- Openamp-system-reference - lists.openampproject.org

by Chan King Choy, Nathalie

Adding a 30 min sync in the alternate weeks between our main sync, leading up to the November webinar. November 2022 webinar planning document in OpenAMP Google Drive (limited access)<https://docs.google.com/document/d/1aje8wThv7pKdJ5tvX0SNncJtK1YtigFgyXpxhvO…> [https://st2.zoom.us/static/6.2.8460/image/new/ZoomLogo_110_25.png]<https://zoom.us/> Hi there, Nathalie Chan King Choy is inviting you to a scheduled Zoom meeting. Join Zoom Meeting<https://xilinx.zoom.us/j/91782323144?pwd=SDAxbUZLSWQ1akEvek9tY0dUcWZsZz09&f…> Phone one-tap: US: +17209289299,,91782323144#,,,,,,0#,,8593217354#<tel:+17209289299,,91782323144#,,,,,,0#,,8593217354#> or +19292056099,,91782323144#,,,,,,0#,,8593217354#<tel:+19292056099,,91782323144#,,,,,,0#,,8593217354#> Meeting URL: https://xilinx.zoom.us/j/91782323144?pwd=SDAxbUZLSWQ1akEvek9tY0dUcWZsZz09&f… Meeting ID: 917 8232 3144 Passcode: fp$3QJysed Join by Telephone For higher quality, dial a number based on your current location. Dial: US: +1 720 928 9299 or +1 929 205 6099 or +1 669 900 6833 or 833 548 0276 (Toll Free) or 833 548 0282 (Toll Free) or 877 853 5257 (Toll Free) or 888 475 4499 (Toll Free) India: +91 226 480 2722 or +91 22 71 279 525 or +91 406 480 2722 or +91 446 480 2722 or +91 806 480 2722 or +91 80 71 279 440 or +91 116 480 2722 or +91 22 48 798 004 or +91 224 879 8012 or +91 225 097 2744 or +91 225 097 2745 or 000 800 001 4002 (Toll Free) or 000 800 050 5050 (Toll Free) Ireland: +353 1 653 3895 or +353 6 163 9031 or +353 1 536 9320 or 1800 943 965 (Toll Free) or 1800 949 238 (Toll Free) or 1800 901 561 (Toll Free) Singapore: +65 3158 7288 or +65 3165 1065 or 800 101 3814 (Toll Free) or 800 852 6054 (Toll Free) or 1800 407 5602 (Toll Free) Meeting ID: 917 8232 3144 Passcode: 8593217354 International numbers<https://xilinx.zoom.us/u/acpYowtZ5> Join from a Video Conference room system Meeting ID: 917 8232 3144 Passcode: 8593217354 US: 91782323144(a)global.zoomcrc.com<mailto:91782323144@global.zoomcrc.com> Passcode: 8593217354 APAC: 91782323144(a)global.zoomcrc.com<mailto:91782323144@global.zoomcrc.com> Passcode: 8593217354 India: 91782323144(a)global.zoomcrc.com<mailto:91782323144@global.zoomcrc.com> Passcode: 8593217354 Europe: 91782323144(a)global.zoomcrc.com<mailto:91782323144@global.zoomcrc.com> Passcode: 8593217354 SIP: 91782323144(a)zoomcrc.com<mailto:91782323144@zoomcrc.com> Passcode: 8593217354

2 years, 11 months

1
0
0 0

Calling for agenda: 2022-09-14 OpenAMP System Reference

by Chan King Choy, Nathalie

Hi all, Please let us know if you have topics for this Wednesday's call. Here's what I have so far: * Nov webinar: * Planning: Does anything else need to go in the plan<https://docs.google.com/document/d/1aje8wThv7pKdJ5tvX0SNncJtK1YtigFgyXpxhvO…>? (access limited, lmk if you need access) * Discuss work breakdown. Flesh out the tasks & owners. * If you want to get a head start, you can go to the plan<https://docs.google.com/document/d/1aje8wThv7pKdJ5tvX0SNncJtK1YtigFgyXpxhvO…> in Google Drive and start adding sub-bullets to the work breakdown * Individual updates Thanks & regards, Nathalie

2 years, 11 months

1
0
0 0

Calling for agenda: 2022-08-31 OpenAMP System Reference call

by Chan King Choy, Nathalie

Hi all, Please let us know if you have specific topics for this week's OpenAMP System Reference call. Here's what I have so far: * Is 1st half of November still looking good for a demo? * November demo: * What to convey? * What should it include? * What work needs to happen to support that? * Individual updates For the November demo discussion, here is an outline that can be helpful (we don't have to use it, but it can be a starting point): Who is our target audience? Target audience e.g. System architect e.g. RTOS vendor ??? Need/pain/application How could they benefit from OpenAMP? What is their alternative to OpenAMP? Why makes OpenAMP better suited to their needs? Webinar plan * What is expected state of audience coming into the webinar? * What is desired state of the audience after viewing the webinar? * Call to action: * What should they do next? * Why should they do this? (i.e. what's the benefit for them?) * Top 3 key takeaways * What are the sections of the webinar? For example: * Intro (Why OpenAMP slides) + what's in today's demo * Demo 1: ?? * Demo 2: ?? * CTA: * Try it out, here's where to find the stuff & how to give feedback * Get involved, here's how * Summary of takeaways/Conclusion Work breakdown * Getting the webinar scheduled/advertised * Slides * Demo 1 * Demo 2 * Demo N * Repositories * Documentation * Any other top level tasks to break down, or better way to categorize the tasks? Thanks & regards, Nathalie

2 years, 11 months

1
1
0 0

2022-08-17 OpenAMP System Reference notes and action items

by Chan King Choy, Nathalie

Hi all, Notes from yesterday's call can be found at https://github.com/OpenAMP/openamp-system-reference/wiki/OpenAMP-System-Ref… . Action items from this call: * @arnaud.pouliquen@st.com<mailto:arnaud.pouliquen@st.com>: Check if ST has parts w/ hard-wired Ethernet. * @Bill Mills<mailto:bill.mills@linaro.org>: ask if others in Linaro know more on Zephyr certification project status. * @Chan King Choy, Nathalie<mailto:nathalie.chan-king-choy@amd.com>: Send email to Board when we should meet again & put social channels on the agenda * @Milea, Danut Gabriel (Danut)<mailto:Danut.Milea@windriver.com>: The required mods to use UART for R5 are in one of our Zephyr repos. In a DT overlay somewhere. Can check & share. Best regards, Nathalie

3 years

2
1
0 0

2022-07-13 and 2022-06-29 notes posted

by Chan King Choy, Nathalie

Hi all, The notes & action items are now posted at https://github.com/OpenAMP/openamp-system-reference/wiki/OpenAMP-System-Ref… for today's call, as well as the 6/29 call which I forgot to post before >_< Best regards, Nathalie

3 years, 1 month

1
0
0 0

Calling for agenda: 2022-06-29 OpenAMP System Reference bi-weekly

by Chan King Choy, Nathalie

Hi all, Calling for agenda items. Here's what I have so far: * Individual updates * 7/6 call: Dan will be OOO 7/6 & other US folks might be taking holidays that week due to the long weekend. Options: * A) Keep it? * B) Push out the 7/6 call by a week, reconvene on 7/13 * C) Cancel 7/6, reconvene on 7/20 Thanks & regards, Nathalie

3 years, 1 month

1
0
0 0

Re: [PATCH v8 6/6] drivers: remoteproc: Add Xilinx r5 remoteproc driver

by Tanmay Shah

Hi Mathieu, I agree to all of the comments. I will address them in next revision. Please find my response below for some comments: Thanks, Tanmay On 6/8/22 10:37 AM, Mathieu Poirier wrote: > On Thu, Jun 02, 2022 at 01:38:34PM -0700, Tanmay Shah wrote: >> This driver enables r5f dual core Real time Processing Unit subsystem >> available on Xilinx Zynq Ultrascale MPSoC Platform. RPU subsystem >> (cluster) can be configured in different modes e.g. split mode in which >> two r5f cores work independent of each other and lock-step mode in which >> both r5f cores execute same code clock-for-clock and notify if the >> result is different. >> >> The Xilinx r5 Remoteproc Driver boots the RPU cores via calls to the Xilinx >> Platform Management Unit that handles the R5 configuration, memory access >> and R5 lifecycle management. The interface to this manager is done in this >> driver via zynqmp_pm_* function calls. >> >> Signed-off-by: Ben Levinsky <ben.levinsky(a)xilinx.com> >> Signed-off-by: Tanmay Shah <tanmay.shah(a)xilinx.com> >> --- >> >> Changes in v8: >> - None >> >> Changes in v7: >> - None >> >> Changes in v6: >> - None >> >> Changes in v5: >> - None >> >> Changes in v4: >> - Remove redundant header files >> - use dev_err_probe() to report errors during probe >> - Fix missing check on error code returned by zynqmp_r5_add_rproc_core() >> - Fix memory leaks all over the driver when resource allocation fails for any core >> - make cluster mode check only at one place >> - remove redundant initialization of variable >> - remove redundant use of of_node_put() >> - Fix Comment format problem >> - Assign offset of zynqmp_tcm_banks instead of duplicating it >> - Add tcm and memory regions rproc carveouts during prepare instead of parse_fw >> - Remove rproc_mem_entry object from r5_core >> - Use put_device() and rproc_del() APIs to fix memory leaks >> - Replace pr_* with dev_*. This was missed in v3, fix now. >> - Use "GPL" instead of "GPL v2" in MODULE_LICENSE macro. This was suggested by checkpatch script. >> >> Changes in v3: >> - Fix checkpatch script indentation warning >> - Remove unused variable from xilinx remoteproc driver >> - use C style comments, i.e /*...*/ >> - Remove redundant debug information which can be derived using /proc/device-tree >> - Fix multilined comment format >> - s/"final fot TCM"/"final for TCM" >> - Function devm_kzalloc() does not return an code on error, just NULL. >> Remove redundant error check for this function throughout the driver. >> - Fix RPU mode configuration and add documentation accordingly >> - Get rid of the indentations to match function documentation style with rest of the driver >> - Fix memory leak by only using r5_rproc->priv and not replace it with new instance >> - Use 'i' for the outer loop and 'j' for the inner one as per convention >> - Remove redundant error and NULL checks throughout the driver >> - Use devm_kcalloc() when more than one element is required >> - Add memory-regions carveouts during driver probe instead of parse_fw call >> This removes redundant copy of reserved_mem object in r5_core structure. >> - Fix memory leak by using of_node_put() >> - Fix indentation of tcm_mem_map function args >> - Remove redundant init of variables >> - Initialize tcm bank size variable for lockstep mode >> - Replace u32 with phys_addr_t for variable stroing memory bank address >> - Add documentation of TCM behavior in lockstep mode >> - Use dev_get_drvdata instead of platform driver API >> - Remove info level messages >> - Fix checkpatch.pl warnings >> - Add documentation for the Xilinx r5f platform to understand driver design >> drivers/remoteproc/Kconfig | 12 + >> drivers/remoteproc/Makefile | 1 + >> drivers/remoteproc/xlnx_r5_remoteproc.c | 1045 +++++++++++++++++++++++ >> 3 files changed, 1058 insertions(+) >> create mode 100644 drivers/remoteproc/xlnx_r5_remoteproc.c >> >> diff --git a/drivers/remoteproc/Kconfig b/drivers/remoteproc/Kconfig >> index 166019786653..5637a71c0677 100644 >> --- a/drivers/remoteproc/Kconfig >> +++ b/drivers/remoteproc/Kconfig >> @@ -352,6 +352,18 @@ config TI_K3_R5_REMOTEPROC >> It's safe to say N here if you're not interested in utilizing >> a slave processor. >> >> +config XLNX_R5_REMOTEPROC >> + tristate "Xilinx R5 remoteproc support" >> + depends on PM && ARCH_ZYNQMP >> + depends on ZYNQMP_FIRMWARE >> + select RPMSG_VIRTIO >> + select ZYNQMP_IPI_MBOX >> + help >> + Say y or m here to support Xilinx R5 remote processors via the remote >> + processor framework. >> + >> + It's safe to say N if not interested in using RPU r5f cores. >> + >> endif # REMOTEPROC >> >> endmenu >> diff --git a/drivers/remoteproc/Makefile b/drivers/remoteproc/Makefile >> index 5478c7cb9e07..91314a9b43ce 100644 >> --- a/drivers/remoteproc/Makefile >> +++ b/drivers/remoteproc/Makefile >> @@ -38,3 +38,4 @@ obj-$(CONFIG_ST_SLIM_REMOTEPROC) += st_slim_rproc.o >> obj-$(CONFIG_STM32_RPROC) += stm32_rproc.o >> obj-$(CONFIG_TI_K3_DSP_REMOTEPROC) += ti_k3_dsp_remoteproc.o >> obj-$(CONFIG_TI_K3_R5_REMOTEPROC) += ti_k3_r5_remoteproc.o >> +obj-$(CONFIG_XLNX_R5_REMOTEPROC) += xlnx_r5_remoteproc.o >> diff --git a/drivers/remoteproc/xlnx_r5_remoteproc.c b/drivers/remoteproc/xlnx_r5_remoteproc.c >> new file mode 100644 >> index 000000000000..394b3469463c >> --- /dev/null >> +++ b/drivers/remoteproc/xlnx_r5_remoteproc.c >> @@ -0,0 +1,1045 @@ >> +// SPDX-License-Identifier: GPL-2.0 >> +/* >> + * ZynqMP R5 Remote Processor driver >> + * >> + */ >> + >> +#include <dt-bindings/power/xlnx-zynqmp-power.h> >> +#include <linux/dma-mapping.h> >> +#include <linux/firmware/xlnx-zynqmp.h> >> +#include <linux/kernel.h> >> +#include <linux/module.h> >> +#include <linux/of_address.h> >> +#include <linux/of_platform.h> >> +#include <linux/of_reserved_mem.h> >> +#include <linux/platform_device.h> >> +#include <linux/remoteproc.h> >> +#include <linux/slab.h> >> + >> +#include "remoteproc_internal.h" >> + >> +/* settings for RPU cluster mode */ >> +enum zynqmp_r5_cluster_mode { >> + SPLIT_MODE = 0, /* When cores run as separate processor */ >> + LOCKSTEP_MODE = 1, /* cores execute same code in lockstep,clk-for-clk */ >> + SINGLE_CPU_MODE = 2, /* core0 is held in reset and only core1 runs */ >> +}; >> + >> +/** >> + * struct mem_bank_data - Memory Bank description >> + * >> + * @addr: Start address of memory bank >> + * @size: Size of Memory bank >> + * @pm_domain_id: Power-domains id of memory bank for firmware to turn on/off >> + * @bank_name: name of the bank for remoteproc framework >> + */ >> +struct mem_bank_data { >> + phys_addr_t addr; >> + size_t size; >> + u32 pm_domain_id; >> + char *bank_name; >> +}; >> + >> +static const struct mem_bank_data zynqmp_tcm_banks[] = { >> + {0xffe00000UL, 0x10000UL, PD_R5_0_ATCM, "atcm0"}, /* TCM 64KB each */ >> + {0xffe20000UL, 0x10000UL, PD_R5_0_BTCM, "btcm0"}, >> + {0xffe90000UL, 0x10000UL, PD_R5_1_ATCM, "atcm1"}, >> + {0xffeb0000UL, 0x10000UL, PD_R5_1_BTCM, "btcm1"}, >> +}; >> + >> +/** >> + * struct zynqmp_r5_core - ZynqMP R5 core structure >> + * >> + * @dev: device of RPU instance >> + * @np: device node of RPU instance >> + * @tcm_bank_count: number TCM banks accessible to this RPU >> + * @tcm_banks: array of each TCM bank data >> + * @rmem_count: Number of reserved mem regions >> + * @rmem: reserved memory region nodes from device tree >> + * @rproc: rproc handle >> + * @pm_domain_id: RPU CPU power domain id >> + */ >> +struct zynqmp_r5_core { >> + struct device *dev; >> + struct device_node *np; >> + int tcm_bank_count; >> + struct mem_bank_data **tcm_banks; >> + int rmem_count; >> + struct reserved_mem **rmem; >> + struct rproc *rproc; >> + u32 pm_domain_id; >> +}; >> + >> +/** >> + * struct zynqmp_r5_cluster - ZynqMP R5 cluster structure >> + * >> + * @dev: r5f subsystem cluster device node >> + * @mode: cluster mode of type zynqmp_r5_cluster_mode >> + * @core_count: number of r5 cores used for this cluster mode >> + * @r5_cores: Array of pointers pointing to r5 core >> + */ >> +struct zynqmp_r5_cluster { >> + struct device *dev; >> + enum zynqmp_r5_cluster_mode mode; >> + int core_count; >> + struct zynqmp_r5_core **r5_cores; >> +}; >> + >> +/* >> + * zynqmp_r5_set_mode - set RPU operation mode >> + * >> + * set RPU operation mode >> + * >> + * Return: 0 for success, negative value for failure >> + */ >> +static int zynqmp_r5_set_mode(struct zynqmp_r5_core *r5_core, >> + enum rpu_oper_mode fw_reg_val, >> + enum rpu_tcm_comb tcm_mode) >> +{ >> + int ret; >> + >> + ret = zynqmp_pm_set_rpu_mode(r5_core->pm_domain_id, fw_reg_val); >> + if (ret < 0) { >> + dev_err(r5_core->dev, "failed to set RPU mode\n"); >> + return ret; >> + } >> + >> + ret = zynqmp_pm_set_tcm_config(r5_core->pm_domain_id, tcm_mode); >> + if (ret < 0) >> + dev_err(r5_core->dev, "failed to configure TCM\n"); >> + >> + return ret; >> +} >> + >> +/* >> + * zynqmp_r5_rproc_start >> + * @rproc: single R5 core's corresponding rproc instance >> + * >> + * Start R5 Core from designated boot address. >> + * >> + * return 0 on success, otherwise non-zero value on failure >> + */ >> +static int zynqmp_r5_rproc_start(struct rproc *rproc) >> +{ >> + struct zynqmp_r5_core *r5_core = rproc->priv; >> + enum rpu_boot_mem bootmem; >> + int ret; >> + >> + /* >> + * The exception vector pointers (EVP) refer to the base-address of >> + * exception vectors (for reset, IRQ, FIQ, etc). The reset-vector >> + * starts at the base-address and subsequent vectors are on 4-byte >> + * boundaries. >> + * >> + * Exception vectors can start either from 0x0000_0000 (LOVEC) or >> + * from 0xFFFF_0000 (HIVEC) which is mapped in the OCM (On-Chip Memory) >> + * >> + * Usually firmware will put Exception vectors at LOVEC. >> + * >> + * It is not recommend that you change the exception vector. >> + * Changing the EVP to HIVEC will result in increased interrupt latency >> + * and jitter. Also, if the OCM is secured and the Cortex-R5F processor >> + * is non-secured, then the Cortex-R5F processor cannot access the >> + * HIVEC exception vectors in the OCM. >> + */ >> + bootmem = (rproc->bootaddr >= 0xFFFC0000) ? >> + PM_RPU_BOOTMEM_HIVEC : PM_RPU_BOOTMEM_LOVEC; >> + >> + dev_dbg(r5_core->dev, "RPU boot addr 0x%llx from %s.", rproc->bootaddr, >> + bootmem == PM_RPU_BOOTMEM_HIVEC ? "OCM" : "TCM"); >> + >> + ret = zynqmp_pm_request_wake(r5_core->pm_domain_id, 1, >> + bootmem, ZYNQMP_PM_REQUEST_ACK_NO); >> + if (ret) >> + dev_err(r5_core->dev, >> + "failed to start RPU = 0x%x\n", r5_core->pm_domain_id); >> + return ret; >> +} >> + >> +/* >> + * zynqmp_r5_rproc_stop >> + * @rproc: single R5 core's corresponding rproc instance >> + * >> + * Power down R5 Core. >> + * >> + * return 0 on success, otherwise non-zero value on failure >> + */ >> +static int zynqmp_r5_rproc_stop(struct rproc *rproc) >> +{ >> + struct zynqmp_r5_core *r5_core = rproc->priv; >> + int ret; >> + >> + ret = zynqmp_pm_force_pwrdwn(r5_core->pm_domain_id, >> + ZYNQMP_PM_REQUEST_ACK_BLOCKING); >> + if (ret) >> + dev_err(r5_core->dev, "failed to stop remoteproc RPU %d\n", ret); >> + >> + return ret; >> +} >> + >> +/* >> + * zynqmp_r5_mem_region_map >> + * @rproc: single R5 core's corresponding rproc instance >> + * @mem: mem entry to map >> + * >> + * Callback to map va for memory-region's carveout. >> + * >> + * return 0 on success, otherwise non-zero value on failure >> + */ >> +static int zynqmp_r5_mem_region_map(struct rproc *rproc, >> + struct rproc_mem_entry *mem) >> +{ >> + void __iomem *va; >> + >> + va = ioremap_wc(mem->dma, mem->len); >> + if (IS_ERR_OR_NULL(va)) >> + return -ENOMEM; >> + >> + mem->va = (void *)va; >> + >> + return 0; >> +} >> + >> +/* >> + * zynqmp_r5_rproc_mem_unmap >> + * @rproc: single R5 core's corresponding rproc instance >> + * @mem: mem entry to unmap >> + * >> + * Unmap memory-region carveout >> + * >> + * return 0 on success, otherwise non-zero value on failure >> + */ >> +static int zynqmp_r5_mem_region_unmap(struct rproc *rproc, >> + struct rproc_mem_entry *mem) >> +{ >> + iounmap((void __iomem *)mem->va); >> + return 0; >> +} >> + >> +/* >> + * add_mem_regions >> + * @r5_core: single R5 core's corresponding zynqmp_r5_core type instance >> + * @rmem: reserved mem region parsed from dt node >> + * >> + * Construct rproc mem carveouts from carveout provided in >> + * memory-region property >> + * >> + * return 0 on success, otherwise non-zero value on failure >> + */ >> +static int add_mem_regions_carveout(struct rproc *rproc) >> +{ >> + struct zynqmp_r5_core *r5_core; >> + struct reserved_mem *rmem; >> + struct rproc_mem_entry **rproc_mem; >> + int i, num_mem_regions; >> + >> + r5_core = (struct zynqmp_r5_core *)rproc->priv; >> + num_mem_regions = r5_core->rmem_count; >> + >> + /* memory regions not defined */ >> + if (num_mem_regions < 1) >> + return 0; > This can't happen as per zynqmp_r5_get_mem_region_node(). That's right. I will fix this in zynqmp_r5_get_mem_region_node(). As memory-region is not required property, we return successfully if memory regions are not defined. So, zynqmp_r5_get_mem_region_node() should not fail if memory-region property is not found. > >> + >> + rproc_mem = kcalloc(num_mem_regions, >> + sizeof(struct rproc_mem_entry *), GFP_KERNEL); >> + if (!rproc_mem) >> + return -ENOMEM; >> + >> + for (i = 0; i < num_mem_regions; i++) { >> + rmem = r5_core->rmem[i]; >> + >> + /* Register associated reserved memory regions */ >> + rproc_mem[i] = rproc_mem_entry_init(&rproc->dev, NULL, >> + (dma_addr_t)rmem->base, >> + rmem->size, rmem->base, >> + zynqmp_r5_mem_region_map, >> + zynqmp_r5_mem_region_unmap, >> + rmem->name); > Any reason you need to call rproc_mem_entry_init() on the vdev0buffers? Thanks for this pointer. Actually this is remoteproc driver only and I was only considering to add only fw-image memory-region. But, as this is generic function for all the memory regions, I think I should handle other regions as well. I believe there is no code change needed to add virngs. However, for vdev*buffers I will use "rproc_of_resm_mem_entry_init()". >> + if (!rproc_mem[i]) >> + goto release_mem_regions; >> + >> + dev_dbg(&rproc->dev, "reserved mem carveout %s addr=%llx, size=0x%llx", >> + rmem->name, rmem->base, rmem->size); >> + } >> + >> + /* >> + * Add carveouts only if all rproc mem enties are >> + * successfully initialized >> + */ >> + for (i = 0; i < num_mem_regions; i++) >> + rproc_add_carveout(rproc, rproc_mem[i]); >> + >> + kfree(rproc_mem); >> + return 0; >> + >> +release_mem_regions: >> + for (i--; i > -1; i--) >> + kfree(rproc_mem[i]); >> + kfree(rproc_mem); >> + return -ENOMEM; >> +} >> + >> +/* >> + * zynqmp_r5_rproc_mem_unmap >> + * @rproc: single R5 core's corresponding rproc instance >> + * @mem: mem entry to unmap >> + * >> + * Unmap TCM banks when powering down R5 core. >> + * >> + * return 0 on success, otherwise non-zero value on failure >> + */ >> +static int tcm_mem_unmap(struct rproc *rproc, struct rproc_mem_entry *mem) >> +{ >> + iounmap((void __iomem *)mem->va); >> + >> + return 0; >> +} >> + >> +/* >> + * tcm_mem_map >> + * @rproc: single R5 core's corresponding rproc instance >> + * @mem: mem entry to initialize the va and da fields of >> + * >> + * Given TCM bank entry, this callback will set device address for R5 >> + * running on TCM and also setup virtual address for TCM bank >> + * remoteproc carveout. >> + * >> + * return 0 on success, otherwise non-zero value on failure >> + */ >> +static int tcm_mem_map(struct rproc *rproc, >> + struct rproc_mem_entry *mem) >> +{ >> + void __iomem *va; >> + >> + va = ioremap_wc(mem->dma, mem->len); >> + if (IS_ERR_OR_NULL(va)) >> + return -ENOMEM; >> + >> + /* Update memory entry va */ >> + mem->va = (void *)va; >> + >> + /* clear TCMs */ >> + memset_io(va, 0, mem->len); >> + >> + /* >> + * The R5s expect their TCM banks to be at address 0x0 and 0x2000, >> + * while on the Linux side they are at 0xffexxxxx. >> + * >> + * Zero out the high 12 bits of the address. This will give >> + * expected values for TCM Banks 0A and 0B (0x0 and 0x20000). >> + */ >> + mem->da &= 0x000fffff; >> + >> + /* >> + * TCM Banks 1A and 1B still have to be translated. >> + * >> + * Below handle these two banks' absolute addresses (0xffe90000 and >> + * 0xffeb0000) and convert to the expected relative addresses >> + * (0x0 and 0x20000). >> + */ >> + if (mem->da == 0x90000 || mem->da == 0xB0000) >> + mem->da -= 0x90000; >> + >> + /* if translated TCM bank address is not valid report error */ >> + if (mem->da != 0x0 && mem->da != 0x20000) { >> + dev_err(&rproc->dev, "invalid TCM address: %x\n", mem->da); >> + return -EINVAL; >> + } >> + return 0; >> +} >> + >> +static int add_tcm_carveout_split_mode(struct rproc *rproc) >> +{ >> + int i, num_banks, ret; >> + struct rproc_mem_entry **rproc_mem; >> + u32 pm_domain_id; >> + phys_addr_t bank_addr; >> + size_t bank_size; >> + char *bank_name; >> + struct device *dev; >> + struct zynqmp_r5_core *r5_core; >> + >> + r5_core = (struct zynqmp_r5_core *)rproc->priv; >> + >> + dev = r5_core->dev; >> + >> + /* go through zynqmp banks for r5 node */ >> + num_banks = r5_core->tcm_bank_count; >> + if (num_banks <= 0) { >> + dev_err(dev, "need to specify TCM banks\n"); >> + return -EINVAL; >> + } > This check should be done in zynqmp_r5_get_tcm_node(), please remove. Same > comment for add_tcm_carveout_lockstep_mode(). > >> + >> + rproc_mem = kcalloc(num_banks, >> + sizeof(struct rproc_mem_entry *), GFP_KERNEL); >> + if (!rproc_mem) >> + return -ENOMEM; >> + >> + /* >> + * Power-on Each 64KB TCM, >> + * register its address space, map and unmap functions >> + * and add carveouts accordingly >> + */ >> + for (i = 0; i < num_banks; i++) { >> + bank_addr = r5_core->tcm_banks[i]->addr; >> + bank_name = r5_core->tcm_banks[i]->bank_name; >> + bank_size = r5_core->tcm_banks[i]->size; >> + pm_domain_id = r5_core->tcm_banks[i]->pm_domain_id; >> + >> + ret = zynqmp_pm_request_node(pm_domain_id, >> + ZYNQMP_PM_CAPABILITY_ACCESS, 0, >> + ZYNQMP_PM_REQUEST_ACK_BLOCKING); >> + if (ret < 0) { >> + dev_err(dev, "failed to turn on TCM 0x%x", pm_domain_id); >> + goto release_tcm_split; >> + } >> + >> + dev_dbg(dev, "TCM carveout split mode %s addr=%llx, size=0x%lx", >> + bank_name, bank_addr, bank_size); >> + >> + rproc_mem[i] = rproc_mem_entry_init(dev, NULL, bank_addr, >> + bank_size, bank_addr, >> + tcm_mem_map, tcm_mem_unmap, >> + bank_name); >> + if (!rproc_mem[i]) { >> + ret = -ENOMEM; >> + goto release_tcm_split; > If we end up here function zynqmp_pm_release_node() would not be called for the > bank that was powered just above. > >> + } >> + } >> + >> + /* >> + * Add carveouts only if all rproc mem enties are >> + * successfully initialized >> + */ >> + for (i = 0; i < num_banks; i++) >> + rproc_add_carveout(rproc, rproc_mem[i]); >> + >> + kfree(rproc_mem); >> + return 0; >> + >> +release_tcm_split: >> + /* If failed, Turn off all TCM banks turned on before */ >> + for (i--; i > -1; i--) { > I have already commented on this, please apply throughout. > >> + pm_domain_id = r5_core->tcm_banks[i]->pm_domain_id; >> + zynqmp_pm_release_node(pm_domain_id); >> + kfree(rproc_mem[i]); >> + } >> + kfree(rproc_mem); >> + return ret; >> +} >> + >> +static int add_tcm_carveout_lockstep_mode(struct rproc *rproc) >> +{ >> + int i, num_banks, ret; >> + struct rproc_mem_entry *rproc_mem; >> + u32 pm_domain_id; >> + phys_addr_t bank_addr; >> + size_t bank_size = 0; >> + char *bank_name; >> + struct device *dev; >> + struct zynqmp_r5_core *r5_core; >> + >> + r5_core = (struct zynqmp_r5_core *)rproc->priv; >> + dev = r5_core->dev; >> + >> + /* Go through zynqmp banks for r5 node */ >> + num_banks = r5_core->tcm_bank_count; >> + if (num_banks <= 0) { >> + dev_err(dev, "need to specify TCM banks\n"); >> + return -EINVAL; >> + } >> + >> + /* >> + * In lockstep mode, TCM is contiguous memory block >> + * However, each TCM block still needs to be enabled individually. >> + * So, Enable each TCM block individually, but add their size >> + * to create contiguous memory region. >> + */ >> + bank_addr = r5_core->tcm_banks[0]->addr; >> + bank_name = r5_core->tcm_banks[0]->bank_name; >> + >> + for (i = 0; i < num_banks; i++) { >> + bank_size += r5_core->tcm_banks[i]->size; >> + pm_domain_id = r5_core->tcm_banks[i]->pm_domain_id; >> + >> + /* Turn on each TCM bank individually */ >> + ret = zynqmp_pm_request_node(pm_domain_id, >> + ZYNQMP_PM_CAPABILITY_ACCESS, 0, >> + ZYNQMP_PM_REQUEST_ACK_BLOCKING); >> + if (ret < 0) { >> + dev_err(dev, "failed to turn on TCM 0x%x", pm_domain_id); >> + goto release_tcm_lockstep; >> + } >> + } >> + >> + dev_dbg(dev, "TCM add carveout lockstep mode %s addr=0x%llx, size=0x%lx", >> + bank_name, bank_addr, bank_size); >> + >> + /* Register TCM address range, TCM map and unmap functions */ >> + rproc_mem = rproc_mem_entry_init(dev, NULL, bank_addr, >> + bank_size, bank_addr, >> + tcm_mem_map, tcm_mem_unmap, >> + bank_name); >> + if (!rproc_mem) { >> + ret = -ENOMEM; >> + goto release_tcm_lockstep; >> + } >> + >> + /* If registration is success, add carveouts */ >> + rproc_add_carveout(rproc, rproc_mem); >> + >> + return 0; >> + >> +release_tcm_lockstep: >> + /* If failed, Turn off all TCM banks turned on before */ >> + for (i--; i > -1; i--) { >> + pm_domain_id = r5_core->tcm_banks[i]->pm_domain_id; >> + zynqmp_pm_release_node(pm_domain_id); >> + } >> + return ret; >> +} >> + >> +/* >> + * add_tcm_banks() >> + * @rproc: single R5 core's corresponding rproc instance >> + * >> + * Given R5 node in remoteproc instance >> + * allocate remoteproc carveout for TCM memory >> + * needed for firmware to be loaded >> + * >> + * return 0 on success, otherwise non-zero value on failure >> + */ >> +static int add_tcm_banks(struct rproc *rproc) >> +{ >> + struct device *dev; >> + struct zynqmp_r5_cluster *cluster; >> + struct zynqmp_r5_core *r5_core; >> + >> + r5_core = (struct zynqmp_r5_core *)rproc->priv; >> + if (!r5_core) >> + return -EINVAL; >> + >> + dev = r5_core->dev; >> + >> + cluster = dev_get_drvdata(dev->parent); >> + if (!cluster) { >> + dev_err(dev->parent, "Invalid driver data\n"); >> + return -EINVAL; >> + } >> + >> + /* >> + * In lockstep mode TCM banks are one contiguous memory region of 256Kb >> + * In split mode, each TCM bank is 64Kb and not contiguous. >> + * We add memory carveouts accordingly. >> + */ >> + if (cluster->mode == SPLIT_MODE) >> + return add_tcm_carveout_split_mode(rproc); >> + else if (cluster->mode == LOCKSTEP_MODE) >> + return add_tcm_carveout_lockstep_mode(rproc); >> + >> + dev_err(cluster->dev, "invalid cluster mode\n"); >> + return -EINVAL; >> +} >> + >> +/* >> + * zynqmp_r5_parse_fw() >> + * @rproc: single R5 core's corresponding rproc instance >> + * @fw: ptr to firmware to be loaded onto r5 core >> + * >> + * When loading firmware, ensure the necessary carveouts are in remoteproc > There are no checks done to that effect in the function... Thanks. This documentation is old. I will remove above line and fix description. > >> + * >> + * return 0 on success, otherwise non-zero value on failure >> + */ >> +static int zynqmp_r5_parse_fw(struct rproc *rproc, const struct firmware *fw) >> +{ >> + int ret; >> + >> + ret = rproc_elf_load_rsc_table(rproc, fw); >> + if (ret == -EINVAL) { >> + /* >> + * resource table only required for IPC. >> + * if not present, this is not necessarily an error; >> + * for example, loading r5 hello world application >> + * so simply inform user and keep going. >> + */ >> + dev_info(&rproc->dev, "no resource table found.\n"); >> + ret = 0; >> + } >> + return ret; >> +} >> + >> +static int zynqmp_r5_rproc_prepare(struct rproc *rproc) >> +{ >> + int ret = 0; >> + >> + ret = add_tcm_banks(rproc); >> + if (ret) { >> + dev_err(&rproc->dev, "failed to get TCM banks, err %d\n", ret); >> + return ret; >> + } >> + >> + ret = add_mem_regions_carveout(rproc); >> + if (ret) >> + dev_warn(&rproc->dev, "failed to get reserve mem regions %d\n", >> + ret); >> + >> + return 0; >> +} >> + >> +static int zynqmp_r5_rproc_unprepare(struct rproc *rproc) >> +{ >> + struct zynqmp_r5_core *r5_core; >> + int i; >> + u32 pm_domain_id; >> + >> + r5_core = (struct zynqmp_r5_core *)rproc->priv; >> + >> + for (i = 0; i < r5_core->tcm_bank_count; i++) { >> + pm_domain_id = r5_core->tcm_banks[i]->pm_domain_id; >> + if (zynqmp_pm_release_node(pm_domain_id)) >> + dev_warn(r5_core->dev, >> + "can't turn off TCM bank 0x%x", pm_domain_id); >> + } >> + >> + return 0; >> +} >> + >> +static const struct rproc_ops zynqmp_r5_rproc_ops = { >> + .prepare = zynqmp_r5_rproc_prepare, >> + .unprepare = zynqmp_r5_rproc_unprepare, >> + .start = zynqmp_r5_rproc_start, >> + .stop = zynqmp_r5_rproc_stop, >> + .load = rproc_elf_load_segments, >> + .parse_fw = zynqmp_r5_parse_fw, >> + .find_loaded_rsc_table = rproc_elf_find_loaded_rsc_table, >> + .sanity_check = rproc_elf_sanity_check, >> + .get_boot_addr = rproc_elf_get_boot_addr, >> +}; >> + >> +/** >> + * zynqmp_r5_add_rproc_core() >> + * Probes ZynqMP R5 processor device node >> + * this is called for each individual R5 core >> + * >> + * @cdev: Device node of each r5 core >> + * >> + * Return: zynqmp_r5_core object for success, error pointer in case of error. >> + */ >> +static struct zynqmp_r5_core *zynqmp_r5_add_rproc_core(struct device *cdev) >> +{ >> + int ret; >> + struct rproc *r5_rproc; >> + struct zynqmp_r5_core *r5_core; >> + >> + /* Set up DMA mask */ >> + ret = dma_set_coherent_mask(cdev, DMA_BIT_MASK(32)); >> + if (ret) >> + return ERR_PTR(ret); >> + >> + /* Allocate remoteproc instance */ >> + r5_rproc = devm_rproc_alloc(cdev, dev_name(cdev), >> + &zynqmp_r5_rproc_ops, >> + NULL, sizeof(struct zynqmp_r5_core)); >> + if (!r5_rproc) { >> + dev_err(cdev, "failed to allocate memory for rproc instance\n"); >> + return ERR_PTR(-ENOMEM); >> + } >> + >> + r5_rproc->auto_boot = false; >> + r5_core = (struct zynqmp_r5_core *)r5_rproc->priv; >> + r5_core->dev = cdev; >> + r5_core->np = dev_of_node(cdev); >> + if (!r5_core->np) { >> + dev_err(cdev, "can't get device node for r5 core\n"); >> + return ERR_PTR(-EINVAL); >> + } >> + >> + /* Add R5 remoteproc core */ >> + ret = rproc_add(r5_rproc); >> + if (ret) { >> + dev_err(cdev, "failed to add r5 remoteproc\n"); >> + return ERR_PTR(ret); >> + } >> + >> + r5_core->rproc = r5_rproc; >> + return r5_core; >> +} >> + >> +/** >> + * zynqmp_r5_get_tcm_node() >> + * Ideally this function should parse tcm node and store information >> + * in r5_core instance. We will use hardcoded TCM information from >> + * driver for now in this function. >> + * >> + * @cluster: pointer to zynqmp_r5_cluster type object >> + * >> + * Return: 0 for success and error code for failure. >> + */ >> +static int zynqmp_r5_get_tcm_node(struct zynqmp_r5_cluster *cluster) >> +{ >> + int tcm_bank_count, tcm_node; >> + int i, j; >> + struct zynqmp_r5_core *r5_core; >> + struct device *dev = cluster->dev; >> + >> + /* >> + * ToDo: Use predefined TCM address space values from driver until >> + * system-dt spec is not final for TCM >> + */ >> + tcm_bank_count = ARRAY_SIZE(zynqmp_tcm_banks); >> + >> + /* count per core tcm banks */ >> + tcm_bank_count = tcm_bank_count / cluster->core_count; >> + >> + /* >> + * r5 core 0 will use all of TCM banks in lockstep mode. >> + * In split mode, r5 core0 will use 128k and r5 core1 will use another >> + * 128k. Assign TCM banks to each core accordingly >> + */ >> + tcm_node = 0; >> + for (i = 0; i < cluster->core_count; i++) { >> + r5_core = cluster->r5_cores[i]; >> + r5_core->tcm_banks = devm_kcalloc(dev, tcm_bank_count, >> + sizeof(struct mem_bank_data *), >> + GFP_KERNEL); >> + if (!r5_core->tcm_banks) >> + return -ENOMEM; >> + >> + for (j = 0; j < tcm_bank_count; j++) { >> + /* >> + * Use pre-defined TCM reg values. >> + * Eventually this should be replaced by values >> + * parsed from dts. >> + */ >> + r5_core->tcm_banks[j] = >> + (struct mem_bank_data *)&zynqmp_tcm_banks[tcm_node]; >> + tcm_node++; >> + } >> + >> + r5_core->tcm_bank_count = tcm_bank_count; >> + } >> + >> + return 0; >> +} >> + >> +/** >> + * zynqmp_r5_get_mem_region_node() >> + * parse memory-region property from dt node and add >> + * memory region carveouts >> + * >> + * @r5_core: pointer to zynqmp_r5_core type object >> + * >> + * Return: 0 for success and error code for failure. >> + */ >> +static int zynqmp_r5_get_mem_region_node(struct zynqmp_r5_core *r5_core) >> +{ >> + int res_mem_count, i; >> + struct device *dev; >> + struct device_node *np, *rmem_np; >> + struct reserved_mem **rmem; >> + >> + dev = r5_core->dev; >> + >> + np = r5_core->np; >> + >> + res_mem_count = of_property_count_elems_of_size(np, "memory-region", >> + sizeof(phandle)); >> + if (res_mem_count <= 0) { >> + dev_warn(dev, "failed to get memory-region property %d\n", >> + res_mem_count); >> + return -EINVAL; >> + } >> + >> + rmem = devm_kcalloc(dev, res_mem_count, >> + sizeof(struct reserved_mem *), GFP_KERNEL); >> + if (!rmem) >> + return -ENOMEM; >> + >> + for (i = 0; i < res_mem_count; i++) { >> + rmem_np = of_parse_phandle(np, "memory-region", i); >> + if (!rmem_np) >> + goto release_rmem; >> + >> + rmem[i] = of_reserved_mem_lookup(rmem_np); >> + if (!rmem[i]) { >> + of_node_put(rmem_np); >> + goto release_rmem; >> + } >> + >> + of_node_put(rmem_np); >> + } >> + >> + r5_core->rmem_count = res_mem_count; >> + r5_core->rmem = rmem; >> + return 0; >> + >> +release_rmem: >> + for (i--; i > -1; i--) >> + kfree(rmem[i]); >> + devm_kfree(dev, rmem); >> + return -ENOMEM; >> +} >> + >> +/* >> + * zynqmp_r5_core_init() >> + * Create and initialize zynqmp_r5_core type object >> + * >> + * @cluster: pointer to zynqmp_r5_cluster type object >> + * >> + * Return: 0 for success and error code for failure. >> + */ >> +static int zynqmp_r5_core_init(struct zynqmp_r5_cluster *cluster, >> + enum rpu_oper_mode fw_reg_val, int tcm_mode) >> +{ >> + int ret, i; >> + struct zynqmp_r5_core *r5_core; >> + struct device *dev = cluster->dev; >> + >> + ret = zynqmp_r5_get_tcm_node(cluster); >> + if (ret < 0) { >> + dev_err(dev, "can't get tcm node, err %d\n", ret); >> + return ret; >> + } >> + >> + for (i = 0; i < cluster->core_count; i++) { >> + r5_core = cluster->r5_cores[i]; >> + >> + ret = zynqmp_r5_get_mem_region_node(r5_core); >> + if (ret) >> + dev_warn(dev, "memory-region prop failed %d\n", ret); >> + >> + /* Initialize r5 cores with power-domains parsed from dts */ >> + ret = of_property_read_u32_index(r5_core->np, "power-domains", >> + 1, &r5_core->pm_domain_id); >> + if (ret) { >> + dev_err(dev, "failed to get power-domains property\n"); >> + return ret; >> + } >> + >> + ret = zynqmp_r5_set_mode(r5_core, fw_reg_val, tcm_mode); >> + if (ret) { >> + dev_err(dev, "failed to set r5 cluster mode %d, err %d\n", >> + cluster->mode, ret); >> + return ret; >> + } >> + } >> + >> + return 0; >> +} >> + >> +/* >> + * zynqmp_r5_cluster_init() >> + * Create and initialize zynqmp_r5_cluster type object >> + * >> + * @cluster: pointer to zynqmp_r5_cluster type object >> + * >> + * Return: 0 for success and error code for failure. >> + */ >> +static int zynqmp_r5_cluster_init(struct zynqmp_r5_cluster *cluster) >> +{ >> + struct device *dev = cluster->dev; >> + struct device_node *dev_node = dev_of_node(dev); >> + struct device_node *child; >> + struct platform_device *child_pdev; >> + int core_count, ret, i; >> + enum zynqmp_r5_cluster_mode cluster_mode = LOCKSTEP_MODE; >> + struct zynqmp_r5_core **r5_cores; >> + enum rpu_tcm_comb tcm_mode; >> + enum rpu_oper_mode fw_reg_val; >> + >> + ret = of_property_read_u32(dev_node, "xlnx,cluster-mode", &cluster_mode); >> + >> + /* >> + * on success returns 0, if not defined then returns -EINVAL, >> + * In that case, default is LOCKSTEP mode >> + */ >> + if (ret != -EINVAL && ret != 0) { >> + dev_err(dev, "Invalid xlnx,cluster-mode property\n"); >> + return -EINVAL; >> + } >> + >> + /* >> + * For now driver only supports split mode and lockstep mode. >> + * fail driver probe if either of that is not set in dts. >> + */ >> + if (cluster_mode == LOCKSTEP_MODE) { >> + tcm_mode = PM_RPU_TCM_COMB; >> + fw_reg_val = PM_RPU_MODE_LOCKSTEP; >> + } else if (cluster_mode == SPLIT_MODE) { >> + tcm_mode = PM_RPU_TCM_SPLIT; >> + fw_reg_val = PM_RPU_MODE_SPLIT; >> + } else { >> + dev_err(dev, "driver does not support cluster mode %d\n", cluster_mode); >> + return -EINVAL; >> + } >> + >> + /* >> + * Number of cores is decided by number of child nodes of >> + * r5f subsystem node in dts. If Split mode is used in dts >> + * 2 child nodes are expected. >> + * In lockstep mode if two child nodes are available, >> + * only use first child node and consider it as core0 >> + * and ignore core1 dt node. >> + */ >> + core_count = of_get_available_child_count(dev_node); >> + if (core_count <= 0) { >> + dev_err(dev, "Invalid number of r5 cores %d", core_count); >> + return -EINVAL; >> + } else if (cluster_mode == SPLIT_MODE && core_count != 2) { >> + dev_err(dev, "Invalid number of r5 cores for split mode\n"); >> + return -EINVAL; >> + } else if (cluster_mode == LOCKSTEP_MODE && core_count == 2) { >> + dev_warn(dev, "Only r5 core0 will be used\n"); >> + core_count = 1; >> + } >> + >> + r5_cores = kcalloc(core_count, >> + sizeof(struct zynqmp_r5_core *), GFP_KERNEL); >> + if (!r5_cores) >> + return -ENOMEM; >> + >> + i = 0; >> + for_each_available_child_of_node(dev_node, child) { >> + child_pdev = of_find_device_by_node(child); >> + if (!child_pdev) { >> + of_node_put(child); >> + ret = -ENODEV; >> + goto release_r5_cores; >> + } >> + >> + /* create and add remoteproc instance of type struct rproc */ >> + r5_cores[i] = zynqmp_r5_add_rproc_core(&child_pdev->dev); >> + if (IS_ERR(r5_cores[i])) { >> + of_node_put(child); >> + ret = PTR_ERR(r5_cores[i]); >> + goto release_r5_cores; >> + } >> + >> + i++; >> + >> + /* >> + * If two child nodes are available in dts in lockstep mode, >> + * then ignore second child node. >> + */ >> + if (i == core_count) { >> + of_node_put(child); >> + break; >> + } >> + } >> + >> + cluster->mode = cluster_mode; >> + cluster->core_count = core_count; >> + cluster->r5_cores = r5_cores; >> + >> + ret = zynqmp_r5_core_init(cluster, fw_reg_val, tcm_mode); >> + if (ret < 0) { >> + dev_err(dev, "failed to init r5 core err %d\n", ret); >> + cluster->core_count = 0; >> + cluster->r5_cores = NULL; >> + goto release_r5_cores; >> + } >> + >> + return 0; >> + >> +release_r5_cores: >> + for (i--; i > -1; i--) { >> + put_device(r5_cores[i]->dev); >> + rproc_del(r5_cores[i]->rproc); >> + } >> + kfree(r5_cores); >> + return ret; >> +} >> + >> +static void zynqmp_r5_core_exit(struct zynqmp_r5_core *r5_core) >> +{ >> + /* release r5_core device */ >> + put_device(r5_core->dev); >> + >> + rproc_del(r5_core->rproc); >> +} >> + >> +static void zynqmp_r5_cluster_exit(void *data) >> +{ >> + struct platform_device *pdev = (struct platform_device *)data; >> + struct zynqmp_r5_cluster *cluster; >> + int i; >> + >> + cluster = (struct zynqmp_r5_cluster *)platform_get_drvdata(pdev); >> + if (!cluster) >> + return; >> + >> + for (i = 0; i < cluster->core_count; i++) { >> + zynqmp_r5_core_exit(cluster->r5_cores[i]); >> + cluster->r5_cores[i] = NULL; >> + } >> + >> + kfree(cluster->r5_cores); >> + kfree(cluster); >> + platform_set_drvdata(pdev, NULL); >> +} >> + >> +/* >> + * zynqmp_r5_remoteproc_probe() >> + * >> + * @pdev: domain platform device for R5 cluster >> + * >> + * called when driver is probed, for each R5 core specified in DT, >> + * setup as needed to do remoteproc-related operations >> + * >> + * Return: 0 for success, negative value for failure. >> + */ >> +static int zynqmp_r5_remoteproc_probe(struct platform_device *pdev) >> +{ >> + int ret; >> + struct zynqmp_r5_cluster *cluster; >> + struct device *dev = &pdev->dev; >> + >> + cluster = kzalloc(sizeof(*cluster), GFP_KERNEL); >> + if (!cluster) >> + return -ENOMEM; >> + >> + cluster->dev = dev; >> + >> + ret = devm_of_platform_populate(dev); >> + if (ret) { >> + dev_err_probe(dev, ret, "failed to populate platform dev\n"); >> + kfree(cluster); >> + return ret; >> + } >> + >> + /* wire in so each core can be cleaned up at driver remove */ >> + platform_set_drvdata(pdev, cluster); >> + >> + ret = zynqmp_r5_cluster_init(cluster); >> + if (ret) { >> + zynqmp_r5_cluster_exit(pdev); >> + dev_err_probe(dev, ret, "Invalid r5f subsystem device tree\n"); >> + return ret; >> + } >> + >> + ret = devm_add_action_or_reset(dev, zynqmp_r5_cluster_exit, pdev); > I am still not convinced you need to use the devm_add_action_or_reset() API as > opposed to a simple platform_driver::remove function, but this is something we > can tackle in another revision. Ok for now, I will keep it as it is. > I am done reviewing this driver. > > Thanks, > Mathieu > >> + if (ret) >> + return ret; >> + >> + return 0; >> +} >> + >> +/* Match table for OF platform binding */ >> +static const struct of_device_id zynqmp_r5_remoteproc_match[] = { >> + { .compatible = "xlnx,zynqmp-r5fss", }, >> + { /* end of list */ }, >> +}; >> +MODULE_DEVICE_TABLE(of, zynqmp_r5_remoteproc_match); >> + >> +static struct platform_driver zynqmp_r5_remoteproc_driver = { >> + .probe = zynqmp_r5_remoteproc_probe, >> + .driver = { >> + .name = "zynqmp_r5_remoteproc", >> + .of_match_table = zynqmp_r5_remoteproc_match, >> + }, >> +}; >> +module_platform_driver(zynqmp_r5_remoteproc_driver); >> + >> +MODULE_DESCRIPTION("Xilinx R5F remote processor driver"); >> +MODULE_AUTHOR("Xilinx Inc."); >> +MODULE_LICENSE("GPL"); >> -- >> 2.25.1 >>

3 years, 2 months

1
0
0 0

Re: [PATCH v8 6/6] drivers: remoteproc: Add Xilinx r5 remoteproc driver

by Tanmay Shah

Hi Mathieu, Thanks for reviews. I will address all the comments in this email in next revision. Thanks, Tanmay On 6/7/22 9:52 AM, Mathieu Poirier wrote: > Hi Tanmay, > > On Thu, Jun 02, 2022 at 01:38:34PM -0700, Tanmay Shah wrote: >> This driver enables r5f dual core Real time Processing Unit subsystem >> available on Xilinx Zynq Ultrascale MPSoC Platform. RPU subsystem >> (cluster) can be configured in different modes e.g. split mode in which >> two r5f cores work independent of each other and lock-step mode in which >> both r5f cores execute same code clock-for-clock and notify if the >> result is different. >> >> The Xilinx r5 Remoteproc Driver boots the RPU cores via calls to the Xilinx >> Platform Management Unit that handles the R5 configuration, memory access >> and R5 lifecycle management. The interface to this manager is done in this >> driver via zynqmp_pm_* function calls. >> >> Signed-off-by: Ben Levinsky <ben.levinsky(a)xilinx.com> >> Signed-off-by: Tanmay Shah <tanmay.shah(a)xilinx.com> >> --- >> >> Changes in v8: >> - None >> >> Changes in v7: >> - None >> >> Changes in v6: >> - None >> >> Changes in v5: >> - None >> >> Changes in v4: >> - Remove redundant header files >> - use dev_err_probe() to report errors during probe >> - Fix missing check on error code returned by zynqmp_r5_add_rproc_core() >> - Fix memory leaks all over the driver when resource allocation fails for any core >> - make cluster mode check only at one place >> - remove redundant initialization of variable >> - remove redundant use of of_node_put() >> - Fix Comment format problem >> - Assign offset of zynqmp_tcm_banks instead of duplicating it >> - Add tcm and memory regions rproc carveouts during prepare instead of parse_fw >> - Remove rproc_mem_entry object from r5_core >> - Use put_device() and rproc_del() APIs to fix memory leaks >> - Replace pr_* with dev_*. This was missed in v3, fix now. >> - Use "GPL" instead of "GPL v2" in MODULE_LICENSE macro. This was suggested by checkpatch script. >> >> Changes in v3: >> - Fix checkpatch script indentation warning >> - Remove unused variable from xilinx remoteproc driver >> - use C style comments, i.e /*...*/ >> - Remove redundant debug information which can be derived using /proc/device-tree >> - Fix multilined comment format >> - s/"final fot TCM"/"final for TCM" >> - Function devm_kzalloc() does not return an code on error, just NULL. >> Remove redundant error check for this function throughout the driver. >> - Fix RPU mode configuration and add documentation accordingly >> - Get rid of the indentations to match function documentation style with rest of the driver >> - Fix memory leak by only using r5_rproc->priv and not replace it with new instance >> - Use 'i' for the outer loop and 'j' for the inner one as per convention >> - Remove redundant error and NULL checks throughout the driver >> - Use devm_kcalloc() when more than one element is required >> - Add memory-regions carveouts during driver probe instead of parse_fw call >> This removes redundant copy of reserved_mem object in r5_core structure. >> - Fix memory leak by using of_node_put() >> - Fix indentation of tcm_mem_map function args >> - Remove redundant init of variables >> - Initialize tcm bank size variable for lockstep mode >> - Replace u32 with phys_addr_t for variable stroing memory bank address >> - Add documentation of TCM behavior in lockstep mode >> - Use dev_get_drvdata instead of platform driver API >> - Remove info level messages >> - Fix checkpatch.pl warnings >> - Add documentation for the Xilinx r5f platform to understand driver design >> drivers/remoteproc/Kconfig | 12 + >> drivers/remoteproc/Makefile | 1 + >> drivers/remoteproc/xlnx_r5_remoteproc.c | 1045 +++++++++++++++++++++++ >> 3 files changed, 1058 insertions(+) >> create mode 100644 drivers/remoteproc/xlnx_r5_remoteproc.c >> >> diff --git a/drivers/remoteproc/Kconfig b/drivers/remoteproc/Kconfig >> index 166019786653..5637a71c0677 100644 >> --- a/drivers/remoteproc/Kconfig >> +++ b/drivers/remoteproc/Kconfig >> @@ -352,6 +352,18 @@ config TI_K3_R5_REMOTEPROC >> It's safe to say N here if you're not interested in utilizing >> a slave processor. >> >> +config XLNX_R5_REMOTEPROC >> + tristate "Xilinx R5 remoteproc support" >> + depends on PM && ARCH_ZYNQMP >> + depends on ZYNQMP_FIRMWARE >> + select RPMSG_VIRTIO >> + select ZYNQMP_IPI_MBOX >> + help >> + Say y or m here to support Xilinx R5 remote processors via the remote >> + processor framework. >> + >> + It's safe to say N if not interested in using RPU r5f cores. >> + >> endif # REMOTEPROC >> >> endmenu >> diff --git a/drivers/remoteproc/Makefile b/drivers/remoteproc/Makefile >> index 5478c7cb9e07..91314a9b43ce 100644 >> --- a/drivers/remoteproc/Makefile >> +++ b/drivers/remoteproc/Makefile >> @@ -38,3 +38,4 @@ obj-$(CONFIG_ST_SLIM_REMOTEPROC) += st_slim_rproc.o >> obj-$(CONFIG_STM32_RPROC) += stm32_rproc.o >> obj-$(CONFIG_TI_K3_DSP_REMOTEPROC) += ti_k3_dsp_remoteproc.o >> obj-$(CONFIG_TI_K3_R5_REMOTEPROC) += ti_k3_r5_remoteproc.o >> +obj-$(CONFIG_XLNX_R5_REMOTEPROC) += xlnx_r5_remoteproc.o >> diff --git a/drivers/remoteproc/xlnx_r5_remoteproc.c b/drivers/remoteproc/xlnx_r5_remoteproc.c >> new file mode 100644 >> index 000000000000..394b3469463c >> --- /dev/null >> +++ b/drivers/remoteproc/xlnx_r5_remoteproc.c >> @@ -0,0 +1,1045 @@ >> +// SPDX-License-Identifier: GPL-2.0 >> +/* >> + * ZynqMP R5 Remote Processor driver >> + * >> + */ >> + >> +#include <dt-bindings/power/xlnx-zynqmp-power.h> >> +#include <linux/dma-mapping.h> >> +#include <linux/firmware/xlnx-zynqmp.h> >> +#include <linux/kernel.h> >> +#include <linux/module.h> >> +#include <linux/of_address.h> >> +#include <linux/of_platform.h> >> +#include <linux/of_reserved_mem.h> >> +#include <linux/platform_device.h> >> +#include <linux/remoteproc.h> >> +#include <linux/slab.h> >> + >> +#include "remoteproc_internal.h" >> + >> +/* settings for RPU cluster mode */ >> +enum zynqmp_r5_cluster_mode { >> + SPLIT_MODE = 0, /* When cores run as separate processor */ >> + LOCKSTEP_MODE = 1, /* cores execute same code in lockstep,clk-for-clk */ >> + SINGLE_CPU_MODE = 2, /* core0 is held in reset and only core1 runs */ >> +}; >> + >> +/** >> + * struct mem_bank_data - Memory Bank description >> + * >> + * @addr: Start address of memory bank >> + * @size: Size of Memory bank >> + * @pm_domain_id: Power-domains id of memory bank for firmware to turn on/off >> + * @bank_name: name of the bank for remoteproc framework >> + */ >> +struct mem_bank_data { >> + phys_addr_t addr; >> + size_t size; >> + u32 pm_domain_id; >> + char *bank_name; >> +}; >> + >> +static const struct mem_bank_data zynqmp_tcm_banks[] = { >> + {0xffe00000UL, 0x10000UL, PD_R5_0_ATCM, "atcm0"}, /* TCM 64KB each */ >> + {0xffe20000UL, 0x10000UL, PD_R5_0_BTCM, "btcm0"}, >> + {0xffe90000UL, 0x10000UL, PD_R5_1_ATCM, "atcm1"}, >> + {0xffeb0000UL, 0x10000UL, PD_R5_1_BTCM, "btcm1"}, >> +}; >> + >> +/** >> + * struct zynqmp_r5_core - ZynqMP R5 core structure >> + * >> + * @dev: device of RPU instance >> + * @np: device node of RPU instance >> + * @tcm_bank_count: number TCM banks accessible to this RPU >> + * @tcm_banks: array of each TCM bank data >> + * @rmem_count: Number of reserved mem regions >> + * @rmem: reserved memory region nodes from device tree >> + * @rproc: rproc handle >> + * @pm_domain_id: RPU CPU power domain id >> + */ >> +struct zynqmp_r5_core { >> + struct device *dev; >> + struct device_node *np; >> + int tcm_bank_count; >> + struct mem_bank_data **tcm_banks; >> + int rmem_count; >> + struct reserved_mem **rmem; >> + struct rproc *rproc; >> + u32 pm_domain_id; >> +}; >> + >> +/** >> + * struct zynqmp_r5_cluster - ZynqMP R5 cluster structure >> + * >> + * @dev: r5f subsystem cluster device node >> + * @mode: cluster mode of type zynqmp_r5_cluster_mode >> + * @core_count: number of r5 cores used for this cluster mode >> + * @r5_cores: Array of pointers pointing to r5 core >> + */ >> +struct zynqmp_r5_cluster { >> + struct device *dev; >> + enum zynqmp_r5_cluster_mode mode; >> + int core_count; >> + struct zynqmp_r5_core **r5_cores; >> +}; >> + >> +/* >> + * zynqmp_r5_set_mode - set RPU operation mode >> + * >> + * set RPU operation mode >> + * >> + * Return: 0 for success, negative value for failure >> + */ >> +static int zynqmp_r5_set_mode(struct zynqmp_r5_core *r5_core, >> + enum rpu_oper_mode fw_reg_val, >> + enum rpu_tcm_comb tcm_mode) >> +{ >> + int ret; >> + >> + ret = zynqmp_pm_set_rpu_mode(r5_core->pm_domain_id, fw_reg_val); >> + if (ret < 0) { >> + dev_err(r5_core->dev, "failed to set RPU mode\n"); >> + return ret; >> + } >> + >> + ret = zynqmp_pm_set_tcm_config(r5_core->pm_domain_id, tcm_mode); >> + if (ret < 0) >> + dev_err(r5_core->dev, "failed to configure TCM\n"); >> + >> + return ret; >> +} >> + >> +/* >> + * zynqmp_r5_rproc_start >> + * @rproc: single R5 core's corresponding rproc instance >> + * >> + * Start R5 Core from designated boot address. >> + * >> + * return 0 on success, otherwise non-zero value on failure >> + */ >> +static int zynqmp_r5_rproc_start(struct rproc *rproc) >> +{ >> + struct zynqmp_r5_core *r5_core = rproc->priv; >> + enum rpu_boot_mem bootmem; >> + int ret; >> + >> + /* >> + * The exception vector pointers (EVP) refer to the base-address of >> + * exception vectors (for reset, IRQ, FIQ, etc). The reset-vector >> + * starts at the base-address and subsequent vectors are on 4-byte >> + * boundaries. >> + * >> + * Exception vectors can start either from 0x0000_0000 (LOVEC) or >> + * from 0xFFFF_0000 (HIVEC) which is mapped in the OCM (On-Chip Memory) >> + * >> + * Usually firmware will put Exception vectors at LOVEC. >> + * >> + * It is not recommend that you change the exception vector. >> + * Changing the EVP to HIVEC will result in increased interrupt latency >> + * and jitter. Also, if the OCM is secured and the Cortex-R5F processor >> + * is non-secured, then the Cortex-R5F processor cannot access the >> + * HIVEC exception vectors in the OCM. >> + */ >> + bootmem = (rproc->bootaddr >= 0xFFFC0000) ? >> + PM_RPU_BOOTMEM_HIVEC : PM_RPU_BOOTMEM_LOVEC; >> + >> + dev_dbg(r5_core->dev, "RPU boot addr 0x%llx from %s.", rproc->bootaddr, >> + bootmem == PM_RPU_BOOTMEM_HIVEC ? "OCM" : "TCM"); >> + >> + ret = zynqmp_pm_request_wake(r5_core->pm_domain_id, 1, >> + bootmem, ZYNQMP_PM_REQUEST_ACK_NO); >> + if (ret) >> + dev_err(r5_core->dev, >> + "failed to start RPU = 0x%x\n", r5_core->pm_domain_id); >> + return ret; >> +} >> + >> +/* >> + * zynqmp_r5_rproc_stop >> + * @rproc: single R5 core's corresponding rproc instance >> + * >> + * Power down R5 Core. >> + * >> + * return 0 on success, otherwise non-zero value on failure >> + */ >> +static int zynqmp_r5_rproc_stop(struct rproc *rproc) >> +{ >> + struct zynqmp_r5_core *r5_core = rproc->priv; >> + int ret; >> + >> + ret = zynqmp_pm_force_pwrdwn(r5_core->pm_domain_id, >> + ZYNQMP_PM_REQUEST_ACK_BLOCKING); >> + if (ret) >> + dev_err(r5_core->dev, "failed to stop remoteproc RPU %d\n", ret); >> + >> + return ret; >> +} >> + >> +/* >> + * zynqmp_r5_mem_region_map >> + * @rproc: single R5 core's corresponding rproc instance >> + * @mem: mem entry to map >> + * >> + * Callback to map va for memory-region's carveout. >> + * >> + * return 0 on success, otherwise non-zero value on failure >> + */ >> +static int zynqmp_r5_mem_region_map(struct rproc *rproc, >> + struct rproc_mem_entry *mem) >> +{ >> + void __iomem *va; >> + >> + va = ioremap_wc(mem->dma, mem->len); >> + if (IS_ERR_OR_NULL(va)) >> + return -ENOMEM; >> + >> + mem->va = (void *)va; >> + >> + return 0; >> +} >> + >> +/* >> + * zynqmp_r5_rproc_mem_unmap >> + * @rproc: single R5 core's corresponding rproc instance >> + * @mem: mem entry to unmap >> + * >> + * Unmap memory-region carveout >> + * >> + * return 0 on success, otherwise non-zero value on failure >> + */ >> +static int zynqmp_r5_mem_region_unmap(struct rproc *rproc, >> + struct rproc_mem_entry *mem) >> +{ >> + iounmap((void __iomem *)mem->va); >> + return 0; >> +} >> + >> +/* >> + * add_mem_regions >> + * @r5_core: single R5 core's corresponding zynqmp_r5_core type instance >> + * @rmem: reserved mem region parsed from dt node >> + * >> + * Construct rproc mem carveouts from carveout provided in >> + * memory-region property >> + * >> + * return 0 on success, otherwise non-zero value on failure >> + */ >> +static int add_mem_regions_carveout(struct rproc *rproc) >> +{ >> + struct zynqmp_r5_core *r5_core; >> + struct reserved_mem *rmem; >> + struct rproc_mem_entry **rproc_mem; >> + int i, num_mem_regions; >> + >> + r5_core = (struct zynqmp_r5_core *)rproc->priv; >> + num_mem_regions = r5_core->rmem_count; >> + >> + /* memory regions not defined */ >> + if (num_mem_regions < 1) >> + return 0; >> + >> + rproc_mem = kcalloc(num_mem_regions, >> + sizeof(struct rproc_mem_entry *), GFP_KERNEL); >> + if (!rproc_mem) >> + return -ENOMEM; >> + >> + for (i = 0; i < num_mem_regions; i++) { >> + rmem = r5_core->rmem[i]; >> + >> + /* Register associated reserved memory regions */ >> + rproc_mem[i] = rproc_mem_entry_init(&rproc->dev, NULL, >> + (dma_addr_t)rmem->base, >> + rmem->size, rmem->base, >> + zynqmp_r5_mem_region_map, >> + zynqmp_r5_mem_region_unmap, >> + rmem->name); >> + if (!rproc_mem[i]) >> + goto release_mem_regions; >> + >> + dev_dbg(&rproc->dev, "reserved mem carveout %s addr=%llx, size=0x%llx", >> + rmem->name, rmem->base, rmem->size); >> + } >> + >> + /* >> + * Add carveouts only if all rproc mem enties are >> + * successfully initialized >> + */ >> + for (i = 0; i < num_mem_regions; i++) >> + rproc_add_carveout(rproc, rproc_mem[i]); >> + >> + kfree(rproc_mem); >> + return 0; >> + >> +release_mem_regions: >> + for (i--; i > -1; i--) >> + kfree(rproc_mem[i]); >> + kfree(rproc_mem); >> + return -ENOMEM; >> +} >> + >> +/* >> + * zynqmp_r5_rproc_mem_unmap >> + * @rproc: single R5 core's corresponding rproc instance >> + * @mem: mem entry to unmap >> + * >> + * Unmap TCM banks when powering down R5 core. >> + * >> + * return 0 on success, otherwise non-zero value on failure >> + */ >> +static int tcm_mem_unmap(struct rproc *rproc, struct rproc_mem_entry *mem) >> +{ >> + iounmap((void __iomem *)mem->va); >> + >> + return 0; >> +} >> + >> +/* >> + * tcm_mem_map >> + * @rproc: single R5 core's corresponding rproc instance >> + * @mem: mem entry to initialize the va and da fields of >> + * >> + * Given TCM bank entry, this callback will set device address for R5 >> + * running on TCM and also setup virtual address for TCM bank >> + * remoteproc carveout. >> + * >> + * return 0 on success, otherwise non-zero value on failure >> + */ >> +static int tcm_mem_map(struct rproc *rproc, >> + struct rproc_mem_entry *mem) >> +{ >> + void __iomem *va; >> + >> + va = ioremap_wc(mem->dma, mem->len); >> + if (IS_ERR_OR_NULL(va)) >> + return -ENOMEM; >> + >> + /* Update memory entry va */ >> + mem->va = (void *)va; >> + >> + /* clear TCMs */ >> + memset_io(va, 0, mem->len); >> + >> + /* >> + * The R5s expect their TCM banks to be at address 0x0 and 0x2000, >> + * while on the Linux side they are at 0xffexxxxx. >> + * >> + * Zero out the high 12 bits of the address. This will give >> + * expected values for TCM Banks 0A and 0B (0x0 and 0x20000). >> + */ >> + mem->da &= 0x000fffff; >> + >> + /* >> + * TCM Banks 1A and 1B still have to be translated. >> + * >> + * Below handle these two banks' absolute addresses (0xffe90000 and >> + * 0xffeb0000) and convert to the expected relative addresses >> + * (0x0 and 0x20000). >> + */ >> + if (mem->da == 0x90000 || mem->da == 0xB0000) >> + mem->da -= 0x90000; >> + >> + /* if translated TCM bank address is not valid report error */ >> + if (mem->da != 0x0 && mem->da != 0x20000) { >> + dev_err(&rproc->dev, "invalid TCM address: %x\n", mem->da); >> + return -EINVAL; >> + } >> + return 0; >> +} >> + >> +static int add_tcm_carveout_split_mode(struct rproc *rproc) >> +{ >> + int i, num_banks, ret; >> + struct rproc_mem_entry **rproc_mem; >> + u32 pm_domain_id; >> + phys_addr_t bank_addr; >> + size_t bank_size; >> + char *bank_name; >> + struct device *dev; >> + struct zynqmp_r5_core *r5_core; >> + >> + r5_core = (struct zynqmp_r5_core *)rproc->priv; >> + >> + dev = r5_core->dev; >> + >> + /* go through zynqmp banks for r5 node */ >> + num_banks = r5_core->tcm_bank_count; >> + if (num_banks <= 0) { >> + dev_err(dev, "need to specify TCM banks\n"); >> + return -EINVAL; >> + } >> + >> + rproc_mem = kcalloc(num_banks, >> + sizeof(struct rproc_mem_entry *), GFP_KERNEL); >> + if (!rproc_mem) >> + return -ENOMEM; >> + >> + /* >> + * Power-on Each 64KB TCM, >> + * register its address space, map and unmap functions >> + * and add carveouts accordingly >> + */ >> + for (i = 0; i < num_banks; i++) { >> + bank_addr = r5_core->tcm_banks[i]->addr; >> + bank_name = r5_core->tcm_banks[i]->bank_name; >> + bank_size = r5_core->tcm_banks[i]->size; >> + pm_domain_id = r5_core->tcm_banks[i]->pm_domain_id; >> + >> + ret = zynqmp_pm_request_node(pm_domain_id, >> + ZYNQMP_PM_CAPABILITY_ACCESS, 0, >> + ZYNQMP_PM_REQUEST_ACK_BLOCKING); >> + if (ret < 0) { >> + dev_err(dev, "failed to turn on TCM 0x%x", pm_domain_id); >> + goto release_tcm_split; >> + } >> + >> + dev_dbg(dev, "TCM carveout split mode %s addr=%llx, size=0x%lx", >> + bank_name, bank_addr, bank_size); >> + >> + rproc_mem[i] = rproc_mem_entry_init(dev, NULL, bank_addr, >> + bank_size, bank_addr, >> + tcm_mem_map, tcm_mem_unmap, >> + bank_name); >> + if (!rproc_mem[i]) { >> + ret = -ENOMEM; >> + goto release_tcm_split; >> + } >> + } >> + >> + /* >> + * Add carveouts only if all rproc mem enties are >> + * successfully initialized >> + */ >> + for (i = 0; i < num_banks; i++) >> + rproc_add_carveout(rproc, rproc_mem[i]); >> + >> + kfree(rproc_mem); >> + return 0; >> + >> +release_tcm_split: >> + /* If failed, Turn off all TCM banks turned on before */ >> + for (i--; i > -1; i--) { >> + pm_domain_id = r5_core->tcm_banks[i]->pm_domain_id; >> + zynqmp_pm_release_node(pm_domain_id); >> + kfree(rproc_mem[i]); >> + } >> + kfree(rproc_mem); >> + return ret; >> +} >> + >> +static int add_tcm_carveout_lockstep_mode(struct rproc *rproc) >> +{ >> + int i, num_banks, ret; >> + struct rproc_mem_entry *rproc_mem; >> + u32 pm_domain_id; >> + phys_addr_t bank_addr; >> + size_t bank_size = 0; >> + char *bank_name; >> + struct device *dev; >> + struct zynqmp_r5_core *r5_core; >> + >> + r5_core = (struct zynqmp_r5_core *)rproc->priv; >> + dev = r5_core->dev; >> + >> + /* Go through zynqmp banks for r5 node */ >> + num_banks = r5_core->tcm_bank_count; >> + if (num_banks <= 0) { >> + dev_err(dev, "need to specify TCM banks\n"); >> + return -EINVAL; >> + } >> + >> + /* >> + * In lockstep mode, TCM is contiguous memory block >> + * However, each TCM block still needs to be enabled individually. >> + * So, Enable each TCM block individually, but add their size >> + * to create contiguous memory region. >> + */ >> + bank_addr = r5_core->tcm_banks[0]->addr; >> + bank_name = r5_core->tcm_banks[0]->bank_name; >> + >> + for (i = 0; i < num_banks; i++) { >> + bank_size += r5_core->tcm_banks[i]->size; >> + pm_domain_id = r5_core->tcm_banks[i]->pm_domain_id; >> + >> + /* Turn on each TCM bank individually */ >> + ret = zynqmp_pm_request_node(pm_domain_id, >> + ZYNQMP_PM_CAPABILITY_ACCESS, 0, >> + ZYNQMP_PM_REQUEST_ACK_BLOCKING); >> + if (ret < 0) { >> + dev_err(dev, "failed to turn on TCM 0x%x", pm_domain_id); >> + goto release_tcm_lockstep; >> + } >> + } >> + >> + dev_dbg(dev, "TCM add carveout lockstep mode %s addr=0x%llx, size=0x%lx", >> + bank_name, bank_addr, bank_size); >> + >> + /* Register TCM address range, TCM map and unmap functions */ >> + rproc_mem = rproc_mem_entry_init(dev, NULL, bank_addr, >> + bank_size, bank_addr, >> + tcm_mem_map, tcm_mem_unmap, >> + bank_name); >> + if (!rproc_mem) { >> + ret = -ENOMEM; >> + goto release_tcm_lockstep; >> + } >> + >> + /* If registration is success, add carveouts */ >> + rproc_add_carveout(rproc, rproc_mem); >> + >> + return 0; >> + >> +release_tcm_lockstep: >> + /* If failed, Turn off all TCM banks turned on before */ >> + for (i--; i > -1; i--) { >> + pm_domain_id = r5_core->tcm_banks[i]->pm_domain_id; >> + zynqmp_pm_release_node(pm_domain_id); >> + } >> + return ret; >> +} >> + >> +/* >> + * add_tcm_banks() >> + * @rproc: single R5 core's corresponding rproc instance >> + * >> + * Given R5 node in remoteproc instance >> + * allocate remoteproc carveout for TCM memory >> + * needed for firmware to be loaded >> + * >> + * return 0 on success, otherwise non-zero value on failure >> + */ >> +static int add_tcm_banks(struct rproc *rproc) >> +{ >> + struct device *dev; >> + struct zynqmp_r5_cluster *cluster; >> + struct zynqmp_r5_core *r5_core; >> + >> + r5_core = (struct zynqmp_r5_core *)rproc->priv; >> + if (!r5_core) >> + return -EINVAL; >> + >> + dev = r5_core->dev; >> + >> + cluster = dev_get_drvdata(dev->parent); >> + if (!cluster) { >> + dev_err(dev->parent, "Invalid driver data\n"); >> + return -EINVAL; >> + } >> + >> + /* >> + * In lockstep mode TCM banks are one contiguous memory region of 256Kb >> + * In split mode, each TCM bank is 64Kb and not contiguous. >> + * We add memory carveouts accordingly. >> + */ >> + if (cluster->mode == SPLIT_MODE) >> + return add_tcm_carveout_split_mode(rproc); >> + else if (cluster->mode == LOCKSTEP_MODE) >> + return add_tcm_carveout_lockstep_mode(rproc); >> + >> + dev_err(cluster->dev, "invalid cluster mode\n"); >> + return -EINVAL; >> +} >> + >> +/* >> + * zynqmp_r5_parse_fw() >> + * @rproc: single R5 core's corresponding rproc instance >> + * @fw: ptr to firmware to be loaded onto r5 core >> + * >> + * When loading firmware, ensure the necessary carveouts are in remoteproc >> + * >> + * return 0 on success, otherwise non-zero value on failure >> + */ >> +static int zynqmp_r5_parse_fw(struct rproc *rproc, const struct firmware *fw) >> +{ >> + int ret; >> + >> + ret = rproc_elf_load_rsc_table(rproc, fw); >> + if (ret == -EINVAL) { >> + /* >> + * resource table only required for IPC. >> + * if not present, this is not necessarily an error; >> + * for example, loading r5 hello world application >> + * so simply inform user and keep going. >> + */ >> + dev_info(&rproc->dev, "no resource table found.\n"); >> + ret = 0; >> + } >> + return ret; >> +} >> + >> +static int zynqmp_r5_rproc_prepare(struct rproc *rproc) >> +{ >> + int ret = 0; >> + >> + ret = add_tcm_banks(rproc); >> + if (ret) { >> + dev_err(&rproc->dev, "failed to get TCM banks, err %d\n", ret); >> + return ret; >> + } >> + >> + ret = add_mem_regions_carveout(rproc); >> + if (ret) >> + dev_warn(&rproc->dev, "failed to get reserve mem regions %d\n", >> + ret); >> + >> + return 0; >> +} >> + >> +static int zynqmp_r5_rproc_unprepare(struct rproc *rproc) >> +{ >> + struct zynqmp_r5_core *r5_core; >> + int i; >> + u32 pm_domain_id; >> + >> + r5_core = (struct zynqmp_r5_core *)rproc->priv; >> + >> + for (i = 0; i < r5_core->tcm_bank_count; i++) { >> + pm_domain_id = r5_core->tcm_banks[i]->pm_domain_id; >> + if (zynqmp_pm_release_node(pm_domain_id)) >> + dev_warn(r5_core->dev, >> + "can't turn off TCM bank 0x%x", pm_domain_id); >> + } >> + >> + return 0; >> +} >> + >> +static const struct rproc_ops zynqmp_r5_rproc_ops = { >> + .prepare = zynqmp_r5_rproc_prepare, >> + .unprepare = zynqmp_r5_rproc_unprepare, >> + .start = zynqmp_r5_rproc_start, >> + .stop = zynqmp_r5_rproc_stop, >> + .load = rproc_elf_load_segments, >> + .parse_fw = zynqmp_r5_parse_fw, >> + .find_loaded_rsc_table = rproc_elf_find_loaded_rsc_table, >> + .sanity_check = rproc_elf_sanity_check, >> + .get_boot_addr = rproc_elf_get_boot_addr, >> +}; >> + >> +/** >> + * zynqmp_r5_add_rproc_core() >> + * Probes ZynqMP R5 processor device node >> + * this is called for each individual R5 core >> + * >> + * @cdev: Device node of each r5 core >> + * >> + * Return: zynqmp_r5_core object for success, error pointer in case of error. >> + */ >> +static struct zynqmp_r5_core *zynqmp_r5_add_rproc_core(struct device *cdev) >> +{ >> + int ret; >> + struct rproc *r5_rproc; >> + struct zynqmp_r5_core *r5_core; >> + >> + /* Set up DMA mask */ >> + ret = dma_set_coherent_mask(cdev, DMA_BIT_MASK(32)); >> + if (ret) >> + return ERR_PTR(ret); >> + >> + /* Allocate remoteproc instance */ >> + r5_rproc = devm_rproc_alloc(cdev, dev_name(cdev), >> + &zynqmp_r5_rproc_ops, >> + NULL, sizeof(struct zynqmp_r5_core)); > Not sure why devm_rproc_alloc() is used when rproc_add() is used below and > rproc_del() in zynqmp_r5_core_exit(). > >> + if (!r5_rproc) { >> + dev_err(cdev, "failed to allocate memory for rproc instance\n"); >> + return ERR_PTR(-ENOMEM); >> + } >> + >> + r5_rproc->auto_boot = false; >> + r5_core = (struct zynqmp_r5_core *)r5_rproc->priv; >> + r5_core->dev = cdev; >> + r5_core->np = dev_of_node(cdev); >> + if (!r5_core->np) { >> + dev_err(cdev, "can't get device node for r5 core\n"); >> + return ERR_PTR(-EINVAL); >> + } >> + >> + /* Add R5 remoteproc core */ >> + ret = rproc_add(r5_rproc); >> + if (ret) { >> + dev_err(cdev, "failed to add r5 remoteproc\n"); >> + return ERR_PTR(ret); >> + } >> + >> + r5_core->rproc = r5_rproc; >> + return r5_core; >> +} >> + >> +/** >> + * zynqmp_r5_get_tcm_node() >> + * Ideally this function should parse tcm node and store information >> + * in r5_core instance. We will use hardcoded TCM information from >> + * driver for now in this function. >> + * >> + * @cluster: pointer to zynqmp_r5_cluster type object >> + * >> + * Return: 0 for success and error code for failure. >> + */ >> +static int zynqmp_r5_get_tcm_node(struct zynqmp_r5_cluster *cluster) >> +{ >> + int tcm_bank_count, tcm_node; >> + int i, j; >> + struct zynqmp_r5_core *r5_core; >> + struct device *dev = cluster->dev; > struct device *dev = cluster->dev; > struct zynqmp_r5_core *r5_core; > int tcm_bank_count, tcm_node; > int i, j; > >> + >> + /* >> + * ToDo: Use predefined TCM address space values from driver until > s/until/while > >> + * system-dt spec is not final for TCM >> + */ >> + tcm_bank_count = ARRAY_SIZE(zynqmp_tcm_banks); >> + >> + /* count per core tcm banks */ >> + tcm_bank_count = tcm_bank_count / cluster->core_count; >> + >> + /* >> + * r5 core 0 will use all of TCM banks in lockstep mode. >> + * In split mode, r5 core0 will use 128k and r5 core1 will use another >> + * 128k. Assign TCM banks to each core accordingly >> + */ >> + tcm_node = 0; >> + for (i = 0; i < cluster->core_count; i++) { >> + r5_core = cluster->r5_cores[i]; >> + r5_core->tcm_banks = devm_kcalloc(dev, tcm_bank_count, >> + sizeof(struct mem_bank_data *), >> + GFP_KERNEL); >> + if (!r5_core->tcm_banks) >> + return -ENOMEM; >> + >> + for (j = 0; j < tcm_bank_count; j++) { >> + /* >> + * Use pre-defined TCM reg values. >> + * Eventually this should be replaced by values >> + * parsed from dts. >> + */ >> + r5_core->tcm_banks[j] = >> + (struct mem_bank_data *)&zynqmp_tcm_banks[tcm_node]; >> + tcm_node++; >> + } >> + >> + r5_core->tcm_bank_count = tcm_bank_count; >> + } >> + >> + return 0; >> +} >> + >> +/** >> + * zynqmp_r5_get_mem_region_node() >> + * parse memory-region property from dt node and add >> + * memory region carveouts >> + * >> + * @r5_core: pointer to zynqmp_r5_core type object >> + * >> + * Return: 0 for success and error code for failure. >> + */ >> +static int zynqmp_r5_get_mem_region_node(struct zynqmp_r5_core *r5_core) >> +{ >> + int res_mem_count, i; >> + struct device *dev; >> + struct device_node *np, *rmem_np; >> + struct reserved_mem **rmem; >> + >> + dev = r5_core->dev; >> + > Extra newline. > >> + np = r5_core->np; >> + >> + res_mem_count = of_property_count_elems_of_size(np, "memory-region", >> + sizeof(phandle)); >> + if (res_mem_count <= 0) { >> + dev_warn(dev, "failed to get memory-region property %d\n", >> + res_mem_count); >> + return -EINVAL; >> + } >> + >> + rmem = devm_kcalloc(dev, res_mem_count, >> + sizeof(struct reserved_mem *), GFP_KERNEL); >> + if (!rmem) >> + return -ENOMEM; >> + >> + for (i = 0; i < res_mem_count; i++) { >> + rmem_np = of_parse_phandle(np, "memory-region", i); >> + if (!rmem_np) >> + goto release_rmem; >> + >> + rmem[i] = of_reserved_mem_lookup(rmem_np); >> + if (!rmem[i]) { >> + of_node_put(rmem_np); >> + goto release_rmem; >> + } >> + >> + of_node_put(rmem_np); >> + } >> + >> + r5_core->rmem_count = res_mem_count; >> + r5_core->rmem = rmem; >> + return 0; >> + >> +release_rmem: >> + for (i--; i > -1; i--) > s/"i > -1"/"i >= 0" > >> + kfree(rmem[i]); >> + devm_kfree(dev, rmem); >> + return -ENOMEM; >> +} >> + >> +/* >> + * zynqmp_r5_core_init() >> + * Create and initialize zynqmp_r5_core type object >> + * >> + * @cluster: pointer to zynqmp_r5_cluster type object >> + * >> + * Return: 0 for success and error code for failure. >> + */ >> +static int zynqmp_r5_core_init(struct zynqmp_r5_cluster *cluster, >> + enum rpu_oper_mode fw_reg_val, int tcm_mode) > I would have expected @tcm_mode to be of type "enum rpu_tcm_comp". > >> +{ >> + int ret, i; >> + struct zynqmp_r5_core *r5_core; >> + struct device *dev = cluster->dev; >> + >> + ret = zynqmp_r5_get_tcm_node(cluster); >> + if (ret < 0) { >> + dev_err(dev, "can't get tcm node, err %d\n", ret); >> + return ret; >> + } >> + >> + for (i = 0; i < cluster->core_count; i++) { >> + r5_core = cluster->r5_cores[i]; >> + >> + ret = zynqmp_r5_get_mem_region_node(r5_core); >> + if (ret) >> + dev_warn(dev, "memory-region prop failed %d\n", ret); >> + >> + /* Initialize r5 cores with power-domains parsed from dts */ >> + ret = of_property_read_u32_index(r5_core->np, "power-domains", >> + 1, &r5_core->pm_domain_id); >> + if (ret) { >> + dev_err(dev, "failed to get power-domains property\n"); >> + return ret; >> + } >> + >> + ret = zynqmp_r5_set_mode(r5_core, fw_reg_val, tcm_mode); >> + if (ret) { >> + dev_err(dev, "failed to set r5 cluster mode %d, err %d\n", >> + cluster->mode, ret); >> + return ret; >> + } >> + } >> + >> + return 0; >> +} >> + >> +/* >> + * zynqmp_r5_cluster_init() >> + * Create and initialize zynqmp_r5_cluster type object >> + * >> + * @cluster: pointer to zynqmp_r5_cluster type object >> + * >> + * Return: 0 for success and error code for failure. >> + */ >> +static int zynqmp_r5_cluster_init(struct zynqmp_r5_cluster *cluster) >> +{ >> + struct device *dev = cluster->dev; >> + struct device_node *dev_node = dev_of_node(dev); >> + struct device_node *child; >> + struct platform_device *child_pdev; >> + int core_count, ret, i; >> + enum zynqmp_r5_cluster_mode cluster_mode = LOCKSTEP_MODE; >> + struct zynqmp_r5_core **r5_cores; >> + enum rpu_tcm_comb tcm_mode; >> + enum rpu_oper_mode fw_reg_val; >> + >> + ret = of_property_read_u32(dev_node, "xlnx,cluster-mode", &cluster_mode); >> + >> + /* >> + * on success returns 0, if not defined then returns -EINVAL, >> + * In that case, default is LOCKSTEP mode >> + */ >> + if (ret != -EINVAL && ret != 0) { >> + dev_err(dev, "Invalid xlnx,cluster-mode property\n"); >> + return -EINVAL; >> + } >> + >> + /* >> + * For now driver only supports split mode and lockstep mode. >> + * fail driver probe if either of that is not set in dts. >> + */ >> + if (cluster_mode == LOCKSTEP_MODE) { >> + tcm_mode = PM_RPU_TCM_COMB; >> + fw_reg_val = PM_RPU_MODE_LOCKSTEP; >> + } else if (cluster_mode == SPLIT_MODE) { >> + tcm_mode = PM_RPU_TCM_SPLIT; >> + fw_reg_val = PM_RPU_MODE_SPLIT; >> + } else { >> + dev_err(dev, "driver does not support cluster mode %d\n", cluster_mode); >> + return -EINVAL; >> + } >> + >> + /* >> + * Number of cores is decided by number of child nodes of >> + * r5f subsystem node in dts. If Split mode is used in dts >> + * 2 child nodes are expected. >> + * In lockstep mode if two child nodes are available, >> + * only use first child node and consider it as core0 >> + * and ignore core1 dt node. >> + */ >> + core_count = of_get_available_child_count(dev_node); >> + if (core_count <= 0) { >> + dev_err(dev, "Invalid number of r5 cores %d", core_count); >> + return -EINVAL; >> + } else if (cluster_mode == SPLIT_MODE && core_count != 2) { >> + dev_err(dev, "Invalid number of r5 cores for split mode\n"); >> + return -EINVAL; >> + } else if (cluster_mode == LOCKSTEP_MODE && core_count == 2) { >> + dev_warn(dev, "Only r5 core0 will be used\n"); >> + core_count = 1; >> + } >> + >> + r5_cores = kcalloc(core_count, >> + sizeof(struct zynqmp_r5_core *), GFP_KERNEL); >> + if (!r5_cores) >> + return -ENOMEM; >> + >> + i = 0; >> + for_each_available_child_of_node(dev_node, child) { >> + child_pdev = of_find_device_by_node(child); > Here a reference to child_pdev->dev is taken... > >> + if (!child_pdev) { >> + of_node_put(child); >> + ret = -ENODEV; >> + goto release_r5_cores; >> + } >> + >> + /* create and add remoteproc instance of type struct rproc */ >> + r5_cores[i] = zynqmp_r5_add_rproc_core(&child_pdev->dev); >> + if (IS_ERR(r5_cores[i])) { >> + of_node_put(child); > If devm_rproc_alloc() fails in zynqmp_r5_add_rproc_core() the reference to > child_pdev-dev taken above is not released. > > >> + ret = PTR_ERR(r5_cores[i]); >> + goto release_r5_cores; >> + } >> + >> + i++; >> + >> + /* >> + * If two child nodes are available in dts in lockstep mode, >> + * then ignore second child node. >> + */ >> + if (i == core_count) { >> + of_node_put(child); >> + break; >> + } >> + } >> + >> + cluster->mode = cluster_mode; >> + cluster->core_count = core_count; >> + cluster->r5_cores = r5_cores; >> + >> + ret = zynqmp_r5_core_init(cluster, fw_reg_val, tcm_mode); >> + if (ret < 0) { >> + dev_err(dev, "failed to init r5 core err %d\n", ret); >> + cluster->core_count = 0; >> + cluster->r5_cores = NULL; >> + goto release_r5_cores; >> + } >> + >> + return 0; >> + >> +release_r5_cores: >> + for (i--; i > -1; i--) { > s/"i > -1"/"i >= 0" > >> + put_device(r5_cores[i]->dev); >> + rproc_del(r5_cores[i]->rproc); >> + } >> + kfree(r5_cores); >> + return ret; >> +} >> + >> +static void zynqmp_r5_core_exit(struct zynqmp_r5_core *r5_core) >> +{ >> + /* release r5_core device */ >> + put_device(r5_core->dev); >> + >> + rproc_del(r5_core->rproc); >> +} >> + >> +static void zynqmp_r5_cluster_exit(void *data) >> +{ >> + struct platform_device *pdev = (struct platform_device *)data; >> + struct zynqmp_r5_cluster *cluster; >> + int i; >> + >> + cluster = (struct zynqmp_r5_cluster *)platform_get_drvdata(pdev); >> + if (!cluster) >> + return; >> + >> + for (i = 0; i < cluster->core_count; i++) { >> + zynqmp_r5_core_exit(cluster->r5_cores[i]); > I don't see a need for this extra function. Just call put_device() and > rproc_del() here. > >> + cluster->r5_cores[i] = NULL; > Not sure this is useful since both cluster->r5_cores and cluster are freed > below. > > I will review zynqmp_r5_rproc_ops and up tomorrow and Thursday. > > Thanks, > Mathieu > >> + } >> + >> + kfree(cluster->r5_cores); >> + kfree(cluster); >> + platform_set_drvdata(pdev, NULL); >> +} >> + >> +/* >> + * zynqmp_r5_remoteproc_probe() >> + * >> + * @pdev: domain platform device for R5 cluster >> + * >> + * called when driver is probed, for each R5 core specified in DT, >> + * setup as needed to do remoteproc-related operations >> + * >> + * Return: 0 for success, negative value for failure. >> + */ >> +static int zynqmp_r5_remoteproc_probe(struct platform_device *pdev) >> +{ >> + int ret; >> + struct zynqmp_r5_cluster *cluster; >> + struct device *dev = &pdev->dev; >> + >> + cluster = kzalloc(sizeof(*cluster), GFP_KERNEL); >> + if (!cluster) >> + return -ENOMEM; >> + >> + cluster->dev = dev; >> + >> + ret = devm_of_platform_populate(dev); >> + if (ret) { >> + dev_err_probe(dev, ret, "failed to populate platform dev\n"); >> + kfree(cluster); >> + return ret; >> + } >> + >> + /* wire in so each core can be cleaned up at driver remove */ >> + platform_set_drvdata(pdev, cluster); >> + >> + ret = zynqmp_r5_cluster_init(cluster); >> + if (ret) { >> + zynqmp_r5_cluster_exit(pdev); >> + dev_err_probe(dev, ret, "Invalid r5f subsystem device tree\n"); >> + return ret; >> + } >> + >> + ret = devm_add_action_or_reset(dev, zynqmp_r5_cluster_exit, pdev); >> + if (ret) >> + return ret; >> + >> + return 0; >> +} >> + >> +/* Match table for OF platform binding */ >> +static const struct of_device_id zynqmp_r5_remoteproc_match[] = { >> + { .compatible = "xlnx,zynqmp-r5fss", }, >> + { /* end of list */ }, >> +}; >> +MODULE_DEVICE_TABLE(of, zynqmp_r5_remoteproc_match); >> + >> +static struct platform_driver zynqmp_r5_remoteproc_driver = { >> + .probe = zynqmp_r5_remoteproc_probe, >> + .driver = { >> + .name = "zynqmp_r5_remoteproc", >> + .of_match_table = zynqmp_r5_remoteproc_match, >> + }, >> +}; >> +module_platform_driver(zynqmp_r5_remoteproc_driver); >> + >> +MODULE_DESCRIPTION("Xilinx R5F remote processor driver"); >> +MODULE_AUTHOR("Xilinx Inc."); >> +MODULE_LICENSE("GPL"); >> -- >> 2.25.1 >>

3 years, 2 months

1
0
0 0

Re: [PATCH v8 1/6] dt-bindings: remoteproc: Add Xilinx RPU subsystem bindings

by Tanmay Shah

On 6/9/22 10:33 AM, Rob Herring wrote: > On Tue, Jun 07, 2022 at 10:32:48AM -0700, Tanmay Shah wrote: >> Hi Rob, >> >> Ping for reviews. > No need to ping me. You can check status of your patch here: > > https://patchwork.ozlabs.org/project/devicetree-bindings/list/ > > If it is listed, Krzysztof or I will get to it. Thanks for the link. This is good to know. > Rob

3 years, 2 months

1
0
0 0

Re: [PATCH v8 1/6] dt-bindings: remoteproc: Add Xilinx RPU subsystem bindings

by Tanmay Shah

HI Rob, Thanks for reviews. Please find my comments below. On 6/9/22 10:41 AM, Rob Herring wrote: > On Fri, Jun 03, 2022 at 08:33:13AM +0200, Peter Korsgaard wrote: >>>>>>> "Tanmay" == Tanmay Shah <tanmay.shah(a)xilinx.com> writes: >> Hi, >> >> > Xilinx ZynqMP platform has dual-core ARM Cortex R5 Realtime Processing >> > Unit(RPU) subsystem. This patch adds dt-bindings for RPU subsystem >> > (cluster). >> >> > Signed-off-by: Tanmay Shah <tanmay.shah(a)xilinx.com> >> > --- >> >> > Changes in v8: >> > - Add 'items:' for sram property >> >> > Changes in v7: >> > - Add minItems in sram property >> >> > Changes in v6: >> > - Add maxItems to sram and memory-region property >> >> > Changes in v5: >> > - Add constraints of the possible values of xlnx,cluster-mode property >> > - fix description of power-domains property for r5 core >> > - Remove reg, address-cells and size-cells properties as it is not required >> > - Fix description of mboxes property >> > - Add description of each memory-region and remove old .txt binding link >> > reference in the description >> >> > Changes in v4: >> > - Add memory-region, mboxes and mbox-names properties in example >> >> > Changes in v3: >> > - None >> >> > .../bindings/remoteproc/xlnx,r5f-rproc.yaml | 130 ++++++++++++++++++ >> > include/dt-bindings/power/xlnx-zynqmp-power.h | 6 + >> > 2 files changed, 138 insertions(+) >> > create mode 100644 Documentation/devicetree/bindings/remoteproc/xlnx,r5f-rproc.yaml >> >> > diff --git a/Documentation/devicetree/bindings/remoteproc/xlnx,r5f-rproc.yaml b/Documentation/devicetree/bindings/remoteproc/xlnx,r5f-rproc.yaml >> > new file mode 100644 >> > index 000000000000..adfe05ff157a >> > --- /dev/null >> > +++ b/Documentation/devicetree/bindings/remoteproc/xlnx,r5f-rproc.yaml >> > @@ -0,0 +1,132 @@ >> > +# SPDX-License-Identifier: (GPL-2.0-only or BSD-2-Clause) >> > +%YAML 1.2 >> > +--- >> > +$id: http://devicetree.org/schemas/remoteproc/xlnx,r5f-rproc.yaml# >> > +$schema: http://devicetree.org/meta-schemas/core.yaml# >> > + >> > +title: Xilinx R5F processor subsystem >> > + >> > +maintainers: >> > + - Ben Levinsky <ben.levinsky(a)xilinx.com> >> > + - Tanmay Shah <tanmay.shah(a)xilinx.com> >> > + >> > +description: | >> > + The Xilinx platforms include a pair of Cortex-R5F processors (RPU) for >> > + real-time processing based on the Cortex-R5F processor core from ARM. >> > + The Cortex-R5F processor implements the Arm v7-R architecture and includes a >> > + floating-point unit that implements the Arm VFPv3 instruction set. >> > + >> > +properties: >> > + compatible: >> > + const: xlnx,zynqmp-r5fss >> > + >> > + xlnx,cluster-mode: >> > + $ref: /schemas/types.yaml#/definitions/uint32 >> > + enum: [0, 1, 2] >> >> A textual mode ("dual", "lock-step", "single") would be more readable. >> >> >> > + description: | >> > + The RPU MPCore can operate in split mode(Dual-processor performance), Safety >> >> space missing before "(Dual-processor" >> >> >> > + lock-step mode(Both RPU cores execute the same code in lock-step, >> > + clock-for-clock) or Single CPU mode (RPU core 0 can be held in reset while >> >> "can be" sounds a bit weak, perhaps "is" >> >> >> > + core 1 runs normally). The processor does not support dynamic configuration. >> > + Switching between modes is only permitted immediately after a processor reset. >> > + If set to 1 then lockstep mode and if 0 then split mode. >> > + If set to 2 then single CPU mode. When not defined, default will be lockstep mode. >> >> This looks a bit confusing. If you decide to stick to the numerical >> modes, then at least list them in numerical order, E.G.: >> >> 0: split >> 1: lockstep >> 2: single I think the description is fine. for readability I can just add numerical order as above. > The bigger issue has been multiple Cortex-R5 bindings all doing their > own thing for this feature which is not vendor specific (except TI has > their own extra mode or something). How TCMs are described is the > other issue. I've just stopped caring because no one listens. Values used for cluster-mode property for xlnx platform is same as ti's cluster mode property. 0: split, 1: lockstep and 2: single-cpu mode. It is matching with ti,k3-r5 bindings. However, default value for some of the ti platforms are different. xlnx platforms have single-cpu mode as well. It will take some time to design system-dt spec for TCM. So, for now I use hard-code values of TCM from driver. TCM bindings for linux will be posted once system-dt spec is accepted. Meanwhile we would like to upstream rest of the bindings and driver. > >>> + >> > +patternProperties: >> > + "^r5f-[a-f0-9]+$": >> > + type: object >> > + description: | >> > + The RPU is located in the Low Power Domain of the Processor Subsystem. >> > + Each processor includes separate L1 instruction and data caches and >> > + tightly coupled memories (TCM). System memory is cacheable, but the TCM >> > + memory space is non-cacheable. >> > + >> > + Each RPU contains one 64KB memory and two 32KB memories that >> > + are accessed via the TCM A and B port interfaces, for a total of 128KB >> > + per processor. In lock-step mode, the processor has access to 256KB of >> > + TCM memory. >> > + >> > + properties: >> > + compatible: >> > + const: xlnx,zynqmp-r5f >> > + >> > + power-domains: >> > + description: RPU core PM domain specifier >> > + maxItems: 1 >> >> A bit more detail would be good, E.G. something like arm/cpus.yaml does: >> >> List of phandles and PM domain specifiers, as defined by bindings of the >> PM domain provider (see also ../power_domain.txt). >> >> And the phandle-array ref. > Both suggestions are wrong. We don't need common properties re-described > by every user. They already have a type definition too, so we don't need > to repeat phandle-array ref either. > > The existing description is not needed either as it doesn't provide any > information specific to this binding. You only need to describe each > entry when there is more than 1 entry because we need to know what each > entry is and the order. I will remove current description as well. Thanks. > >>> + >> > + mboxes: >> > + minItems: 1 >> > + items: >> > + - description: mailbox channel to send data to RPU >> > + - description: mailbox channel to receive data from RPU >> > + >> > + mbox-names: >> > + minItems: 1 >> > + items: >> > + - const: tx >> > + - const: rx >> >> And here as well for mailbox/mailbox.txt > Nope! > >> >> > + >> > + sram: >> > + $ref: /schemas/types.yaml#/definitions/phandle-array >> > + minItems: 1 >> > + maxItems: 8 >> > + items: >> > + maxItems: 1 >> > + description: | >> > + phandles to one or more reserved on-chip SRAM regions. Other than TCM, >> > + the RPU can execute instructions and access data from, the OCM memory, >> > + the main DDR memory, and other system memories. >> >> Drop the comma after "from" >> >> >> > + >> > + The regions should be defined as child nodes of the respective SRAM >> > + node, and should be defined as per the generic bindings in, >> >> Drop the comma after "in" >> >> >> > + Documentation/devicetree/bindings/sram/sram.yaml >> > + >> > + memory-region: >> > + description: | >> > + List of phandles to the reserved memory regions associated with the >> > + remoteproc device. This is variable and describes the memories shared with >> > + the remote processor (e.g. remoteproc firmware and carveouts, rpmsg >> > + vrings, ...). This reserved memory region will be allocated on DDR memory. >> >> s/on DDR/in DDR/ >> >> > + minItems: 1 >> > + maxItems: 8 >> > + items: >> > + - description: region used for RPU firmware image section >> > + - description: vdev buffer >> > + - description: vring0 >> > + - description: vring1 >> > + additionalItems: true >> > + >> > + required: >> > + - compatible >> > + - power-domains >> > + >> > + unevaluatedProperties: false >> > + >> > +required: >> > + - compatible >> > + >> > +additionalProperties: false >> > + >> > +examples: >> > + - | >> > + r5fss: r5fss { >> > + compatible = "xlnx,zynqmp-r5fss"; >> > + xlnx,cluster-mode = <1>; >> > + >> > + r5f-0 { >> > + compatible = "xlnx,zynqmp-r5f"; >> > + power-domains = <&zynqmp_firmware 0x7>; >> > + memory-region = <&rproc_0_fw_image>, <&rpu0vdev0buffer>, <&rpu0vdev0vring0>, <&rpu0vdev0vring1>; >> > + mboxes = <&ipi_mailbox_rpu0 0>, <&ipi_mailbox_rpu0 1>; >> > + mbox-names = "tx", "rx"; >> > + }; >> > + >> > + r5f-1 { >> > + compatible = "xlnx,zynqmp-r5f"; >> > + power-domains = <&zynqmp_firmware 0x8>; >> > + memory-region = <&rproc_1_fw_image>, <&rpu1vdev0buffer>, <&rpu1vdev0vring0>, <&rpu1vdev0vring1>; >> > + mboxes = <&ipi_mailbox_rpu1 0>, <&ipi_mailbox_rpu1 1>; >> > + mbox-names = "tx", "rx"; >> > + }; >> > + }; >> > +... >> > diff --git a/include/dt-bindings/power/xlnx-zynqmp-power.h b/include/dt-bindings/power/xlnx-zynqmp-power.h >> > index 0d9a412fd5e0..618024cbb20d 100644 >> > --- a/include/dt-bindings/power/xlnx-zynqmp-power.h >> > +++ b/include/dt-bindings/power/xlnx-zynqmp-power.h >> > @@ -6,6 +6,12 @@ >> > #ifndef _DT_BINDINGS_ZYNQMP_POWER_H >> > #define _DT_BINDINGS_ZYNQMP_POWER_H >> >> > +#define PD_RPU_0 7 >> > +#define PD_RPU_1 8 >> > +#define PD_R5_0_ATCM 15 >> > +#define PD_R5_0_BTCM 16 >> > +#define PD_R5_1_ATCM 17 >> > +#define PD_R5_1_BTCM 18 >> > #define PD_USB_0 22 >> > #define PD_USB_1 23 >> > #define PD_TTC_0 24 >> > -- >> >> > 2.25.1 >> >> >> -- >> Bye, Peter Korsgaard >>

3 years, 2 months

1
0
0 0

2025

2024

2023

2022

2021

Openamp-system-reference