just to add to my previous message:
after 14 hours of uptime with both wifi interfaces enabled, but no
devices connected to them total available memory fell to 44MB, free
60MB. checked again after it passed 20 hours it was the same so no
further decrease.
all processes had max 2% of memory consump
> On Jan 8, 2020, at 3:10 PM, Hauke Mehrtens wrote:
>
> On 1/8/20 7:24 AM, Petr Štetiar wrote:
>> Hauke Mehrtens [2020-01-07 23:21:19]:
>>
>> Hi,
>>
>> thanks for your work.
>>
Hauke Mehrtens (6):
buildsystem: Make PIE ASLR option tristate
dnsmasq: Activate PIE by default
>
Sent from my iPhone
> On Jan 8, 2020, at 2:50 PM, Tom Psyborg wrote:
>
> On 08/01/2020, Rosen Penev wrote:
>> From: Chuanhong Guo
>>
>> This new function make batch processing of network packets possible,
>> which slightly improves performance.
>>
>> Signed-off-by: Chuanhong Guo
>> Teste
On 1/8/20 7:24 AM, Petr Štetiar wrote:
> Hauke Mehrtens [2020-01-07 23:21:19]:
>
> Hi,
>
> thanks for your work.
>
>>> Hauke Mehrtens (6):
>>> buildsystem: Make PIE ASLR option tristate
>>> dnsmasq: Activate PIE by default
>>> dropbear: Activate PIE by default
>>> hostapd: Activate PIE
On 08/01/2020, Rosen Penev wrote:
> From: Chuanhong Guo
>
> This new function make batch processing of network packets possible,
> which slightly improves performance.
>
> Signed-off-by: Chuanhong Guo
> Tested-by: Rosen Penev
> ---
> around a 20mbps improvement is measured on a TP-LINK Archer
On Wed, Jan 8, 2020 at 2:35 PM Rosen Penev wrote:
>
> NASM builds completely broken files with pentium targets. This causes
> ffmpeg to fail compilation. Instead of working around the problem in the
> ffmpeg Makefile, it's easier to just not build for broken platforms.
>
> Not even objdump can fig
* systemd: update documentation URL
* global: bump copyright
Usual house keeping.
* Makefile: DEBUG_TOOLS -> DEBUG and document
* Makefile: port static analysis check
* dns-hatchet: adjust path for new repo layout
* Makefile: rework automatic version.h mangling
These are some important-ish clean
* socket: mark skbs as not on list when receiving via gro
Certain drivers will pass gro skbs to udp, at which point the udp driver
simply iterates through them and passes them off to encap_rcv, which is
where we pick up. At the moment, we're not attempting to coalesce these
into bundles, but we al
NASM builds completely broken files with pentium targets. This causes
ffmpeg to fail compilation. Instead of working around the problem in the
ffmpeg Makefile, it's easier to just not build for broken platforms.
Not even objdump can figure out what the generated files are.
Signed-off-by: Rosen Pe
From: Chuanhong Guo
This new function make batch processing of network packets possible,
which slightly improves performance.
Signed-off-by: Chuanhong Guo
Tested-by: Rosen Penev
---
around a 20mbps improvement is measured on a TP-LINK Archer C7v2
.../net/ethernet/atheros/ag71xx/ag71xx_main.c
FYI
https://forum.openwrt.org/t/refresh-project-identity-new-openwrt-project-logo
___
openwrt-devel mailing list
openwrt-devel@lists.openwrt.org
https://lists.openwrt.org/mailman/listinfo/openwrt-devel
u-boot splits nand factory firmware at 2M offset, flash the first
part as kernel into spi nor and the other part as ubi into nand
flash. With previous commit increasing kernel size to 4M, generated
factory firmware is broken because ubi is at 4M offset.
This commit reduces kernel size definition t
> -Original Message-
> From: openwrt-devel [mailto:openwrt-devel-boun...@lists.openwrt.org] On
> Behalf Of Adrian Schmutzler
> Sent: Sonntag, 5. Januar 2020 17:43
> To: openwrt-devel@lists.openwrt.org
> Subject: [OpenWrt-Devel] [PATCH 5/5] kernel: remove obsolete kernel version
> switches
>
13 matches
Mail list logo
