Booting this image: https://download.freebsd.org/ftp/snapshots/arm64/aarch64/ISO-IMAGES/13.0/FreeBSD-13.0-CURRENT-arm64-aarch64-RPI3-20201105-ef87bd449eb.img.xz on RPI3 produces SDHC errors. See: https://lists.freebsd.org/pipermail/freebsd-arm/2020-November/022692.html It seemed to start with r366106 and was supposed to be fixed by rS367268 but still seems to be happening. The proposed patch from the arm email list didn't help.
For the record, here's the full boot log: Consoles: EFI console Reading loader env vars from /efi/freebsd/loader.env Setting currdev to disk0p1: FreeBSD/arm64 EFI loader, Revision 1.1 (Thu Nov 5 07:53:09 UTC 2020 root@releng1.nyi.freebsd.org) Command line arguments: loader.efi Image base: 0x39e17000 EFI version: 2.80 EFI Firmware: Das U-Boot (rev 8224.1792) Console: comconsole (0) Load Path: /efi\boot\bootaa64.efi Load Device: /VenHw(e61d73b9-a384-4acc-aeab-82e828f3628b)/SD(0)/SD(0)/HD(1,0x01,0,0x81f,0x18fa8) Trying ESP: /VenHw(e61d73b9-a384-4acc-aeab-82e828f3628b)/SD(0)/SD(0)/HD(1,0x01,0,0x81f,0x18fa8) Setting currdev to disk0p1: Trying: /VenHw(e61d73b9-a384-4acc-aeab-82e828f3628b)/SD(0)/SD(0)/HD(2,0x01,0,0x197c7,0x5e6821) Setting currdev to disk0p2: Loading /boot/defaults/loader.conf Loading /boot/defaults/loader.conf Loading /boot/device.hints Loading /boot/loader.conf Loading /boot/loader.conf.local Loading kernel... /boot/kernel/kernel text=0x2a8 text=0x87ade8 text=0x1efa4c data=0x193940 data=0x0+0x541146 syms=[0x8+0x116598+0x8+0x13b120] Loading configured modules... /boot/kernel/umodem.ko text=0x2120 text=0x1390 data=0x6e0+0x10 syms=[0x8+0xf48+0x8+0xb6e] loading required module 'ucom' /boot/kernel/ucom.ko text=0x21a0 text=0x2e20 data=0x880+0x858 syms=[0x8+0x11a0+0x8+0xb2c] can't find '/etc/hostid' can't find '/boot/entropy' Hit [Enter] to boot immediately, or any other key for command prompt. Booting [/boot/kernel/kernel]... Using DTB provided by EFI at 0x7ef6000. EFI framebuffer information: addr, size 0x3eaf0000, 0x10a800 dimensions 656 x 416 stride 656 masks 0x00ff0000, 0x0000ff00, 0x000000ff, 0xff000000 ---<<BOOT>>--- KDB: debugger backends: ddb KDB: current backend: ddb Copyright (c) 1992-2020 The FreeBSD Project. Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994 The Regents of the University of California. All rights reserved. FreeBSD is a registered trademark of The FreeBSD Foundation. FreeBSD 13.0-CURRENT #0 ef87bd449eb-c254225(main): Thu Nov 5 08:56:01 UTC 2020 root@releng1.nyi.freebsd.org:/usr/obj/usr/src/arm64.aarch64/sys/GENERIC arm64 FreeBSD clang version 11.0.0 (git@github.com:llvm/llvm-project.git llvmorg-11.0.0-0-g176249bd673) WARNING: WITNESS option enabled, expect reduced performance. VT(efifb): resolution 656x416 module firmware already present! real memory = 993849344 (947 MB) avail memory = 943214592 (899 MB) Starting CPU 1 (1) Starting CPU 2 (2) Starting CPU 3 (3) FreeBSD/SMP: Multiprocessor System Detected: 4 CPUs arc4random: WARNING: initial seeding bypassed the cryptographic random device because it was not yet seeded and the knob 'bypass_before_seeding' was enabled. random: entropy device external interface MAP 39f3f000 mode 2 pages 1 MAP 39f45000 mode 2 pages 2 MAP 3b350000 mode 2 pages 16 MAP 3f100000 mode 0 pages 1 WARNING: Device "kbd" is Giant locked and may be deleted before FreeBSD 13.0. kbd0 at kbdmux0 WARNING: Device "openfirm" is Giant locked and may be deleted before FreeBSD 13.0. ofwbus0: <Open Firmware Device Tree> simplebus0: <Flattened device tree simple bus> on ofwbus0 ofw_clkbus0: <OFW clocks bus> on ofwbus0 clk_fixed0: <Fixed clock> on ofw_clkbus0 clk_fixed1: <Fixed clock> on ofw_clkbus0 regfix0: <Fixed Regulator> on ofwbus0 regfix1: <Fixed Regulator> on ofwbus0 psci0: <ARM Power State Co-ordination Interface Driver> on ofwbus0 lintc0: <BCM2836 Interrupt Controller> mem 0x40000000-0x400000ff on simplebus0 intc0: <BCM2835 Interrupt Controller> mem 0x7e00b200-0x7e00b3ff irq 48 on simplebus0 gpio0: <BCM2708/2835 GPIO controller> mem 0x7e200000-0x7e2000b3 irq 24,25 on simplebus0 gpiobus0: <OFW GPIO bus> on gpio0 mbox0: <BCM2835 VideoCore Mailbox> mem 0x7e00b880-0x7e00b8bf irq 23 on simplebus0 bcm2835_firmware0: <BCM2835 Firmware> on simplebus0 gpio1: <Raspberry Pi Firmware GPIO controller> on bcm2835_firmware0 gpiobus1: <GPIO bus> on gpio1 generic_timer0: <ARMv7 Generic Timer> irq 1,2,3,4 on ofwbus0 Timecounter "ARM MPCore Timecounter" frequency 19200000 Hz quality 1000 Event timer "ARM MPCore Eventtimer" frequency 19200000 Hz quality 1000 usb_nop_xceiv0: <USB NOP PHY> on ofwbus0 bcm_dma0: <BCM2835 DMA Controller> mem 0x7e007000-0x7e007eff irq 6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21 on simplebus0 bcmwd0: <BCM2708/2835 Watchdog> mem 0x7e100000-0x7e100113,0x7e00a000-0x7e00a023 on simplebus0 bcm2835_clkman0: <BCM283x Clock Manager> mem 0x7e101000-0x7e102fff on simplebus0 bcmrng0: <Broadcom BCM2835 RNG> mem 0x7e104000-0x7e10400f irq 22 on simplebus0 gpioc0: <GPIO controller> on gpio0 uart0: <PrimeCell UART (PL011)> mem 0x7e201000-0x7e2011ff irq 26 on simplebus0 uart0: console (115200,n,8,1) spi0: <BCM2708/2835 SPI controller> mem 0x7e204000-0x7e2041ff irq 28 on simplebus0 spibus0: <OFW SPI bus> on spi0 spibus0: <unknown card> at cs 0 mode 0 spibus0: <unknown card> at cs 1 mode 0 iichb0: <BCM2708/2835 BSC controller> mem 0x7e804000-0x7e804fff irq 40 on simplebus0 bcm283x_dwcotg0: <DWC OTG 2.0 integrated USB controller (bcm283x)> mem 0x7e980000-0x7e98ffff,0x7e006000-0x7e006fff irq 46,47 on simplebus0 usbus1 on bcm283x_dwcotg0 sdhci_bcm0: <Broadcom 2708 SDHCI controller> mem 0x7e300000-0x7e3000ff irq 50 on simplebus0 mmc0: <MMC/SD bus> on sdhci_bcm0 gpioc1: <GPIO controller> on gpio1 fb0: <BCM2835 VT framebuffer driver> on simplebus0 fb0: keeping existing fb bpp of 32 fbd0 on fb0 WARNING: Device "fb" is Giant locked and may be deleted before FreeBSD 13.0. VT: Replacing driver "efifb" with new "fb". fb0: 656x416(656x416@0,0) 32bpp fb0: fbswap: 1, pitch 2624, base rx3eaf0000, screen_size 1091584 pmu0: <Performance Monitoring Unit> irq 0 on ofwbus0 cpulist0: <Open Firmware CPU Group> on ofwbus0 cpu0: <Open Firmware CPU> on cpulist0 bcm2835_cpufreq0: <CPU Frequency Control> on cpu0 cpu1: <Open Firmware CPU> on cpulist0 cpu2: <Open Firmware CPU> on cpulist0 cpu3: <Open Firmware CPU> on cpulist0 gpioled0: <GPIO LEDs> on ofwbus0 gpioled0: <led0> failed to map pin lock order reversal: (sleepable after non-sleepable) 1st 0xffff000000c14ad8 LED mtx (LED mtx, sleep mutex) @ /usr/src/sys/dev/led/led.c:298 2nd 0xffffa00000849810 Raspberry Pi firmware gpio (Raspberry Pi firmware gpio, sx) @ /usr/src/sys/arm/broadcom/bcm2835/raspberrypi_gpio.c:252 lock order LED mtx -> Raspberry Pi firmware gpio attempted at: #0 0xffff000000568374 at witness_checkorder+0xc54 #1 0xffff000000505c84 at _sx_xlock+0x7c #2 0xffff0000007dd6a4 at rpi_fw_gpio_pin_set+0xe8 #3 0xffff000000295afc at led_create_state+0x158 #4 0xffff000000267448 at gpioled_attach+0x290 #5 0xffff0000005345cc at device_attach+0x400 #6 0xffff000000534134 at device_probe_and_attach+0x7c #7 0xffff00000053631c at bus_generic_new_pass+0xf8 #8 0xffff0000005362cc at bus_generic_new_pass+0xa8 #9 0xffff0000005362cc at bus_generic_new_pass+0xa8 #10 0xffff0000005316b0 at bus_set_pass+0x4c #11 0xffff00000048c57c at mi_startup+0x12c uma_zalloc_debug: zone "malloc-64" with the following non-sleepable locks held: exclusive sleep mutex LED mtx (LED mtx) r = 0 (0xffff000000c14ad8) locked @ /usr/src/sys/dev/led/led.c:298 stack backtrace: #0 0xffff000000568694 at witness_debugger+0x64 #1 0xffff000000569824 at witness_warn+0x3ec #2 0xffff000000772194 at uma_zalloc_debug+0x2c #3 0xffff000000771bf8 at uma_zalloc_arg+0x2c #4 0xffff0000004d1848 at malloc+0xf0 #5 0xffff0000007d2864 at bcm2835_firmware_property+0x44 #6 0xffff0000007dd6bc at rpi_fw_gpio_pin_set+0x100 #7 0xffff000000295afc at led_create_state+0x158 #8 0xffff000000267448 at gpioled_attach+0x290 #9 0xffff0000005345cc at device_attach+0x400 #10 0xffff000000534134 at device_probe_and_attach+0x7c #11 0xffff00000053631c at bus_generic_new_pass+0xf8 #12 0xffff0000005362cc at bus_generic_new_pass+0xa8 #13 0xffff0000005362cc at bus_generic_new_pass+0xa8 #14 0xffff0000005316b0 at bus_set_pass+0x4c #15 0xffff00000048c57c at mi_startup+0x12c uma_zalloc_debug: zone "malloc-16" with the following non-sleepable locks held: exclusive sleep mutex LED mtx (LED mtx) r = 0 (0xffff000000c14ad8) locked @ /usr/src/sys/dev/led/led.c:298 stack backtrace: #0 0xffff000000568694 at witness_debugger+0x64 #1 0xffff000000569824 at witness_warn+0x3ec #2 0xffff000000772194 at uma_zalloc_debug+0x2c #3 0xffff000000771bf8 at uma_zalloc_arg+0x2c #4 0xffff0000004d1848 at malloc+0xf0 #5 0xffff0000007ea740 at bounce_bus_dmamem_alloc+0x50 #6 0xffff0000007d533c at bcm2835_mbox_property+0xdc #7 0xffff0000007d2898 at bcm2835_firmware_property+0x78 #8 0xffff0000007dd6bc at rpi_fw_gpio_pin_set+0x100 #9 0xffff000000295afc at led_create_state+0x158 #10 0xffff000000267448 at gpioled_attach+0x290 #11 0xffff0000005345cc at device_attach+0x400 #12 0xffff000000534134 at device_probe_and_attach+0x7c #13 0xffff00000053631c at bus_generic_new_pass+0xf8 #14 0xffff0000005362cc at bus_generic_new_pass+0xa8 #15 0xffff0000005362cc at bus_generic_new_pass+0xa8 #16 0xffff0000005316b0 at bus_set_pass+0x4c #17 0xffff00000048c57c at mi_startup+0x12c uma_zalloc_debug: zone "malloc-128" with the following non-sleepable locks held: exclusive sleep mutex LED mtx (LED mtx) r = 0 (0xffff000000c14ad8) locked @ /usr/src/sys/dev/led/led.c:298 stack backtrace: #0 0xffff000000568694 at witness_debugger+0x64 #1 0xffff000000569824 at witness_warn+0x3ec #2 0xffff000000772194 at uma_zalloc_debug+0x2c #3 0xffff000000771bf8 at uma_zalloc_arg+0x2c #4 0xffff0000004d1848 at malloc+0xf0 #5 0xffff0000007ea788 at bounce_bus_dmamem_alloc+0x98 #6 0xffff0000007d533c at bcm2835_mbox_property+0xdc #7 0xffff0000007d2898 at bcm2835_firmware_property+0x78 #8 0xffff0000007dd6bc at rpi_fw_gpio_pin_set+0x100 #9 0xffff000000295afc at led_create_state+0x158 #10 0xffff000000267448 at gpioled_attach+0x290 #11 0xffff0000005345cc at device_attach+0x400 #12 0xffff000000534134 at device_probe_and_attach+0x7c #13 0xffff00000053631c at bus_generic_new_pass+0xf8 #14 0xffff0000005362cc at bus_generic_new_pass+0xa8 #15 0xffff0000005362cc at bus_generic_new_pass+0xa8 #16 0xffff0000005316b0 at bus_set_pass+0x4c #17 0xffff00000048c57c at mi_startup+0x12c uma_zalloc_debug: zone "malloc-64" with the following non-sleepable locks held: exclusive sleep mutex LED mtx (LED mtx) r = 0 (0xffff000000c14ad8) locked @ /usr/src/sys/dev/led/led.c:298 stack backtrace: #0 0xffff000000568694 at witness_debugger+0x64 #1 0xffff000000569824 at witness_warn+0x3ec #2 0xffff000000772194 at uma_zalloc_debug+0x2c #3 0xffff000000771bf8 at uma_zalloc_arg+0x2c #4 0xffff0000004d1848 at malloc+0xf0 #5 0xffff0000007ea8d8 at bounce_bus_dmamem_alloc+0x1e8 #6 0xffff0000007d533c at bcm2835_mbox_property+0xdc #7 0xffff0000007d2898 at bcm2835_firmware_property+0x78 #8 0xffff0000007dd6bc at rpi_fw_gpio_pin_set+0x100 #9 0xffff000000295afc at led_create_state+0x158 #10 0xffff000000267448 at gpioled_attach+0x290 #11 0xffff0000005345cc at device_attach+0x400 #12 0xffff000000534134 at device_probe_and_attach+0x7c #13 0xffff00000053631c at bus_generic_new_pass+0xf8 #14 0xffff0000005362cc at bus_generic_new_pass+0xa8 #15 0xffff0000005362cc at bus_generic_new_pass+0xa8 #16 0xffff0000005316b0 at bus_set_pass+0x4c #17 0xffff00000048c57c at mi_startup+0x12c cryptosoft0: <software crypto> Timecounters tick every 1.000 msec usbus1: 480Mbps High Speed USB v2.0 iicbus0: <OFW I2C bus> on iichb0 iic0: <I2C generic I/O> on iicbus0 ugen1.1: <DWCOTG OTG Root HUB> at usbus1 uhub0 on usbus1 uhub0: <DWCOTG OTG Root HUB, class 9/0, rev 2.00/1.00, addr 1> on usbus1 mmcsd0: 32GB <SDHC SD32G 8.5 SN F1970496 MFG 09/2020 by 3 SD> at mmc0 50.0MHz/4bit/65535-block bcm2835_cpufreq0: ARM 600MHz, Core 250MHz, SDRAM 400MHz, Turbo OFF Release APs...done CPU 0: ARM Cortex-A53 r0p4 affinity: 0 Cache Type = <64 byte D-cacheline,64 byte I-cacheline,VIPT ICache,64 byte ERG,64 byte CWG> Trying to mount root from ufs:/dev/ufs/rootfs [rw]... Instruction Set Attributes 0 = <CRC32> Instruction Set Attributes 1 = <> Processor Features 0 = <AdvSIMD,FP,EL3 32,EL2 32,EL1 32,EL0 32> Processor Features 1 = <> Memory Model Features 0 = <TGran4,TGran64,SNSMem,BigEnd,16bit ASID,1TB PA> Memory Model Features 1 = <8bit VMID> Memory Model Features 2 = <32bit CCIDX,48bit VA> Debug Features 0 = <2 CTX BKPTs,4 Watchpoints,6 Breakpoints,PMUv3,Debugv8> Debug Features 1 = <> Auxiliary Features 0 = <> Auxiliary Features 1 = <> CPU 1: ARM Cortex-A53 r0p4 affinity: 1 CPU 2: ARM Cortex-A53 r0p4 affinity: 2 CPU 3: ARM Cortex-A53 r0p4 affinity: 3 WARNING: WITNESS option enabled, expect reduced performance. Warning: no time-of-day clock registered, system time will not be set accurately Dual Console: Serial Primary, Video Secondary uhub0: 1 port with 1 removable, self powered Growing root partition to fill device random: randomdev_wait_until_seeded unblock wait ugen1.2: <vendor 0x0424 product 0x9514> at usbus1 uhub1 on uhub0 uhub1: <vendor 0x0424 product 0x9514, class 9/0, rev 2.00/2.00, addr 2> on usbus1 uhub1: MTT enabled uhub1: 5 ports with 4 removable, self powered random: unblocking device. GEOM_PART: mmcsd0s2 was automatically resized. Use `gpart commit mmcsd0s2` to save changes or `gpart undo mmcsd0s2` to revert them. mmcsd0s2 resized mmcsd0s2a resized gpart: arg0 'ufs/rootfs': Invalid argument mmcsd0: Error indicated: 6 NO MEMORY grordfs: read emmcsd0: Error indicated: 1 Timeout mmcsd0: Error indicated: 1 Timeout mmcsd0: Error indicated: 1 Timeout mmcsd0: Error indicated: 1 Timeout rror: 5129216: Operation not supported sdhci_bcm0-slot0: Got data interrupt 0x00600000, but there is no active command. sdhci_bcm0-slot0: ============== REGISTER DUMP ============== sdhci_bcm0-slot0: Sys addr: 0x00000000 | Version: 0x00009902 sdhci_bcm0-slot0: Blk size: 0x00000200 | Blk cnt: 0x00000010 sdhci_bcm0-slot0: Argument: 0x000197d7 | Trn mode: 0x00000036 sdhci_bcm0-slot0: Present: 0x010f0000 | Host ctl: 0x00000003 sdhci_bcm0-slot0: Power: 0x0000000f | Blk gap: 0x00000000 sdhci_bcm0-slot0: Wake-up: 0x00000000 | Clock: 0x00000207 sdhci_bcm0-slot0: Timeout: 0x0000000e | Int stat: 0x00000000 sdhci_bcm0-slot0: Int enab: 0x01ff00bb | Sig enab: 0x01ff00bb sdhci_bcm0-slot0: AC12 err: 0x00000000 | Host ctl2:0x00000000 ugen1.3: <vendor 0x0424 product 0xec00> at usbus1 sdhci_bcm0-slot0: Caps: 0x00000000 | Caps2: 0x00000000 smsc0 on uhub1 sdhci_bcm0-slot0: Max curr: 0x00000001 | ADMA err: 0x00000000 smsc0: <vendor 0x0424 product 0xec00, rev 2.00/2.00, addr 3> on usbus1 sdhci_bcm0-slot0: ADMA addr:0x00000000 | Slot int: 0x00000000 sdhci_bcm0-slot0: =========================================== sdhci_bcm0-slot0: Got data interrupt 0x00600000, but there is no active command. smsc0: chip 0xec00, rev. 0002 sdhci_bcm0-slot0: ============== REGISTER DUMP ============== sdhci_bcm0-slot0: Sys addr: 0x00000000 | Version: 0x00009902 sdhci_bcm0-slot0: Blk size: 0x00000200 | Blk cnt: 0x00000010 sdhci_bcm0-slot0: Argument: 0x000197c7 | Trn mode: 0x00000036 miibus0: <MII bus>sdhci_bcm0-slot0: Present: 0x010f0000 | Host ctl: 0x00000002 on smsc0 sdhci_bcm0-slot0: Power: 0x0000000f | Blk gap: 0x00000000 smscphy0: <SMC LAN8700 10/100 interface>sdhci_bcm0-slot0: Wake-up: 0x00000000 | Clock: 0x00000207 PHY 1sdhci_bcm0-slot0: Timeout: 0x0000000e | Int stat: 0x00000000 on miibus0 sdhci_bcm0-slot0: Int enab: 0x01ff00bb | Sig enab: 0x01ff00bb smscphy0: sdhci_bcm0-slot0: AC12 err: 0x00000000 | Host ctl2:0x00000000 10baseTsdhci_bcm0-slot0: Caps: 0x00000000 | Caps2: 0x00000000 , 10baseT-FDXsdhci_bcm0-slot0: Max curr: 0x00000001 | ADMA err: 0x00000000 , 100baseTXsdhci_bcm0-slot0: ADMA addr:0x00000000 | Slot int: 0x00000000 , 100baseTX-FDXsdhci_bcm0-slot0: =========================================== , autosdhci_bcm0-slot0: Got data interrupt 0x00600000, but there is no active command. sdhci_bcm0-slot0: ============== REGISTER DUMP ============== ue0: <USB Ethernet> on smsc0 sdhci_bcm0-slot0: Sys addr: 0x00000000 | Version: 0x00009902 ue0: Ethernet address: b8:27:eb:e7:f1:dc sdhci_bcm0-slot0: Blk size: 0x00000200 | Blk cnt: 0x00000010 sdhci_bcm0-slot0: Argument: 0x000199c7 | Trn mode: 0x00000036 sdhci_bcm0-slot0: Present: 0x010f0000 | Host ctl: 0x00000003 sdhci_bcm0-slot0: Power: 0x0000000f | Blk gap: 0x00000000 sdhci_bcm0-slot0: Wake-up: 0x00000000 | Clock: 0x00000207 sdhci_bcm0-slot0: Timeout: 0x0000000e | Int stat: 0x00000000 sdhci_bcm0-slot0: Int enab: 0x01ff00bb | Sig enab: 0x01ff00bb sdhci_bcm0-slot0: AC12 err: 0x00000000 | Host ctl2:0x00000000 sdhci_bcm0-slot0: Caps: 0x00000000 | Caps2: 0x00000000 sdhci_bcm0-slot0: Max curr: 0x00000001 | ADMA err: 0x00000000 sdhci_bcm0-slot0: ADMA addr:0x00000000 | Slot int: 0x00000000 sdhci_bcm0-slot0: =========================================== sdhci_bcm0-slot0: Got data interrupt 0x00600000, but there is no active command. sdhci_bcm0-slot0: ============== REGISTER DUMP ============== sdhci_bcm0-slot0: Sys addr: 0x00000000 | Version: 0x00009902 sdhci_bcm0-slot0: Blk size: 0x00000200 | Blk cnt: 0x00000001 sdhci_bcm0-slot0: Argument: 0x03b71ffe | Trn mode: 0x00000012 sdhci_bcm0-slot0: Present: 0x010f0000 | Host ctl: 0x00000003 sdhci_bcm0-slot0: Power: 0x0000000f | Blk gap: 0x00000000 sdhci_bcm0-slot0: Wake-up: 0x00000000 | Clock: 0x00000207 sdhci_bcm0-slot0: Timeout: 0x0000000e | Int stat: 0x00000000 sdhci_bcm0-slot0: Int enab: 0x01ff00bb | Sig enab: 0x01ff00bb sdhci_bcm0-slot0: AC12 err: 0x00000000 | Host ctl2:0x00000000 sdhci_bcm0-slot0: Caps: 0x00000000 | Caps2: 0x00000000 sdhci_bcm0-slot0: Max curr: 0x00000001 | ADMA err: 0x00000000 sdhci_bcm0-slot0: ADMA addr:0x00000000 | Slot int: 0x00000000 sdhci_bcm0-slot0: =========================================== No suitable dump device was found. eval: /sbin/md5: Input/output error Setting hostuuid: 30303030-3030-3030-3930-653766316463. Setting hostid: 0x. sysctl: invalid unsigned long '0x' eval: /sbin/swapon: Input/output error Starting file system chesdhci_bcm0-slot0: Got data interrupt 0x00600000, but there is no active command. sdhci_bcm0-slot0: ============== REGISTER DUMP ============== sdhci_bcm0-slot0: Sys addr: 0x00000000 | Version: 0x00009902 sdhci_bcm0-slot0: Blk size: 0x00000200 | Blk cnt: 0x00000028 sdhci_bcm0-slot0: Argument: 0x004b6400 | Trn mode: 0x00000036 sdhci_bcm0-slot0: Present: 0x010f0000 | Host ctl: 0x00000003 sdhci_bcm0-slot0: Power: 0x0000000f | Blk gap: 0x00000000 sdhci_bcm0-slot0: Wake-up: 0x00000000 | Clock: 0x00000207 sdhci_bcm0-slot0: Timeout: 0x0000000e | Int stat: 0x00000000 sdhci_bcm0-slot0: Int enab: 0x01ff00bb | Sig enab: 0x01ff00bb sdhci_bcm0-slot0: AC12 err: 0x00000000 | Host ctl2:0x00000000 sdhci_bcm0-slot0: Caps: 0x00000000 | Caps2: 0x00000000 sdhci_bcm0-slot0: Max curr: 0x00000001 | ADMA err: 0x00000000 sdhci_bcm0-slot0: ADMA addr:0x00000000 | Slot int: 0x00000000 sdhci_bcm0-slot0: =========================================== cks: eval: fsck: Input/output error Unknown error 126; help! ERROR: ABORTING BOOT (sending SIGTERM to parent)! sdhci_bcm0-slot0: Got data interrupt 0x00600000, but there is no active command. sdhci_bcm0-slot0: ============== REGISTER DUMP ============== sdhci_bcm0-slot0: Sys addr: 0x00000000 | Version: 0x00009902 sdhci_bcm0-slot0: Blk size: 0x00000200 | Blk cnt: 0x00000040 sdhci_bcm0-slot0: Argument: 0x00019980 | Trn mode: 0x00000036 sdhci_bcm0-slot0: Present: 0x010f0000 | Host ctl: 0x00000002 sdhci_bcm0-slot0: Power: 0x0000000f | Blk gap: 0x00000000 sdhci_bcm0-slot0: Wake-up: 0x00000000 | Clock: 0x00000207 sdhci_bcm0-slot0: Timeout: 0x0000000e | Int stat: 0x00000000 sdhci_bcm0-slot0: Int enab: 0x01ff00bb | Sig enab: 0x01ff00bb sdhci_bcm0-slot0: AC12 err: 0x00000000 | Host ctl2:0x00000000 sdhci_bcm0-slot0: Caps: 0x00000000 | Caps2: 0x00000000 sdhci_bcm0-slot0: Max curr: 0x00000001 | ADMA err: 0x00000000 sdhci_bcm0-slot0: 94ADMA addr:0x00000000 | Slot int: 0x00000000 ugen1.4: <Realtek 802.11n WLAN Adapter> at usbus1 sdhci_bcm0-slot0: =========================================== nit 1 - - /bin/sh on /etc/rc terminated abnormally, going to single user mode sdhci_bcm0-slot0: Got data interrupt 0x00600000, but there is no active command. sdhci_bcm0-slot0: ============== REGISTER DUMP ============== sdhci_bcm0-slot0: Sys addr: 0x00000000 | Version: 0x00009902 sdhci_bcm0-slot0: Blk size: 0x00000200 | Blk cnt: 0x00000008 sdhci_bcm0-slot0: Argument: 0x005f8218 | Trn mode: 0x00000036 sdhci_bcm0-slot0: Present: 0x010f0000 | Host ctl: 0x00000003 sdhci_bcm0-slot0: Power: 0x0000000f | Blk gap: 0x00000000 sdhci_bcm0-slot0: Wake-up: 0x00000000 | Clock: 0x00000207 sdhci_bcm0-slot0: Timeout: 0x0000000e | Int stat: 0x00000000 sdhci_bcm0-slot0: Int enab: 0x01ff00bb | Sig enab: 0x01ff00bb sdhci_bcm0-slot0: AC12 err: 0x00000000 | Host ctl2:0x00000000 sdhci_bcm0-slot0: Caps: 0x00000000 | Caps2: 0x00000000 sdhci_bcm0-slot0: Max curr: 0x00000001 | ADMA err: 0x00000000 sdhci_bcm0-slot0: ADMA addr:0x00000000 | Slot int: 0x00000000 sdhci_bcm0-slot0: =========================================== inpx scanner failed Growing root partition to fill device mmcsd0: Error indicated: 1 Timeout mmcsd0: Error indicated: 1 Timeout mmcsd0: Error indicated: 1 Timeout mmcsd0: Error indicated: 1 Timeout mmcsd0: Error indicated: 1 Timeout gpart: Operation not supported gpart: Operation not supported gpart: arg0 'ufs/rootfs': Invalid argument growfs: unable to read superblock: Input/output error No suitable dump device was found. eval: /sbin/md5: Input/output error Setting hostuuid: 30303030-3030-3030-3930-653766316463. Setting hostid: 0x. ugen1.5: <Logitech USB Receiver> at usbus1 ukbd0 on uhub1 ukbd0: <Logitech USB Receiver, class 0/0, rev 2.00/29.01, addr 5> on usbus1 invalid unsigned long '0x' kbd1 at ukbd0 eval: /swapon: Input/output error Fast boot: skipping disk checks. mount: /dev/ufs/rootfs: Operation not supported Mounting root filesystem rw failed, startup aborted ERROR: ABORTING BOOT (sending SIGTERM to parent)! 2020-11-05T10:05:22.765870+00:00 - init 1 - - /bin/sh on /etc/rc terminated abnormally, going to single user mode input in flex scanner failed Growing root partition to fill device gpart: Operation not supported gpart: Operation not supported gpart: arg0 'ufs/rootfs': Invalid argument growfs: unable to read superblock: Input/output error No suitable dump device was found. eval: /sbin/md5: Input/output error Setting hostuuid: 30303030-3030-3030-3930-653766316463. Setting hostid: 0x. sysctl: invalid unsigned long '0x' eval: /sbin/swapon: Input/output error Fast boot: skipping disk checks. WARNING: / was not properly dismounted fsync: giving up on dirty (error = 45) 0xffffa00000d5f7c0: type VCHR usecount 1, writecount 0, refcount 25 seqc users 0 rdev 0xffffa00000f7bc00 hold count flags () flags () v_object 0xffffa00000837b58 ref 0 pages 179 cleanbuf 22 dirtybuf 1 lock type mntfs: EXCL by thread 0xffffa00000ff6580 (pid 171, mount, tid 100084) mount: /dev/ufs/rootfs: Operation not supported Mounting root filesystem rw failed, startup aborted ERROR: ABORTING BOOT (sending SIGTERM to parent)! 2020-11-05T10:05:23.396207+00:00 - init 1 - - /bin/sh on /etc/rc terminated abnormally, going to single user mode input in flex scanner failed Growing root partition to fill device mmcsd0: Error indicated: 1 Timeout gpaOperation not supported mmcsd0: Error indicated: 1 Timeout gparOperation not supported gpart: arg0 'ufs/rootfs': Invalid argument mmcsd0: Error indicated: 1 Timeout growunable to read superblock: Input/output error No suitable dump device was found. mmcsd0: Error indicated: 1 Timeout mmcsd0: Error indicated: 1 Timeout eval: /sbin/md5: Input/output error Setting hostuuid: 30303030-3030-3030-3930-653766316463. Setting hostid: 0x. sysctl: invalid unsigned long '0x' eval: /sbin/swapon: Input/output error Fast boot: skipping disk checks. WARNING: / was not properly dismounted
I looked at this a bit. For whatever reason, growfs is triggering a DMA transfer of length 512 on a buffer that is not cacheline size-aligned. Because both the beginning and end of the buffer address range are not cacheline size-aligned, the loop in _bus_dmamap_count_pages() concludes that it needs 2 pages in order to bounce the request, but clearly that's not true since the entire request is only 512 bytes.
(In reply to Mark Johnston from comment #2) ... and the broadcom sdhci driver is using a busdma tag with a maximum transfer size of 512 bytes, so it didn't reserve enough pages, and bus_dmamap_load() fails.
BTW, I'm not sure why the driver fails so catastrophically after the initial failure. Normal I/O should still work fine since it's sufficiently aligned. Without knowing anything about these drivers it feels like the mmc driver or the broadcom driver have some error handling bug.
This patch: https://github.com/strejda/freebsd/commit/f7b80551d7e765d3b39f8ef9ed10cd18662dacc6 seems to fix it for me
Steve, I afraid that above patch is wrong. Mark, because I cannot reproduce this issue, I need additional round of explanation :) I think that you analysis is slightly unprecise. I think (if I’m able to read the code correctly) that_bus_dmamap_count_pages() returns 1 for unaligned buffer *if whole buffer fits into one page*. It returns 2 * if buffer is unaligned and if it crossing page boundary. This looks like correct behavior for me. IMHO, problem is that we should reflect this fact in pre-allocation of bounced pages - we short of 1 page in this computation (https://svnweb.freebsd.org/base/head/sys/arm64/arm64/busdma_bounce.c?annotate=367984#l311 or https://svnweb.freebsd.org/base/head/sys/arm64/arm64/busdma_bounce.c?annotate=367984#l449 ) But arm code does this correctly (https://svnweb.freebsd.org/base/head/sys/arm/arm/busdma_machdep.c?annotate=365706#l546 or https://svnweb.freebsd.org/base/head/sys/arm/arm/busdma_machdep.c?annotate=365706#l672 ). Did I miss something? Thanks, Michal
(In reply to Michal Meloun from comment #6) > It returns 2 * if buffer is unaligned and if it crossing page boundary. This looks like correct behavior for me. So to be clear, I'm not familiar with busdma. Why do we need two pages to bounce a 512 buffer? Does busdma need to preserve the property that the original buffer crosses a page boundary?
Current busdma implementation may want to preserve page offset for bounced buffers (see BUS_DMA_KEEP_PG_OFFSET). Probably because of this, the all implementations splits source buffer on page boundaries, then process each fragment separately and at end bounced fragments are checked for physical continuity and eventually joined together.
(In reply to Michal Meloun from comment #8) But here we are bouncing because the buffer is not aligned to a cache line, so we cannot preserve the page offset anyway. Maybe I am misunderstanding what BUS_DMA_KEEP_PG_OFFSET means though. Allocating two pages will work of course, but I do not see why it is required in this case.
We can hold page offset although buffer is not cacheline aligned. The cacheline alignment is necessary for generic buffer because we must ensure that other party doesn’t write simultaneously to other part of cacheline (thus making full cacheline dirty). But by putting bounced buffer to any location of fully owned page satisfies this condition – nobody can write to unused parts of start/tail of cacheline. But we slightly diverging. All implementation for counting bounced pages and loading buffers are now +/- consistent, although suboptimal. I’m not ready to change all these – so I want to change allocation of bounced pages. This is part where individual implementation differs, and where arm64 implementation is clearly not consistent with buffer loading/ page counting.
(In reply to Michal Meloun from comment #10) That seems like a reasonable argument to me. Thanks for explaining.
A commit references this bug: Author: mmel Date: Wed Dec 16 14:36:58 UTC 2020 New revision: 368697 URL: https://svnweb.freebsd.org/changeset/base/368697 Log: Allocate right number of pages for the bounced buffers crossing the page. One of the disadvantages of our current busdma code is the fact that we process the bounced buffer in a page-by-page manner. This means that the short (subpage) buffer allocated across page boundaries is bounced to 2 separate pages. This suboptimal behavior is consistent across all platforms and can be related to (probably unimplementable or incompatible with bouncing) BUS_DMA_KEEP_PG_OFFSET flag. Therefore, allocate one additional page to be fully comply with this requirement. Discused with: markj PR: 251018 Changes: head/sys/arm64/arm64/busdma_bounce.c
^Triage: committed back in 2020.