have my image now up and running
the reason is ?
mt7615_add_interface and mt7615_mcu_set_dev_info seems not to be called
this should be called, the funciton is used to add interface (ifconfig wlan0)
So can you see AP-STA-CONNECTED now?
have my image now up and running
the reason is ?
mt7615_add_interface and mt7615_mcu_set_dev_info seems not to be called
this should be called, the funciton is used to add interface (ifconfig wlan0)
So can you see AP-STA-CONNECTED now?
ok, they get called when try to bring the interface up…not at boottime
[ 862.311122] DEBUG: Passed mt7615_add_interface 69
[ 862.319666] DEBUG: Passed mt7615_add_interface 76
[ 862.325457] DEBUG: Passed mt7615_add_interface 83
[ 862.330584] DEBUG: Passed mt7615_mcu_set_dev_info 685
[ 862.336083] DEBUG: Passed mt7615_mcu_set_dev_info 691
[ 862.341534] DEBUG: Passed __mt7615_mcu_set_dev_info 649
[ 862.349177] DEBUG: Passed __mt7615_mcu_set_dev_info 652
[ 862.354790] DEBUG: Passed __mt7615_mcu_set_dev_info 654
[ 862.360382] DEBUG: Passed __mt7615_mcu_set_dev_info 662
[ 862.366078] DEBUG: Passed mt7615_add_interface 89 ret:0
[ 862.371490] DEBUG: Passed mt7615_add_interface 93
[ 862.376446] DEBUG: Passed mt7615_add_interface 99
[ 862.381400] DEBUG: Passed mt7615_add_interface 103
[ 862.386444] DEBUG: Passed mt7615_add_interface 105
So what is the current progress?
Can you ping to sta now or test it through iperf now? but i still think your efuse is not completed, we need those RF parameter.
i cannot bring wlan0 up and hostapd needs interface up
i rewrite my card with my image because i had strange things with jacks image (syslog cannot be deleted and shows old entries)
in my image i also needed a udev-rule to get rid of the “new” interface-naming:
/etc/udev/rules.d/70-persistent-net.rules
SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", DRIVERS=="mt7615e", NAME="wlan0"
eep-array looks right?
I have no idea. I will find a card with empty efuse to test again.
eep-array looks right?
No
eeprom.txt (5.2 KB)
or @frank-w you can try to copy the content to data[off], where off starting from 52
you mean assign your values to data[off - MT_EE_NIC_CONF_0].val?
any way to write directly to eeprom and set these values?
yes. since your eeprom is almost empty you can assign my values if you want.
I don’t know the way to write directly to eeprom maybe bpiteam know it.
But i’m not sure if this could work or not as every card has its own value
seems not working imho the card needs some of this data for bootup @jackzeng, can i write eeprom of mt7615-card? mt7615_eep.txt (130,2 KB)
this is how i did it:
drivers/net/wireless/mediatek/mt76_new/mt7615/mcu.c
547 const char eepdata[] = { //from 52
548 68,0,0,96,1,1,0,0,
549 0,0,0,0,0,0,68,0,
550 4,0,0,0,0,0,0,0,
551 0,0,224,1,32,32,0,179,
552 64,182,64,196,35,128,129,128,
553 192,195,35,130,129,129,192,196,
554 35,129,128,128,64,196,35,128,
555 129,128,64,197,32,196,128,192,
556 196,32,128,128,192,195,32,128,
557 128,64,195,32,128,128,64,194,
558 32,128,128,64,194,32,128,128,
559 64,194,32,128,128,192,194,32,
560 128,128,64,195,32,128,128,64,
561 195,32,128,128,64,195,32,128,
562 128,192,194,32,128,128,64,194,
563 32,128,128,192,193,32,128,128,
564 64,193,32,128,128,192,192,32,
565 128,128,198,198,196,196,196,0,
566 0,194,0,194,193,129,129,129,
567 133,194,194,0,130,130,130,129,
568 0,197,197,197,0,0,195,0,
569 195,195,130,130,130,131,132,195,
570 195,195,130,130,130,0,0,130,
571 130,0,0,0,133,0,0,0,
572 180,0,194,135
573 };
574
575 skb = mt7615_mcu_msg_alloc(NULL, size + sizeof(req_hdr));
576 memcpy(skb_put(skb, sizeof(req_hdr)), &req_hdr, sizeof(req_hdr));
577 data = (struct req_data *)skb_put(skb, size);
578 memset(data, 0, size);
579
580 for (off = MT_EE_NIC_CONF_0; off < __MT_EE_MAX; off++)
581 {
582 if (off>=52 && off<=247) eep[off]=eepdata[off-52];
583 printk(KERN_ALERT "DEBUG: Passed %s %d eep[%d]:%d\n",__FUNCTION__,__LINE__,off,eep[off]);
584
585 data[off - MT_EE_NIC_CONF_0].val = eep[off];
586 }
587 return mt7615_mcu_msg_send(dev, skb, MCU_EXT_CMD_EFUSE_BUFFER_MODE,
https://github.com/Nossiac/mtk-openwrt-feeds/issues/55 here a eeprom-file is dumped in openwrt from a /proc/mtd device (i have none, but maybe a kernel-option is missing), i found MT7615E1_EEPROM.bin (1 KB)
openwrt 5ghz
channel=36
hw_mode=a
ieee80211n=1
R2 VHT40 1*nss
hw_mode=a
channel=36
ieee80211n=1
ieee80211ac=1
require_ht=1
require_vht=1
minstrel rate control :
tried my mt7615-card on bpi-r2, but it is not visible in lspci…used same cable as it is recognized on r64…very strange,
also my mt7612 is not recognized through cable (works without)
is possible, that r2 gives too less power on pci-bus for using with extender-cable?
R2 VHT80
interface=wlp1s0
driver=nl80211
ssid=126test
hw_mode=a
channel=36
ieee80211n=1
ieee80211ac=1
require_ht=1
require_vht=1
vht_capab=[SHORT-GI-80][RXLDPC]
ht_capab=[HT40+][SHORT-GI-20][SHORT-GI-40][LDPC]
vht_oper_chwidth=1
vht_oper_centr_freq_seg0_idx=42
country_code=US
minstrel rate control rc_stat
20190303 update
vht80 2*nss
IxChariot test:
Bidirectional:
RX:
TX (need to fix drop issue):
I’ve posted the v1 to mainline. The driver fully supports AP, station and monitor mode. https://patchwork.kernel.org/project/linux-mediatek/list/?series=97035 https://patchwork.kernel.org/project/linux-mediatek/list/?series=97041
The host driver has been applied to mainline.
Also add support for mtd eeprom parsing - https://patchwork.kernel.org/patch/10947385/
i test that driver, and when encryption is enabled, the speed drops almost 10 times
Thanks. will fix it.
MT7615 module have support BPI-R2, BPI-R64, and We have tested the parametric performance of module 7615 in the laboratory.To achieve the optimal
BPI-R64+MT7615 test wifi function with openwrt
I have been successful in using the SD card version of the img file to boot into a working system. I would like to use the EMMC.img version and have the router boot and function out of emmc. Can you guide me how to flash the EMMC.img version? I tried using the following options in u-boot menu:
The TFTP transfer was successful, but got errors.
In case of “System Load Linux Kernel”, I got
Bytes transferred = 20191432 (13418c8 hex)
get filesize 0x13418c8
[do_filesize_check] partition_size: 0x1e00000 , filesize: 0x13418c8
[do_read_image_blks] Image format error,neither FIT image nor old image.
Bad Magic Number.
mmc0 is current device
mmc0 operation
dev_num = 0
MMC write: dev # 0, block # 4096, count 98304 ... 98304 blocks write: OK
mmc0 is current device
mmc0 operation
dev_num = 0
MMC read: dev # 0, block # 4096, count 1 ... 1 blocks read: OK
[do_read_image_blks] Image format error,neither FIT image nor old image.
Bad Magic Number.
mmc0 operation
dev_num = 0
MMC read: dev # 0, block # 4096, count 0 ... 0 blocks read: OK
bootm flag=0, states=70f
Wrong Image Format for bootm command
ERROR: can't get kernel image!
In case of “Load flashimage”, I got
Bytes transferred = 20191432 (13418c8 hex)
get filesize 0x13418c8
[do_filesize_check] partition_size: 0x1400000 , filesize: 0x13418c8
mmc0 is current device
mmc0 operation
dev_num = 0
MMC write: dev # 0, block # 0, count 40960 ... 40960 blocks write: OK
dev_num = 0
Writing to MMC(0)... offset=1310720
size=4096
mmc->write_bl_len=512
done
invaild env!!!
Thanks