode and b/iwlwifi-8265-31.ucode differ
[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 833 bytes --]
^ permalink raw reply related
* [PATCH 1/2] cfg80211: don't print log output for building shipped-certs
From: Arnd Bergmann @ 2017-10-13 12:04 UTC (permalink / raw)
To: Johannes Berg
Cc: Arnd Bergmann, David S. Miller, linux-wireless, netdev,
linux-kernel
Building an allmodconfig kernel with 'make -s' now prints a single line:
GEN net/wireless/shipped-certs.c
Using '$(kecho)' here will skip the output with 'make -s' but
otherwise keeps printing it, which is consistent with how we
handle all the other output.
Fixes: 90a53e4432b1 ("cfg80211: implement regdb signature checking")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
net/wireless/Makefile | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/net/wireless/Makefile b/net/wireless/Makefile
index 219baea57e4e..e585f3f71f77 100644
--- a/net/wireless/Makefile
+++ b/net/wireless/Makefile
@@ -23,7 +23,7 @@ cfg80211-y += extra-certs.o
endif
$(obj)/shipped-certs.c: $(wildcard $(srctree)/$(src)/certs/*.x509)
- @echo " GEN $@"
+ @$(kecho) " GEN $@"
@echo '#include "reg.h"' > $@
@echo 'const u8 shipped_regdb_certs[] = {' >> $@
@for f in $^ ; do hexdump -v -e '1/1 "0x%.2x," "\n"' < $$f >> $@ ; done
@@ -32,7 +32,7 @@ $(obj)/shipped-certs.c: $(wildcard $(srctree)/$(src)/certs/*.x509)
$(obj)/extra-certs.c: $(CONFIG_CFG80211_EXTRA_REGDB_KEYDIR:"%"=%) \
$(wildcard $(CONFIG_CFG80211_EXTRA_REGDB_KEYDIR:"%"=%)/*.x509)
- @echo " GEN $@"
+ @$(kecho) " GEN $@"
@echo '#include "reg.h"' > $@
@echo 'const u8 extra_regdb_certs[] = {' >> $@
@for f in $^ ; do test -f $$f && hexdump -v -e '1/1 "0x%.2x," "\n"' < $$f >> $@ || true ; done
--
2.9.0
^ permalink raw reply related
* [PATCH 2/2] cfg80211: fix CFG80211_EXTRA_REGDB_KEYDIR typo
From: Arnd Bergmann @ 2017-10-13 12:04 UTC (permalink / raw)
To: Johannes Berg
Cc: Arnd Bergmann, David S. Miller, linux-wireless, netdev,
linux-kernel
In-Reply-To: <20171013120446.4079666-1-arnd@arndb.de>
The missing CONFIG_ prefix means this macro is never defined,
leading to a possible Kbuild warning:
net/wireless/reg.c:666:20: error: 'load_keys_from_buffer' defined but not used [-Werror=unused-function]
static void __init load_keys_from_buffer(const u8 *p, unsigned int buflen)
When we use the correct symbol, the warning also goes away.
Fixes: 90a53e4432b1 ("cfg80211: implement regdb signature checking")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
net/wireless/reg.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/net/wireless/reg.c b/net/wireless/reg.c
index 58319c82ecb3..3871998059de 100644
--- a/net/wireless/reg.c
+++ b/net/wireless/reg.c
@@ -723,7 +723,7 @@ static int __init load_builtin_regdb_keys(void)
#ifdef CONFIG_CFG80211_USE_KERNEL_REGDB_KEYS
load_keys_from_buffer(shipped_regdb_certs, shipped_regdb_certs_len);
#endif
-#ifdef CFG80211_EXTRA_REGDB_KEYDIR
+#ifdef CONFIG_CFG80211_EXTRA_REGDB_KEYDIR
if (CONFIG_CFG80211_EXTRA_REGDB_KEYDIR[0] != '\0')
load_keys_from_buffer(extra_regdb_certs, extra_regdb_certs_len);
#endif
--
2.9.0
^ permalink raw reply related
* Re: ath10k: Store coverage-class in case firmware is not booted.
From: Kalle Valo @ 2017-10-13 12:08 UTC (permalink / raw)
To: Ben Greear; +Cc: linux-wireless, ath10k, Ben Greear
In-Reply-To: <1507068707-14050-1-git-send-email-greearb@candelatech.com>
Ben Greear <greearb@candelatech.com> wrote:
> This way, we can apply the values when the NIC does come up.
>
> Signed-off-by: Ben Greear <greearb@candelatech.com>
> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
Patch applied to ath-next branch of ath.git, thanks.
0a590a38778d ath10k: store coverage-class in case firmware is not booted
--
https://patchwork.kernel.org/patch/9983585/
https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches
^ permalink raw reply
* Re: [PATCH 2/2] cfg80211: fix CFG80211_EXTRA_REGDB_KEYDIR typo
From: Johannes Berg @ 2017-10-13 12:12 UTC (permalink / raw)
To: Arnd Bergmann; +Cc: David S. Miller, linux-wireless, netdev, linux-kernel
In-Reply-To: <20171013120446.4079666-2-arnd@arndb.de>
Thanks, both applied.
johannes
^ permalink raw reply
* Re: [2/2] ath10k: handle tdls peer events
From: Kalle Valo @ 2017-10-13 12:18 UTC (permalink / raw)
To: mpubbise; +Cc: ath10k, linux-wireless, Manikanta Pubbisetty
In-Reply-To: <1507618202-25473-3-git-send-email-mpubbise@qti.qualcomm.com>
mpubbise@qti.qualcomm.com wrote:
> Handle tdls peer events from the target. TDLS events for the peer
> could be discover, teardown, etc. As of now, adding the logic to
> handle tdls teardown events alone.
>
> Teardown due to peer traffic indication(PTR) timeout is one such
> teardown event from the target.
>
> Tested this change on QCA9888 with 10.4-3.5.1-00018 fw version.
>
> Signed-off-by: Manikanta Pubbisetty <mpubbise@qti.qualcomm.com>
> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
This patch introduces new warnings:
$ ath10k-check
drivers/net/wireless/ath/ath10k/wmi.c:4470:39: warning: incorrect type in argument 2 (different base types)
drivers/net/wireless/ath/ath10k/wmi.c:4470:39: expected int [signed] vdev_id
drivers/net/wireless/ath/ath10k/wmi.c:4470:39: got restricted __le32 [usertype] vdev_id
drivers/net/wireless/ath/ath10k/wmi.c:4481:27: warning: restricted __le32 degrades to integer
drivers/net/wireless/ath/ath10k/wmi.c:4481:27: warning: restricted __le32 degrades to integer
drivers/net/wireless/ath/ath10k/wmi.c:4481:27: warning: restricted __le32 degrades to integer
drivers/net/wireless/ath/ath10k/wmi.c:4492:48: warning: incorrect type in argument 2 (different base types)
drivers/net/wireless/ath/ath10k/wmi.c:4492:48: expected unsigned int [unsigned] [usertype] vdev_id
drivers/net/wireless/ath/ath10k/wmi.c:4492:48: got restricted __le32 [usertype] vdev_id
drivers/net/wireless/ath/ath10k/wmi.c:4479:19: warning: restricted __le32 degrades to integer
drivers/net/wireless/ath/ath10k/wmi.c:4460:6: warning: symbol 'ath10k_wmi_handle_tdls_peer_event' was not declared. Should it be static?
--
https://patchwork.kernel.org/patch/9995131/
https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches
^ permalink raw reply
* [PATCH] mac80211: don't track HT capability changes
From: Luca Coelho @ 2017-10-13 12:26 UTC (permalink / raw)
To: johannes; +Cc: linux-wireless, Johannes Berg
From: Johannes Berg <johannes.berg@intel.com>
The code here (more or less accidentally) tracks the HT capability of
the AP when connected, and we found at least one AP that erroneously
toggles its 20/40 capability bit when changing between 20/40 MHz. The
connection to the AP is then broken because we set the 40 MHz disable
flag based on this, as soon as it switches to 20 MHz, but because the
flag then changed, we disconnect.
I'd be inclined to just ignore this issue, since we then reconnect
while the AP is in 20 MHz mode and never use 40 MHz with it again,
but this code is a bit strange anyway - we don't use the capabilities
for anything else.
Change the code to simply not track the HT capabilities at all, which
assumes that the AP at least sets 20/40 capability when operating in
40 MHz (or higher). If not, rate scaling might end up using only the
narrower bandwidth.
The new behaviour also mirrors what VHT does, where we only check the
VHT operation.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
net/mac80211/mlme.c | 14 +++-----------
1 file changed, 3 insertions(+), 11 deletions(-)
diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index ee5ca1bc5a20..e4ededa1909d 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -145,7 +145,6 @@ static u32
ieee80211_determine_chantype(struct ieee80211_sub_if_data *sdata,
struct ieee80211_supported_band *sband,
struct ieee80211_channel *channel,
- const struct ieee80211_ht_cap *ht_cap,
const struct ieee80211_ht_operation *ht_oper,
const struct ieee80211_vht_operation *vht_oper,
struct cfg80211_chan_def *chandef, bool tracking)
@@ -163,20 +162,13 @@ ieee80211_determine_chantype(struct ieee80211_sub_if_data *sdata,
chandef->center_freq1 = channel->center_freq;
chandef->center_freq2 = 0;
- if (!ht_cap || !ht_oper || !sta_ht_cap.ht_supported) {
+ if (!ht_oper || !sta_ht_cap.ht_supported) {
ret = IEEE80211_STA_DISABLE_HT | IEEE80211_STA_DISABLE_VHT;
goto out;
}
chandef->width = NL80211_CHAN_WIDTH_20;
- if (!(ht_cap->cap_info &
- cpu_to_le16(IEEE80211_HT_CAP_SUP_WIDTH_20_40))) {
- ret = IEEE80211_STA_DISABLE_40MHZ;
- vht_chandef = *chandef;
- goto out;
- }
-
ht_cfreq = ieee80211_channel_to_frequency(ht_oper->primary_chan,
channel->band);
/* check that channel matches the right operating channel */
@@ -344,7 +336,7 @@ static int ieee80211_config_bw(struct ieee80211_sub_if_data *sdata,
/* calculate new channel (type) based on HT/VHT operation IEs */
flags = ieee80211_determine_chantype(sdata, sband, chan,
- ht_cap, ht_oper, vht_oper,
+ ht_oper, vht_oper,
&chandef, true);
/*
@@ -4312,7 +4304,7 @@ static int ieee80211_prep_channel(struct ieee80211_sub_if_data *sdata,
ifmgd->flags |= ieee80211_determine_chantype(sdata, sband,
cbss->channel,
- ht_cap, ht_oper, vht_oper,
+ ht_oper, vht_oper,
&chandef, false);
sdata->needed_rx_chains = min(ieee80211_ht_vht_rx_chains(sdata, cbss),
--
2.14.2
^ permalink raw reply related
* Re: [PATCH] ath10k: spectral: Simplify error checking
From: Kalle Valo @ 2017-10-13 12:28 UTC (permalink / raw)
To: Christos Gkekas
Cc: ath10k@lists.infradead.org, linux-wireless@vger.kernel.org,
netdev@vger.kernel.org, linux-kernel@vger.kernel.org
In-Reply-To: <1507491026-28690-1-git-send-email-chris.gekas@gmail.com>
Christos Gkekas <chris.gekas@gmail.com> writes:
> Variable val is unsigned so checking whether it is less than zero is
> redundant.
>
> Signed-off-by: Christos Gkekas <chris.gekas@gmail.com>
> ---
> drivers/net/wireless/ath/ath10k/spectral.c | 5 +----
> 1 file changed, 1 insertion(+), 4 deletions(-)
>
> diff --git a/drivers/net/wireless/ath/ath10k/spectral.c b/drivers/net/wir=
eless/ath/ath10k/spectral.c
> index dd9cc09..1867937 100644
> --- a/drivers/net/wireless/ath/ath10k/spectral.c
> +++ b/drivers/net/wireless/ath/ath10k/spectral.c
> @@ -403,10 +403,7 @@ static ssize_t write_file_spectral_count(struct file=
*file,
> return -EFAULT;
> =20
> buf[len] =3D '\0';
> - if (kstrtoul(buf, 0, &val))
> - return -EINVAL;
> -
> - if (val < 0 || val > 255)
> + if (kstrtoul(buf, 0, &val) || val > 255)
> return -EINVAL;
Removing the check for negative is correct but I don't think you are
simplifying anything, on the contrary it's harder to read. Please keep
the two if statements separate.
--=20
Kalle Valo=
^ permalink raw reply
* Re: [PATCH v2] ath10k: Retry pci probe on failure.
From: Kalle Valo @ 2017-10-13 12:41 UTC (permalink / raw)
To: greearb@candelatech.com
Cc: linux-wireless@vger.kernel.org, ath10k@lists.infradead.org
In-Reply-To: <1507068826-14677-1-git-send-email-greearb@candelatech.com>
greearb@candelatech.com writes:
> From: Ben Greear <greearb@candelatech.com>
>
> This works around a problem we see when sometimes the wifi NIC does
> not respond the first time. This seems to happen especially often on
> some of the 9984 NICs in mid-range platforms.
>
> Signed-off-by: Ben Greear <greearb@candelatech.com>
[...]
> -static int ath10k_pci_probe(struct pci_dev *pdev,
> - const struct pci_device_id *pci_dev)
> +static int __ath10k_pci_probe(struct pci_dev *pdev,
> + const struct pci_device_id *pci_dev)
> {
> int ret =3D 0;
> struct ath10k *ar;
> @@ -3672,6 +3672,22 @@ static int ath10k_pci_probe(struct pci_dev *pdev,
> return ret;
> }
> =20
> +static int ath10k_pci_probe(struct pci_dev *pdev,
> + const struct pci_device_id *pci_dev)
> +{
> + int cnt =3D 0;
> + int rv;
> + do {
> + rv =3D __ath10k_pci_probe(pdev, pci_dev);
> + if (rv =3D=3D 0)
> + return rv;
> + pr_err("ath10k: failed to probe PCI : %d, retry-count: %d\n", rv, cnt)=
;
> + mdelay(10); /* let the ath10k firmware gerbil take a small break */
> + } while (cnt++ < 10);
> + return rv;
> +}
This is a sledgehammer approach and it causes reload for all error
cases, like when hardware is broken or memory allocation is failing.
When the problem happens does it always fail at the the same place? Is
it hw reset or something else? It's better to retry the invidiual action
than to do this hack. Or is it just some more delay needed somewhere?
--=20
Kalle Valo=
^ permalink raw reply
* Re: [PATCH] ath9k: debug: Simplify error checking
From: Kalle Valo @ 2017-10-13 12:49 UTC (permalink / raw)
To: Christos Gkekas
Cc: QCA ath9k Development, linux-wireless, netdev, linux-kernel
In-Reply-To: <1507498152-4058-1-git-send-email-chris.gekas@gmail.com>
Christos Gkekas <chris.gekas@gmail.com> writes:
> Variable val is unsigned so checking whether it is less than zero is
> redundant.
>
> Signed-off-by: Christos Gkekas <chris.gekas@gmail.com>
> ---
> drivers/net/wireless/ath/ath9k/debug.c | 5 +----
> 1 file changed, 1 insertion(+), 4 deletions(-)
>
> diff --git a/drivers/net/wireless/ath/ath9k/debug.c b/drivers/net/wireless/ath/ath9k/debug.c
> index 01fa301..3b93c23 100644
> --- a/drivers/net/wireless/ath/ath9k/debug.c
> +++ b/drivers/net/wireless/ath/ath9k/debug.c
> @@ -1164,10 +1164,7 @@ static ssize_t write_file_tpc(struct file *file, const char __user *user_buf,
> return -EFAULT;
>
> buf[len] = '\0';
> - if (kstrtoul(buf, 0, &val))
> - return -EINVAL;
> -
> - if (val < 0 || val > 1)
> + if (kstrtoul(buf, 0, &val) || val > 1)
> return -EINVAL;
Same as with the ath10k patch, please keep the two if statements
separate.
--
Kalle Valo
^ permalink raw reply
* [PATCH 0/9] ] iwlwifi: updates intended for v4.15 2017-10-13
From: Luca Coelho @ 2017-10-13 13:24 UTC (permalink / raw)
To: kvalo; +Cc: linux-wireless, Luca Coelho
From: Luca Coelho <luciano.coelho@intel.com>
Hi,
Here's the third set of patches intended for 4.15. A few small fixes,
but mostly just code cleanups.
These are the changes:
* Allocate reorder buffer dynamically to save memory;
* Fix a FW dump problem in the A000 family;
* Fix for a statistics gathering issue;
* Sort the list of 9000 devices to make it easier to find entries;
* A couple of cleanups in the FW dump code;
* Remove some unnecessary variables and fields and calculations;
As usual, I'm pushing this to a pending branch, for kbuild bot, and
will send a pull-request later.
Please review.
Cheers,
Luca.
Beni Lev (1):
iwlwifi: mvm: allow reading UMAC error data from SMEM in A000 devices
Johannes Berg (3):
iwlwifi: mvm: allocate reorder buffer according to need
iwlwifi: mvm: pass baid_data to iwl_mvm_release_frames()
iwlwifi: pcie: remove set but not used variable tcph
Liad Kaufman (1):
iwlwifi: mvm: add missing lq_color
Luca Coelho (3):
iwlwifi: mvm: move umac_error_event_table validity check to where it's
set
iwlwifi: define minimum valid address for umac_error_event_table in
cfg
iwlwifi: pcie: sort IDs for the 9000 series for easier comparisons
Sara Sharon (1):
iwlwifi: mvm: remove duplicated fields in mvm reorder buffer
drivers/net/wireless/intel/iwlwifi/cfg/8000.c | 3 +-
drivers/net/wireless/intel/iwlwifi/cfg/9000.c | 3 +-
drivers/net/wireless/intel/iwlwifi/cfg/a000.c | 3 +-
drivers/net/wireless/intel/iwlwifi/fw/api/tx.h | 4 +-
drivers/net/wireless/intel/iwlwifi/iwl-config.h | 1 +
drivers/net/wireless/intel/iwlwifi/mvm/fw.c | 20 +++++-
drivers/net/wireless/intel/iwlwifi/mvm/mvm.h | 43 +++++++++---
drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c | 49 ++++++++-----
drivers/net/wireless/intel/iwlwifi/mvm/sta.c | 44 +++++++++---
drivers/net/wireless/intel/iwlwifi/mvm/tx.c | 3 +-
drivers/net/wireless/intel/iwlwifi/mvm/utils.c | 17 ++---
drivers/net/wireless/intel/iwlwifi/pcie/drv.c | 84 +++++++++++------------
drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c | 5 +-
13 files changed, 176 insertions(+), 103 deletions(-)
--
2.14.2
^ permalink raw reply
* [PATCH 1/9] iwlwifi: mvm: allocate reorder buffer according to need
From: Luca Coelho @ 2017-10-13 13:24 UTC (permalink / raw)
To: kvalo; +Cc: linux-wireless, Johannes Berg, Luca Coelho
In-Reply-To: <20171013132502.10473-1-luca@coelho.fi>
From: Johannes Berg <johannes.berg@intel.com>
Now that we may have up to 256 entries per reorder buffer, and possibly up
to 16 queues, we can use a LOT of memory for this (64k for each station).
Allocate it according to what we need, which is of course much less for HT
stations (only 16k at a max of 16 queues).
However, this comes at the expense of complicating the code a bit to
calculate the right entry structure to use for each frame.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
---
drivers/net/wireless/intel/iwlwifi/mvm/mvm.h | 39 +++++++++++++++++++++++----
drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c | 30 ++++++++++++++-------
drivers/net/wireless/intel/iwlwifi/mvm/sta.c | 38 +++++++++++++++++++++++---
3 files changed, 90 insertions(+), 17 deletions(-)
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 7fa7849367ef..f008d5f4c585 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -590,8 +590,6 @@ enum iwl_mvm_tdls_cs_state {
* @last_amsdu: track last ASMDU SN for duplication detection
* @last_sub_index: track ASMDU sub frame index for duplication detection
* @tid: the tid
- * @entries: list of skbs stored
- * @reorder_time: time the packet was stored in the reorder buffer
* @reorder_timer: timer for frames are in the reorder buffer. For AMSDU
* it is the time of last received sub-frame
* @removed: prevent timer re-arming
@@ -608,8 +606,6 @@ struct iwl_mvm_reorder_buffer {
u16 last_amsdu;
u8 last_sub_index;
u8 tid;
- struct sk_buff_head entries[IEEE80211_MAX_AMPDU_BUF];
- unsigned long reorder_time[IEEE80211_MAX_AMPDU_BUF];
struct timer_list reorder_timer;
bool removed;
bool valid;
@@ -617,16 +613,39 @@ struct iwl_mvm_reorder_buffer {
struct iwl_mvm *mvm;
} ____cacheline_aligned_in_smp;
+/**
+ * struct _iwl_mvm_reorder_buf_entry - reorder buffer entry per-queue/per-seqno
+ * @frames: list of skbs stored
+ * @reorder_time: time the packet was stored in the reorder buffer
+ */
+struct _iwl_mvm_reorder_buf_entry {
+ struct sk_buff_head frames;
+ unsigned long reorder_time;
+};
+
+/* make this indirection to get the aligned thing */
+struct iwl_mvm_reorder_buf_entry {
+ struct _iwl_mvm_reorder_buf_entry e;
+}
+#ifndef __CHECKER__
+/* sparse doesn't like this construct: "bad integer constant expression" */
+__aligned(roundup_pow_of_two(sizeof(struct _iwl_mvm_reorder_buf_entry)))
+#endif
+;
+
/**
* struct iwl_mvm_baid_data - BA session data
* @sta_id: station id
* @tid: tid of the session
* @baid baid of the session
* @timeout: the timeout set in the addba request
+ * @entries_per_queue: # of buffers per queue, this actually gets
+ * aligned up to avoid cache line sharing between queues
* @last_rx: last rx jiffies, updated only if timeout passed from last update
* @session_timer: timer to check if BA session expired, runs at 2 * timeout
* @mvm: mvm pointer, needed for timer context
* @reorder_buf: reorder buffer, allocated per queue
+ * @reorder_buf_data: data
*/
struct iwl_mvm_baid_data {
struct rcu_head rcu_head;
@@ -634,12 +653,22 @@ struct iwl_mvm_baid_data {
u8 tid;
u8 baid;
u16 timeout;
+ u16 entries_per_queue;
unsigned long last_rx;
struct timer_list session_timer;
struct iwl_mvm *mvm;
- struct iwl_mvm_reorder_buffer reorder_buf[];
+ struct iwl_mvm_reorder_buffer reorder_buf[IWL_MAX_RX_HW_QUEUES];
+ struct iwl_mvm_reorder_buf_entry entries[];
};
+static inline struct iwl_mvm_baid_data *
+iwl_mvm_baid_data_from_reorder_buf(struct iwl_mvm_reorder_buffer *buf)
+{
+ return (void *)((u8 *)buf -
+ offsetof(struct iwl_mvm_baid_data, reorder_buf) -
+ sizeof(*buf) * buf->queue);
+}
+
/*
* enum iwl_mvm_queue_status - queue status
* @IWL_MVM_QUEUE_FREE: the queue is not allocated nor reserved
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
index 836c6cf4b369..14c2d366f256 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
@@ -410,6 +410,11 @@ static void iwl_mvm_release_frames(struct iwl_mvm *mvm,
struct iwl_mvm_reorder_buffer *reorder_buf,
u16 nssn)
{
+ struct iwl_mvm_baid_data *baid_data =
+ iwl_mvm_baid_data_from_reorder_buf(reorder_buf);
+ struct iwl_mvm_reorder_buf_entry *entries =
+ &baid_data->entries[reorder_buf->queue *
+ baid_data->entries_per_queue];
u16 ssn = reorder_buf->head_sn;
lockdep_assert_held(&reorder_buf->lock);
@@ -420,7 +425,7 @@ static void iwl_mvm_release_frames(struct iwl_mvm *mvm,
while (iwl_mvm_is_sn_less(ssn, nssn, reorder_buf->buf_size)) {
int index = ssn % reorder_buf->buf_size;
- struct sk_buff_head *skb_list = &reorder_buf->entries[index];
+ struct sk_buff_head *skb_list = &entries[index].e.frames;
struct sk_buff *skb;
ssn = ieee80211_sn_inc(ssn);
@@ -443,11 +448,11 @@ static void iwl_mvm_release_frames(struct iwl_mvm *mvm,
if (reorder_buf->num_stored && !reorder_buf->removed) {
u16 index = reorder_buf->head_sn % reorder_buf->buf_size;
- while (skb_queue_empty(&reorder_buf->entries[index]))
+ while (skb_queue_empty(&entries[index].e.frames))
index = (index + 1) % reorder_buf->buf_size;
/* modify timer to match next frame's expiration time */
mod_timer(&reorder_buf->reorder_timer,
- reorder_buf->reorder_time[index] + 1 +
+ entries[index].e.reorder_time + 1 +
RX_REORDER_BUF_TIMEOUT_MQ);
} else {
del_timer(&reorder_buf->reorder_timer);
@@ -457,6 +462,10 @@ static void iwl_mvm_release_frames(struct iwl_mvm *mvm,
void iwl_mvm_reorder_timer_expired(unsigned long data)
{
struct iwl_mvm_reorder_buffer *buf = (void *)data;
+ struct iwl_mvm_baid_data *baid_data =
+ iwl_mvm_baid_data_from_reorder_buf(buf);
+ struct iwl_mvm_reorder_buf_entry *entries =
+ &baid_data->entries[buf->queue * baid_data->entries_per_queue];
int i;
u16 sn = 0, index = 0;
bool expired = false;
@@ -472,7 +481,7 @@ void iwl_mvm_reorder_timer_expired(unsigned long data)
for (i = 0; i < buf->buf_size ; i++) {
index = (buf->head_sn + i) % buf->buf_size;
- if (skb_queue_empty(&buf->entries[index])) {
+ if (skb_queue_empty(&entries[index].e.frames)) {
/*
* If there is a hole and the next frame didn't expire
* we want to break and not advance SN
@@ -480,7 +489,8 @@ void iwl_mvm_reorder_timer_expired(unsigned long data)
cont = false;
continue;
}
- if (!cont && !time_after(jiffies, buf->reorder_time[index] +
+ if (!cont &&
+ !time_after(jiffies, entries[index].e.reorder_time +
RX_REORDER_BUF_TIMEOUT_MQ))
break;
@@ -513,7 +523,7 @@ void iwl_mvm_reorder_timer_expired(unsigned long data)
* accordingly to this frame.
*/
mod_timer(&buf->reorder_timer,
- buf->reorder_time[index] +
+ entries[index].e.reorder_time +
1 + RX_REORDER_BUF_TIMEOUT_MQ);
}
spin_unlock(&buf->lock);
@@ -608,6 +618,7 @@ static bool iwl_mvm_reorder(struct iwl_mvm *mvm,
u8 tid = *ieee80211_get_qos_ctl(hdr) & IEEE80211_QOS_CTL_TID_MASK;
u8 sub_frame_idx = desc->amsdu_info &
IWL_RX_MPDU_AMSDU_SUBFRAME_IDX_MASK;
+ struct iwl_mvm_reorder_buf_entry *entries;
int index;
u16 nssn, sn;
u8 baid;
@@ -658,6 +669,7 @@ static bool iwl_mvm_reorder(struct iwl_mvm *mvm,
IWL_RX_MPDU_REORDER_SN_SHIFT;
buffer = &baid_data->reorder_buf[queue];
+ entries = &baid_data->entries[queue * baid_data->entries_per_queue];
spin_lock_bh(&buffer->lock);
@@ -713,7 +725,7 @@ static bool iwl_mvm_reorder(struct iwl_mvm *mvm,
* If it is the same SN then if the subframe index is incrementing it
* is the same AMSDU - otherwise it is a retransmission.
*/
- tail = skb_peek_tail(&buffer->entries[index]);
+ tail = skb_peek_tail(&entries[index].e.frames);
if (tail && !amsdu)
goto drop;
else if (tail && (sn != buffer->last_amsdu ||
@@ -721,9 +733,9 @@ static bool iwl_mvm_reorder(struct iwl_mvm *mvm,
goto drop;
/* put in reorder buffer */
- __skb_queue_tail(&buffer->entries[index], skb);
+ __skb_queue_tail(&entries[index].e.frames, skb);
buffer->num_stored++;
- buffer->reorder_time[index] = jiffies;
+ entries[index].e.reorder_time = jiffies;
if (amsdu) {
buffer->last_amsdu = sn;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
index 3711f226220c..a26cf8e89036 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
@@ -2104,6 +2104,8 @@ static void iwl_mvm_free_reorder(struct iwl_mvm *mvm,
int j;
struct iwl_mvm_reorder_buffer *reorder_buf =
&data->reorder_buf[i];
+ struct iwl_mvm_reorder_buf_entry *entries =
+ &data->entries[i * data->entries_per_queue];
spin_lock_bh(&reorder_buf->lock);
if (likely(!reorder_buf->num_stored)) {
@@ -2119,7 +2121,7 @@ static void iwl_mvm_free_reorder(struct iwl_mvm *mvm,
WARN_ON(1);
for (j = 0; j < reorder_buf->buf_size; j++)
- __skb_queue_purge(&reorder_buf->entries[j]);
+ __skb_queue_purge(&entries[j].e.frames);
/*
* Prevent timer re-arm. This prevents a very far fetched case
* where we timed out on the notification. There may be prior
@@ -2144,6 +2146,8 @@ static void iwl_mvm_init_reorder_buffer(struct iwl_mvm *mvm,
for (i = 0; i < mvm->trans->num_rx_queues; i++) {
struct iwl_mvm_reorder_buffer *reorder_buf =
&data->reorder_buf[i];
+ struct iwl_mvm_reorder_buf_entry *entries =
+ &data->entries[i * data->entries_per_queue];
int j;
reorder_buf->num_stored = 0;
@@ -2161,7 +2165,7 @@ static void iwl_mvm_init_reorder_buffer(struct iwl_mvm *mvm,
reorder_buf->tid = data->tid;
reorder_buf->valid = false;
for (j = 0; j < reorder_buf->buf_size; j++)
- __skb_queue_head_init(&reorder_buf->entries[j]);
+ __skb_queue_head_init(&entries[j].e.frames);
}
}
@@ -2182,16 +2186,44 @@ int iwl_mvm_sta_rx_agg(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
}
if (iwl_mvm_has_new_rx_api(mvm) && start) {
+ u16 reorder_buf_size = buf_size * sizeof(baid_data->entries[0]);
+
+ /* sparse doesn't like the __align() so don't check */
+#ifndef __CHECKER__
+ /*
+ * The division below will be OK if either the cache line size
+ * can be divided by the entry size (ALIGN will round up) or if
+ * if the entry size can be divided by the cache line size, in
+ * which case the ALIGN() will do nothing.
+ */
+ BUILD_BUG_ON(SMP_CACHE_BYTES % sizeof(baid_data->entries[0]) &&
+ sizeof(baid_data->entries[0]) % SMP_CACHE_BYTES);
+#endif
+
+ /*
+ * Upward align the reorder buffer size to fill an entire cache
+ * line for each queue, to avoid sharing cache lines between
+ * different queues.
+ */
+ reorder_buf_size = ALIGN(reorder_buf_size, SMP_CACHE_BYTES);
+
/*
* Allocate here so if allocation fails we can bail out early
* before starting the BA session in the firmware
*/
baid_data = kzalloc(sizeof(*baid_data) +
mvm->trans->num_rx_queues *
- sizeof(baid_data->reorder_buf[0]),
+ reorder_buf_size,
GFP_KERNEL);
if (!baid_data)
return -ENOMEM;
+
+ /*
+ * This division is why we need the above BUILD_BUG_ON(),
+ * if that doesn't hold then this will not be right.
+ */
+ baid_data->entries_per_queue =
+ reorder_buf_size / sizeof(baid_data->entries[0]);
}
cmd.mac_id_n_color = cpu_to_le32(mvm_sta->mac_id_n_color);
--
2.14.2
^ permalink raw reply related
* [PATCH 3/9] iwlwifi: mvm: pass baid_data to iwl_mvm_release_frames()
From: Luca Coelho @ 2017-10-13 13:24 UTC (permalink / raw)
To: kvalo; +Cc: linux-wireless, Johannes Berg, Luca Coelho
In-Reply-To: <20171013132502.10473-1-luca@coelho.fi>
From: Johannes Berg <johannes.berg@intel.com>
All callers of iwl_mvm_release_frames() already have the baid_data
pointer, so we don't need to (re)calculate it inside the function.
Just pass it instead.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
---
drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c | 16 ++++++++--------
1 file changed, 8 insertions(+), 8 deletions(-)
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
index fea7b965ea49..8e4f3b0a12cd 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
@@ -407,11 +407,10 @@ static bool iwl_mvm_is_sn_less(u16 sn1, u16 sn2, u16 buffer_size)
static void iwl_mvm_release_frames(struct iwl_mvm *mvm,
struct ieee80211_sta *sta,
struct napi_struct *napi,
+ struct iwl_mvm_baid_data *baid_data,
struct iwl_mvm_reorder_buffer *reorder_buf,
u16 nssn)
{
- struct iwl_mvm_baid_data *baid_data =
- iwl_mvm_baid_data_from_reorder_buf(reorder_buf);
struct iwl_mvm_reorder_buf_entry *entries =
&baid_data->entries[reorder_buf->queue *
baid_data->entries_per_queue];
@@ -515,7 +514,7 @@ void iwl_mvm_reorder_timer_expired(unsigned long data)
sta_id, sn);
iwl_mvm_event_frame_timeout_callback(buf->mvm, mvmsta->vif,
sta, baid_data->tid);
- iwl_mvm_release_frames(buf->mvm, sta, NULL, buf, sn);
+ iwl_mvm_release_frames(buf->mvm, sta, NULL, baid_data, buf, sn);
rcu_read_unlock();
} else {
/*
@@ -555,7 +554,7 @@ static void iwl_mvm_del_ba(struct iwl_mvm *mvm, int queue,
/* release all frames that are in the reorder buffer to the stack */
spin_lock_bh(&reorder_buf->lock);
- iwl_mvm_release_frames(mvm, sta, NULL, reorder_buf,
+ iwl_mvm_release_frames(mvm, sta, NULL, ba_data, reorder_buf,
ieee80211_sn_add(reorder_buf->head_sn,
reorder_buf->buf_size));
spin_unlock_bh(&reorder_buf->lock);
@@ -683,7 +682,7 @@ static bool iwl_mvm_reorder(struct iwl_mvm *mvm,
}
if (ieee80211_is_back_req(hdr->frame_control)) {
- iwl_mvm_release_frames(mvm, sta, napi, buffer, nssn);
+ iwl_mvm_release_frames(mvm, sta, napi, baid_data, buffer, nssn);
goto drop;
}
@@ -698,7 +697,8 @@ static bool iwl_mvm_reorder(struct iwl_mvm *mvm,
buffer->buf_size)) {
u16 min_sn = ieee80211_sn_less(sn, nssn) ? sn : nssn;
- iwl_mvm_release_frames(mvm, sta, napi, buffer, min_sn);
+ iwl_mvm_release_frames(mvm, sta, napi, baid_data, buffer,
+ min_sn);
}
/* drop any oudated packets */
@@ -755,7 +755,7 @@ static bool iwl_mvm_reorder(struct iwl_mvm *mvm,
* release notification with up to date NSSN.
*/
if (!amsdu || last_subframe)
- iwl_mvm_release_frames(mvm, sta, napi, buffer, nssn);
+ iwl_mvm_release_frames(mvm, sta, napi, baid_data, buffer, nssn);
spin_unlock_bh(&buffer->lock);
return true;
@@ -1075,7 +1075,7 @@ void iwl_mvm_rx_frame_release(struct iwl_mvm *mvm, struct napi_struct *napi,
reorder_buf = &ba_data->reorder_buf[queue];
spin_lock_bh(&reorder_buf->lock);
- iwl_mvm_release_frames(mvm, sta, napi, reorder_buf,
+ iwl_mvm_release_frames(mvm, sta, napi, ba_data, reorder_buf,
le16_to_cpu(release->nssn));
spin_unlock_bh(&reorder_buf->lock);
--
2.14.2
^ permalink raw reply related
* [PATCH 2/9] iwlwifi: mvm: remove duplicated fields in mvm reorder buffer
From: Luca Coelho @ 2017-10-13 13:24 UTC (permalink / raw)
To: kvalo; +Cc: linux-wireless, Sara Sharon, Luca Coelho
In-Reply-To: <20171013132502.10473-1-luca@coelho.fi>
From: Sara Sharon <sara.sharon@intel.com>
The reason station id and tid fields are both in baid data and
in the reorder buffer per queue is that we couldn't access the
baid_data in the reorder timer functions.
Now that we do some pointer math and access it anyway, those
fields can be removed.
This save some space and some code.
Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
---
drivers/net/wireless/intel/iwlwifi/mvm/mvm.h | 4 ----
drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c | 7 ++++---
drivers/net/wireless/intel/iwlwifi/mvm/sta.c | 6 +-----
3 files changed, 5 insertions(+), 12 deletions(-)
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index f008d5f4c585..5e37ec9e786e 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -585,11 +585,9 @@ enum iwl_mvm_tdls_cs_state {
* @head_sn: reorder window head sn
* @num_stored: number of mpdus stored in the buffer
* @buf_size: the reorder buffer size as set by the last addba request
- * @sta_id: sta id of this reorder buffer
* @queue: queue of this reorder buffer
* @last_amsdu: track last ASMDU SN for duplication detection
* @last_sub_index: track ASMDU sub frame index for duplication detection
- * @tid: the tid
* @reorder_timer: timer for frames are in the reorder buffer. For AMSDU
* it is the time of last received sub-frame
* @removed: prevent timer re-arming
@@ -601,11 +599,9 @@ struct iwl_mvm_reorder_buffer {
u16 head_sn;
u16 num_stored;
u8 buf_size;
- u8 sta_id;
int queue;
u16 last_amsdu;
u8 last_sub_index;
- u8 tid;
struct timer_list reorder_timer;
bool removed;
bool valid;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
index 14c2d366f256..fea7b965ea49 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
@@ -503,17 +503,18 @@ void iwl_mvm_reorder_timer_expired(unsigned long data)
if (expired) {
struct ieee80211_sta *sta;
struct iwl_mvm_sta *mvmsta;
+ u8 sta_id = baid_data->sta_id;
rcu_read_lock();
- sta = rcu_dereference(buf->mvm->fw_id_to_mac_id[buf->sta_id]);
+ sta = rcu_dereference(buf->mvm->fw_id_to_mac_id[sta_id]);
mvmsta = iwl_mvm_sta_from_mac80211(sta);
/* SN is set to the last expired frame + 1 */
IWL_DEBUG_HT(buf->mvm,
"Releasing expired frames for sta %u, sn %d\n",
- buf->sta_id, sn);
+ sta_id, sn);
iwl_mvm_event_frame_timeout_callback(buf->mvm, mvmsta->vif,
- sta, buf->tid);
+ sta, baid_data->tid);
iwl_mvm_release_frames(buf->mvm, sta, NULL, buf, sn);
rcu_read_unlock();
} else {
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
index a26cf8e89036..9d0110ee6e54 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
@@ -2137,7 +2137,6 @@ static void iwl_mvm_free_reorder(struct iwl_mvm *mvm,
}
static void iwl_mvm_init_reorder_buffer(struct iwl_mvm *mvm,
- u32 sta_id,
struct iwl_mvm_baid_data *data,
u16 ssn, u8 buf_size)
{
@@ -2161,8 +2160,6 @@ static void iwl_mvm_init_reorder_buffer(struct iwl_mvm *mvm,
spin_lock_init(&reorder_buf->lock);
reorder_buf->mvm = mvm;
reorder_buf->queue = i;
- reorder_buf->sta_id = sta_id;
- reorder_buf->tid = data->tid;
reorder_buf->valid = false;
for (j = 0; j < reorder_buf->buf_size; j++)
__skb_queue_head_init(&entries[j].e.frames);
@@ -2294,8 +2291,7 @@ int iwl_mvm_sta_rx_agg(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
mod_timer(&baid_data->session_timer,
TU_TO_EXP_TIME(timeout * 2));
- iwl_mvm_init_reorder_buffer(mvm, mvm_sta->sta_id,
- baid_data, ssn, buf_size);
+ iwl_mvm_init_reorder_buffer(mvm, baid_data, ssn, buf_size);
/*
* protect the BA data with RCU to cover a case where our
* internal RX sync mechanism will timeout (not that it's
--
2.14.2
^ permalink raw reply related
* [PATCH 4/9] iwlwifi: mvm: allow reading UMAC error data from SMEM in A000 devices
From: Luca Coelho @ 2017-10-13 13:24 UTC (permalink / raw)
To: kvalo; +Cc: linux-wireless, Beni Lev, Luca Coelho
In-Reply-To: <20171013132502.10473-1-luca@coelho.fi>
From: Beni Lev <beni.lev@intel.com>
Currently, UMAC error data reading is restricted to DCCM.
A000 NICs use SMEM for this data.
Signed-off-by: Beni Lev <beni.lev@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
---
drivers/net/wireless/intel/iwlwifi/mvm/utils.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/utils.c b/drivers/net/wireless/intel/iwlwifi/mvm/utils.c
index 2da1b088ac01..4ade688db63b 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/utils.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/utils.c
@@ -459,7 +459,8 @@ static void iwl_mvm_dump_umac_error_log(struct iwl_mvm *mvm)
base = mvm->umac_error_event_table;
- if (base < 0x800000) {
+ if (base < (mvm->trans->cfg->device_family == IWL_DEVICE_FAMILY_A000 ?
+ 0x400000 : 0x800000)) {
IWL_ERR(mvm,
"Not valid error log pointer 0x%08X for %s uCode\n",
base,
--
2.14.2
^ permalink raw reply related
* [PATCH 5/9] iwlwifi: mvm: move umac_error_event_table validity check to where it's set
From: Luca Coelho @ 2017-10-13 13:24 UTC (permalink / raw)
To: kvalo; +Cc: linux-wireless, Luca Coelho
In-Reply-To: <20171013132502.10473-1-luca@coelho.fi>
From: Luca Coelho <luciano.coelho@intel.com>
There's no point in checking the validity of the
umac_error_event_table pointer every time we generate a dump. It's
cleaner to do so when we read the value, namely when we receive the
alive data.
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
---
drivers/net/wireless/intel/iwlwifi/mvm/fw.c | 13 +++++++++++--
drivers/net/wireless/intel/iwlwifi/mvm/utils.c | 18 ++++--------------
2 files changed, 15 insertions(+), 16 deletions(-)
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
index f476882291ae..70d0505a176e 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
@@ -200,10 +200,19 @@ static bool iwl_alive_fn(struct iwl_notif_wait_data *notif_wait,
mvm->umac_error_event_table = le32_to_cpu(umac->error_info_addr);
+ if (mvm->umac_error_event_table <
+ (mvm->trans->cfg->device_family == IWL_DEVICE_FAMILY_A000 ?
+ 0x400000 : 0x800000))
+ IWL_ERR(mvm,
+ "Not valid error log pointer 0x%08X for %s uCode\n",
+ mvm->umac_error_event_table,
+ (mvm->fwrt.cur_fw_img == IWL_UCODE_INIT) ?
+ "Init" : "RT");
+ else
+ mvm->support_umac_log = true;
+
alive_data->scd_base_addr = le32_to_cpu(lmac1->scd_base_ptr);
alive_data->valid = status == IWL_ALIVE_STATUS_OK;
- if (mvm->umac_error_event_table)
- mvm->support_umac_log = true;
IWL_DEBUG_FW(mvm,
"Alive ucode status 0x%04x revision 0x%01X 0x%01X\n",
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/utils.c b/drivers/net/wireless/intel/iwlwifi/mvm/utils.c
index 4ade688db63b..d46115e2d69e 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/utils.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/utils.c
@@ -455,21 +455,12 @@ static void iwl_mvm_dump_umac_error_log(struct iwl_mvm *mvm)
{
struct iwl_trans *trans = mvm->trans;
struct iwl_umac_error_event_table table;
- u32 base;
- base = mvm->umac_error_event_table;
-
- if (base < (mvm->trans->cfg->device_family == IWL_DEVICE_FAMILY_A000 ?
- 0x400000 : 0x800000)) {
- IWL_ERR(mvm,
- "Not valid error log pointer 0x%08X for %s uCode\n",
- base,
- (mvm->fwrt.cur_fw_img == IWL_UCODE_INIT)
- ? "Init" : "RT");
+ if (!mvm->support_umac_log)
return;
- }
- iwl_trans_read_mem_bytes(trans, base, &table, sizeof(table));
+ iwl_trans_read_mem_bytes(trans, mvm->umac_error_event_table, &table,
+ sizeof(table));
if (ERROR_START_OFFSET <= table.valid * ERROR_ELEM_SIZE) {
IWL_ERR(trans, "Start IWL Error Log Dump:\n");
@@ -609,8 +600,7 @@ void iwl_mvm_dump_nic_error_log(struct iwl_mvm *mvm)
if (mvm->error_event_table[1])
iwl_mvm_dump_lmac_error_log(mvm, mvm->error_event_table[1]);
- if (mvm->support_umac_log)
- iwl_mvm_dump_umac_error_log(mvm);
+ iwl_mvm_dump_umac_error_log(mvm);
}
int iwl_mvm_find_free_queue(struct iwl_mvm *mvm, u8 sta_id, u8 minq, u8 maxq)
--
2.14.2
^ permalink raw reply related
* [PATCH 6/9] iwlwifi: define minimum valid address for umac_error_event_table in cfg
From: Luca Coelho @ 2017-10-13 13:24 UTC (permalink / raw)
To: kvalo; +Cc: linux-wireless, Luca Coelho
In-Reply-To: <20171013132502.10473-1-luca@coelho.fi>
From: Luca Coelho <luciano.coelho@intel.com>
We now have two different minimum valid values for
umac_error_event_table. To avoid hardcoding the minimum value in the
driver, add a value to cfg where it can be read from.
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
---
drivers/net/wireless/intel/iwlwifi/cfg/8000.c | 3 ++-
drivers/net/wireless/intel/iwlwifi/cfg/9000.c | 3 ++-
drivers/net/wireless/intel/iwlwifi/cfg/a000.c | 3 ++-
drivers/net/wireless/intel/iwlwifi/iwl-config.h | 1 +
drivers/net/wireless/intel/iwlwifi/mvm/fw.c | 17 +++++++++++------
5 files changed, 18 insertions(+), 9 deletions(-)
diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/8000.c b/drivers/net/wireless/intel/iwlwifi/cfg/8000.c
index 2e6c52664cee..32d6aa25b9a9 100644
--- a/drivers/net/wireless/intel/iwlwifi/cfg/8000.c
+++ b/drivers/net/wireless/intel/iwlwifi/cfg/8000.c
@@ -165,7 +165,8 @@ static const struct iwl_tt_params iwl8000_tt_params = {
.thermal_params = &iwl8000_tt_params, \
.apmg_not_supported = true, \
.ext_nvm = true, \
- .dbgc_supported = true
+ .dbgc_supported = true, \
+ .min_umac_error_event_table = 0x800000
#define IWL_DEVICE_8000 \
IWL_DEVICE_8000_COMMON, \
diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/9000.c b/drivers/net/wireless/intel/iwlwifi/cfg/9000.c
index 2babe0a1f18b..94d717bffbbe 100644
--- a/drivers/net/wireless/intel/iwlwifi/cfg/9000.c
+++ b/drivers/net/wireless/intel/iwlwifi/cfg/9000.c
@@ -149,7 +149,8 @@ static const struct iwl_tt_params iwl9000_tt_params = {
.mac_addr_from_csr = true, \
.rf_id = true, \
.ext_nvm = true, \
- .dbgc_supported = true
+ .dbgc_supported = true, \
+ .min_umac_error_event_table = 0x800000
const struct iwl_cfg iwl9160_2ac_cfg = {
.name = "Intel(R) Dual Band Wireless AC 9160",
diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/a000.c b/drivers/net/wireless/intel/iwlwifi/cfg/a000.c
index 076f8d6d1124..f27eeba3c87c 100644
--- a/drivers/net/wireless/intel/iwlwifi/cfg/a000.c
+++ b/drivers/net/wireless/intel/iwlwifi/cfg/a000.c
@@ -135,7 +135,8 @@ static const struct iwl_ht_params iwl_a000_ht_params = {
.gen2 = true, \
.ext_nvm = true, \
.dbgc_supported = true, \
- .tx_cmd_queue_size = 32
+ .tx_cmd_queue_size = 32, \
+ .min_umac_error_event_table = 0x400000
const struct iwl_cfg iwla000_2ac_cfg_hr = {
.name = "Intel(R) Dual Band Wireless AC a000",
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-config.h b/drivers/net/wireless/intel/iwlwifi/iwl-config.h
index b9f3b350fe34..175bab54df7f 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-config.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-config.h
@@ -384,6 +384,7 @@ struct iwl_cfg {
u8 max_vht_ampdu_exponent;
u8 ucode_api_max;
u8 ucode_api_min;
+ u32 min_umac_error_event_table;
};
/*
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
index 70d0505a176e..0296df625cd5 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
@@ -176,6 +176,7 @@ static bool iwl_alive_fn(struct iwl_notif_wait_data *notif_wait,
struct iwl_lmac_alive *lmac1;
struct iwl_lmac_alive *lmac2 = NULL;
u16 status;
+ u32 umac_error_event_table;
if (iwl_rx_packet_payload_len(pkt) == sizeof(*palive)) {
palive = (void *)pkt->data;
@@ -198,18 +199,22 @@ static bool iwl_alive_fn(struct iwl_notif_wait_data *notif_wait,
mvm->sf_space.addr = le32_to_cpu(lmac1->st_fwrd_addr);
mvm->sf_space.size = le32_to_cpu(lmac1->st_fwrd_size);
- mvm->umac_error_event_table = le32_to_cpu(umac->error_info_addr);
+ umac_error_event_table = le32_to_cpu(umac->error_info_addr);
- if (mvm->umac_error_event_table <
- (mvm->trans->cfg->device_family == IWL_DEVICE_FAMILY_A000 ?
- 0x400000 : 0x800000))
+ if (!umac_error_event_table) {
+ mvm->support_umac_log = false;
+ } else if (umac_error_event_table >=
+ mvm->trans->cfg->min_umac_error_event_table) {
+ mvm->support_umac_log = true;
+ mvm->umac_error_event_table = umac_error_event_table;
+ } else {
IWL_ERR(mvm,
"Not valid error log pointer 0x%08X for %s uCode\n",
mvm->umac_error_event_table,
(mvm->fwrt.cur_fw_img == IWL_UCODE_INIT) ?
"Init" : "RT");
- else
- mvm->support_umac_log = true;
+ mvm->support_umac_log = false;
+ }
alive_data->scd_base_addr = le32_to_cpu(lmac1->scd_base_ptr);
alive_data->valid = status == IWL_ALIVE_STATUS_OK;
--
2.14.2
^ permalink raw reply related
* [PATCH 7/9] iwlwifi: mvm: add missing lq_color
From: Luca Coelho @ 2017-10-13 13:25 UTC (permalink / raw)
To: kvalo; +Cc: linux-wireless, Liad Kaufman, Luca Coelho
In-Reply-To: <20171013132502.10473-1-luca@coelho.fi>
From: Liad Kaufman <liad.kaufman@intel.com>
In the compressed BA notif, the driver didn't parse out
the LQ color, so statistics for the rates tried were
always thrown out. Add it so it gets correctly used.
While at it, fix the name of the relevant field in the
struct.
Fixes: c46e7724bfe9 ("iwlwifi: mvm: support new BA notification response")
Signed-off-by: Liad Kaufman <liad.kaufman@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
---
drivers/net/wireless/intel/iwlwifi/fw/api/tx.h | 4 ++--
drivers/net/wireless/intel/iwlwifi/mvm/tx.c | 3 ++-
2 files changed, 4 insertions(+), 3 deletions(-)
diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/tx.h b/drivers/net/wireless/intel/iwlwifi/fw/api/tx.h
index 14ad9fb895f9..f5d5ba7e37ec 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/tx.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/tx.h
@@ -710,7 +710,7 @@ enum iwl_mvm_ba_resp_flags {
* @reduced_txp: power reduced according to TPC. This is the actual value and
* not a copy from the LQ command. Thus, if not the first rate was used
* for Tx-ing then this value will be set to 0 by FW.
- * @initial_rate: TLC rate info, initial rate index, TLC table color
+ * @tlc_rate_info: TLC rate info, initial rate index, TLC table color
* @retry_cnt: retry count
* @query_byte_cnt: SCD query byte count
* @query_frame_cnt: SCD query frame count
@@ -730,7 +730,7 @@ struct iwl_mvm_compressed_ba_notif {
__le32 flags;
u8 sta_id;
u8 reduced_txp;
- u8 initial_rate;
+ u8 tlc_rate_info;
u8 retry_cnt;
__le32 query_byte_cnt;
__le16 query_frame_cnt;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
index 172b5e63d3fb..df3c21c5f722 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
@@ -1746,6 +1746,7 @@ void iwl_mvm_rx_ba_notif(struct iwl_mvm *mvm, struct iwl_rx_cmd_buffer *rxb)
if (iwl_mvm_has_new_tx_api(mvm)) {
struct iwl_mvm_compressed_ba_notif *ba_res =
(void *)pkt->data;
+ u8 lq_color = TX_RES_RATE_TABLE_COL_GET(ba_res->tlc_rate_info);
int i;
sta_id = ba_res->sta_id;
@@ -1754,7 +1755,7 @@ void iwl_mvm_rx_ba_notif(struct iwl_mvm *mvm, struct iwl_rx_cmd_buffer *rxb)
ba_info.status.tx_time =
(u16)le32_to_cpu(ba_res->wireless_time);
ba_info.status.status_driver_data[0] =
- (void *)(uintptr_t)ba_res->reduced_txp;
+ RS_DRV_DATA_PACK(lq_color, ba_res->reduced_txp);
if (!le16_to_cpu(ba_res->tfd_cnt))
goto out;
--
2.14.2
^ permalink raw reply related
* [PATCH 8/9] iwlwifi: pcie: sort IDs for the 9000 series for easier comparisons
From: Luca Coelho @ 2017-10-13 13:25 UTC (permalink / raw)
To: kvalo; +Cc: linux-wireless, Luca Coelho
In-Reply-To: <20171013132502.10473-1-luca@coelho.fi>
From: Luca Coelho <luciano.coelho@intel.com>
It's hard to find values that are missing in the list, so sorting the
values and comparing them makes it much easier. To simplify this
task, sort the devices in the list.
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
---
drivers/net/wireless/intel/iwlwifi/pcie/drv.c | 84 +++++++++++++--------------
1 file changed, 42 insertions(+), 42 deletions(-)
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
index ce7254ec0514..aa3c07192624 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
@@ -512,65 +512,65 @@ static const struct pci_device_id iwl_hw_card_ids[] = {
{IWL_PCI_DEVICE(0x24FD, 0x0012, iwl8275_2ac_cfg)},
/* 9000 Series */
- {IWL_PCI_DEVICE(0x271B, 0x0010, iwl9160_2ac_cfg)},
- {IWL_PCI_DEVICE(0x271B, 0x0014, iwl9160_2ac_cfg)},
- {IWL_PCI_DEVICE(0x271B, 0x0210, iwl9160_2ac_cfg)},
{IWL_PCI_DEVICE(0x2526, 0x0000, iwl9260_2ac_cfg)},
{IWL_PCI_DEVICE(0x2526, 0x0010, iwl9260_2ac_cfg)},
{IWL_PCI_DEVICE(0x2526, 0x0014, iwl9260_2ac_cfg)},
- {IWL_PCI_DEVICE(0x2526, 0xA014, iwl9260_2ac_cfg)},
- {IWL_PCI_DEVICE(0x2526, 0x4010, iwl9260_2ac_cfg)},
+ {IWL_PCI_DEVICE(0x2526, 0x0030, iwl9560_2ac_cfg)},
+ {IWL_PCI_DEVICE(0x2526, 0x0034, iwl9560_2ac_cfg)},
+ {IWL_PCI_DEVICE(0x2526, 0x0038, iwl9560_2ac_cfg)},
+ {IWL_PCI_DEVICE(0x2526, 0x003C, iwl9560_2ac_cfg)},
+ {IWL_PCI_DEVICE(0x2526, 0x0060, iwl9460_2ac_cfg)},
+ {IWL_PCI_DEVICE(0x2526, 0x0064, iwl9460_2ac_cfg)},
+ {IWL_PCI_DEVICE(0x2526, 0x00A0, iwl9460_2ac_cfg)},
+ {IWL_PCI_DEVICE(0x2526, 0x00A4, iwl9460_2ac_cfg)},
{IWL_PCI_DEVICE(0x2526, 0x0210, iwl9260_2ac_cfg)},
{IWL_PCI_DEVICE(0x2526, 0x0214, iwl9260_2ac_cfg)},
+ {IWL_PCI_DEVICE(0x2526, 0x0230, iwl9560_2ac_cfg)},
+ {IWL_PCI_DEVICE(0x2526, 0x0234, iwl9560_2ac_cfg)},
+ {IWL_PCI_DEVICE(0x2526, 0x0238, iwl9560_2ac_cfg)},
+ {IWL_PCI_DEVICE(0x2526, 0x023C, iwl9560_2ac_cfg)},
+ {IWL_PCI_DEVICE(0x2526, 0x0260, iwl9460_2ac_cfg)},
+ {IWL_PCI_DEVICE(0x2526, 0x02A0, iwl9460_2ac_cfg)},
+ {IWL_PCI_DEVICE(0x2526, 0x02A4, iwl9460_2ac_cfg)},
+ {IWL_PCI_DEVICE(0x2526, 0x1030, iwl9560_2ac_cfg)},
{IWL_PCI_DEVICE(0x2526, 0x1410, iwl9270_2ac_cfg)},
+ {IWL_PCI_DEVICE(0x2526, 0x1420, iwl9460_2ac_cfg)},
{IWL_PCI_DEVICE(0x2526, 0x1610, iwl9270_2ac_cfg)},
- {IWL_PCI_DEVICE(0x9DF0, 0x0A10, iwl9460_2ac_cfg)},
+ {IWL_PCI_DEVICE(0x2526, 0x4010, iwl9260_2ac_cfg)},
+ {IWL_PCI_DEVICE(0x2526, 0x4030, iwl9560_2ac_cfg)},
+ {IWL_PCI_DEVICE(0x2526, 0x40A4, iwl9460_2ac_cfg)},
+ {IWL_PCI_DEVICE(0x2526, 0xA014, iwl9260_2ac_cfg)},
+ {IWL_PCI_DEVICE(0x271B, 0x0010, iwl9160_2ac_cfg)},
+ {IWL_PCI_DEVICE(0x271B, 0x0014, iwl9160_2ac_cfg)},
+ {IWL_PCI_DEVICE(0x271B, 0x0210, iwl9160_2ac_cfg)},
+ {IWL_PCI_DEVICE(0x30DC, 0x0060, iwl9460_2ac_cfg)},
+ {IWL_PCI_DEVICE(0x31DC, 0x0030, iwl9560_2ac_cfg)},
+ {IWL_PCI_DEVICE(0x31DC, 0x0034, iwl9560_2ac_cfg)},
+ {IWL_PCI_DEVICE(0x31DC, 0x0038, iwl9560_2ac_cfg)},
+ {IWL_PCI_DEVICE(0x31DC, 0x003C, iwl9560_2ac_cfg)},
+ {IWL_PCI_DEVICE(0x31DC, 0x0060, iwl9460_2ac_cfg)},
+ {IWL_PCI_DEVICE(0x9DF0, 0x0000, iwl9460_2ac_cfg)},
{IWL_PCI_DEVICE(0x9DF0, 0x0010, iwl9460_2ac_cfg)},
+ {IWL_PCI_DEVICE(0x9DF0, 0x0030, iwl9560_2ac_cfg)},
+ {IWL_PCI_DEVICE(0x9DF0, 0x0034, iwl9560_2ac_cfg)},
+ {IWL_PCI_DEVICE(0x9DF0, 0x0038, iwl9560_2ac_cfg)},
+ {IWL_PCI_DEVICE(0x9DF0, 0x003C, iwl9560_2ac_cfg)},
+ {IWL_PCI_DEVICE(0x9DF0, 0x0060, iwl9460_2ac_cfg)},
{IWL_PCI_DEVICE(0x9DF0, 0x0210, iwl9460_2ac_cfg)},
- {IWL_PCI_DEVICE(0x9DF0, 0x0410, iwl9460_2ac_cfg)},
- {IWL_PCI_DEVICE(0x9DF0, 0x0610, iwl9460_2ac_cfg)},
{IWL_PCI_DEVICE(0x9DF0, 0x0310, iwl9460_2ac_cfg)},
- {IWL_PCI_DEVICE(0x9DF0, 0x0000, iwl9460_2ac_cfg)},
+ {IWL_PCI_DEVICE(0x9DF0, 0x0410, iwl9460_2ac_cfg)},
{IWL_PCI_DEVICE(0x9DF0, 0x0510, iwl9460_2ac_cfg)},
- {IWL_PCI_DEVICE(0x9DF0, 0x2010, iwl9460_2ac_cfg)},
- {IWL_PCI_DEVICE(0x2526, 0x1420, iwl9460_2ac_cfg)},
+ {IWL_PCI_DEVICE(0x9DF0, 0x0610, iwl9460_2ac_cfg)},
{IWL_PCI_DEVICE(0x9DF0, 0x0710, iwl9460_2ac_cfg)},
+ {IWL_PCI_DEVICE(0x9DF0, 0x0A10, iwl9460_2ac_cfg)},
+ {IWL_PCI_DEVICE(0x9DF0, 0x2010, iwl9460_2ac_cfg)},
{IWL_PCI_DEVICE(0x9DF0, 0x2A10, iwl9460_2ac_cfg)},
- {IWL_PCI_DEVICE(0x30DC, 0x0060, iwl9460_2ac_cfg)},
- {IWL_PCI_DEVICE(0x2526, 0x0060, iwl9460_2ac_cfg)},
- {IWL_PCI_DEVICE(0x2526, 0x0260, iwl9460_2ac_cfg)},
- {IWL_PCI_DEVICE(0x2526, 0x0064, iwl9460_2ac_cfg)},
- {IWL_PCI_DEVICE(0x2526, 0x00A4, iwl9460_2ac_cfg)},
- {IWL_PCI_DEVICE(0x2526, 0x40A4, iwl9460_2ac_cfg)},
- {IWL_PCI_DEVICE(0x2526, 0x02A4, iwl9460_2ac_cfg)},
- {IWL_PCI_DEVICE(0x2526, 0x00A0, iwl9460_2ac_cfg)},
- {IWL_PCI_DEVICE(0x2526, 0x02A0, iwl9460_2ac_cfg)},
- {IWL_PCI_DEVICE(0x9DF0, 0x0060, iwl9460_2ac_cfg)},
- {IWL_PCI_DEVICE(0xA370, 0x0060, iwl9460_2ac_cfg)},
- {IWL_PCI_DEVICE(0x31DC, 0x0060, iwl9460_2ac_cfg)},
- {IWL_PCI_DEVICE(0x2526, 0x0030, iwl9560_2ac_cfg)},
- {IWL_PCI_DEVICE(0x2526, 0x4030, iwl9560_2ac_cfg)},
- {IWL_PCI_DEVICE(0x2526, 0x0230, iwl9560_2ac_cfg)},
- {IWL_PCI_DEVICE(0x2526, 0x0234, iwl9560_2ac_cfg)},
- {IWL_PCI_DEVICE(0x2526, 0x0238, iwl9560_2ac_cfg)},
- {IWL_PCI_DEVICE(0x2526, 0x023C, iwl9560_2ac_cfg)},
- {IWL_PCI_DEVICE(0x9DF0, 0x0030, iwl9560_2ac_cfg)},
{IWL_PCI_DEVICE(0xA370, 0x0030, iwl9560_2ac_cfg)},
- {IWL_PCI_DEVICE(0x31DC, 0x0030, iwl9560_2ac_cfg)},
- {IWL_PCI_DEVICE(0x2526, 0x1030, iwl9560_2ac_cfg)},
- {IWL_PCI_DEVICE(0xA370, 0x1030, iwl9560_2ac_cfg)},
- {IWL_PCI_DEVICE(0x9DF0, 0x0034, iwl9560_2ac_cfg)},
{IWL_PCI_DEVICE(0xA370, 0x0034, iwl9560_2ac_cfg)},
- {IWL_PCI_DEVICE(0x31DC, 0x0034, iwl9560_2ac_cfg)},
- {IWL_PCI_DEVICE(0x2526, 0x0038, iwl9560_2ac_cfg)},
- {IWL_PCI_DEVICE(0x2526, 0x003C, iwl9560_2ac_cfg)},
- {IWL_PCI_DEVICE(0x9DF0, 0x0038, iwl9560_2ac_cfg)},
{IWL_PCI_DEVICE(0xA370, 0x0038, iwl9560_2ac_cfg)},
- {IWL_PCI_DEVICE(0x31DC, 0x0038, iwl9560_2ac_cfg)},
- {IWL_PCI_DEVICE(0x9DF0, 0x003C, iwl9560_2ac_cfg)},
{IWL_PCI_DEVICE(0xA370, 0x003C, iwl9560_2ac_cfg)},
- {IWL_PCI_DEVICE(0x31DC, 0x003C, iwl9560_2ac_cfg)},
- {IWL_PCI_DEVICE(0x2526, 0x0034, iwl9560_2ac_cfg)},
+ {IWL_PCI_DEVICE(0xA370, 0x0060, iwl9460_2ac_cfg)},
+ {IWL_PCI_DEVICE(0xA370, 0x1030, iwl9560_2ac_cfg)},
/* a000 Series */
{IWL_PCI_DEVICE(0x2720, 0x0A10, iwla000_2ac_cfg_hr_cdb)},
--
2.14.2
^ permalink raw reply related
* [PATCH 9/9] iwlwifi: pcie: remove set but not used variable tcph
From: Luca Coelho @ 2017-10-13 13:25 UTC (permalink / raw)
To: kvalo; +Cc: linux-wireless, Johannes Berg, Luca Coelho
In-Reply-To: <20171013132502.10473-1-luca@coelho.fi>
From: Johannes Berg <johannes.berg@intel.com>
This variable is never used, so remove the code to set it.
After this, the variable 'iph' also has the same fate.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
---
drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c | 5 +----
1 file changed, 1 insertion(+), 4 deletions(-)
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c b/drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c
index 79e4c73a9709..16b345f54ff0 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c
@@ -289,8 +289,7 @@ static int iwl_pcie_gen2_build_amsdu(struct iwl_trans *trans,
struct sk_buff *csum_skb = NULL;
unsigned int tb_len;
dma_addr_t tb_phys;
- struct tcphdr *tcph;
- u8 *iph, *subf_hdrs_start = hdr_page->pos;
+ u8 *subf_hdrs_start = hdr_page->pos;
total_len -= data_left;
@@ -312,8 +311,6 @@ static int iwl_pcie_gen2_build_amsdu(struct iwl_trans *trans,
* as MAC header.
*/
tso_build_hdr(skb, hdr_page->pos, &tso, data_left, !total_len);
- iph = hdr_page->pos + 8;
- tcph = (void *)(iph + ip_hdrlen);
hdr_page->pos += snap_ip_tcp_hdrlen;
--
2.14.2
^ permalink raw reply related
* Re: pull-request: wireless-drivers 2017-10-13
From: David Miller @ 2017-10-13 15:43 UTC (permalink / raw)
To: kvalo; +Cc: linux-wireless, netdev, linux-kernel
In-Reply-To: <878tgfpl0l.fsf@kamboji.qca.qualcomm.com>
From: Kalle Valo <kvalo@codeaurora.org>
Date: Fri, 13 Oct 2017 10:25:14 +0300
> here's a pull request to net tree, more info in the signed tag below.
> Please let me know if there are any problems.
Pulled, thanks Kalle.
^ permalink raw reply
* Re: [PATCH v2] ath10k: Retry pci probe on failure.
From: Adrian Chadd @ 2017-10-13 15:50 UTC (permalink / raw)
To: Kalle Valo
Cc: greearb@candelatech.com, linux-wireless@vger.kernel.org,
ath10k@lists.infradead.org
In-Reply-To: <87a80vnrsb.fsf@kamboji.qca.qualcomm.com>
On 13 October 2017 at 05:41, Kalle Valo <kvalo@qca.qualcomm.com> wrote:
> greearb@candelatech.com writes:
>
>> From: Ben Greear <greearb@candelatech.com>
>>
>> This works around a problem we see when sometimes the wifi NIC does
>> not respond the first time. This seems to happen especially often on
>> some of the 9984 NICs in mid-range platforms.
>>
>> Signed-off-by: Ben Greear <greearb@candelatech.com>
>
> [...]
>
>> -static int ath10k_pci_probe(struct pci_dev *pdev,
>> - const struct pci_device_id *pci_dev)
>> +static int __ath10k_pci_probe(struct pci_dev *pdev,
>> + const struct pci_device_id *pci_dev)
>> {
>> int ret = 0;
>> struct ath10k *ar;
>> @@ -3672,6 +3672,22 @@ static int ath10k_pci_probe(struct pci_dev *pdev,
>> return ret;
>> }
>>
>> +static int ath10k_pci_probe(struct pci_dev *pdev,
>> + const struct pci_device_id *pci_dev)
>> +{
>> + int cnt = 0;
>> + int rv;
>> + do {
>> + rv = __ath10k_pci_probe(pdev, pci_dev);
>> + if (rv == 0)
>> + return rv;
>> + pr_err("ath10k: failed to probe PCI : %d, retry-count: %d\n", rv, cnt);
>> + mdelay(10); /* let the ath10k firmware gerbil take a small break */
>> + } while (cnt++ < 10);
>> + return rv;
>> +}
>
> This is a sledgehammer approach and it causes reload for all error
> cases, like when hardware is broken or memory allocation is failing.
>
> When the problem happens does it always fail at the the same place? Is
> it hw reset or something else? It's better to retry the invidiual action
> than to do this hack. Or is it just some more delay needed somewhere?
I am seeing WMI timeouts during initial firmware load and wait on
QCA9984 + BCM7444S SoC.
My guess is the WMI wakeup time is not "right" enough and needs to be
extended a little bit.
But then, I have played a lot of whackamole with WMI timeouts during
my loooong porting effort..
-adrian
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox