zephyr/soc/xtensa/intel_adsp/tools/cavstool.py

821 lines
29 KiB
Python
Raw Normal View History

#!/usr/bin/env python3
# Copyright(c) 2022 Intel Corporation. All rights reserved.
# SPDX-License-Identifier: Apache-2.0
import os
import sys
import struct
import logging
import asyncio
import time
import subprocess
import ctypes
import mmap
import argparse
import socketserver
import threading
import netifaces
# Global variable use to sync between log and request services.
# When it is true, the adsp is able to start running.
start_output = False
lock = threading.Lock()
HOST = None
PORT_LOG = 9999
PORT_REQ = PORT_LOG + 1
BUF_SIZE = 4096
CMD_LOG_START = "start_log"
CMD_LOG_STOP = "stop_log"
CMD_DOWNLOAD = "download"
logging.basicConfig()
log = logging.getLogger("cavs-fw")
log.setLevel(logging.INFO)
PAGESZ = 4096
HUGEPAGESZ = 2 * 1024 * 1024
HUGEPAGE_FILE = "/dev/hugepages/cavs-fw-dma.tmp."
# SRAM windows. Each appears in a 128k region starting at 512k.
#
# Window 0 is the FW_STATUS area, and 4k after that the IPC "outbox"
# Window 1 is the IPC "inbox" (host-writable memory, just 384 bytes currently)
# Window 2 is unused by this script
# Window 3 is winstream-formatted log output
OUTBOX_OFFSET = (512 + (0 * 128)) * 1024 + 4096
INBOX_OFFSET = (512 + (1 * 128)) * 1024
WINSTREAM_OFFSET = (512 + (3 * 128)) * 1024
class HDAStream:
# creates an hda stream with at 2 buffers of buf_len
def __init__(self, stream_id: int):
self.stream_id = stream_id
self.base = hdamem + 0x0080 + (stream_id * 0x20)
log.info(f"Mapping registers for hda stream {self.stream_id} at base {self.base:x}")
self.hda = Regs(hdamem)
self.hda.GCAP = 0x0000
self.hda.GCTL = 0x0008
self.hda.DPLBASE = 0x0070
self.hda.DPUBASE = 0x0074
self.hda.SPBFCH = 0x0700
self.hda.SPBFCTL = 0x0704
self.hda.PPCH = 0x0800
self.hda.PPCTL = 0x0804
self.hda.PPSTS = 0x0808
self.hda.SPIB = 0x0708 + stream_id*0x08
self.hda.freeze()
self.regs = Regs(self.base)
self.regs.CTL = 0x00
self.regs.STS = 0x03
self.regs.LPIB = 0x04
self.regs.CBL = 0x08
self.regs.LVI = 0x0c
self.regs.FIFOW = 0x0e
self.regs.FIFOS = 0x10
self.regs.FMT = 0x12
self.regs.FIFOL= 0x14
self.regs.BDPL = 0x18
self.regs.BDPU = 0x1c
self.regs.freeze()
self.dbg0 = Regs(hdamem + 0x0084 + (0x20*stream_id))
self.dbg0.DPIB = 0x00
self.dbg0.EFIFOS = 0x10
self.dbg0.freeze()
self.reset()
def __del__(self):
self.reset()
def config(self, buf_len: int):
log.info(f"Configuring stream {self.stream_id}")
self.buf_len = buf_len
log.info("Allocating huge page and setting up buffers")
self.mem, self.hugef, self.buf_list_addr, self.pos_buf_addr, self.n_bufs = self.setup_buf(buf_len)
log.info("Setting buffer list, length, and stream id and traffic priority bit")
self.regs.CTL = ((self.stream_id & 0xFF) << 20) | (1 << 18) # must be set to something other than 0?
self.regs.BDPU = (self.buf_list_addr >> 32) & 0xffffffff
self.regs.BDPL = self.buf_list_addr & 0xffffffff
self.regs.CBL = buf_len
self.regs.LVI = self.n_bufs - 1
self.debug()
log.info(f"Configured stream {self.stream_id}")
def write(self, data):
bufl = min(len(data), self.buf_len)
log.info(f"Writing data to stream {self.stream_id}, len {bufl}, SPBFCTL {self.hda.SPBFCTL:x}, SPIB {self.hda.SPIB}")
self.mem[0:bufl] = data[0:bufl]
self.mem[bufl:bufl+bufl] = data[0:bufl]
self.hda.SPBFCTL |= (1 << self.stream_id)
self.hda.SPIB += bufl
log.info(f"Wrote data to stream {self.stream_id}, SPBFCTL {self.hda.SPBFCTL:x}, SPIB {self.hda.SPIB}")
def start(self):
log.info(f"Starting stream {self.stream_id}, CTL {self.regs.CTL:x}")
self.regs.CTL |= 2
log.info(f"Started stream {self.stream_id}, CTL {self.regs.CTL:x}")
def stop(self):
log.info(f"Stopping stream {self.stream_id}, CTL {self.regs.CTL:x}")
self.regs.CTL &= 2
time.sleep(0.1)
self.regs.CTL |= 1
log.info(f"Stopped stream {self.stream_id}, CTL {self.regs.CTL:x}")
def setup_buf(self, buf_len: int):
(mem, phys_addr, hugef) = map_phys_mem(self.stream_id)
log.info(f"Mapped 2M huge page at 0x{phys_addr:x} for buf size ({buf_len})")
# create two buffers in the page of buf_len and mark them
# in a buffer descriptor list for the hardware to use
buf0_len = buf_len
buf1_len = buf_len
bdl_off = buf0_len + buf1_len
# bdl is 2 (64bits, 16 bytes) per entry, we have two
mem[bdl_off:bdl_off + 32] = struct.pack("<QQQQ",
phys_addr,
buf0_len,
phys_addr + buf0_len,
buf1_len)
dpib_off = bdl_off+32
# ensure buffer is initialized, sanity
for i in range(0, buf_len*2):
mem[i] = 0
log.info("Filled the buffer descriptor list (BDL) for DMA.")
return (mem, hugef, phys_addr + bdl_off, phys_addr+dpib_off, 2)
def debug(self):
log.info("HDA %d: PPROC %d, CTL 0x%x, LPIB 0x%x, BDPU 0x%x, BDPL 0x%x, CBL 0x%x, LVI 0x%x",
self.stream_id, (hda.PPCTL >> self.stream_id) & 1, self.regs.CTL, self.regs.LPIB, self.regs.BDPU,
self.regs.BDPL, self.regs.CBL, self.regs.LVI)
log.info(" FIFOW %d, FIFOS %d, FMT %x, FIFOL %d, DPIB %d, EFIFOS %d",
self.regs.FIFOW & 0x7, self.regs.FIFOS, self.regs.FMT, self.regs.FIFOL, self.dbg0.DPIB, self.dbg0.EFIFOS)
log.info(" status: FIFORDY %d, DESE %d, FIFOE %d, BCIS %d",
(self.regs.STS >> 5) & 1, (self.regs.STS >> 4) & 1, (self.regs.STS >> 3) & 1, (self.regs.STS >> 2) & 1)
def reset(self):
# Turn DMA off and reset the stream. Clearing START first is a
# noop per the spec, but absolutely required for stability.
# Apparently the reset doesn't stop the stream, and the next load
# starts before it's ready and kills the load (and often the DSP).
# The sleep too is required, on at least one board (a fast
# chromebook) putting the two writes next each other also hangs
# the DSP!
log.info(f"Resetting stream {self.stream_id}")
self.debug()
self.regs.CTL &= ~2 # clear START
time.sleep(0.1)
# set enter reset bit
self.regs.CTL = 1
while (self.regs.CTL & 1) == 0: pass
# clear enter reset bit to exit reset
self.regs.CTL = 0
while (self.regs.CTL & 1) == 1: pass
log.info(f"Disable SPIB and set position 0 of stream {self.stream_id}")
self.hda.SPBFCTL = 0
self.hda.SPIB = 0
#log.info("Setting dma position buffer and enable it")
#self.hda.DPUBASE = self.pos_buf_addr >> 32 & 0xffffffff
#self.hda.DPLBASE = self.pos_buf_addr & 0xfffffff0 | 1
log.info(f"Enabling dsp capture (PROCEN) of stream {self.stream_id}")
self.hda.PPCTL |= (1 << self.stream_id)
self.debug()
log.info(f"Reset stream {self.stream_id}")
def map_regs():
p = runx(f"grep -iPl 'PCI_CLASS=40(10|38)0' /sys/bus/pci/devices/*/uevent")
pcidir = os.path.dirname(p)
soc/intel_adsp: Fixups for cavs18 SMP boot This platform was stale for a long time and got a little left behind. Basic OS stuff was working but secondary core bringup didn't. It has a slightly different set of choices from the "weird hardware quirks" menu: + Like cAVS 1.5, it boots from a ROM that needs a short delay after power-up before it can receive the startup IDC. + But, like 2.5 and unlike 1.5, it doesn't start running until the PWRCTL bit for the core gets set by DSP software (1.5 gets launched by the host). So the delay needed to move down a bit. + It wants that PWRCTL bit to be set last, after CLKCTL enables the clock. (Which makes sense I guess: EE classes always tell you hold circuits in reset while an initial clock propagates). Not sure why it was in the reverse order originally; this way works for everything. + The ROM likes to scribble on the interrupt controller and mask its own IDC interrupts after we've already set it up. They have to be unmasked. We had code to do this already, thinking it was a workaround for legecy SOF code (that we never actually located). Now I'm thinking it was this behavior all along being detected by SOF's more extensive hardware CI. Take out the test and do it always, it's like nine instructions. + The host/loader-side behavior is a mix of 1.5 and 2.5. It won't actually start the secondary cores under host command, but it does need to see bits set for them in ADSPCS for the DSP-initiated power-up to work (2.5 would just ignore all but core 0's bits). + Also, like 1.5, it needs the host DMA stream to be explicitly stopped (and not just reset) or else further loads will be unstable. Note that the loader changes now require more logic than just "1.5 or not", so the platform detection has been enhanced to fully categorize the device based on PCI ID (not quite: we don't have any 2.0 platform hardware, so I left that alone for now). Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
2022-01-21 16:22:37 -08:00
# Platform/quirk detection. ID lists cribbed from the SOF kernel driver
global cavs15, cavs18, cavs25
did = int(open(f"{pcidir}/device").read().rstrip(), 16)
cavs15 = did in [ 0x5a98, 0x1a98, 0x3198 ]
cavs18 = did in [ 0x9dc8, 0xa348, 0x02c8, 0x06c8, 0xa3f0 ]
cavs25 = did in [ 0xa0c8, 0x43c8, 0x4b55, 0x4b58, 0x7ad0, 0x51c8 ]
# Check sysfs for a loaded driver and remove it
if os.path.exists(f"{pcidir}/driver"):
mod = os.path.basename(os.readlink(f"{pcidir}/driver/module"))
found_msg = f"Existing driver \"{mod}\" found"
if args.log_only:
log.info(found_msg)
else:
log.warning(found_msg + ", unloading module")
runx(f"rmmod -f {mod}")
# Disengage runtime power management so the kernel doesn't put it to sleep
with open(f"{pcidir}/power/control", "w") as ctrl:
ctrl.write("on")
# Make sure PCI memory space access and busmastering are enabled.
# Also disable interrupts so as not to confuse the kernel.
with open(f"{pcidir}/config", "wb+") as cfg:
cfg.seek(4)
cfg.write(b'\x06\x04')
# Standard HD Audio Registers
global hdamem
(hdamem, _) = bar_map(pcidir, 0)
hda = Regs(hdamem)
hda.GCAP = 0x0000
hda.GCTL = 0x0008
hda.SPBFCTL = 0x0704
hda.PPCTL = 0x0804
# Find the ID of the first output stream
hda_ostream_id = (hda.GCAP >> 8) & 0x0f # number of input streams
log.info(f"Selected output stream {hda_ostream_id} (GCAP = 0x{hda.GCAP:x})")
hda.SD_SPIB = 0x0708 + (8 * hda_ostream_id)
hda.freeze()
# Standard HD Audio Stream Descriptor
sd = Regs(hdamem + 0x0080 + (hda_ostream_id * 0x20))
sd.CTL = 0x00
sd.CBL = 0x08
sd.LVI = 0x0c
sd.BDPL = 0x18
sd.BDPU = 0x1c
sd.freeze()
# Intel Audio DSP Registers
global bar4_mmap
(bar4_mem, bar4_mmap) = bar_map(pcidir, 4)
dsp = Regs(bar4_mem)
dsp.ADSPCS = 0x00004
dsp.HIPCTDR = 0x00040 if cavs15 else 0x000c0
dsp.HIPCTDA = 0x000c4 # 1.8+ only
dsp.HIPCTDD = 0x00044 if cavs15 else 0x000c8
dsp.HIPCIDR = 0x00048 if cavs15 else 0x000d0
dsp.HIPCIDA = 0x000d4 # 1.8+ only
dsp.HIPCIDD = 0x0004c if cavs15 else 0x000d8
dsp.SRAM_FW_STATUS = 0x80000 # Start of first SRAM window
dsp.freeze()
soc/intel_adsp: Fixups for cavs18 SMP boot This platform was stale for a long time and got a little left behind. Basic OS stuff was working but secondary core bringup didn't. It has a slightly different set of choices from the "weird hardware quirks" menu: + Like cAVS 1.5, it boots from a ROM that needs a short delay after power-up before it can receive the startup IDC. + But, like 2.5 and unlike 1.5, it doesn't start running until the PWRCTL bit for the core gets set by DSP software (1.5 gets launched by the host). So the delay needed to move down a bit. + It wants that PWRCTL bit to be set last, after CLKCTL enables the clock. (Which makes sense I guess: EE classes always tell you hold circuits in reset while an initial clock propagates). Not sure why it was in the reverse order originally; this way works for everything. + The ROM likes to scribble on the interrupt controller and mask its own IDC interrupts after we've already set it up. They have to be unmasked. We had code to do this already, thinking it was a workaround for legecy SOF code (that we never actually located). Now I'm thinking it was this behavior all along being detected by SOF's more extensive hardware CI. Take out the test and do it always, it's like nine instructions. + The host/loader-side behavior is a mix of 1.5 and 2.5. It won't actually start the secondary cores under host command, but it does need to see bits set for them in ADSPCS for the DSP-initiated power-up to work (2.5 would just ignore all but core 0's bits). + Also, like 1.5, it needs the host DMA stream to be explicitly stopped (and not just reset) or else further loads will be unstable. Note that the loader changes now require more logic than just "1.5 or not", so the platform detection has been enhanced to fully categorize the device based on PCI ID (not quite: we don't have any 2.0 platform hardware, so I left that alone for now). Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
2022-01-21 16:22:37 -08:00
return (hda, sd, dsp, hda_ostream_id)
def setup_dma_mem(fw_bytes):
(mem, phys_addr, _) = map_phys_mem(hda_ostream_id)
mem[0:len(fw_bytes)] = fw_bytes
log.info("Mapped 2M huge page at 0x%x to contain %d bytes of firmware"
% (phys_addr, len(fw_bytes)))
# HDA requires at least two buffers be defined, but we don't care about
# boundaries because it's all a contiguous region. Place a vestigial
# 128-byte (minimum size and alignment) buffer after the main one, and put
# the 4-entry BDL list into the final 128 bytes of the page.
buf0_len = HUGEPAGESZ - 2 * 128
buf1_len = 128
bdl_off = buf0_len + buf1_len
mem[bdl_off:bdl_off + 32] = struct.pack("<QQQQ",
phys_addr, buf0_len,
phys_addr + buf0_len, buf1_len)
log.info("Filled the buffer descriptor list (BDL) for DMA.")
return (phys_addr + bdl_off, 2)
global_mmaps = [] # protect mmap mappings from garbage collection!
# Maps 2M of contiguous memory using a single page from hugetlbfs,
# then locates its physical address for use as a DMA buffer.
def map_phys_mem(stream_id):
# Make sure hugetlbfs is mounted (not there on chromeos)
os.system("mount | grep -q hugetlbfs ||"
+ " (mkdir -p /dev/hugepages; "
+ " mount -t hugetlbfs hugetlbfs /dev/hugepages)")
# Ensure the kernel has enough budget for one new page
free = int(runx("awk '/HugePages_Free/ {print $2}' /proc/meminfo"))
if free == 0:
tot = 1 + int(runx("awk '/HugePages_Total/ {print $2}' /proc/meminfo"))
os.system(f"echo {tot} > /proc/sys/vm/nr_hugepages")
hugef_name = HUGEPAGE_FILE + str(stream_id)
hugef = open(hugef_name, "w+")
hugef.truncate(HUGEPAGESZ)
mem = mmap.mmap(hugef.fileno(), HUGEPAGESZ)
log.info("type of mem is %s", str(type(mem)))
global_mmaps.append(mem)
os.unlink(hugef_name)
# Find the local process address of the mapping, then use that to extract
# the physical address from the kernel's pagemap interface. The physical
# page frame number occupies the bottom bits of the entry.
mem[0] = 0 # Fault the page in so it has an address!
vaddr = ctypes.addressof(ctypes.c_int.from_buffer(mem))
vpagenum = vaddr >> 12
pagemap = open("/proc/self/pagemap", "rb")
pagemap.seek(vpagenum * 8)
pent = pagemap.read(8)
paddr = (struct.unpack("Q", pent)[0] & ((1 << 55) - 1)) * PAGESZ
pagemap.close()
return (mem, paddr, hugef)
# Maps a PCI BAR and returns the in-process address
def bar_map(pcidir, barnum):
f = open(pcidir + "/resource" + str(barnum), "r+")
mm = mmap.mmap(f.fileno(), os.fstat(f.fileno()).st_size)
global_mmaps.append(mm)
log.info("Mapped PCI bar %d of length %d bytes."
% (barnum, os.fstat(f.fileno()).st_size))
return (ctypes.addressof(ctypes.c_int.from_buffer(mm)), mm)
# Syntactic sugar to make register block definition & use look nice.
# Instantiate from a base address, assign offsets to (uint32) named registers as
# fields, call freeze(), then the field acts as a direct alias for the register!
class Regs:
def __init__(self, base_addr):
vars(self)["base_addr"] = base_addr
vars(self)["ptrs"] = {}
vars(self)["frozen"] = False
def freeze(self):
vars(self)["frozen"] = True
def __setattr__(self, name, val):
if not self.frozen and name not in self.ptrs:
addr = self.base_addr + val
self.ptrs[name] = ctypes.c_uint32.from_address(addr)
else:
self.ptrs[name].value = val
def __getattr__(self, name):
return self.ptrs[name].value
def runx(cmd):
return subprocess.check_output(cmd, shell=True).decode().rstrip()
def load_firmware(fw_file):
try:
fw_bytes = open(fw_file, "rb").read()
except Exception as e:
log.error(f"Could not read firmware file: `{fw_file}'")
log.error(e)
sys.exit(1)
(magic, sz) = struct.unpack("4sI", fw_bytes[0:8])
if magic == b'XMan':
log.info(f"Trimming {sz} bytes of extended manifest")
fw_bytes = fw_bytes[sz:len(fw_bytes)]
# This actually means "enable access to BAR4 registers"!
hda.PPCTL |= (1 << 30) # GPROCEN, "global processing enable"
log.info("Resetting HDA device")
hda.GCTL = 0
while hda.GCTL & 1: pass
hda.GCTL = 1
while not hda.GCTL & 1: pass
log.info("Powering down DSP cores")
dsp.ADSPCS = 0xffff
while dsp.ADSPCS & 0xff000000: pass
log.info(f"Configuring HDA stream {hda_ostream_id} to transfer firmware image")
(buf_list_addr, num_bufs) = setup_dma_mem(fw_bytes)
sd.CTL = 1
while (sd.CTL & 1) == 0: pass
sd.CTL = 0
while (sd.CTL & 1) == 1: pass
sd.CTL = (1 << 20) # Set stream ID to anything non-zero
sd.BDPU = (buf_list_addr >> 32) & 0xffffffff
sd.BDPL = buf_list_addr & 0xffffffff
sd.CBL = len(fw_bytes)
sd.LVI = num_bufs - 1
hda.PPCTL |= (1 << hda_ostream_id)
# SPIB ("Software Position In Buffer") is an Intel HDA extension
# that puts a transfer boundary into the stream beyond which the
# other side will not read. The ROM wants to poll on a "buffer
# full" bit on the other side that only works with this enabled.
hda.SPBFCTL |= (1 << hda_ostream_id)
hda.SD_SPIB = len(fw_bytes)
soc/intel_adsp: Fixups for cavs18 SMP boot This platform was stale for a long time and got a little left behind. Basic OS stuff was working but secondary core bringup didn't. It has a slightly different set of choices from the "weird hardware quirks" menu: + Like cAVS 1.5, it boots from a ROM that needs a short delay after power-up before it can receive the startup IDC. + But, like 2.5 and unlike 1.5, it doesn't start running until the PWRCTL bit for the core gets set by DSP software (1.5 gets launched by the host). So the delay needed to move down a bit. + It wants that PWRCTL bit to be set last, after CLKCTL enables the clock. (Which makes sense I guess: EE classes always tell you hold circuits in reset while an initial clock propagates). Not sure why it was in the reverse order originally; this way works for everything. + The ROM likes to scribble on the interrupt controller and mask its own IDC interrupts after we've already set it up. They have to be unmasked. We had code to do this already, thinking it was a workaround for legecy SOF code (that we never actually located). Now I'm thinking it was this behavior all along being detected by SOF's more extensive hardware CI. Take out the test and do it always, it's like nine instructions. + The host/loader-side behavior is a mix of 1.5 and 2.5. It won't actually start the secondary cores under host command, but it does need to see bits set for them in ADSPCS for the DSP-initiated power-up to work (2.5 would just ignore all but core 0's bits). + Also, like 1.5, it needs the host DMA stream to be explicitly stopped (and not just reset) or else further loads will be unstable. Note that the loader changes now require more logic than just "1.5 or not", so the platform detection has been enhanced to fully categorize the device based on PCI ID (not quite: we don't have any 2.0 platform hardware, so I left that alone for now). Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
2022-01-21 16:22:37 -08:00
# Start DSP. Host needs to provide power to all cores on 1.5
# (which also starts them) and 1.8 (merely gates power, DSP also
# has to set PWRCTL). The bits for cores other than 0 are ignored
# on 2.5 where the DSP has full control.
log.info(f"Starting DSP, ADSPCS = 0x{dsp.ADSPCS:x}")
soc/intel_adsp: Fixups for cavs18 SMP boot This platform was stale for a long time and got a little left behind. Basic OS stuff was working but secondary core bringup didn't. It has a slightly different set of choices from the "weird hardware quirks" menu: + Like cAVS 1.5, it boots from a ROM that needs a short delay after power-up before it can receive the startup IDC. + But, like 2.5 and unlike 1.5, it doesn't start running until the PWRCTL bit for the core gets set by DSP software (1.5 gets launched by the host). So the delay needed to move down a bit. + It wants that PWRCTL bit to be set last, after CLKCTL enables the clock. (Which makes sense I guess: EE classes always tell you hold circuits in reset while an initial clock propagates). Not sure why it was in the reverse order originally; this way works for everything. + The ROM likes to scribble on the interrupt controller and mask its own IDC interrupts after we've already set it up. They have to be unmasked. We had code to do this already, thinking it was a workaround for legecy SOF code (that we never actually located). Now I'm thinking it was this behavior all along being detected by SOF's more extensive hardware CI. Take out the test and do it always, it's like nine instructions. + The host/loader-side behavior is a mix of 1.5 and 2.5. It won't actually start the secondary cores under host command, but it does need to see bits set for them in ADSPCS for the DSP-initiated power-up to work (2.5 would just ignore all but core 0's bits). + Also, like 1.5, it needs the host DMA stream to be explicitly stopped (and not just reset) or else further loads will be unstable. Note that the loader changes now require more logic than just "1.5 or not", so the platform detection has been enhanced to fully categorize the device based on PCI ID (not quite: we don't have any 2.0 platform hardware, so I left that alone for now). Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
2022-01-21 16:22:37 -08:00
dsp.ADSPCS = 0xff0000 if not cavs25 else 0x01fefe
while (dsp.ADSPCS & 0x1000000) == 0: pass
# Wait for the ROM to boot and signal it's ready. This short
# sleep seems to be needed; if we're banging on the memory window
# during initial boot (before/while the window control registers
# are configured?) the DSP hardware will hang fairly reliably.
log.info("Wait for ROM startup")
time.sleep(0.1)
while (dsp.SRAM_FW_STATUS >> 24) != 5: pass
# Send the DSP an IPC message to tell the device how to boot.
# Note: with cAVS 1.8+ the ROM receives the stream argument as an
# index within the array of output streams (and we always use the
# first one by construction). But with 1.5 it's the HDA index,
# and depends on the number of input streams on the device.
stream_idx = hda_ostream_id if cavs15 else 0
ipcval = ( (1 << 31) # BUSY bit
| (0x01 << 24) # type = PURGE_FW
| (1 << 14) # purge_fw = 1
| (stream_idx << 9)) # dma_id
log.info(f"Sending IPC command, HIPIDR = 0x{ipcval:x}")
dsp.HIPCIDR = ipcval
log.info(f"Starting DMA, FW_STATUS = 0x{dsp.SRAM_FW_STATUS:x}")
sd.CTL |= 2 # START flag
wait_fw_entered()
soc/intel_adsp: Fixups for cavs18 SMP boot This platform was stale for a long time and got a little left behind. Basic OS stuff was working but secondary core bringup didn't. It has a slightly different set of choices from the "weird hardware quirks" menu: + Like cAVS 1.5, it boots from a ROM that needs a short delay after power-up before it can receive the startup IDC. + But, like 2.5 and unlike 1.5, it doesn't start running until the PWRCTL bit for the core gets set by DSP software (1.5 gets launched by the host). So the delay needed to move down a bit. + It wants that PWRCTL bit to be set last, after CLKCTL enables the clock. (Which makes sense I guess: EE classes always tell you hold circuits in reset while an initial clock propagates). Not sure why it was in the reverse order originally; this way works for everything. + The ROM likes to scribble on the interrupt controller and mask its own IDC interrupts after we've already set it up. They have to be unmasked. We had code to do this already, thinking it was a workaround for legecy SOF code (that we never actually located). Now I'm thinking it was this behavior all along being detected by SOF's more extensive hardware CI. Take out the test and do it always, it's like nine instructions. + The host/loader-side behavior is a mix of 1.5 and 2.5. It won't actually start the secondary cores under host command, but it does need to see bits set for them in ADSPCS for the DSP-initiated power-up to work (2.5 would just ignore all but core 0's bits). + Also, like 1.5, it needs the host DMA stream to be explicitly stopped (and not just reset) or else further loads will be unstable. Note that the loader changes now require more logic than just "1.5 or not", so the platform detection has been enhanced to fully categorize the device based on PCI ID (not quite: we don't have any 2.0 platform hardware, so I left that alone for now). Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
2022-01-21 16:22:37 -08:00
# Turn DMA off and reset the stream. Clearing START first is a
# noop per the spec, but absolutely required for stability.
# Apparently the reset doesn't stop the stream, and the next load
# starts before it's ready and kills the load (and often the DSP).
# The sleep too is required, on at least one board (a fast
# chromebook) putting the two writes next each other also hangs
# the DSP!
sd.CTL &= ~2 # clear START
time.sleep(0.1)
sd.CTL |= 1
log.info(f"cAVS firmware load complete")
def wait_fw_entered():
log.info("Waiting for firmware handoff, FW_STATUS = 0x%x", dsp.SRAM_FW_STATUS)
for _ in range(200):
alive = dsp.SRAM_FW_STATUS & ((1 << 28) - 1) == 5 # "FW_ENTERED"
if alive:
break
time.sleep(0.01)
if not alive:
log.warning("Load failed? FW_STATUS = 0x%x", dsp.SRAM_FW_STATUS)
# This SHOULD be just "mem[start:start+length]", but slicing an mmap
# array seems to be unreliable on one of my machines (python 3.6.9 on
# Ubuntu 18.04). Read out bytes individually.
def win_read(start, length):
try:
return b''.join(bar4_mmap[WINSTREAM_OFFSET + x].to_bytes(1, 'little')
for x in range(start, start + length))
except IndexError as ie:
# A FW in a bad state may cause winstream garbage
log.error("IndexError in bar4_mmap[%d + %d]", WINSTREAM_OFFSET, start)
log.error("bar4_mmap.size()=%d", bar4_mmap.size())
raise ie
def win_hdr():
return struct.unpack("<IIII", win_read(0, 16))
# Python implementation of the same algorithm in sys_winstream_read(),
# see there for details.
def winstream_read(last_seq):
while True:
(wlen, start, end, seq) = win_hdr()
if last_seq == 0:
last_seq = seq if args.no_history else (seq - ((end - start) % wlen))
if seq == last_seq or start == end:
return (seq, "")
behind = seq - last_seq
if behind > ((end - start) % wlen):
return (seq, "")
copy = (end - behind) % wlen
suffix = min(behind, wlen - copy)
result = win_read(16 + copy, suffix)
if suffix < behind:
result += win_read(16, behind - suffix)
(wlen, start1, end, seq1) = win_hdr()
if start1 == start and seq1 == seq:
# Best effort attempt at decoding, replacing unusable characters
# Found to be useful when it really goes wrong
return (seq, result.decode("utf-8", "replace"))
async def ipc_delay_done():
await asyncio.sleep(0.1)
dsp.HIPCTDA = 1<<31
ipc_timestamp = 0
# Super-simple command language, driven by the test code on the DSP
def ipc_command(data, ext_data):
send_msg = False
done = True
log.debug ("ipc data %d, ext_data %x", data, ext_data)
if data == 0: # noop, with synchronous DONE
pass
elif data == 1: # async command: signal DONE after a delay (on 1.8+)
if not cavs15:
done = False
asyncio.ensure_future(ipc_delay_done())
elif data == 2: # echo back ext_data as a message command
send_msg = True
elif data == 3: # set ADSPCS
dsp.ADSPCS = ext_data
elif data == 4: # echo back microseconds since last timestamp command
global ipc_timestamp
t = round(time.time() * 1e6)
ext_data = t - ipc_timestamp
ipc_timestamp = t
send_msg = True
elif data == 5: # copy word at outbox[ext_data >> 16] to inbox[ext_data & 0xffff]
src = OUTBOX_OFFSET + 4 * (ext_data >> 16)
dst = INBOX_OFFSET + 4 * (ext_data & 0xffff)
for i in range(4):
bar4_mmap[dst + i] = bar4_mmap[src + i]
elif data == 6: # HDA RESET (init if not exists)
stream_id = ext_data & 0xff
if stream_id in hda_streams:
hda_streams[stream_id].reset()
else:
hda_str = HDAStream(stream_id)
hda_streams[stream_id] = hda_str
elif data == 7: # HDA CONFIG
stream_id = ext_data & 0xFF
buf_len = ext_data >> 8 & 0xFFFF
hda_str = hda_streams[stream_id]
hda_str.config(buf_len)
elif data == 8: # HDA START
stream_id = ext_data & 0xFF
hda_streams[stream_id].start()
hda_streams[stream_id].mem.seek(0)
elif data == 9: # HDA STOP
stream_id = ext_data & 0xFF
hda_streams[stream_id].stop()
elif data == 10: # HDA VALIDATE
stream_id = ext_data & 0xFF
hda_str = hda_streams[stream_id]
hda_str.debug()
is_ramp_data = True
hda_str.mem.seek(0)
for (i, val) in enumerate(hda_str.mem.read(256)):
if i != val:
is_ramp_data = False
# log.info("stream[%d][%d]: %d", stream_id, i, val) # debug helper
log.info("Is ramp data? " + str(is_ramp_data))
ext_data = int(is_ramp_data)
log.info(f"Ext data to send back on ramp status {ext_data}")
send_msg = True
elif data == 11: # HDA HOST OUT SEND
stream_id = ext_data & 0xff
buf = bytearray(256)
for i in range(0, 256):
buf[i] = i
hda_streams[stream_id].write(buf)
elif data == 12: # HDA PRINT
log.info("Doing HDA Print")
stream_id = ext_data & 0xFF
buf_len = ext_data >> 8 & 0xFFFF
hda_str = hda_streams[stream_id]
pos = hda_str.mem.tell()
buf_data = hda_str.mem.read(buf_len).decode("utf-8", "replace")
log.info(f"DSP LOG MSG (idx: {pos}, len: {buf_len}): {buf_data}")
pos = hda_str.mem.tell()
if pos >= hda_str.buf_len*2:
log.info(f"Wrapping log reader, pos {pos} len {hda_str.buf_len}")
hda_str.mem.seek(0)
else:
log.warning(f"cavstool: Unrecognized IPC command 0x{data:x} ext 0x{ext_data:x}")
dsp.HIPCTDR = 1<<31 # Ack local interrupt, also signals DONE on v1.5
if cavs18:
time.sleep(0.01) # Needed on 1.8, or the command below won't send!
if done and not cavs15:
dsp.HIPCTDA = 1<<31 # Signal done
if send_msg:
dsp.HIPCIDD = ext_data
dsp.HIPCIDR = (1<<31) | ext_data
async def _main(server):
#TODO this bit me, remove the globals, write a little FirmwareLoader class or something to contain.
global hda, sd, dsp, hda_ostream_id, hda_streams
try:
soc/intel_adsp: Fixups for cavs18 SMP boot This platform was stale for a long time and got a little left behind. Basic OS stuff was working but secondary core bringup didn't. It has a slightly different set of choices from the "weird hardware quirks" menu: + Like cAVS 1.5, it boots from a ROM that needs a short delay after power-up before it can receive the startup IDC. + But, like 2.5 and unlike 1.5, it doesn't start running until the PWRCTL bit for the core gets set by DSP software (1.5 gets launched by the host). So the delay needed to move down a bit. + It wants that PWRCTL bit to be set last, after CLKCTL enables the clock. (Which makes sense I guess: EE classes always tell you hold circuits in reset while an initial clock propagates). Not sure why it was in the reverse order originally; this way works for everything. + The ROM likes to scribble on the interrupt controller and mask its own IDC interrupts after we've already set it up. They have to be unmasked. We had code to do this already, thinking it was a workaround for legecy SOF code (that we never actually located). Now I'm thinking it was this behavior all along being detected by SOF's more extensive hardware CI. Take out the test and do it always, it's like nine instructions. + The host/loader-side behavior is a mix of 1.5 and 2.5. It won't actually start the secondary cores under host command, but it does need to see bits set for them in ADSPCS for the DSP-initiated power-up to work (2.5 would just ignore all but core 0's bits). + Also, like 1.5, it needs the host DMA stream to be explicitly stopped (and not just reset) or else further loads will be unstable. Note that the loader changes now require more logic than just "1.5 or not", so the platform detection has been enhanced to fully categorize the device based on PCI ID (not quite: we don't have any 2.0 platform hardware, so I left that alone for now). Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
2022-01-21 16:22:37 -08:00
(hda, sd, dsp, hda_ostream_id) = map_regs()
except Exception as e:
log.error("Could not map device in sysfs; run as root?")
log.error(e)
sys.exit(1)
log.info(f"Detected cAVS {'1.5' if cavs15 else '1.8+'} hardware")
if args.log_only:
wait_fw_entered()
else:
if not fw_file:
log.error("Firmware file argument missing")
sys.exit(1)
load_firmware(fw_file)
time.sleep(0.1)
if not args.quiet:
adsp_log("--\n", server)
hda_streams = dict()
last_seq = 0
while start_output is True:
await asyncio.sleep(0.03)
(last_seq, output) = winstream_read(last_seq)
if output:
adsp_log(output, server)
if dsp.HIPCTDR & 0x80000000:
ipc_command(dsp.HIPCTDR & ~0x80000000, dsp.HIPCTDD)
if dsp.HIPCIDA & 0x80000000:
dsp.HIPCIDA = 1<<31 # must ACK any DONE interrupts that arrive!
class adsp_request_handler(socketserver.BaseRequestHandler):
"""
The request handler class for control the actions of server.
"""
def receive_fw(self, filename):
try:
with open(fw_file,'wb') as f:
cnt = 0
log.info("Receiving...")
while True:
l = self.request.recv(BUF_SIZE)
ll = len(l)
cnt = cnt + ll
if not l:
break
else:
f.write(l)
except Exception as e:
log.error(f"Get exception {e} during FW transfer.")
return 1
log.info(f"Done Receiving {cnt}.")
def handle(self):
global start_output, fw_file
cmd = self.request.recv(BUF_SIZE)
log.info(f"{self.client_address[0]} wrote: {cmd}")
action = cmd.decode("utf-8")
log.debug(f'load {action}')
if action == CMD_DOWNLOAD:
self.request.sendall(cmd)
recv_fn = self.request.recv(BUF_SIZE)
log.info(f"{self.client_address[0]} wrote: {recv_fn}")
try:
tmp_file = recv_fn.decode("utf-8")
except UnicodeDecodeError:
tmp_file = "zephyr.ri.decode_error"
log.info(f'did not receive a correct filename')
lock.acquire()
fw_file = tmp_file
ret = self.receive_fw(fw_file)
if not ret:
start_output = True
lock.release()
log.debug(f'{recv_fn}, {fw_file}, {start_output}')
elif action == CMD_LOG_STOP:
self.request.sendall(cmd)
lock.acquire()
start_output = False
if fw_file:
os.remove(fw_file)
fw_file = None
lock.release()
else:
log.error("incorrect load communitcation!")
class adsp_log_handler(socketserver.BaseRequestHandler):
"""
The log handler class for grabbing output messages of server.
"""
def run_adsp(self):
self.loop = asyncio.get_event_loop()
self.loop.run_until_complete(_main(self))
def handle(self):
global start_output, fw_file
cmd = self.request.recv(BUF_SIZE)
log.info(f"{self.client_address[0]} wrote: {cmd}")
action = cmd.decode("utf-8")
log.debug(f'monitor {action}')
if action == CMD_LOG_START:
self.request.sendall(cmd)
log.info(f"Waiting for instruction...")
while start_output is False:
time.sleep(1)
log.info(f"Loaded FW {fw_file} and running...")
if os.path.exists(fw_file):
self.run_adsp()
self.request.sendall("service complete.".encode())
log.info("service complete.")
else:
log.error("cannot find the FW file")
lock.acquire()
fw_file = None
start_output = False
lock.release()
else:
log.error("incorrect monitor communitcation!")
def adsp_log(output, server):
if server:
server.request.sendall(output.encode("utf-8"))
else:
sys.stdout.write(output)
sys.stdout.flush()
def get_host_ip():
"""
Helper tool use to detect host's serving ip address.
"""
interfaces = netifaces.interfaces()
for i in interfaces:
if i != "lo":
try:
netifaces.ifaddresses(i)
ip = netifaces.ifaddresses(i)[netifaces.AF_INET][0]['addr']
log.info (f"Use interface {i}, IP address: {ip}")
except Exception:
log.info(f"Ignore the interface {i} which is not activated.")
return ip
ap = argparse.ArgumentParser(description="DSP loader/logger tool")
ap.add_argument("-q", "--quiet", action="store_true",
help="No loader output, just DSP logging")
ap.add_argument("-l", "--log-only", action="store_true",
help="Don't load firmware, just show log output")
ap.add_argument("-n", "--no-history", action="store_true",
help="No current log buffer at start, just new output")
ap.add_argument("-s", "--server-addr",
help="No current log buffer at start, just new output")
ap.add_argument("fw_file", nargs="?", help="Firmware file")
args = ap.parse_args()
if args.quiet:
log.setLevel(logging.WARN)
if args.fw_file:
fw_file = args.fw_file
else:
fw_file = None
if args.server_addr:
HOST = args.server_addr
else:
HOST = get_host_ip()
if __name__ == "__main__":
# When fw_file is assigned or in log_only mode, it will
# not serve as a daemon. That mean it just run load
# firmware or read the log directly.
if args.fw_file or args.log_only:
start_output = True
try:
asyncio.get_event_loop().run_until_complete(_main(None))
except KeyboardInterrupt:
start_output = False
finally:
sys.exit(0)
# Launch the command request service
socketserver.TCPServer.allow_reuse_address = True
req_server = socketserver.TCPServer((HOST, PORT_REQ), adsp_request_handler)
req_t = threading.Thread(target=req_server.serve_forever, daemon=True)
# Activate the log service which output adsp execution
with socketserver.TCPServer((HOST, PORT_LOG), adsp_log_handler) as log_server:
try:
log.info("Req server start...")
req_t.start()
log.info("Log server start...")
log_server.serve_forever()
except KeyboardInterrupt:
lock.acquire()
start_output = False
lock.release()
log_server.shutdown()
req_server.shutdown()