Summary:
- During Wi-Fi stress test, I noticed that ASSERT() happened
in gs2200m_hal_write(). Actually, this happens in 20-40 hours.
- In this case, resp[1] was always 0x0 which is not defined
in the spec.
- In the previous implementation, retry was only done in case of
WR_RESP_NOK. (0x13)
- This commit changes this condition and it reties in case of
0x0 as well
Impact:
- gs2200m driver
Testing:
- Tested with spresense:wifi_smp and stm32f4discovery:wifi
Signed-off-by: Masayuki Ishikawa <Masayuki.Ishikawa@jp.sony.com>
Summary:
- During testing iperf -s with lc823450-xgevk:rndis,
I noticed that hard fault happens
- Finally, I found that group_kill_children() is not
not protected by a critical section
- This commit fixes this issue
Impact:
- SMP only
Testing:
- Tested with iperf with the following configurations
- lc823450-xgevk:rndis, spresense:rndis_smp
- Tested with ostest with the following configurations
- lc823450-xgevk:rndis, esp32-devkitc:smp (QEMU)
- sabre-6quad:smp (QEMU), maix-bit:smp (QEMU)
- spresense:rndis_smp, sim:smp
Signed-off-by: Masayuki Ishikawa <Masayuki.Ishikawa@jp.sony.com>
Port C was not working because the GPIO pin count was
wrong. The 48 pin packages has 34 GPIO (Not counting PH0 & PH1)
It is GPIOA GPIO B (sans PB11) and GPIOC PC13-PC15
The calls via RTC API weren't fast enough for the edge case
of minimum counter value, resulting in the timer never
expiring as the counter had already passed the compare value.
This now uses direct register access functions and also
gets the latest counter value in edge case.
This function is already declared in include/nuttx/init.h include this
file instead.
Signed-off-by: Abdelatif Guettouche <abdelatif.guettouche@espressif.com>
Don't declare riscv_addregion if CONFIG_MM_REGIONS is < 1, so we won't
have to provide a dummy stub for every chip.
Also rename the function from up_addregion to riscv_addregion since it's
not exported outside the arch directory.
Signed-off-by: Abdelatif Guettouche <abdelatif.guettouche@espressif.com>
The MCAN driver private structure has been renamed to struct sam_mcan_s,
but some functions reference sam_can_s. There are missing defines
of return variable in some functions.
Signed-off-by: Pavel Pisa <ppisa@pikron.com>
Summary:
- This commit adds iperf to rndis and rndis_smp defcofigs
Impact:
- rndis and rndis_smp
Testing:
- Tested with iperf
Signed-off-by: Masayuki Ishikawa <Masayuki.Ishikawa@jp.sony.com>
this change to support receive the udp data from the specified
port without obtaining the address.
e.g: disable the Bootstrap flag on dhcpc handshake
Reference:
RFC1542: Clarifications and Extensions for the Bootstrap Protocol.
Signed-off-by: chao.an <anchao@xiaomi.com>
The BaseBoard carries Teensy 4.1 SBC. The complete design
of the baseboard (teensy_bb-1.tdb) is available at
https://gitlab.com/pikron/projects/imxrt-devel/-/tree/master/hw
Attempt is to provide configuration which enables most
of interfaces of the BaseBoard and allows runtime applications
loading over TFTP.
Signed-off-by: Pavel Pisa <ppisa@pikron.com>
The ARMv7 specifies minimal alignment to be 8 for EABI. Ensure
that sections are placed with even greater minimal alignment.
Sections final alignment should be adjusted according
to alignment of their inputs, but for some reasons some
combinations lead to .ARM.exidx overlap with .data.
When alignment is increased then the direct fetch of the
fill location
_eronly = ABSOLUTE(.);
does not match with final LMA placement of the .data section.
Use constructs which should result in matching .data LMA
address for all cases
_eronly = ABSOLUTE(.);
Signed-off-by: Pavel Pisa <ppisa@pikron.com>
The BSP MAke.defs line
LDFLAGS += -Map=$(TOPDIR)/nuttx.map
results in absolute path in nuttx-export/scripts/Make.defs
which would mean that each application rebuild would attempt
to modify file inside original build of NuttX system.
Even worse, it prevents linking by GCC, because -Map=xxx/nuttx.map
works directly in LD, it would require -Wl,-Map= for GCC.
Fixed name of map file prevents to define map file by build
system using exported NuttX libraries and link kit.
Change is done to allow OMK template based applications
development for another NuttX BSP
https://github.com/ppisa/nuttx-devel/tree/master/nuttx-omk-template
Signed-off-by: Pavel Pisa <ppisa@pikron.com>
The ifstate is already set in
enc_ifup -> enc_reset -> enc_linkstatus
The ifstate after this call is either ENCSTATE_RUNNING or ENCSTATE_UP
If ifstate is ENCSTATE_RUNNING, it would be wrong to set it to ENCSTATE_UP;
this would lead to enc_txavail never invoking the driver callback, causing
very long latencies in sending.
Signed-off-by: Jukka Laitinen <jukkax@ssrc.tii.ae>
Summary:
- This commit fixes a compile error in imx_enet.c
with CONFIG_DEBUG_ASSERTIONS=y
Impact:
- None
Testing:
- Tested with sabre-6quad:netnsh with QEMU
Signed-off-by: Masayuki Ishikawa <Masayuki.Ishikawa@jp.sony.com>
Summary:
- This commit adds DEBUGASSERT() to check the IOB size
Impact:
- None
Testing:
- Tested with sabre-6quad:netnsh with QEMU
Signed-off-by: Masayuki Ishikawa <Masayuki.Ishikawa@jp.sony.com>
Summary:
- This commit makes the iperf work with CONFIG_NET_TCP_WRITE_BUFFERS
- Since the iperf sends a large packet (e.g. 16KB), CONFIG_IOB_NBUFFERS
must be increased enough to hold the packet
- To improve performance CONFIG_IMX_ENET_NTXBUFFERS was changed to
the default (i.e. 2)
Impact:
- sabre-6quad:netnsh only
Testing:
- Tested with iperf with QEMU
Signed-off-by: Masayuki Ishikawa <Masayuki.Ishikawa@jp.sony.com>
