Files
linux-nvgpu/drivers/gpu/nvgpu/gm206/acr_gm206.c
Terje Bergstrom 697fe17dd6 gpu: nvgpu: Suppress error msg from VBIOS overlay
Suppress error message when nvgpu tries to load VBIOS overlay, but
one is not found. This situation is normal. This is done by moving
gk20a_request_firmware() to be nvgpu generic function
nvgpu_request_firmware(), and adding a NO_WARN flag to it.

Introduce also a NO_SOC flag to suppress attempt to load firmware
from SoC specific directory in addition to the chip specific
directory. Use it for dGPU firmware files.

Bug 200236777

Change-Id: I0294d3308f029a6a6d3c2effa579d5f69a91e418
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/1223840
(cherry picked from commit cca44c3f010f15918cdd2259c15170ba1917828a)
Reviewed-on: http://git-master/r/1233353
GVS: Gerrit_Virtual_Submit
2016-10-09 13:03:35 -07:00

336 lines
11 KiB
C

/*
* Copyright (c) 2015-2016, NVIDIA CORPORATION. All rights reserved.
*
* This program is free software; you can redistribute it and/or modify it
* under the terms and conditions of the GNU General Public License,
* version 2, as published by the Free Software Foundation.
*
* This program is distributed in the hope 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/delay.h> /* for mdelay */
#include <linux/firmware.h>
#include <linux/clk.h>
#include <linux/module.h>
#include <linux/debugfs.h>
#include <linux/dma-mapping.h>
#include <linux/io.h>
#include "gk20a/gk20a.h"
#include "gk20a/pmu_gk20a.h"
#include "gk20a/semaphore_gk20a.h"
#include "hw_pwr_gm206.h"
#include "acr.h"
#include "acr_gm206.h"
#include "nvgpu_common.h"
/*Defines*/
#define gm206_dbg_pmu(fmt, arg...) \
gk20a_dbg(gpu_dbg_pmu, fmt, ##arg)
/* Both size and address of WPR need to be 128K-aligned */
#define WPR_ALIGNMENT 0x20000
#define DGPU_WPR 0x10000000 /* start from 256MB location at VIDMEM */
#define DGPU_WPR_SIZE 0x100000
static int gm206_pmu_populate_loader_cfg(struct gk20a *g,
void *lsfm, u32 *p_bl_gen_desc_size);
static int gm206_flcn_populate_bl_dmem_desc(struct gk20a *g,
void *lsfm, u32 *p_bl_gen_desc_size, u32 falconid);
static int gm206_bootstrap_hs_flcn(struct gk20a *g);
void gm206_wpr_info(struct gk20a *g, struct wpr_carveout_info *inf)
{
inf->wpr_base = DGPU_WPR;
inf->size = DGPU_WPR_SIZE;
}
static void flcn64_set_dma(struct falc_u64 *dma_addr, u64 value)
{
dma_addr->lo |= u64_lo32(value);
dma_addr->hi |= u64_lo32(value);
}
int gm206_alloc_blob_space(struct gk20a *g,
size_t size, struct mem_desc *mem)
{
struct wpr_carveout_info wpr_inf;
if (mem->size)
return 0;
g->ops.pmu.get_wpr(g, &wpr_inf);
return gk20a_gmmu_alloc_attr_vid_at(g, 0, wpr_inf.size, mem,
wpr_inf.wpr_base);
}
void gm206_init_secure_pmu(struct gpu_ops *gops)
{
gm20b_init_secure_pmu(gops);
gops->pmu.prepare_ucode = prepare_ucode_blob;
gops->pmu.pmu_setup_hw_and_bootstrap = gm206_bootstrap_hs_flcn;
gops->pmu.get_wpr = gm206_wpr_info;
gops->pmu.alloc_blob_space = gm206_alloc_blob_space;
gops->pmu.pmu_populate_loader_cfg = gm206_pmu_populate_loader_cfg;
gops->pmu.flcn_populate_bl_dmem_desc = gm206_flcn_populate_bl_dmem_desc;
}
static int gm206_pmu_populate_loader_cfg(struct gk20a *g,
void *lsfm, u32 *p_bl_gen_desc_size)
{
struct wpr_carveout_info wpr_inf;
struct pmu_gk20a *pmu = &g->pmu;
struct lsfm_managed_ucode_img_v1 *p_lsfm =
(struct lsfm_managed_ucode_img_v1 *)lsfm;
struct flcn_ucode_img *p_img = &(p_lsfm->ucode_img);
struct loader_config_v1 *ldr_cfg =
&(p_lsfm->bl_gen_desc.loader_cfg_v1);
u64 addr_base;
struct pmu_ucode_desc *desc;
u64 addr_code, addr_data;
u32 addr_args;
if (p_img->desc == NULL) /*This means its a header based ucode,
and so we do not fill BL gen desc structure*/
return -EINVAL;
desc = p_img->desc;
/*
Calculate physical and virtual addresses for various portions of
the PMU ucode image
Calculate the 32-bit addresses for the application code, application
data, and bootloader code. These values are all based on IM_BASE.
The 32-bit addresses will be the upper 32-bits of the virtual or
physical addresses of each respective segment.
*/
addr_base = p_lsfm->lsb_header.ucode_off;
g->ops.pmu.get_wpr(g, &wpr_inf);
addr_base += wpr_inf.wpr_base;
gm206_dbg_pmu("pmu loader cfg u32 addrbase %x\n", (u32)addr_base);
/*From linux*/
addr_code = u64_lo32((addr_base +
desc->app_start_offset +
desc->app_resident_code_offset));
gm206_dbg_pmu("app start %d app res code off %d\n",
desc->app_start_offset, desc->app_resident_code_offset);
addr_data = u64_lo32((addr_base +
desc->app_start_offset +
desc->app_resident_data_offset));
gm206_dbg_pmu("app res data offset%d\n",
desc->app_resident_data_offset);
gm206_dbg_pmu("bl start off %d\n", desc->bootloader_start_offset);
addr_args = ((pwr_falcon_hwcfg_dmem_size_v(
gk20a_readl(g, pwr_falcon_hwcfg_r())))
<< GK20A_PMU_DMEM_BLKSIZE2);
addr_args -= g->ops.pmu_ver.get_pmu_cmdline_args_size(pmu);
gm206_dbg_pmu("addr_args %x\n", addr_args);
/* Populate the loader_config state*/
ldr_cfg->dma_idx = GK20A_PMU_DMAIDX_UCODE;
flcn64_set_dma(&ldr_cfg->code_dma_base, addr_code);
ldr_cfg->code_size_total = desc->app_size;
ldr_cfg->code_size_to_load = desc->app_resident_code_size;
ldr_cfg->code_entry_point = desc->app_imem_entry;
flcn64_set_dma(&ldr_cfg->data_dma_base, addr_data);
ldr_cfg->data_size = desc->app_resident_data_size;
flcn64_set_dma(&ldr_cfg->overlay_dma_base, addr_code);
/* Update the argc/argv members*/
ldr_cfg->argc = 1;
ldr_cfg->argv = addr_args;
*p_bl_gen_desc_size = sizeof(struct loader_config_v1);
g->acr.pmu_args = addr_args;
return 0;
}
static int gm206_flcn_populate_bl_dmem_desc(struct gk20a *g,
void *lsfm, u32 *p_bl_gen_desc_size, u32 falconid)
{
struct wpr_carveout_info wpr_inf;
struct lsfm_managed_ucode_img_v1 *p_lsfm =
(struct lsfm_managed_ucode_img_v1 *)lsfm;
struct flcn_ucode_img *p_img = &(p_lsfm->ucode_img);
struct flcn_bl_dmem_desc_v1 *ldr_cfg =
&(p_lsfm->bl_gen_desc.bl_dmem_desc_v1);
u64 addr_base;
struct pmu_ucode_desc *desc;
u64 addr_code, addr_data;
if (p_img->desc == NULL) /*This means its a header based ucode,
and so we do not fill BL gen desc structure*/
return -EINVAL;
desc = p_img->desc;
/*
Calculate physical and virtual addresses for various portions of
the PMU ucode image
Calculate the 32-bit addresses for the application code, application
data, and bootloader code. These values are all based on IM_BASE.
The 32-bit addresses will be the upper 32-bits of the virtual or
physical addresses of each respective segment.
*/
addr_base = p_lsfm->lsb_header.ucode_off;
g->ops.pmu.get_wpr(g, &wpr_inf);
addr_base += wpr_inf.wpr_base;
gm206_dbg_pmu("gen loader cfg %x u32 addrbase %x ID\n", (u32)addr_base,
p_lsfm->wpr_header.falcon_id);
addr_code = u64_lo32((addr_base +
desc->app_start_offset +
desc->app_resident_code_offset));
addr_data = u64_lo32((addr_base +
desc->app_start_offset +
desc->app_resident_data_offset));
gm206_dbg_pmu("gen cfg %x u32 addrcode %x & data %x load offset %xID\n",
(u32)addr_code, (u32)addr_data, desc->bootloader_start_offset,
p_lsfm->wpr_header.falcon_id);
/* Populate the LOADER_CONFIG state */
memset((void *) ldr_cfg, 0, sizeof(struct flcn_bl_dmem_desc_v1));
ldr_cfg->ctx_dma = GK20A_PMU_DMAIDX_UCODE;
flcn64_set_dma(&ldr_cfg->code_dma_base, addr_code);
ldr_cfg->non_sec_code_size = desc->app_resident_code_size;
flcn64_set_dma(&ldr_cfg->data_dma_base, addr_data);
ldr_cfg->data_size = desc->app_resident_data_size;
ldr_cfg->code_entry_point = desc->app_imem_entry;
*p_bl_gen_desc_size = sizeof(struct flcn_bl_dmem_desc_v1);
return 0;
}
/*Loads ACR bin to FB mem and bootstraps PMU with bootloader code
* start and end are addresses of ucode blob in non-WPR region*/
static int gm206_bootstrap_hs_flcn(struct gk20a *g)
{
struct mm_gk20a *mm = &g->mm;
struct vm_gk20a *vm = &mm->pmu.vm;
int i, err = 0;
u64 *acr_dmem;
u32 img_size_in_bytes = 0;
u32 status;
struct acr_desc *acr = &g->acr;
const struct firmware *acr_fw = acr->acr_fw;
struct flcn_bl_dmem_desc_v1 *bl_dmem_desc = &acr->bl_dmem_desc_v1;
u32 *acr_ucode_header_t210_load;
u32 *acr_ucode_data_t210_load;
struct wpr_carveout_info wpr_inf;
gm206_dbg_pmu("");
if (!acr_fw) {
/*First time init case*/
acr_fw = nvgpu_request_firmware(g,
GM20B_HSBIN_PMU_UCODE_IMAGE,
NVGPU_REQUEST_FIRMWARE_NO_SOC);
if (!acr_fw) {
gk20a_err(dev_from_gk20a(g), "pmu ucode get fail");
return -ENOENT;
}
acr->acr_fw = acr_fw;
acr->hsbin_hdr = (struct bin_hdr *)acr_fw->data;
acr->fw_hdr = (struct acr_fw_header *)(acr_fw->data +
acr->hsbin_hdr->header_offset);
acr_ucode_data_t210_load = (u32 *)(acr_fw->data +
acr->hsbin_hdr->data_offset);
acr_ucode_header_t210_load = (u32 *)(acr_fw->data +
acr->fw_hdr->hdr_offset);
img_size_in_bytes = ALIGN((acr->hsbin_hdr->data_size), 256);
/* Lets patch the signatures first.. */
if (acr_ucode_patch_sig(g, acr_ucode_data_t210_load,
(u32 *)(acr_fw->data +
acr->fw_hdr->sig_prod_offset),
(u32 *)(acr_fw->data +
acr->fw_hdr->sig_dbg_offset),
(u32 *)(acr_fw->data +
acr->fw_hdr->patch_loc),
(u32 *)(acr_fw->data +
acr->fw_hdr->patch_sig)) < 0) {
gk20a_err(dev_from_gk20a(g), "patch signatures fail");
err = -1;
goto err_release_acr_fw;
}
err = gk20a_gmmu_alloc_map_sys(vm, img_size_in_bytes,
&acr->acr_ucode);
if (err) {
err = -ENOMEM;
goto err_release_acr_fw;
}
g->ops.pmu.get_wpr(g, &wpr_inf);
acr_dmem = (u64 *)
&(((u8 *)acr_ucode_data_t210_load)[
acr_ucode_header_t210_load[2]]);
acr->acr_dmem_desc = (struct flcn_acr_desc *)((u8 *)(
acr->acr_ucode.cpu_va) + acr_ucode_header_t210_load[2]);
((struct flcn_acr_desc *)acr_dmem)->nonwpr_ucode_blob_start =
wpr_inf.wpr_base;
((struct flcn_acr_desc *)acr_dmem)->nonwpr_ucode_blob_size =
wpr_inf.size;
((struct flcn_acr_desc *)acr_dmem)->regions.no_regions = 1;
((struct flcn_acr_desc *)acr_dmem)->wpr_offset = 0;
((struct flcn_acr_desc *)acr_dmem)->wpr_region_id = 1;
((struct flcn_acr_desc *)acr_dmem)->regions.region_props[
0].region_id = 1;
((struct flcn_acr_desc *)acr_dmem)->regions.region_props[
0].start_addr = wpr_inf.wpr_base >> 8;
((struct flcn_acr_desc *)acr_dmem)->regions.region_props[
0].end_addr = (wpr_inf.wpr_base + wpr_inf.size) >> 8;
for (i = 0; i < (img_size_in_bytes/4); i++) {
((u32 *)acr->acr_ucode.cpu_va)[i] =
acr_ucode_data_t210_load[i];
}
/*
* In order to execute this binary, we will be using
* a bootloader which will load this image into PMU IMEM/DMEM.
* Fill up the bootloader descriptor for PMU HAL to use..
* TODO: Use standard descriptor which the generic bootloader is
* checked in.
*/
bl_dmem_desc->signature[0] = 0;
bl_dmem_desc->signature[1] = 0;
bl_dmem_desc->signature[2] = 0;
bl_dmem_desc->signature[3] = 0;
bl_dmem_desc->ctx_dma = GK20A_PMU_DMAIDX_VIRT;
flcn64_set_dma(&bl_dmem_desc->code_dma_base,
acr->acr_ucode.gpu_va);
bl_dmem_desc->non_sec_code_off = acr_ucode_header_t210_load[0];
bl_dmem_desc->non_sec_code_size = acr_ucode_header_t210_load[1];
bl_dmem_desc->sec_code_off = acr_ucode_header_t210_load[5];
bl_dmem_desc->sec_code_size = acr_ucode_header_t210_load[6];
bl_dmem_desc->code_entry_point = 0; /* Start at 0th offset */
flcn64_set_dma(&bl_dmem_desc->data_dma_base,
acr->acr_ucode.gpu_va +
(acr_ucode_header_t210_load[2]));
bl_dmem_desc->data_size = acr_ucode_header_t210_load[3];
} else
acr->acr_dmem_desc->nonwpr_ucode_blob_size = 0;
status = pmu_exec_gen_bl(g, bl_dmem_desc, 1);
if (status != 0) {
err = status;
goto err_free_ucode_map;
}
return 0;
err_free_ucode_map:
gk20a_gmmu_unmap_free(vm, &acr->acr_ucode);
err_release_acr_fw:
release_firmware(acr_fw);
acr->acr_fw = NULL;
return err;
}