[LEDE-DEV] Fwd: [PATCH] ath9k: Protect queue draining by rcu_read_lock()

I suggest that this gets merged to OpenWRT/LEDE before this makes mainline and the long term support branches of the upstream Kernel. Nick -- Forwarded message -- From: Toke Høiland-Jørgensen Date: Fri, Feb 2, 2018 at 10:36 AM Subject: [PATCH] ath9k: Protect queue

Re: [LEDE-DEV] [OpenWrt-Devel] [PATCH 0/5] x86: add support for microcode update

Hi, There's a backport to GCC 5.5 here: https://github.com/hjl-tools/gcc/commits/hjl/indirect/gcc-5-branch/master Cheers, Nick ___ Lede-dev mailing list Lede-dev@lists.infradead.org http://lists.infradead.org/mailman/listinfo/lede-dev

Re: [LEDE-DEV] [OpenWrt-Devel] [PATCH 0/5] x86: add support for microcode update

Should be resolved in 4.4.113 and 4.9.78 that are now available. Cheers, Nick ___ Lede-dev mailing list Lede-dev@lists.infradead.org http://lists.infradead.org/mailman/listinfo/lede-dev

Re: [LEDE-DEV] [OpenWrt-Devel] [PATCH 0/5] x86: add support for microcode update

Yup, this affects both 4.4.112 and 4.9.77 A fix for this has been queued by Greg for both 4.4.113 and 4.9.78 It does not affect 4.14.14 Cheers, Nick ___ Lede-dev mailing list Lede-dev@lists.infradead.org

Re: [LEDE-DEV] [OpenWrt-Devel] [PATCH 0/5] x86: add support for microcode update

Hi, Thanks! I have done so and this is now queued for 4.9.y and 4.4.y https://git.kernel.org/pub/scm/linux/kernel/git/stable/stable-queue.git/commit/?id=d28663e54705568523979ab8097b9672ee79b17f

Re: [LEDE-DEV] [OpenWrt-Devel] [PATCH 0/5] x86: add support for microcode update

> > https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable.git/commit/arch/x86/mm/kaiser.c?h=v4.9.75=8018307a45a90ab2eecfd03d48b7efb31707df37 > > I already wrote this in a reply on Jan 8th... > >> Cheers, >> >> Nick >> >> On Thu, Ja

Re: [LEDE-DEV] [OpenWrt-Devel] [PATCH 0/5] x86: add support for microcode update

nel/git/stable/linux-stable.git/commit/?h=v4.9.77=43fe95308d276bdfd133f5951cc25565e39982ec Can we backport this? Cheers, Nick On Thu, Jan 18, 2018 at 9:15 PM, Hauke Mehrtens <ha...@hauke-m.de> wrote: > On 01/18/2018 01:51 PM, Nick Lowe wrote: >> Does an update to the Kernel, 4.9.7

Re: [LEDE-DEV] [OpenWrt-Devel] [PATCH 0/5] x86: add support for microcode update

Does an update to the Kernel, 4.9.77 and 4.14.14 need to be made to properly address this? There are fixes to mitigate Spectre. ___ Lede-dev mailing list Lede-dev@lists.infradead.org http://lists.infradead.org/mailman/listinfo/lede-dev

Re: [LEDE-DEV] ath9k - WARN_ON origskb->len <= FCS_LEN - Driver bug?

PM, Nick Lowe <nick.l...@gmail.com> wrote: > Hi all, > > I am seeing a WARN_ON being hit frequently with ath9k after updating > to OpenWrt SNAPSHOT r5763-89233a8. The code generating the warning > seems to point that this is a driver bug. > > rx.c: >

[LEDE-DEV] ath9k - WARN_ON origskb->len <= FCS_LEN - Driver bug?

Hi all, I am seeing a WARN_ON being hit frequently with ath9k after updating to OpenWrt SNAPSHOT r5763-89233a8. The code generating the warning seems to point that this is a driver bug. rx.c: if (ieee80211_hw_check(>hw, RX_INCLUDES_FCS)) { if (unlikely(origskb->len <=

Re: [LEDE-DEV] [PATCH v1] kernel: bump 4.9 to 4.9.75

Sorry, it was back ported to 4.9.75 :-) On Mon, Jan 8, 2018 at 12:34 PM, Nick Lowe <nick.l...@gmail.com> wrote: > It looks like PCID and INVPCID support is likely to be back ported to > the 4.9.76 Linux Kernel: https://patchwork.kernel.org/pa

Re: [LEDE-DEV] [PATCH v1] kernel: bump 4.9 to 4.9.75

It looks like PCID and INVPCID support is likely to be back ported to the 4.9.76 Linux Kernel: https://patchwork.kernel.org/patch/10143225/ ___ Lede-dev mailing list Lede-dev@lists.infradead.org http://lists.infradead.org/mailman/listinfo/lede-dev

Re: [LEDE-DEV] [PATCH v1] kernel: bump 4.9 to 4.9.75

Hi all, I am a moron, I missed: +skip: + if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD) + goto disable; It uses a different path... Regards, NIck On Mon, Jan 8, 2018 at 11:27 AM, Nick Lowe <nick.l...@gmail.com> wrote: > Hi Kevin, > > I am not following :-) For AM

Re: [LEDE-DEV] [PATCH v1] kernel: bump 4.9 to 4.9.75

Hi Kevin, I am not following :-) For AMD, there should not be a tradeoff as there is no need for the page table isolation. Regards, Nick On Mon, Jan 8, 2018 at 11:21 AM, Kevin Darbyshire-Bryant <ke...@darbyshire-bryant.me.uk> wrote: > > >> On 8 Jan 2018, at 11:12,

Re: [LEDE-DEV] [PATCH v1] kernel: bump 4.9 to 4.9.75

Agreed. So this will seemingly regress something like an APU2 and therefore probably should not be merged to LEDE as-is? Nick On Mon, Jan 8, 2018 at 11:08 AM, Kevin Darbyshire-Bryant <ke...@darbyshire-bryant.me.uk> wrote: > > >> On 8 Jan 2018, at 11:04, Nick Lowe <nick.

Re: [LEDE-DEV] [PATCH v1] kernel: bump 4.9 to 4.9.75

Hi, Where in 4.9.75 does it avoid the page table isolation mitigation for AMD? Committed to the 4.14 and 4.15 branch is: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?h=v4.15-rc7=694d99d40972f12e59a3696effee8a376b79d7c8 X86_BUG_CPU_INSECURE becomes

[LEDE-DEV] KRACK (Key Reinstallation Attacks), now patched upstream in hostapd

Referencing https://www.krackattacks.com/ Patches can be found at: https://w1.fi/security/2017-1/ and an advisory at http://seclists.org/oss-sec/2017/q4/83 Patches can also be found committed at https://w1.fi/cgit/hostap/log/ I think we need to get these patches integrated in to LEDE asap, or

Re: [LEDE-DEV] [PATCH] Adding acct_interval to wireless configuration

Hi Yury, Have you tried returning an Acct-Interim-Interval attribute in the Access-Accept packets sent by your RADIUS server, which should configure this on a per client/station basis? Cheers, Nick ___ Lede-dev mailing list

Re: [LEDE-DEV] i386_geode packages not used

Hi, Do we even need a specific build for Geode? I would consider making the generic x86 build target i686 with -march=i686 and -mtune=i686 with the already fixed version of binutils (GAS) that LEDE uses, which does not make bad assumption over the presence of the NOPL instruction. (There was a

Re: [LEDE-DEV] hostapd: Default disable b rates, add legacy_rates

Hi, In practice, this should stop most stations from using the non-OFDM rates and should also retain compatibility with g or n stations that were to still make some use of barker coding/CCK modulation when sending some frames. I think that this is a preferable approach therefore for default

Re: [LEDE-DEV] [RFC] Disable 802.11b rates by default and set supported/basic rates appropriately for 2.4 GHz BSSes based on require_mode

I have sent separately via GitHub. ___ Lede-dev mailing list Lede-dev@lists.infradead.org http://lists.infradead.org/mailman/listinfo/lede-dev

[LEDE-DEV] [RFC] Disable 802.11b rates by default and set supported/basic rates appropriately for 2.4 GHz BSSes based on require_mode

be filtered to include OFDM rates only. Signed-off-by: Nick Lowe <nick.l...@gmail.com> --- package/network/services/hostapd/files/hostapd.sh | 79 +++ 1 file changed, 68 insertions(+), 11 deletions(-) diff --git a/package/network/services/hostapd/files/hostapd.sh b/package/n

[LEDE-DEV] x86 and x86-64 builds broken

Hi all, The x86 and x86-64 builds are currently broken: http://phase1.builds.lede-project.org/builders/x86%2F64 This appears to be as a result of the TPM support patches that were merged. Thanks, Nick ___ Lede-dev mailing list

Re: [LEDE-DEV] Commit "luci-mod-admin-full: place 80211w options behind a check for wpad-full" causes 802.11w to be hidden with full hostapd

By what does this have to do with wpad, I mean that there should be no dependency on EAP features being present for 802.11w to work. Nick ___ Lede-dev mailing list Lede-dev@lists.infradead.org http://lists.infradead.org/mailman/listinfo/lede-dev

[LEDE-DEV] Commit "luci-mod-admin-full: place 80211w options behind a check for wpad-full" causes 802.11w to be hidden with full hostapd

Hi, The patch committed here does not seem to be correct as it hides 802.11w support with the full hostapd package: https://git.lede-project.org/?p=project/luci.git;a=commit;h=07e01d094eb25e1f036e85b8cfc5aceccc56003c With the hostapd package installed, not mini, the output of hostapd -veap is:

Re: [LEDE-DEV] ath9k stability in LEDE master branch - kernel panic every ~5 minutes

Hi all, With the candidate fix included, LEDE Reboot SNAPSHOT r3386-0799de6 / LuCI Master (git-17.041.67116-1834eff) does not crash for me with airtime fairness enabled. Huge thanks! Regards, Nick ___ Lede-dev mailing list

Re: [LEDE-DEV] ath9k stability in LEDE master branch - kernel panic every ~5 minutes

Hi Dave, The hardware platform is 2 x AR9390 manufactured by Senao on an apu2c4. 01:00.0 "Network controller" "Qualcomm Atheros" "AR93xx Wireless Network Adapter" -r01 "Qualcomm Atheros" "Device 0205" 05:00.0 "Network controller" "Qualcomm Atheros" "AR93xx Wireless Network Adapter" -r01

Re: [LEDE-DEV] ath9k stability in LEDE master branch - kernel panic every ~5 minutes

Hi all, This is triggered by the airtime fairness patches for ath9k being present. When this is disabled via: echo 0 > /sys/kernel/debug/ieee80211/phy0/ath9k/airtime_flags echo 0 > /sys/kernel/debug/ieee80211/phy1/ath9k/airtime_flags ... these kernel panics do not occur. I perhaps erroneously

Re: [LEDE-DEV] ath9k stability in LEDE master branch - kernel panic every ~5 minutes

Thanks, I have reported this as a bug. https://bugs.lede-project.org/index.php?do=details_id=473 ___ Lede-dev mailing list Lede-dev@lists.infradead.org http://lists.infradead.org/mailman/listinfo/lede-dev