| Message ID | 1392318930-8771-3-git-send-email-liam.r.girdwood@linux.intel.com (mailing list archive) |
|---|---|
| State | Accepted |
| Delegated to: | Mark Brown |
| Headers | show |
At Thu, 13 Feb 2014 19:15:28 +0000, Liam Girdwood wrote: > > Provide services for Intel SST drivers to load SST modular firmware. > > SST Firmware can be made up of several modules. These modules can exist > within any of the compatible SST memory blocks. Provide a generic memory > block and firmware module manager that can be used with any SST firmware > and core. > > Signed-off-by: Liam Girdwood <liam.r.girdwood@linux.intel.com> > --- > sound/soc/intel/sst-firmware.c | 805 +++++++++++++++++++++++++++++++++++++++++ > 1 file changed, 805 insertions(+) > create mode 100644 sound/soc/intel/sst-firmware.c > > diff --git a/sound/soc/intel/sst-firmware.c b/sound/soc/intel/sst-firmware.c > new file mode 100644 > index 0000000..8126b46 > --- /dev/null > +++ b/sound/soc/intel/sst-firmware.c > @@ -0,0 +1,805 @@ > +/* > + * Intel SST Firmware Loader > + * > + * Copyright (C) 2013, Intel Corporation. All rights reserved. > + * > + * This program is free software; you can redistribute it and/or > + * modify it under the terms of the GNU General Public License version > + * 2 as published by the Free Software Foundation. > + * > + * This program is distributed in the hope that it will be useful, > + * but WITHOUT ANY WARRANTY; without even the implied warranty of > + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the > + * GNU General Public License for more details. > + * > + */ > + > +#include <linux/kernel.h> > +#include <linux/slab.h> > +#include <linux/sched.h> > +#include <linux/firmware.h> > +#include <linux/export.h> > +#include <linux/platform_device.h> > +#include <linux/dma-mapping.h> > +#include <linux/dmaengine.h> > +#include <linux/pci.h> > + > +/* supported DMA engine drivers */ > +#include <linux/dw_dmac.h> > + > +#include <asm/page.h> > +#include <asm/pgtable.h> > + > +#include "sst-dsp.h" > +#include "sst-dsp-priv.h" > + > +#define SST_DMA_RESOURCES 2 > +#define SST_DSP_DMA_MAX_BURST 0x3 > + > +struct sst_dma { > + struct sst_dsp *sst; > + > + struct platform_device *dma_dev; > + struct resource dma_resource[SST_DMA_RESOURCES]; > + struct dma_async_tx_descriptor *desc; > + struct dma_chan *ch; > +}; > + > +static void sst_memcpy32(void *dest, void *src, int bytes) > +{ > + int i; > + > + /* copy one 32 bit word at a time as 64 bit access is not supported */ > + for (i = 0; i < bytes; i += 4) > + memcpy_toio(dest + i, src + i, 4); > +} > + > +static void sst_dma_transfer_complete(void *arg) > +{ > + struct sst_dsp *sst = (struct sst_dsp *)arg; > + > + dev_dbg(sst->dev, "DMA: callback\n"); > +} > + > +int sst_dsp_dma_copy(struct sst_dsp *sst, dma_addr_t src_addr, > + dma_addr_t dest_addr, size_t size) > +{ > + struct dma_async_tx_descriptor *desc; > + struct sst_dma *dma = sst->dma; > + > + if (dma->ch == NULL) { > + dev_err(sst->dev, "error: no DMA channel\n"); > + return -ENODEV; > + } > + > + dev_dbg(sst->dev, "DMA: src: 0x%lx dest 0x%lx size %zu\n", > + (unsigned long)src_addr, (unsigned long)dest_addr, size); > + > + desc = dma->ch->device->device_prep_dma_memcpy(dma->ch, dest_addr, > + src_addr, size, DMA_CTRL_ACK); > + if (!desc){ > + dev_err(sst->dev, "error: dma prep memcpy failed\n"); > + return -EINVAL; > + } > + > + desc->callback = sst_dma_transfer_complete; > + desc->callback_param = sst; > + > + desc->tx_submit(desc); > + dma_wait_for_async_tx(desc); > + > + return 0; > +} > +EXPORT_SYMBOL_GPL(sst_dsp_dma_copy); > + > +static bool dma_chan_filter(struct dma_chan *chan, void *param) > +{ > + struct sst_dsp *dsp = (struct sst_dsp *)param; > + struct sst_dma *dma = dsp->dma; > + > + /* only accept channels from this device */ > + if (chan->device->dev != &dma->dma_dev->dev) > + return false; > + > + /* todo: add chan_id testing */ > + return true; > +} > + > +int sst_dsp_dma_get_channel(struct sst_dsp *dsp, int chan_id) > +{ > + struct sst_dma *dma = dsp->dma; > + struct dma_slave_config slave; > + dma_cap_mask_t mask; > + int ret; > + > + /* The Intel MID DMA engine driver needs the slave config set but > + * Synopsis DMA engine driver safely ignores the slave config */ > + dma_cap_zero(mask); > + dma_cap_set(DMA_SLAVE, mask); > + dma_cap_set(DMA_MEMCPY, mask); > + > + dma->ch = dma_request_channel(mask, dma_chan_filter, dsp); > + if (dma->ch == NULL) { > + dev_err(dsp->dev, "error: DMA request channel failed\n"); > + return -EIO; > + } > + > + memset(&slave, 0, sizeof(slave)); > + slave.direction = DMA_MEM_TO_DEV; > + slave.src_addr_width = > + slave.dst_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES; > + slave.src_maxburst = slave.dst_maxburst = SST_DSP_DMA_MAX_BURST; > + > + ret = dmaengine_slave_config(dma->ch, &slave); > + if (ret) { > + dev_err(dsp->dev, "error: unable to set DMA slave config %d\n", > + ret); > + dma_release_channel(dma->ch); > + dma->ch = NULL; > + } > + > + return ret; > +} > +EXPORT_SYMBOL_GPL(sst_dsp_dma_get_channel); > + > +void sst_dsp_dma_put_channel(struct sst_dsp *dsp) > +{ > + struct sst_dma *dma = dsp->dma; > + > + dma_release_channel(dma->ch); > + dma->ch = NULL; > +} > +EXPORT_SYMBOL_GPL(sst_dsp_dma_put_channel); > + > +/* platform data for DesignWare DMA Engine */ > +static struct dw_dma_platform_data dw_pdata = { > + .chan_allocation_order = CHAN_ALLOCATION_ASCENDING, > + .chan_priority = CHAN_PRIORITY_ASCENDING, > +}; > + > +int sst_dma_new(struct sst_dsp *sst) > +{ > + struct sst_pdata *sst_pdata = sst->pdata; > + struct sst_dma *dma; > + const char *dma_dev_name; > + size_t dma_pdata_size; > + void *dma_pdata; > + > + /* configure the correct platform data for whatever DMA engine > + * is attached to the ADSP IP. */ > + switch (sst->pdata->dma_engine) { > + case SST_DMA_TYPE_DW: > + dma_pdata = &dw_pdata; > + dma_pdata_size = sizeof(dw_pdata); > + dma_dev_name = "dw_dmac"; > + break; > + case SST_DMA_TYPE_MID: > + dma_pdata = NULL; > + dma_pdata_size = 0; > + dma_dev_name = "Intel MID DMA"; > + break; > + default: > + dev_err(sst->dev, "error: invalid DMA engine %d\n", > + sst->pdata->dma_engine); > + return -EINVAL; > + } > + > + dma = devm_kzalloc(sst->dev, sizeof(struct sst_dma), GFP_KERNEL); > + if (!dma) > + return -ENOMEM; > + > + dma->sst = sst; > + sst->dma = dma; > + > + dma->dma_resource[0].start = sst->addr.lpe_base + > + sst_pdata->dma_base; > + dma->dma_resource[0].end = sst->addr.lpe_base + > + sst_pdata->dma_base + > + sst_pdata->dma_size; > + dma->dma_resource[0].flags = IORESOURCE_MEM; > + dma->dma_resource[1].start = sst_pdata->irq; > + dma->dma_resource[1].end = sst_pdata->irq; > + dma->dma_resource[1].flags = IORESOURCE_IRQ; > + > + /* now register DMA engine device */ > + dma->dma_dev = platform_device_register_resndata(sst->dev, > + dma_dev_name, -1, dma->dma_resource, 2, > + dma_pdata, dma_pdata_size); > + > + if (dma->dma_dev == NULL) { > + dev_err(sst->dev, "error: DMA device register failed\n"); > + return -ENODEV; > + } > + > + sst->fw_use_dma = true; > + return 0; > +} > +EXPORT_SYMBOL(sst_dma_new); > + > +void sst_dma_free(struct sst_dma *dma) > +{ > + if (dma->ch) > + dma_release_channel(dma->ch); > + platform_device_unregister(dma->dma_dev); > +} > +EXPORT_SYMBOL(sst_dma_free); > + > +/* create new generic firmware object */ > +struct sst_fw *sst_fw_new(struct sst_dsp *dsp, > + const struct firmware *fw, void *private) > +{ > + struct sst_fw *sst_fw; > + int err; > + > + if (!dsp->ops->parse_fw) > + return NULL; > + > + sst_fw = kzalloc(sizeof(*sst_fw), GFP_KERNEL); > + if (sst_fw == NULL) > + return NULL; > + > + sst_fw->dsp = dsp; > + sst_fw->private = private; > + sst_fw->size = fw->size; > + > + err = dma_coerce_mask_and_coherent(dsp->dev, DMA_BIT_MASK(32)); > + if (err < 0) { > + kfree(sst_fw); > + return NULL; > + } > + > + /* allocate DMA buffer to store FW data */ > + sst_fw->dma_buf = dma_alloc_coherent(dsp->dev, sst_fw->size, > + &sst_fw->dmable_fw_paddr, GFP_DMA); > + if (!sst_fw->dma_buf) { > + dev_err(dsp->dev, "error: DMA alloc failed\n"); > + kfree(sst_fw); > + return NULL; > + } > + > + /* copy FW data to DMA-able memory */ > + memcpy((void *)sst_fw->dma_buf, (void *)fw->data, fw->size); > + release_firmware(fw); Should fw object be really released here? For example, if this function returns NULL after this point, the caller doesn't know whether fw was released or not. It'd be more consistent to release it in the caller side, or always release no matter whether the function succeeds or not. > + > + if (dsp->fw_use_dma) { > + err = sst_dsp_dma_get_channel(dsp, 0); > + if (err < 0) > + goto chan_err; > + } > + > + /* call core specific FW paser to load FW data into DSP */ > + err = dsp->ops->parse_fw(sst_fw); > + if (err < 0) { > + dev_err(dsp->dev, "error: parse fw failed %d\n", err); > + goto parse_err; > + } > + > + if (dsp->fw_use_dma) > + sst_dsp_dma_put_channel(dsp); > + > + mutex_lock(&dsp->mutex); > + list_add(&sst_fw->list, &dsp->fw_list); > + mutex_unlock(&dsp->mutex); > + > + return sst_fw; > + > +parse_err: > + if (dsp->fw_use_dma) > + sst_dsp_dma_put_channel(dsp); > +chan_err: > + dma_free_coherent(dsp->dev, sst_fw->size, > + sst_fw->dma_buf, > + sst_fw->dmable_fw_paddr); > + kfree(sst_fw); > + return NULL; > +} > +EXPORT_SYMBOL_GPL(sst_fw_new); > + > +/* free single firmware object */ > +void sst_fw_free(struct sst_fw *sst_fw) > +{ > + struct sst_dsp *dsp = sst_fw->dsp; > + > + mutex_lock(&dsp->mutex); > + list_del(&sst_fw->list); > + mutex_unlock(&dsp->mutex); > + > + dma_free_coherent(dsp->dev, sst_fw->size, sst_fw->dma_buf, > + sst_fw->dmable_fw_paddr); > + kfree(sst_fw); > +} > +EXPORT_SYMBOL_GPL(sst_fw_free); > + > +/* free all firmware objects */ > +void sst_fw_free_all(struct sst_dsp *dsp) > +{ > + struct sst_fw *sst_fw, *t; > + > + mutex_lock(&dsp->mutex); > + list_for_each_entry_safe(sst_fw, t, &dsp->fw_list, list) { > + > + list_del(&sst_fw->list); > + dma_free_coherent(dsp->dev, sst_fw->size, sst_fw->dma_buf, > + sst_fw->dmable_fw_paddr); > + kfree(sst_fw); > + } > + mutex_unlock(&dsp->mutex); > +} > +EXPORT_SYMBOL_GPL(sst_fw_free_all); > + > +/* create a new SST generic module from FW template */ > +struct sst_module *sst_module_new(struct sst_fw *sst_fw, > + struct sst_module_template *template, void *private) > +{ > + struct sst_dsp *dsp = sst_fw->dsp; > + struct sst_module *sst_module; > + > + sst_module = kzalloc(sizeof(*sst_module), GFP_KERNEL); > + if (sst_module == NULL) > + return NULL; > + > + sst_module->id = template->id; > + sst_module->dsp = dsp; > + sst_module->sst_fw = sst_fw; > + > + memcpy(&sst_module->s, &template->s, sizeof(struct sst_module_data)); > + memcpy(&sst_module->p, &template->p, sizeof(struct sst_module_data)); > + > + INIT_LIST_HEAD(&sst_module->block_list); > + > + mutex_lock(&dsp->mutex); > + list_add(&sst_module->list, &dsp->module_list); > + mutex_unlock(&dsp->mutex); > + > + return sst_module; > +} > +EXPORT_SYMBOL_GPL(sst_module_new); > + > +/* free firmware module and remove from available list */ > +void sst_module_free(struct sst_module *sst_module) > +{ > + struct sst_dsp *dsp = sst_module->dsp; > + > + mutex_lock(&dsp->mutex); > + list_del(&sst_module->list); > + mutex_unlock(&dsp->mutex); > + > + kfree(sst_module); > +} > +EXPORT_SYMBOL_GPL(sst_module_free); > + > +/* allocate contiguous free DSP blocks - callers hold locks */ > +static int block_alloc_contiguous(struct sst_module *module, > + struct sst_module_data *data, u32 next_offset, int size) > +{ > + struct sst_dsp *dsp = module->dsp; > + struct sst_mem_block *block, *tmp; > + int ret; > + > + /* find first free blocks that can hold module */ > + list_for_each_entry_safe(block, tmp, &dsp->free_block_list, list) { > + > + /* ignore blocks that dont match type */ > + if (block->type != data->type) > + continue; > + > + /* is block next after parent ? */ > + if (next_offset == block->offset) { > + > + /* do we need more blocks */ > + if (size > block->size) { > + ret = block_alloc_contiguous(module, > + data, block->offset + block->size, > + size - block->size); > + if (ret < 0) > + return ret; How many contiguous blocks can be? The recursive call for each one block doesn't look scaling. Takashi > + } > + > + /* add block to module */ > + block->data_type = data->data_type; > + block->bytes_used = block->size; > + list_move(&block->list, &dsp->used_block_list); > + list_add(&block->module_list, &module->block_list); > + dev_dbg(dsp->dev, " module %d added block %d:%d\n", > + module->id, block->type, block->index); > + return 0; > + } > + } > + > + /* free any allocated blocks on failure */ > + list_for_each_entry_safe(block, tmp, &module->block_list, module_list) { > + list_del(&block->module_list); > + list_move(&block->list, &dsp->free_block_list); > + } > + return -ENOMEM; > +} > + > +/* allocate free DSP blocks for module data - callers hold locks */ > +static int block_alloc(struct sst_module *module, > + struct sst_module_data *data) > +{ > + struct sst_dsp *dsp = module->dsp; > + struct sst_mem_block *block, *tmp; > + int ret = 0; > + > + if (data->size == 0) > + return 0; > + > + /* find first free whole blocks that can hold module */ > + list_for_each_entry_safe(block, tmp, &dsp->free_block_list, list) { > + > + /* ignore blocks with wrong type */ > + if (block->type != data->type) > + continue; > + > + if (data->size > block->size) > + continue; > + > + data->offset = block->offset; > + block->data_type = data->data_type; > + block->bytes_used = data->size % block->size; > + list_add(&block->module_list, &module->block_list); > + list_move(&block->list, &dsp->used_block_list); > + dev_dbg(dsp->dev, " *module %d added block %d:%d\n", > + module->id, block->type, block->index); > + return 0; > + } > + > + /* then find free multiple blocks that can hold module */ > + list_for_each_entry_safe(block, tmp, &dsp->free_block_list, list) { > + > + /* ignore blocks with wrong type */ > + if (block->type != data->type) > + continue; > + > + /* do we span > 1 blocks */ > + if (data->size > block->size) { > + ret = block_alloc_contiguous(module, data, > + block->offset + block->size, > + data->size - block->size); > + if (ret == 0) > + return ret; > + } > + } > + > + /* not enough free block space */ > + return -ENOMEM; > +} > + > +/* remove module from memory - callers hold locks */ > +static void block_module_remove(struct sst_module *module) > +{ > + struct sst_mem_block *block, *tmp; > + struct sst_dsp *dsp = module->dsp; > + int err; > + > + /* disable each block */ > + list_for_each_entry(block, &module->block_list, module_list) { > + > + if (block->ops && block->ops->disable) { > + err = block->ops->disable(block); > + if (err < 0) > + dev_err(dsp->dev, > + "error: cant disable block %d:%d\n", > + block->type, block->index); > + } > + } > + > + /* mark each block as free */ > + list_for_each_entry_safe(block, tmp, &module->block_list, module_list) { > + list_del(&block->module_list); > + list_move(&block->list, &dsp->free_block_list); > + } > +} > + > +/* prepare the memory block to receive data from host - callers hold locks */ > +static int block_module_prepare(struct sst_module *module) > +{ > + struct sst_mem_block *block; > + int ret = 0; > + > + /* enable each block so that's it'e ready for module P/S data */ > + list_for_each_entry(block, &module->block_list, module_list) { > + > + if (block->ops && block->ops->enable) > + ret = block->ops->enable(block); > + if (ret < 0) { > + dev_err(module->dsp->dev, > + "error: cant disable block %d:%d\n", > + block->type, block->index); > + goto err; > + } > + } > + return ret; > + > +err: > + list_for_each_entry(block, &module->block_list, module_list) { > + if (block->ops && block->ops->disable) > + block->ops->disable(block); > + } > + return ret; > +} > + > +/* allocate memory blocks for static module addresses - callers hold locks */ > +static int block_alloc_fixed(struct sst_module *module, > + struct sst_module_data *data) > +{ > + struct sst_dsp *dsp = module->dsp; > + struct sst_mem_block *block, *tmp; > + u32 end = data->offset + data->size, block_end; > + int err; > + > + /* only IRAM/DRAM blocks are managed */ > + if (data->type != SST_MEM_IRAM && data->type != SST_MEM_DRAM) > + return 0; > + > + /* are blocks already attached to this module */ > + list_for_each_entry_safe(block, tmp, &module->block_list, module_list) { > + > + /* force compacting mem blocks of the same data_type */ > + if (block->data_type != data->data_type) > + continue; > + > + block_end = block->offset + block->size; > + > + /* find block that holds section */ > + if (data->offset >= block->offset && end < block_end) > + return 0; > + > + /* does block span more than 1 section */ > + if (data->offset >= block->offset && data->offset < block_end) { > + > + err = block_alloc_contiguous(module, data, > + block->offset + block->size, > + data->size - block->size + data->offset - block->offset); > + if (err < 0) > + return -ENOMEM; > + > + /* module already owns blocks */ > + return 0; > + } > + } > + > + /* find first free blocks that can hold section in free list */ > + list_for_each_entry_safe(block, tmp, &dsp->free_block_list, list) { > + block_end = block->offset + block->size; > + > + /* find block that holds section */ > + if (data->offset >= block->offset && end < block_end) { > + > + /* add block */ > + block->data_type = data->data_type; > + list_move(&block->list, &dsp->used_block_list); > + list_add(&block->module_list, &module->block_list); > + return 0; > + } > + > + /* does block span more than 1 section */ > + if (data->offset >= block->offset && data->offset < block_end) { > + > + err = block_alloc_contiguous(module, data, > + block->offset + block->size, > + data->size - block->size); > + if (err < 0) > + return -ENOMEM; > + > + /* add block */ > + block->data_type = data->data_type; > + list_move(&block->list, &dsp->used_block_list); > + list_add(&block->module_list, &module->block_list); > + return 0; > + } > + > + } > + > + return -ENOMEM; > +} > + > +/* Load fixed module data into DSP memory blocks */ > +int sst_module_insert_fixed_block(struct sst_module *module, > + struct sst_module_data *data) > +{ > + struct sst_dsp *dsp = module->dsp; > + struct sst_fw *sst_fw = module->sst_fw; > + int ret; > + > + mutex_lock(&dsp->mutex); > + > + /* alloc blocks that includes this section */ > + ret = block_alloc_fixed(module, data); > + if (ret < 0) { > + dev_err(dsp->dev, > + "error: no free blocks for section at offset 0x%x size 0x%x\n", > + data->offset, data->size); > + mutex_unlock(&dsp->mutex); > + return -ENOMEM; > + } > + > + /* prepare DSP blocks for module copy */ > + ret = block_module_prepare(module); > + if (ret < 0) { > + dev_err(dsp->dev, "error: fw module prepare failed\n"); > + goto err; > + } > + > + /* copy partial module data to blocks */ > + if (dsp->fw_use_dma) { > + ret = sst_dsp_dma_copy(dsp, > + sst_fw->dmable_fw_paddr + data->data_offset, > + dsp->addr.lpe_base + data->offset, data->size); > + if (ret < 0) { > + dev_err(dsp->dev, "error: module copy failed\n"); > + goto err; > + } > + } else > + sst_memcpy32(dsp->addr.lpe + data->offset, data->data, > + data->size); > + > + mutex_unlock(&dsp->mutex); > + return ret; > + > +err: > + block_module_remove(module); > + mutex_unlock(&dsp->mutex); > + return ret; > +} > +EXPORT_SYMBOL_GPL(sst_module_insert_fixed_block); > + > +/* Unload entire module from DSP memory */ > +int sst_block_module_remove(struct sst_module *module) > +{ > + struct sst_dsp *dsp = module->dsp; > + > + mutex_lock(&dsp->mutex); > + block_module_remove(module); > + mutex_unlock(&dsp->mutex); > + return 0; > +} > +EXPORT_SYMBOL_GPL(sst_block_module_remove); > + > +/* register a DSP memory block for use with FW based modules */ > +struct sst_mem_block *sst_mem_block_register(struct sst_dsp *dsp, u32 offset, > + u32 size, enum sst_mem_type type, struct sst_block_ops *ops, u32 index, > + void *private) > +{ > + struct sst_mem_block *block; > + > + block = kzalloc(sizeof(*block), GFP_KERNEL); > + if (block == NULL) > + return NULL; > + > + block->offset = offset; > + block->size = size; > + block->index = index; > + block->type = type; > + block->dsp = dsp; > + block->private = private; > + block->ops = ops; > + > + mutex_lock(&dsp->mutex); > + list_add(&block->list, &dsp->free_block_list); > + mutex_unlock(&dsp->mutex); > + > + return block; > +} > +EXPORT_SYMBOL_GPL(sst_mem_block_register); > + > +/* unregister all DSP memory blocks */ > +void sst_mem_block_unregister_all(struct sst_dsp *dsp) > +{ > + struct sst_mem_block *block, *tmp; > + > + mutex_lock(&dsp->mutex); > + > + /* unregister used blocks */ > + list_for_each_entry_safe(block, tmp, &dsp->used_block_list, list) { > + list_del(&block->list); > + kfree(block); > + } > + > + /* unregister free blocks */ > + list_for_each_entry_safe(block, tmp, &dsp->free_block_list, list) { > + list_del(&block->list); > + kfree(block); > + } > + > + mutex_unlock(&dsp->mutex); > +} > +EXPORT_SYMBOL_GPL(sst_mem_block_unregister_all); > + > +/* allocate scratch buffer blocks */ > +struct sst_module *sst_mem_block_alloc_scratch(struct sst_dsp *dsp) > +{ > + struct sst_module *sst_module, *scratch; > + struct sst_mem_block *block, *tmp; > + u32 block_size; > + int ret = 0; > + > + scratch = kzalloc(sizeof(struct sst_module), GFP_KERNEL); > + if (scratch == NULL) > + return NULL; > + > + mutex_lock(&dsp->mutex); > + > + /* calculate required scratch size */ > + list_for_each_entry(sst_module, &dsp->module_list, list) { > + if (scratch->s.size > sst_module->s.size) > + scratch->s.size = scratch->s.size; > + else > + scratch->s.size = sst_module->s.size; > + } > + > + dev_dbg(dsp->dev, "scratch buffer required is %d bytes\n", > + scratch->s.size); > + > + /* init scratch module */ > + scratch->dsp = dsp; > + scratch->s.type = SST_MEM_DRAM; > + scratch->s.data_type = SST_DATA_S; > + INIT_LIST_HEAD(&scratch->block_list); > + > + /* check free blocks before looking at used blocks for space */ > + if (!list_empty(&dsp->free_block_list)) > + block = list_first_entry(&dsp->free_block_list, > + struct sst_mem_block, list); > + else > + block = list_first_entry(&dsp->used_block_list, > + struct sst_mem_block, list); > + block_size = block->size; > + > + /* allocate blocks for module scratch buffers */ > + dev_dbg(dsp->dev, "allocating scratch blocks\n"); > + ret = block_alloc(scratch, &scratch->s); > + if (ret < 0) { > + dev_err(dsp->dev, "error: can't alloc scratch blocks\n"); > + goto err; > + } > + > + /* assign the same offset of scratch to each module */ > + list_for_each_entry(sst_module, &dsp->module_list, list) > + sst_module->s.offset = scratch->s.offset; > + > + mutex_unlock(&dsp->mutex); > + return scratch; > + > +err: > + list_for_each_entry_safe(block, tmp, &scratch->block_list, module_list) > + list_del(&block->module_list); > + mutex_unlock(&dsp->mutex); > + return NULL; > +} > +EXPORT_SYMBOL_GPL(sst_mem_block_alloc_scratch); > + > +/* free all scratch blocks */ > +void sst_mem_block_free_scratch(struct sst_dsp *dsp, > + struct sst_module *scratch) > +{ > + struct sst_mem_block *block, *tmp; > + > + mutex_lock(&dsp->mutex); > + > + list_for_each_entry_safe(block, tmp, &scratch->block_list, module_list) > + list_del(&block->module_list); > + > + mutex_unlock(&dsp->mutex); > +} > +EXPORT_SYMBOL_GPL(sst_mem_block_free_scratch); > + > +/* get a module from it's unique ID */ > +struct sst_module *sst_module_get_from_id(struct sst_dsp *dsp, u32 id) > +{ > + struct sst_module *module; > + > + mutex_lock(&dsp->mutex); > + > + list_for_each_entry(module, &dsp->module_list, list) { > + if (module->id == id) { > + mutex_unlock(&dsp->mutex); > + return module; > + } > + } > + > + mutex_unlock(&dsp->mutex); > + return NULL; > +} > +EXPORT_SYMBOL_GPL(sst_module_get_from_id); > -- > 1.8.3.2 > > _______________________________________________ > Alsa-devel mailing list > Alsa-devel@alsa-project.org > http://mailman.alsa-project.org/mailman/listinfo/alsa-devel >
On Fri, 2014-02-14 at 09:55 +0100, Takashi Iwai wrote: > At Thu, 13 Feb 2014 19:15:28 +0000, > Liam Girdwood wrote: > > > > + > > +/* create new generic firmware object */ > > +struct sst_fw *sst_fw_new(struct sst_dsp *dsp, > > + const struct firmware *fw, void *private) > > +{ > > + struct sst_fw *sst_fw; > > + int err; > > + > > + if (!dsp->ops->parse_fw) > > + return NULL; > > + > > + sst_fw = kzalloc(sizeof(*sst_fw), GFP_KERNEL); > > + if (sst_fw == NULL) > > + return NULL; > > + > > + sst_fw->dsp = dsp; > > + sst_fw->private = private; > > + sst_fw->size = fw->size; > > + > > + err = dma_coerce_mask_and_coherent(dsp->dev, DMA_BIT_MASK(32)); > > + if (err < 0) { > > + kfree(sst_fw); > > + return NULL; > > + } > > + > > + /* allocate DMA buffer to store FW data */ > > + sst_fw->dma_buf = dma_alloc_coherent(dsp->dev, sst_fw->size, > > + &sst_fw->dmable_fw_paddr, GFP_DMA); > > + if (!sst_fw->dma_buf) { > > + dev_err(dsp->dev, "error: DMA alloc failed\n"); > > + kfree(sst_fw); > > + return NULL; > > + } > > + > > + /* copy FW data to DMA-able memory */ > > + memcpy((void *)sst_fw->dma_buf, (void *)fw->data, fw->size); > > + release_firmware(fw); > > Should fw object be really released here? > For example, if this function returns NULL after this point, the > caller doesn't know whether fw was released or not. It'd be more > consistent to release it in the caller side, or always release no > matter whether the function succeeds or not. Yep, your right. I'll fix this in V2. Fwiw, we only need to take a copy of the FW and store that copy in DMA-able memory (32bit addressable). So the DMA memory will be used for context restore (from resume), DMA firmware loading etc. > > > + > > +/* allocate contiguous free DSP blocks - callers hold locks */ > > +static int block_alloc_contiguous(struct sst_module *module, > > + struct sst_module_data *data, u32 next_offset, int size) > > +{ > > + struct sst_dsp *dsp = module->dsp; > > + struct sst_mem_block *block, *tmp; > > + int ret; > > + > > + /* find first free blocks that can hold module */ > > + list_for_each_entry_safe(block, tmp, &dsp->free_block_list, list) { > > + > > + /* ignore blocks that dont match type */ > > + if (block->type != data->type) > > + continue; > > + > > + /* is block next after parent ? */ > > + if (next_offset == block->offset) { > > + > > + /* do we need more blocks */ > > + if (size > block->size) { > > + ret = block_alloc_contiguous(module, > > + data, block->offset + block->size, > > + size - block->size); > > + if (ret < 0) > > + return ret; > > How many contiguous blocks can be? In theory, the whole DSP memory can be allocated as a contiguous block, but in practice it's only really a few blocks contiguous per module atm. > The recursive call for each one block doesn't look scaling. > It seems to work atm, but I'll update this too for V2. Liam
On Fri, 2014-02-14 at 09:38 +0000, Liam Girdwood wrote: > On Fri, 2014-02-14 at 09:55 +0100, Takashi Iwai wrote: > > At Thu, 13 Feb 2014 19:15:28 +0000, > > Liam Girdwood wrote: > > > > + > > > +/* allocate contiguous free DSP blocks - callers hold locks */ > > > +static int block_alloc_contiguous(struct sst_module *module, > > > + struct sst_module_data *data, u32 next_offset, int size) > > > +{ > > > + struct sst_dsp *dsp = module->dsp; > > > + struct sst_mem_block *block, *tmp; > > > + int ret; > > > + > > > + /* find first free blocks that can hold module */ > > > + list_for_each_entry_safe(block, tmp, &dsp->free_block_list, list) { > > > + > > > + /* ignore blocks that dont match type */ > > > + if (block->type != data->type) > > > + continue; > > > + > > > + /* is block next after parent ? */ > > > + if (next_offset == block->offset) { > > > + > > > + /* do we need more blocks */ > > > + if (size > block->size) { > > > + ret = block_alloc_contiguous(module, > > > + data, block->offset + block->size, > > > + size - block->size); > > > + if (ret < 0) > > > + return ret; > > > > How many contiguous blocks can be? > > In theory, the whole DSP memory can be allocated as a contiguous block, > but in practice it's only really a few blocks contiguous per module atm. > > > The recursive call for each one block doesn't look scaling. > > I've double checked the logic here and added more comments and debug. I may be missing something though, but it does work and is designed to deal with unordered blocks (on address) in the list. The recursion basically checks the whole list each time for the next block in address order. It then checks if the required size > current block size. We decrement size on each by block size and increment address by block size on every call. If the required size < current block size we allocate the block and come back up the stack allocating the other blocks in the sequence. From the debug we get :- > [ 6.733916] haswell-pcm-audio haswell-pcm-audio: module 0 added block 0:4 at offset 0xc0000 > [ 6.733925] haswell-pcm-audio haswell-pcm-audio: module 0 added block 0:3 at offset 0xb8000 > [ 6.733930] haswell-pcm-audio haswell-pcm-audio: module 0 added block 0:2 at offset 0xb0000 > [ 6.733936] haswell-pcm-audio haswell-pcm-audio: module 0 added block 0:1 at offset 0xa8000 > [ 6.810179] haswell-pcm-audio haswell-pcm-audio: module 0 added block 1:1 at offset 0x88000 > [ 6.810189] haswell-pcm-audio haswell-pcm-audio: module 0 added block 1:0 at offset 0x80000 > [ 6.857452] haswell-pcm-audio haswell-pcm-audio: scratch buffer required is 56320 bytes > [ 6.857461] haswell-pcm-audio haswell-pcm-audio: allocating scratch blocks > [ 6.857469] haswell-pcm-audio haswell-pcm-audio: module 0 added block 1:6 at offset 0x70000 > [ 6.860268] haswell-pcm-audio haswell-pcm-audio: FW loaded: type 172 - version: 0.0 build 1 Thanks Liam --------------------------------------------------------------------- Intel Corporation (UK) Limited Registered No. 1134945 (England) Registered Office: Pipers Way, Swindon SN3 1RJ VAT No: 860 2173 47 This e-mail and any attachments may contain confidential material for the sole use of the intended recipient(s). Any review or distribution by others is strictly prohibited. If you are not the intended recipient, please contact the sender and delete all copies.
At Fri, 14 Feb 2014 16:11:15 +0000, Liam Girdwood wrote: > > On Fri, 2014-02-14 at 09:38 +0000, Liam Girdwood wrote: > > On Fri, 2014-02-14 at 09:55 +0100, Takashi Iwai wrote: > > > At Thu, 13 Feb 2014 19:15:28 +0000, > > > Liam Girdwood wrote: > > > > > > + > > > > +/* allocate contiguous free DSP blocks - callers hold locks */ > > > > +static int block_alloc_contiguous(struct sst_module *module, > > > > + struct sst_module_data *data, u32 next_offset, int size) > > > > +{ > > > > + struct sst_dsp *dsp = module->dsp; > > > > + struct sst_mem_block *block, *tmp; > > > > + int ret; > > > > + > > > > + /* find first free blocks that can hold module */ > > > > + list_for_each_entry_safe(block, tmp, &dsp->free_block_list, list) { > > > > + > > > > + /* ignore blocks that dont match type */ > > > > + if (block->type != data->type) > > > > + continue; > > > > + > > > > + /* is block next after parent ? */ > > > > + if (next_offset == block->offset) { > > > > + > > > > + /* do we need more blocks */ > > > > + if (size > block->size) { > > > > + ret = block_alloc_contiguous(module, > > > > + data, block->offset + block->size, > > > > + size - block->size); > > > > + if (ret < 0) > > > > + return ret; > > > > > > How many contiguous blocks can be? > > > > In theory, the whole DSP memory can be allocated as a contiguous block, > > but in practice it's only really a few blocks contiguous per module atm. > > > > > The recursive call for each one block doesn't look scaling. > > > > > I've double checked the logic here and added more comments and debug. I > may be missing something though, but it does work and is designed to > deal with unordered blocks (on address) in the list. > > The recursion basically checks the whole list each time for the next > block in address order. It then checks if the required size > current > block size. We decrement size on each by block size and increment > address by block size on every call. If the required size < current > block size we allocate the block and come back up the stack allocating > the other blocks in the sequence. > > From the debug we get :- > > > [ 6.733916] haswell-pcm-audio haswell-pcm-audio: module 0 added block 0:4 at offset 0xc0000 > > [ 6.733925] haswell-pcm-audio haswell-pcm-audio: module 0 added block 0:3 at offset 0xb8000 > > [ 6.733930] haswell-pcm-audio haswell-pcm-audio: module 0 added block 0:2 at offset 0xb0000 > > [ 6.733936] haswell-pcm-audio haswell-pcm-audio: module 0 added block 0:1 at offset 0xa8000 > > [ 6.810179] haswell-pcm-audio haswell-pcm-audio: module 0 added block 1:1 at offset 0x88000 > > [ 6.810189] haswell-pcm-audio haswell-pcm-audio: module 0 added block 1:0 at offset 0x80000 > > [ 6.857452] haswell-pcm-audio haswell-pcm-audio: scratch buffer required is 56320 bytes > > [ 6.857461] haswell-pcm-audio haswell-pcm-audio: allocating scratch blocks > > [ 6.857469] haswell-pcm-audio haswell-pcm-audio: module 0 added block 1:6 at offset 0x70000 > > [ 6.860268] haswell-pcm-audio haswell-pcm-audio: FW loaded: type 172 - version: 0.0 build 1 Well, the code is a bit confusing because of the comment: static int block_alloc_contiguous(struct sst_module *module, struct sst_module_data *data, u32 next_offset, int size) { ..... /* find first free blocks that can hold module */ list_for_each_entry_safe(block, tmp, &dsp->free_block_list, list) { Actually, this loop doesn't look for the first free blocks but rather looks for the block that matches with the given type and the offset. /* ignore blocks that dont match type */ if (block->type != data->type) continue; /* is block next after parent ? */ if (next_offset == block->offset) { Then, it tries allocating the next block if the size of the found block isn't big enough: /* do we need more blocks */ if (size > block->size) { ret = block_alloc_contiguous(module, data, block->offset + block->size, size - block->size); if (ret < 0) return ret; So, here is the recursion, and it means that at maximum (size / block_size) level recursions may happen. My point is that this doesn't have to be recursion. It can be implemented via a flat loop as simply as well. For example, a code like below would work, too. struct block *find_block(int type, int offset) { struct block *block; list_for_each_entry(block, &free_blocks, list) { if (block->type == type && block->offset == offset) return block; } return NULL; } int alloc_contiguous(int type, int offset, int size) { struct list_head tmp = LIST_HEAD_INIT(tmp); struct block *block; while (size > 0) { block = find_block(type, offset); if (!block) { list_splice(&tmp, &free_list); return -ENOMEM; } list_move_tail(&block->list, &tmp); offset += block->size; size -= block->size; } list_for_each_entry(block, &tmp, list) { block->data_type = xxx; .... } list_splice(&tmp, &used_list); return 0; } BTW, this is still suboptimal; we can remember the last failed point and rescan from there at next. Or, if you manage the sorted list (tree) instead of a plain list, things will be also easier, I suppose. Takashi
On Fri, 2014-02-14 at 18:06 +0100, Takashi Iwai wrote: > At Fri, 14 Feb 2014 16:11:15 +0000, > Liam Girdwood wrote: > > > > On Fri, 2014-02-14 at 09:38 +0000, Liam Girdwood wrote: > > > On Fri, 2014-02-14 at 09:55 +0100, Takashi Iwai wrote: > > > > At Thu, 13 Feb 2014 19:15:28 +0000, > > > > Liam Girdwood wrote: > > > > > > > > + > > > > > +/* allocate contiguous free DSP blocks - callers hold locks */ > > > > > +static int block_alloc_contiguous(struct sst_module *module, > > > > > + struct sst_module_data *data, u32 next_offset, int size) > > > > > +{ > > > > > + struct sst_dsp *dsp = module->dsp; > > > > > + struct sst_mem_block *block, *tmp; > > > > > + int ret; > > > > > + > > > > > + /* find first free blocks that can hold module */ > > > > > + list_for_each_entry_safe(block, tmp, &dsp->free_block_list, list) { > > > > > + > > > > > + /* ignore blocks that dont match type */ > > > > > + if (block->type != data->type) > > > > > + continue; > > > > > + > > > > > + /* is block next after parent ? */ > > > > > + if (next_offset == block->offset) { > > > > > + > > > > > + /* do we need more blocks */ > > > > > + if (size > block->size) { > > > > > + ret = block_alloc_contiguous(module, > > > > > + data, block->offset + block->size, > > > > > + size - block->size); > > > > > + if (ret < 0) > > > > > + return ret; > > > > > > > > How many contiguous blocks can be? > > > > > > In theory, the whole DSP memory can be allocated as a contiguous block, > > > but in practice it's only really a few blocks contiguous per module atm. > > > > > > > The recursive call for each one block doesn't look scaling. > > > > > > > > I've double checked the logic here and added more comments and debug. I > > may be missing something though, but it does work and is designed to > > deal with unordered blocks (on address) in the list. > > > > The recursion basically checks the whole list each time for the next > > block in address order. It then checks if the required size > current > > block size. We decrement size on each by block size and increment > > address by block size on every call. If the required size < current > > block size we allocate the block and come back up the stack allocating > > the other blocks in the sequence. > > > > From the debug we get :- > > > > > [ 6.733916] haswell-pcm-audio haswell-pcm-audio: module 0 added block 0:4 at offset 0xc0000 > > > [ 6.733925] haswell-pcm-audio haswell-pcm-audio: module 0 added block 0:3 at offset 0xb8000 > > > [ 6.733930] haswell-pcm-audio haswell-pcm-audio: module 0 added block 0:2 at offset 0xb0000 > > > [ 6.733936] haswell-pcm-audio haswell-pcm-audio: module 0 added block 0:1 at offset 0xa8000 > > > [ 6.810179] haswell-pcm-audio haswell-pcm-audio: module 0 added block 1:1 at offset 0x88000 > > > [ 6.810189] haswell-pcm-audio haswell-pcm-audio: module 0 added block 1:0 at offset 0x80000 > > > [ 6.857452] haswell-pcm-audio haswell-pcm-audio: scratch buffer required is 56320 bytes > > > [ 6.857461] haswell-pcm-audio haswell-pcm-audio: allocating scratch blocks > > > [ 6.857469] haswell-pcm-audio haswell-pcm-audio: module 0 added block 1:6 at offset 0x70000 > > > [ 6.860268] haswell-pcm-audio haswell-pcm-audio: FW loaded: type 172 - version: 0.0 build 1 > > Well, the code is a bit confusing because of the comment: > I agree, and I had updated the comments for V2. > static int block_alloc_contiguous(struct sst_module *module, > struct sst_module_data *data, u32 next_offset, int size) > { > ..... > /* find first free blocks that can hold module */ > list_for_each_entry_safe(block, tmp, &dsp->free_block_list, list) { > > Actually, this loop doesn't look for the first free blocks but rather > looks for the block that matches with the given type and the offset. > > /* ignore blocks that dont match type */ > if (block->type != data->type) > continue; > > /* is block next after parent ? */ > if (next_offset == block->offset) { > > Then, it tries allocating the next block if the size of the found > block isn't big enough: > > /* do we need more blocks */ > if (size > block->size) { > ret = block_alloc_contiguous(module, > data, block->offset + block->size, > size - block->size); > if (ret < 0) > return ret; > > So, here is the recursion, and it means that at maximum > (size / block_size) level recursions may happen. > > My point is that this doesn't have to be recursion. It can be > implemented via a flat loop as simply as well. For example, a code > like below would work, too. > > struct block *find_block(int type, int offset) > { > struct block *block; > list_for_each_entry(block, &free_blocks, list) { > if (block->type == type && block->offset == offset) > return block; > } > return NULL; > } > > int alloc_contiguous(int type, int offset, int size) > { > struct list_head tmp = LIST_HEAD_INIT(tmp); > struct block *block; > > while (size > 0) { > block = find_block(type, offset); > if (!block) { > list_splice(&tmp, &free_list); > return -ENOMEM; > } > list_move_tail(&block->list, &tmp); > offset += block->size; > size -= block->size; > } > > list_for_each_entry(block, &tmp, list) { > block->data_type = xxx; > .... > } > list_splice(&tmp, &used_list); > return 0; > } > > BTW, this is still suboptimal; we can remember the last failed point > and rescan from there at next. > > Or, if you manage the sorted list (tree) instead of a plain list, > things will be also easier, I suppose. Agree it would be faster if ordered, but this code will be infrequently used so the emphasis was more on ease of development/testing. I'll remove the recursion for V2. Thanks Liam
diff --git a/sound/soc/intel/sst-firmware.c b/sound/soc/intel/sst-firmware.c new file mode 100644 index 0000000..8126b46 --- /dev/null +++ b/sound/soc/intel/sst-firmware.c @@ -0,0 +1,805 @@ +/* + * Intel SST Firmware Loader + * + * Copyright (C) 2013, Intel Corporation. All rights reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License version + * 2 as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + */ + +#include <linux/kernel.h> +#include <linux/slab.h> +#include <linux/sched.h> +#include <linux/firmware.h> +#include <linux/export.h> +#include <linux/platform_device.h> +#include <linux/dma-mapping.h> +#include <linux/dmaengine.h> +#include <linux/pci.h> + +/* supported DMA engine drivers */ +#include <linux/dw_dmac.h> + +#include <asm/page.h> +#include <asm/pgtable.h> + +#include "sst-dsp.h" +#include "sst-dsp-priv.h" + +#define SST_DMA_RESOURCES 2 +#define SST_DSP_DMA_MAX_BURST 0x3 + +struct sst_dma { + struct sst_dsp *sst; + + struct platform_device *dma_dev; + struct resource dma_resource[SST_DMA_RESOURCES]; + struct dma_async_tx_descriptor *desc; + struct dma_chan *ch; +}; + +static void sst_memcpy32(void *dest, void *src, int bytes) +{ + int i; + + /* copy one 32 bit word at a time as 64 bit access is not supported */ + for (i = 0; i < bytes; i += 4) + memcpy_toio(dest + i, src + i, 4); +} + +static void sst_dma_transfer_complete(void *arg) +{ + struct sst_dsp *sst = (struct sst_dsp *)arg; + + dev_dbg(sst->dev, "DMA: callback\n"); +} + +int sst_dsp_dma_copy(struct sst_dsp *sst, dma_addr_t src_addr, + dma_addr_t dest_addr, size_t size) +{ + struct dma_async_tx_descriptor *desc; + struct sst_dma *dma = sst->dma; + + if (dma->ch == NULL) { + dev_err(sst->dev, "error: no DMA channel\n"); + return -ENODEV; + } + + dev_dbg(sst->dev, "DMA: src: 0x%lx dest 0x%lx size %zu\n", + (unsigned long)src_addr, (unsigned long)dest_addr, size); + + desc = dma->ch->device->device_prep_dma_memcpy(dma->ch, dest_addr, + src_addr, size, DMA_CTRL_ACK); + if (!desc){ + dev_err(sst->dev, "error: dma prep memcpy failed\n"); + return -EINVAL; + } + + desc->callback = sst_dma_transfer_complete; + desc->callback_param = sst; + + desc->tx_submit(desc); + dma_wait_for_async_tx(desc); + + return 0; +} +EXPORT_SYMBOL_GPL(sst_dsp_dma_copy); + +static bool dma_chan_filter(struct dma_chan *chan, void *param) +{ + struct sst_dsp *dsp = (struct sst_dsp *)param; + struct sst_dma *dma = dsp->dma; + + /* only accept channels from this device */ + if (chan->device->dev != &dma->dma_dev->dev) + return false; + + /* todo: add chan_id testing */ + return true; +} + +int sst_dsp_dma_get_channel(struct sst_dsp *dsp, int chan_id) +{ + struct sst_dma *dma = dsp->dma; + struct dma_slave_config slave; + dma_cap_mask_t mask; + int ret; + + /* The Intel MID DMA engine driver needs the slave config set but + * Synopsis DMA engine driver safely ignores the slave config */ + dma_cap_zero(mask); + dma_cap_set(DMA_SLAVE, mask); + dma_cap_set(DMA_MEMCPY, mask); + + dma->ch = dma_request_channel(mask, dma_chan_filter, dsp); + if (dma->ch == NULL) { + dev_err(dsp->dev, "error: DMA request channel failed\n"); + return -EIO; + } + + memset(&slave, 0, sizeof(slave)); + slave.direction = DMA_MEM_TO_DEV; + slave.src_addr_width = + slave.dst_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES; + slave.src_maxburst = slave.dst_maxburst = SST_DSP_DMA_MAX_BURST; + + ret = dmaengine_slave_config(dma->ch, &slave); + if (ret) { + dev_err(dsp->dev, "error: unable to set DMA slave config %d\n", + ret); + dma_release_channel(dma->ch); + dma->ch = NULL; + } + + return ret; +} +EXPORT_SYMBOL_GPL(sst_dsp_dma_get_channel); + +void sst_dsp_dma_put_channel(struct sst_dsp *dsp) +{ + struct sst_dma *dma = dsp->dma; + + dma_release_channel(dma->ch); + dma->ch = NULL; +} +EXPORT_SYMBOL_GPL(sst_dsp_dma_put_channel); + +/* platform data for DesignWare DMA Engine */ +static struct dw_dma_platform_data dw_pdata = { + .chan_allocation_order = CHAN_ALLOCATION_ASCENDING, + .chan_priority = CHAN_PRIORITY_ASCENDING, +}; + +int sst_dma_new(struct sst_dsp *sst) +{ + struct sst_pdata *sst_pdata = sst->pdata; + struct sst_dma *dma; + const char *dma_dev_name; + size_t dma_pdata_size; + void *dma_pdata; + + /* configure the correct platform data for whatever DMA engine + * is attached to the ADSP IP. */ + switch (sst->pdata->dma_engine) { + case SST_DMA_TYPE_DW: + dma_pdata = &dw_pdata; + dma_pdata_size = sizeof(dw_pdata); + dma_dev_name = "dw_dmac"; + break; + case SST_DMA_TYPE_MID: + dma_pdata = NULL; + dma_pdata_size = 0; + dma_dev_name = "Intel MID DMA"; + break; + default: + dev_err(sst->dev, "error: invalid DMA engine %d\n", + sst->pdata->dma_engine); + return -EINVAL; + } + + dma = devm_kzalloc(sst->dev, sizeof(struct sst_dma), GFP_KERNEL); + if (!dma) + return -ENOMEM; + + dma->sst = sst; + sst->dma = dma; + + dma->dma_resource[0].start = sst->addr.lpe_base + + sst_pdata->dma_base; + dma->dma_resource[0].end = sst->addr.lpe_base + + sst_pdata->dma_base + + sst_pdata->dma_size; + dma->dma_resource[0].flags = IORESOURCE_MEM; + dma->dma_resource[1].start = sst_pdata->irq; + dma->dma_resource[1].end = sst_pdata->irq; + dma->dma_resource[1].flags = IORESOURCE_IRQ; + + /* now register DMA engine device */ + dma->dma_dev = platform_device_register_resndata(sst->dev, + dma_dev_name, -1, dma->dma_resource, 2, + dma_pdata, dma_pdata_size); + + if (dma->dma_dev == NULL) { + dev_err(sst->dev, "error: DMA device register failed\n"); + return -ENODEV; + } + + sst->fw_use_dma = true; + return 0; +} +EXPORT_SYMBOL(sst_dma_new); + +void sst_dma_free(struct sst_dma *dma) +{ + if (dma->ch) + dma_release_channel(dma->ch); + platform_device_unregister(dma->dma_dev); +} +EXPORT_SYMBOL(sst_dma_free); + +/* create new generic firmware object */ +struct sst_fw *sst_fw_new(struct sst_dsp *dsp, + const struct firmware *fw, void *private) +{ + struct sst_fw *sst_fw; + int err; + + if (!dsp->ops->parse_fw) + return NULL; + + sst_fw = kzalloc(sizeof(*sst_fw), GFP_KERNEL); + if (sst_fw == NULL) + return NULL; + + sst_fw->dsp = dsp; + sst_fw->private = private; + sst_fw->size = fw->size; + + err = dma_coerce_mask_and_coherent(dsp->dev, DMA_BIT_MASK(32)); + if (err < 0) { + kfree(sst_fw); + return NULL; + } + + /* allocate DMA buffer to store FW data */ + sst_fw->dma_buf = dma_alloc_coherent(dsp->dev, sst_fw->size, + &sst_fw->dmable_fw_paddr, GFP_DMA); + if (!sst_fw->dma_buf) { + dev_err(dsp->dev, "error: DMA alloc failed\n"); + kfree(sst_fw); + return NULL; + } + + /* copy FW data to DMA-able memory */ + memcpy((void *)sst_fw->dma_buf, (void *)fw->data, fw->size); + release_firmware(fw); + + if (dsp->fw_use_dma) { + err = sst_dsp_dma_get_channel(dsp, 0); + if (err < 0) + goto chan_err; + } + + /* call core specific FW paser to load FW data into DSP */ + err = dsp->ops->parse_fw(sst_fw); + if (err < 0) { + dev_err(dsp->dev, "error: parse fw failed %d\n", err); + goto parse_err; + } + + if (dsp->fw_use_dma) + sst_dsp_dma_put_channel(dsp); + + mutex_lock(&dsp->mutex); + list_add(&sst_fw->list, &dsp->fw_list); + mutex_unlock(&dsp->mutex); + + return sst_fw; + +parse_err: + if (dsp->fw_use_dma) + sst_dsp_dma_put_channel(dsp); +chan_err: + dma_free_coherent(dsp->dev, sst_fw->size, + sst_fw->dma_buf, + sst_fw->dmable_fw_paddr); + kfree(sst_fw); + return NULL; +} +EXPORT_SYMBOL_GPL(sst_fw_new); + +/* free single firmware object */ +void sst_fw_free(struct sst_fw *sst_fw) +{ + struct sst_dsp *dsp = sst_fw->dsp; + + mutex_lock(&dsp->mutex); + list_del(&sst_fw->list); + mutex_unlock(&dsp->mutex); + + dma_free_coherent(dsp->dev, sst_fw->size, sst_fw->dma_buf, + sst_fw->dmable_fw_paddr); + kfree(sst_fw); +} +EXPORT_SYMBOL_GPL(sst_fw_free); + +/* free all firmware objects */ +void sst_fw_free_all(struct sst_dsp *dsp) +{ + struct sst_fw *sst_fw, *t; + + mutex_lock(&dsp->mutex); + list_for_each_entry_safe(sst_fw, t, &dsp->fw_list, list) { + + list_del(&sst_fw->list); + dma_free_coherent(dsp->dev, sst_fw->size, sst_fw->dma_buf, + sst_fw->dmable_fw_paddr); + kfree(sst_fw); + } + mutex_unlock(&dsp->mutex); +} +EXPORT_SYMBOL_GPL(sst_fw_free_all); + +/* create a new SST generic module from FW template */ +struct sst_module *sst_module_new(struct sst_fw *sst_fw, + struct sst_module_template *template, void *private) +{ + struct sst_dsp *dsp = sst_fw->dsp; + struct sst_module *sst_module; + + sst_module = kzalloc(sizeof(*sst_module), GFP_KERNEL); + if (sst_module == NULL) + return NULL; + + sst_module->id = template->id; + sst_module->dsp = dsp; + sst_module->sst_fw = sst_fw; + + memcpy(&sst_module->s, &template->s, sizeof(struct sst_module_data)); + memcpy(&sst_module->p, &template->p, sizeof(struct sst_module_data)); + + INIT_LIST_HEAD(&sst_module->block_list); + + mutex_lock(&dsp->mutex); + list_add(&sst_module->list, &dsp->module_list); + mutex_unlock(&dsp->mutex); + + return sst_module; +} +EXPORT_SYMBOL_GPL(sst_module_new); + +/* free firmware module and remove from available list */ +void sst_module_free(struct sst_module *sst_module) +{ + struct sst_dsp *dsp = sst_module->dsp; + + mutex_lock(&dsp->mutex); + list_del(&sst_module->list); + mutex_unlock(&dsp->mutex); + + kfree(sst_module); +} +EXPORT_SYMBOL_GPL(sst_module_free); + +/* allocate contiguous free DSP blocks - callers hold locks */ +static int block_alloc_contiguous(struct sst_module *module, + struct sst_module_data *data, u32 next_offset, int size) +{ + struct sst_dsp *dsp = module->dsp; + struct sst_mem_block *block, *tmp; + int ret; + + /* find first free blocks that can hold module */ + list_for_each_entry_safe(block, tmp, &dsp->free_block_list, list) { + + /* ignore blocks that dont match type */ + if (block->type != data->type) + continue; + + /* is block next after parent ? */ + if (next_offset == block->offset) { + + /* do we need more blocks */ + if (size > block->size) { + ret = block_alloc_contiguous(module, + data, block->offset + block->size, + size - block->size); + if (ret < 0) + return ret; + } + + /* add block to module */ + block->data_type = data->data_type; + block->bytes_used = block->size; + list_move(&block->list, &dsp->used_block_list); + list_add(&block->module_list, &module->block_list); + dev_dbg(dsp->dev, " module %d added block %d:%d\n", + module->id, block->type, block->index); + return 0; + } + } + + /* free any allocated blocks on failure */ + list_for_each_entry_safe(block, tmp, &module->block_list, module_list) { + list_del(&block->module_list); + list_move(&block->list, &dsp->free_block_list); + } + return -ENOMEM; +} + +/* allocate free DSP blocks for module data - callers hold locks */ +static int block_alloc(struct sst_module *module, + struct sst_module_data *data) +{ + struct sst_dsp *dsp = module->dsp; + struct sst_mem_block *block, *tmp; + int ret = 0; + + if (data->size == 0) + return 0; + + /* find first free whole blocks that can hold module */ + list_for_each_entry_safe(block, tmp, &dsp->free_block_list, list) { + + /* ignore blocks with wrong type */ + if (block->type != data->type) + continue; + + if (data->size > block->size) + continue; + + data->offset = block->offset; + block->data_type = data->data_type; + block->bytes_used = data->size % block->size; + list_add(&block->module_list, &module->block_list); + list_move(&block->list, &dsp->used_block_list); + dev_dbg(dsp->dev, " *module %d added block %d:%d\n", + module->id, block->type, block->index); + return 0; + } + + /* then find free multiple blocks that can hold module */ + list_for_each_entry_safe(block, tmp, &dsp->free_block_list, list) { + + /* ignore blocks with wrong type */ + if (block->type != data->type) + continue; + + /* do we span > 1 blocks */ + if (data->size > block->size) { + ret = block_alloc_contiguous(module, data, + block->offset + block->size, + data->size - block->size); + if (ret == 0) + return ret; + } + } + + /* not enough free block space */ + return -ENOMEM; +} + +/* remove module from memory - callers hold locks */ +static void block_module_remove(struct sst_module *module) +{ + struct sst_mem_block *block, *tmp; + struct sst_dsp *dsp = module->dsp; + int err; + + /* disable each block */ + list_for_each_entry(block, &module->block_list, module_list) { + + if (block->ops && block->ops->disable) { + err = block->ops->disable(block); + if (err < 0) + dev_err(dsp->dev, + "error: cant disable block %d:%d\n", + block->type, block->index); + } + } + + /* mark each block as free */ + list_for_each_entry_safe(block, tmp, &module->block_list, module_list) { + list_del(&block->module_list); + list_move(&block->list, &dsp->free_block_list); + } +} + +/* prepare the memory block to receive data from host - callers hold locks */ +static int block_module_prepare(struct sst_module *module) +{ + struct sst_mem_block *block; + int ret = 0; + + /* enable each block so that's it'e ready for module P/S data */ + list_for_each_entry(block, &module->block_list, module_list) { + + if (block->ops && block->ops->enable) + ret = block->ops->enable(block); + if (ret < 0) { + dev_err(module->dsp->dev, + "error: cant disable block %d:%d\n", + block->type, block->index); + goto err; + } + } + return ret; + +err: + list_for_each_entry(block, &module->block_list, module_list) { + if (block->ops && block->ops->disable) + block->ops->disable(block); + } + return ret; +} + +/* allocate memory blocks for static module addresses - callers hold locks */ +static int block_alloc_fixed(struct sst_module *module, + struct sst_module_data *data) +{ + struct sst_dsp *dsp = module->dsp; + struct sst_mem_block *block, *tmp; + u32 end = data->offset + data->size, block_end; + int err; + + /* only IRAM/DRAM blocks are managed */ + if (data->type != SST_MEM_IRAM && data->type != SST_MEM_DRAM) + return 0; + + /* are blocks already attached to this module */ + list_for_each_entry_safe(block, tmp, &module->block_list, module_list) { + + /* force compacting mem blocks of the same data_type */ + if (block->data_type != data->data_type) + continue; + + block_end = block->offset + block->size; + + /* find block that holds section */ + if (data->offset >= block->offset && end < block_end) + return 0; + + /* does block span more than 1 section */ + if (data->offset >= block->offset && data->offset < block_end) { + + err = block_alloc_contiguous(module, data, + block->offset + block->size, + data->size - block->size + data->offset - block->offset); + if (err < 0) + return -ENOMEM; + + /* module already owns blocks */ + return 0; + } + } + + /* find first free blocks that can hold section in free list */ + list_for_each_entry_safe(block, tmp, &dsp->free_block_list, list) { + block_end = block->offset + block->size; + + /* find block that holds section */ + if (data->offset >= block->offset && end < block_end) { + + /* add block */ + block->data_type = data->data_type; + list_move(&block->list, &dsp->used_block_list); + list_add(&block->module_list, &module->block_list); + return 0; + } + + /* does block span more than 1 section */ + if (data->offset >= block->offset && data->offset < block_end) { + + err = block_alloc_contiguous(module, data, + block->offset + block->size, + data->size - block->size); + if (err < 0) + return -ENOMEM; + + /* add block */ + block->data_type = data->data_type; + list_move(&block->list, &dsp->used_block_list); + list_add(&block->module_list, &module->block_list); + return 0; + } + + } + + return -ENOMEM; +} + +/* Load fixed module data into DSP memory blocks */ +int sst_module_insert_fixed_block(struct sst_module *module, + struct sst_module_data *data) +{ + struct sst_dsp *dsp = module->dsp; + struct sst_fw *sst_fw = module->sst_fw; + int ret; + + mutex_lock(&dsp->mutex); + + /* alloc blocks that includes this section */ + ret = block_alloc_fixed(module, data); + if (ret < 0) { + dev_err(dsp->dev, + "error: no free blocks for section at offset 0x%x size 0x%x\n", + data->offset, data->size); + mutex_unlock(&dsp->mutex); + return -ENOMEM; + } + + /* prepare DSP blocks for module copy */ + ret = block_module_prepare(module); + if (ret < 0) { + dev_err(dsp->dev, "error: fw module prepare failed\n"); + goto err; + } + + /* copy partial module data to blocks */ + if (dsp->fw_use_dma) { + ret = sst_dsp_dma_copy(dsp, + sst_fw->dmable_fw_paddr + data->data_offset, + dsp->addr.lpe_base + data->offset, data->size); + if (ret < 0) { + dev_err(dsp->dev, "error: module copy failed\n"); + goto err; + } + } else + sst_memcpy32(dsp->addr.lpe + data->offset, data->data, + data->size); + + mutex_unlock(&dsp->mutex); + return ret; + +err: + block_module_remove(module); + mutex_unlock(&dsp->mutex); + return ret; +} +EXPORT_SYMBOL_GPL(sst_module_insert_fixed_block); + +/* Unload entire module from DSP memory */ +int sst_block_module_remove(struct sst_module *module) +{ + struct sst_dsp *dsp = module->dsp; + + mutex_lock(&dsp->mutex); + block_module_remove(module); + mutex_unlock(&dsp->mutex); + return 0; +} +EXPORT_SYMBOL_GPL(sst_block_module_remove); + +/* register a DSP memory block for use with FW based modules */ +struct sst_mem_block *sst_mem_block_register(struct sst_dsp *dsp, u32 offset, + u32 size, enum sst_mem_type type, struct sst_block_ops *ops, u32 index, + void *private) +{ + struct sst_mem_block *block; + + block = kzalloc(sizeof(*block), GFP_KERNEL); + if (block == NULL) + return NULL; + + block->offset = offset; + block->size = size; + block->index = index; + block->type = type; + block->dsp = dsp; + block->private = private; + block->ops = ops; + + mutex_lock(&dsp->mutex); + list_add(&block->list, &dsp->free_block_list); + mutex_unlock(&dsp->mutex); + + return block; +} +EXPORT_SYMBOL_GPL(sst_mem_block_register); + +/* unregister all DSP memory blocks */ +void sst_mem_block_unregister_all(struct sst_dsp *dsp) +{ + struct sst_mem_block *block, *tmp; + + mutex_lock(&dsp->mutex); + + /* unregister used blocks */ + list_for_each_entry_safe(block, tmp, &dsp->used_block_list, list) { + list_del(&block->list); + kfree(block); + } + + /* unregister free blocks */ + list_for_each_entry_safe(block, tmp, &dsp->free_block_list, list) { + list_del(&block->list); + kfree(block); + } + + mutex_unlock(&dsp->mutex); +} +EXPORT_SYMBOL_GPL(sst_mem_block_unregister_all); + +/* allocate scratch buffer blocks */ +struct sst_module *sst_mem_block_alloc_scratch(struct sst_dsp *dsp) +{ + struct sst_module *sst_module, *scratch; + struct sst_mem_block *block, *tmp; + u32 block_size; + int ret = 0; + + scratch = kzalloc(sizeof(struct sst_module), GFP_KERNEL); + if (scratch == NULL) + return NULL; + + mutex_lock(&dsp->mutex); + + /* calculate required scratch size */ + list_for_each_entry(sst_module, &dsp->module_list, list) { + if (scratch->s.size > sst_module->s.size) + scratch->s.size = scratch->s.size; + else + scratch->s.size = sst_module->s.size; + } + + dev_dbg(dsp->dev, "scratch buffer required is %d bytes\n", + scratch->s.size); + + /* init scratch module */ + scratch->dsp = dsp; + scratch->s.type = SST_MEM_DRAM; + scratch->s.data_type = SST_DATA_S; + INIT_LIST_HEAD(&scratch->block_list); + + /* check free blocks before looking at used blocks for space */ + if (!list_empty(&dsp->free_block_list)) + block = list_first_entry(&dsp->free_block_list, + struct sst_mem_block, list); + else + block = list_first_entry(&dsp->used_block_list, + struct sst_mem_block, list); + block_size = block->size; + + /* allocate blocks for module scratch buffers */ + dev_dbg(dsp->dev, "allocating scratch blocks\n"); + ret = block_alloc(scratch, &scratch->s); + if (ret < 0) { + dev_err(dsp->dev, "error: can't alloc scratch blocks\n"); + goto err; + } + + /* assign the same offset of scratch to each module */ + list_for_each_entry(sst_module, &dsp->module_list, list) + sst_module->s.offset = scratch->s.offset; + + mutex_unlock(&dsp->mutex); + return scratch; + +err: + list_for_each_entry_safe(block, tmp, &scratch->block_list, module_list) + list_del(&block->module_list); + mutex_unlock(&dsp->mutex); + return NULL; +} +EXPORT_SYMBOL_GPL(sst_mem_block_alloc_scratch); + +/* free all scratch blocks */ +void sst_mem_block_free_scratch(struct sst_dsp *dsp, + struct sst_module *scratch) +{ + struct sst_mem_block *block, *tmp; + + mutex_lock(&dsp->mutex); + + list_for_each_entry_safe(block, tmp, &scratch->block_list, module_list) + list_del(&block->module_list); + + mutex_unlock(&dsp->mutex); +} +EXPORT_SYMBOL_GPL(sst_mem_block_free_scratch); + +/* get a module from it's unique ID */ +struct sst_module *sst_module_get_from_id(struct sst_dsp *dsp, u32 id) +{ + struct sst_module *module; + + mutex_lock(&dsp->mutex); + + list_for_each_entry(module, &dsp->module_list, list) { + if (module->id == id) { + mutex_unlock(&dsp->mutex); + return module; + } + } + + mutex_unlock(&dsp->mutex); + return NULL; +} +EXPORT_SYMBOL_GPL(sst_module_get_from_id);
Provide services for Intel SST drivers to load SST modular firmware. SST Firmware can be made up of several modules. These modules can exist within any of the compatible SST memory blocks. Provide a generic memory block and firmware module manager that can be used with any SST firmware and core. Signed-off-by: Liam Girdwood <liam.r.girdwood@linux.intel.com> --- sound/soc/intel/sst-firmware.c | 805 +++++++++++++++++++++++++++++++++++++++++ 1 file changed, 805 insertions(+) create mode 100644 sound/soc/intel/sst-firmware.c