[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <DM5PR12MB18356331B588B123FC0E1F56DA7F9@DM5PR12MB1835.namprd12.prod.outlook.com>
Date: Sun, 28 Mar 2021 21:06:47 +0000
From: Gustavo Pimentel <Gustavo.Pimentel@...opsys.com>
To: Greg Kroah-Hartman <gregkh@...uxfoundation.org>
CC: "linux-doc@...r.kernel.org" <linux-doc@...r.kernel.org>,
"linux-pci@...r.kernel.org" <linux-pci@...r.kernel.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
Derek Kiernan <derek.kiernan@...inx.com>,
Dragan Cvetic <dragan.cvetic@...inx.com>,
Arnd Bergmann <arnd@...db.de>,
Andrew Morton <akpm@...ux-foundation.org>,
Jonathan Corbet <corbet@....net>,
Bjorn Helgaas <bhelgaas@...gle.com>,
Krzysztof WilczyĆski <kw@...ux.com>
Subject: RE: [PATCH v7 1/5] misc: Add Synopsys DesignWare xData IP driver
On Sun, Mar 28, 2021 at 13:49:13, Greg Kroah-Hartman
<gregkh@...uxfoundation.org> wrote:
> On Sat, Mar 27, 2021 at 04:06:51AM +0100, Gustavo Pimentel wrote:
> > Add Synopsys DesignWare xData IP driver. This driver enables/disables
> > the PCI traffic generator module pertain to the Synopsys DesignWare
> > prototype.
> >
> > Signed-off-by: Gustavo Pimentel <gustavo.pimentel@...opsys.com>
> > ---
> > drivers/misc/dw-xdata-pcie.c | 401 +++++++++++++++++++++++++++++++++++++++++++
> > 1 file changed, 401 insertions(+)
> > create mode 100644 drivers/misc/dw-xdata-pcie.c
> >
> > diff --git a/drivers/misc/dw-xdata-pcie.c b/drivers/misc/dw-xdata-pcie.c
> > new file mode 100644
> > index 00000000..43fdd35
> > --- /dev/null
> > +++ b/drivers/misc/dw-xdata-pcie.c
> > @@ -0,0 +1,401 @@
> > +// SPDX-License-Identifier: GPL-2.0
> > +/*
> > + * Copyright (c) 2020 Synopsys, Inc. and/or its affiliates.
> > + * Synopsys DesignWare xData driver
> > + *
> > + * Author: Gustavo Pimentel <gustavo.pimentel@...opsys.com>
> > + */
> > +
> > +#include <linux/miscdevice.h>
> > +#include <linux/bitfield.h>
> > +#include <linux/pci-epf.h>
> > +#include <linux/kernel.h>
> > +#include <linux/module.h>
> > +#include <linux/device.h>
> > +#include <linux/bitops.h>
> > +#include <linux/mutex.h>
> > +#include <linux/delay.h>
> > +#include <linux/pci.h>
> > +
> > +#define DW_XDATA_DRIVER_NAME "dw-xdata-pcie"
> > +
> > +#define DW_XDATA_EP_MEM_OFFSET 0x8000000
> > +
> > +struct dw_xdata_pcie_data {
> > + /* xData registers location */
> > + enum pci_barno rg_bar;
> > + off_t rg_off;
> > + size_t rg_sz;
> > +};
> > +
> > +static const struct dw_xdata_pcie_data snps_edda_data = {
> > + /* xData registers location */
> > + .rg_bar = BAR_0,
> > + .rg_off = 0x00000000, /* 0 Kbytes */
> > + .rg_sz = 0x0000012c, /* 300 bytes */
> > +};
> > +
> > +#define STATUS_DONE BIT(0)
> > +
> > +#define CONTROL_DOORBELL BIT(0)
> > +#define CONTROL_IS_WRITE BIT(1)
> > +#define CONTROL_LENGTH(a) FIELD_PREP(GENMASK(13, 2), a)
> > +#define CONTROL_PATTERN_INC BIT(16)
> > +#define CONTROL_NO_ADDR_INC BIT(18)
> > +
> > +#define XPERF_CONTROL_ENABLE BIT(5)
> > +
> > +#define BURST_REPEAT BIT(31)
> > +#define BURST_VALUE 0x1001
> > +
> > +#define PATTERN_VALUE 0x0
> > +
> > +struct dw_xdata_regs {
> > + u32 addr_lsb; /* 0x000 */
> > + u32 addr_msb; /* 0x004 */
> > + u32 burst_cnt; /* 0x008 */
> > + u32 control; /* 0x00c */
> > + u32 pattern; /* 0x010 */
> > + u32 status; /* 0x014 */
> > + u32 RAM_addr; /* 0x018 */
> > + u32 RAM_port; /* 0x01c */
> > + u32 _reserved0[14]; /* 0x020..0x054 */
> > + u32 perf_control; /* 0x058 */
> > + u32 _reserved1[41]; /* 0x05c..0x0fc */
> > + u32 wr_cnt_lsb; /* 0x100 */
> > + u32 wr_cnt_msb; /* 0x104 */
> > + u32 rd_cnt_lsb; /* 0x108 */
> > + u32 rd_cnt_msb; /* 0x10c */
> > +} __packed;
> > +
> > +struct dw_xdata_region {
> > + phys_addr_t paddr; /* physical address */
> > + void __iomem *vaddr; /* virtual address */
> > + size_t sz; /* size */
> > +};
> > +
> > +struct dw_xdata {
> > + struct dw_xdata_region rg_region; /* registers */
> > + size_t max_wr_len; /* max wr xfer len */
> > + size_t max_rd_len; /* max rd xfer len */
> > + struct mutex mutex;
> > + struct pci_dev *pdev;
> > + struct device *dev;
>
> You do not need this 'struct device' pointer at all, please don't store
> it as you are not handling any reference counting correctly.
Agreed.
>
> > + struct miscdevice misc_dev;
> > +};
> > +
> > +static inline struct dw_xdata_regs __iomem *__dw_regs(struct dw_xdata *dw)
> > +{
> > + return dw->rg_region.vaddr;
> > +}
> > +
> > +static void dw_xdata_stop(struct dw_xdata *dw)
> > +{
> > + u32 burst;
> > +
> > + mutex_lock(&dw->mutex);
> > +
> > + burst = readl(&(__dw_regs(dw)->burst_cnt));
> > +
> > + if (burst & BURST_REPEAT) {
> > + burst &= ~(u32)BURST_REPEAT;
> > + writel(burst, &(__dw_regs(dw)->burst_cnt));
> > + }
> > +
> > + mutex_unlock(&dw->mutex);
> > +}
> > +
> > +static void dw_xdata_start(struct dw_xdata *dw, bool write)
> > +{
> > + u32 control, status;
> > +
> > + /* Stop first if xfer in progress */
> > + dw_xdata_stop(dw);
> > +
> > + mutex_lock(&dw->mutex);
> > +
> > + /* Clear status register */
> > + writel(0x0, &(__dw_regs(dw)->status));
> > +
> > + /* Burst count register set for continuous until stopped */
> > + writel(BURST_REPEAT | BURST_VALUE, &(__dw_regs(dw)->burst_cnt));
> > +
> > + /* Pattern register */
> > + writel(PATTERN_VALUE, &(__dw_regs(dw)->pattern));
> > +
> > + /* Control register */
> > + control = CONTROL_DOORBELL | CONTROL_PATTERN_INC | CONTROL_NO_ADDR_INC;
> > + if (write) {
> > + control |= CONTROL_IS_WRITE;
> > + control |= CONTROL_LENGTH(dw->max_wr_len);
> > + } else {
> > + control |= CONTROL_LENGTH(dw->max_rd_len);
> > + }
> > + writel(control, &(__dw_regs(dw)->control));
> > +
> > + /*
> > + * The xData HW block needs about 100 ms to initiate the traffic
> > + * generation according this HW block datasheet.
> > + */
> > + usleep_range(100, 150);
> > +
> > + status = readl(&(__dw_regs(dw)->status));
> > +
> > + mutex_unlock(&dw->mutex);
> > +
> > + if (!(status & STATUS_DONE))
> > + pci_dbg(dw->pdev, "xData: started %s direction\n",
> > + write ? "write" : "read");
> > +}
> > +
> > +static void dw_xdata_perf_meas(struct dw_xdata *dw, u64 *data, bool write)
> > +{
> > + if (write) {
> > + *data = readl(&(__dw_regs(dw)->wr_cnt_msb));
> > + *data <<= 32;
> > + *data |= readl(&(__dw_regs(dw)->wr_cnt_lsb));
> > + } else {
> > + *data = readl(&(__dw_regs(dw)->rd_cnt_msb));
> > + *data <<= 32;
> > + *data |= readl(&(__dw_regs(dw)->rd_cnt_lsb));
> > + }
> > +}
> > +
> > +static u64 dw_xdata_perf_diff(u64 *m1, u64 *m2, u64 time)
> > +{
> > + u64 rate = (*m1 - *m2);
> > +
> > + rate *= (1000 * 1000 * 1000);
> > + rate >>= 20;
> > + rate = DIV_ROUND_CLOSEST_ULL(rate, time);
> > +
> > + return rate;
> > +}
> > +
> > +static void dw_xdata_perf(struct dw_xdata *dw, u64 *rate, bool write)
> > +{
> > + u64 data[2], time[2], diff;
> > +
> > + mutex_lock(&dw->mutex);
> > +
> > + /* First acquisition of current count frames */
> > + writel(0x0, &(__dw_regs(dw)->perf_control));
> > + dw_xdata_perf_meas(dw, &data[0], write);
> > + time[0] = jiffies;
> > + writel((u32)XPERF_CONTROL_ENABLE, &(__dw_regs(dw)->perf_control));
> > +
> > + /*
> > + * Wait 100ms between the 1st count frame acquisition and the 2nd
> > + * count frame acquisition, in order to calculate the speed later
> > + */
> > + mdelay(100);
> > +
> > + /* Second acquisition of current count frames */
> > + writel(0x0, &(__dw_regs(dw)->perf_control));
> > + dw_xdata_perf_meas(dw, &data[1], write);
> > + time[1] = jiffies;
> > + writel((u32)XPERF_CONTROL_ENABLE, &(__dw_regs(dw)->perf_control));
> > +
> > + /*
> > + * Speed calculation
> > + *
> > + * rate = (2nd count frames - 1st count frames) / (time elapsed)
> > + */
> > + diff = jiffies_to_nsecs(time[1] - time[0]);
> > + *rate = dw_xdata_perf_diff(&data[1], &data[0], diff);
> > +
> > + mutex_unlock(&dw->mutex);
> > +
> > + pci_dbg(dw->pdev, "xData: time=%llu us, %s=%llu MB/s\n",
> > + diff, write ? "write" : "read", *rate);
> > +}
> > +
> > +static struct dw_xdata *misc_dev_to_dw(struct miscdevice *misc_dev)
> > +{
> > + return container_of(misc_dev, struct dw_xdata, misc_dev);
> > +}
> > +
> > +static ssize_t write_show(struct device *dev, struct device_attribute *attr,
> > + char *buf)
> > +{
> > + struct miscdevice *misc_dev = dev_get_drvdata(dev);
> > + struct dw_xdata *dw = misc_dev_to_dw(misc_dev);
> > + u64 rate;
> > +
> > + dw_xdata_perf(dw, &rate, true);
> > +
> > + return sysfs_emit(buf, "%llu\n", rate);
> > +}
> > +
> > +static ssize_t write_store(struct device *dev, struct device_attribute *attr,
> > + const char *buf, size_t size)
> > +{
> > + struct miscdevice *misc_dev = dev_get_drvdata(dev);
> > + struct dw_xdata *dw = misc_dev_to_dw(misc_dev);
> > +
> > + pci_dbg(dw->pdev, "xData: requested write transfer\n");
> > +
> > + dw_xdata_start(dw, true);
> > +
> > + return size;
> > +}
> > +
> > +static DEVICE_ATTR_RW(write);
> > +
> > +static ssize_t read_show(struct device *dev, struct device_attribute *attr,
> > + char *buf)
> > +{
> > + struct miscdevice *misc_dev = dev_get_drvdata(dev);
> > + struct dw_xdata *dw = misc_dev_to_dw(misc_dev);
> > + u64 rate;
> > +
> > + dw_xdata_perf(dw, &rate, false);
> > +
> > + return sysfs_emit(buf, "%llu\n", rate);
> > +}
> > +
> > +static ssize_t read_store(struct device *dev, struct device_attribute *attr,
> > + const char *buf, size_t size)
> > +{
> > + struct miscdevice *misc_dev = dev_get_drvdata(dev);
> > + struct dw_xdata *dw = misc_dev_to_dw(misc_dev);
> > +
> > + pci_dbg(dw->pdev, "xData: requested read transfer\n");
>
> dev_dbg() for your misc device, not for your pci device, as that will
> show the proper device that is causing this to happen for.
Ok, I will do a general replacement.
>
> > +
> > + dw_xdata_start(dw, false);
>
> You do not even look at the data written? That feels buggy :(
By data written, you mean the content of buf?
For this particular use case, I don't think that I would need that.
The goal was just to provide a way to trigger/initiate the PCI traffic
generator on the read direction, which doesn't require to have any
particular value, that's why it doesn't check the input value.
On ABI documentation I've given the example "echo 1 >
/sys/class/misc/dw-xdata-pcie/read", but it could be "echo abc >
/sys/class/misc/dw-xdata-pcie/read".
The same applies to "write" and "stop" on the store methods.
Perhaps it might exist a better way to do this kind of operations, any
suggestions?
Of course, I could merge the "write", "read" and "stop" in just one
device attribute as "command", but I think it will be more complex to
understand and work with it.
>
> > +
> > + return size;
> > +}
> > +
> > +static DEVICE_ATTR_RW(read);
> > +
> > +static ssize_t stop_store(struct device *dev, struct device_attribute *attr,
> > + const char *buf, size_t size)
> > +{
> > + struct miscdevice *misc_dev = dev_get_drvdata(dev);
> > + struct dw_xdata *dw = misc_dev_to_dw(misc_dev);
> > +
> > + pci_dbg(dw->pdev, "xData: requested stop any transfer\n");
>
> Same as above.
>
> > +
> > + dw_xdata_stop(dw);
>
> Again, you do not even look at the data?
>
> > +
> > + return size;
> > +}
> > +
> > +static DEVICE_ATTR_WO(stop);
> > +
> > +static struct attribute *xdata_attrs[] = {
> > + &dev_attr_write.attr,
> > + &dev_attr_read.attr,
> > + &dev_attr_stop.attr,
> > + NULL,
> > +};
> > +
> > +ATTRIBUTE_GROUPS(xdata);
> > +
> > +static int dw_xdata_pcie_probe(struct pci_dev *pdev,
> > + const struct pci_device_id *pid)
> > +{
> > + const struct dw_xdata_pcie_data *pdata = (void *)pid->driver_data;
> > + struct dw_xdata *dw;
> > + u64 addr;
> > + int err;
> > +
> > + /* Enable PCI device */
> > + err = pcim_enable_device(pdev);
> > + if (err) {
> > + pci_err(pdev, "enabling device failed\n");
> > + return err;
> > + }
> > +
> > + /* Mapping PCI BAR regions */
> > + err = pcim_iomap_regions(pdev, BIT(pdata->rg_bar), pci_name(pdev));
> > + if (err) {
> > + pci_err(pdev, "xData BAR I/O remapping failed\n");
> > + return err;
> > + }
> > +
> > + pci_set_master(pdev);
> > +
> > + /* Allocate memory */
> > + dw = devm_kzalloc(&pdev->dev, sizeof(*dw), GFP_KERNEL);
> > + if (!dw)
> > + return -ENOMEM;
> > +
> > + /* Data structure initialization */
> > + mutex_init(&dw->mutex);
> > +
> > + dw->rg_region.vaddr = pcim_iomap_table(pdev)[pdata->rg_bar];
> > + if (!dw->rg_region.vaddr)
> > + return -ENOMEM;
> > +
> > + dw->rg_region.vaddr += pdata->rg_off;
> > + dw->rg_region.paddr = pdev->resource[pdata->rg_bar].start;
> > + dw->rg_region.paddr += pdata->rg_off;
> > + dw->rg_region.sz = pdata->rg_sz;
> > +
> > + dw->max_wr_len = pcie_get_mps(pdev);
> > + dw->max_wr_len >>= 2;
> > +
> > + dw->max_rd_len = pcie_get_readrq(pdev);
> > + dw->max_rd_len >>= 2;
> > +
> > + dw->pdev = pdev;
>
> No reference counting?
Since this driver was developed for internal testing purposes and
normally it will be used just with one prototype endpoint, I didn't think
on that, but I'll include that on v8.
>
> > + dw->dev = &pdev->dev;
>
> As mentioned above, this is not needed at all.
Ok.
>
> > +
> > + dw->misc_dev.minor = MISC_DYNAMIC_MINOR;
> > + dw->misc_dev.name = kstrdup(DW_XDATA_DRIVER_NAME, GFP_KERNEL);
>
> Where do you free this memory?
It's not being done, I noticed it after sending the patch series. On v8
this will be fixed.
>
> > + dw->misc_dev.parent = &pdev->dev;
> > + dw->misc_dev.groups = xdata_groups;
> > +
> > + writel(0x0, &(__dw_regs(dw)->RAM_addr));
> > + writel(0x0, &(__dw_regs(dw)->RAM_port));
> > +
> > + addr = dw->rg_region.paddr + DW_XDATA_EP_MEM_OFFSET;
> > + writel(lower_32_bits(addr), &(__dw_regs(dw)->addr_lsb));
> > + writel(upper_32_bits(addr), &(__dw_regs(dw)->addr_msb));
> > + pci_dbg(pdev, "xData: target address = 0x%.16llx\n", addr);
> > +
> > + pci_dbg(pdev, "xData: wr_len = %zu, rd_len = %zu\n",
> > + dw->max_wr_len * 4, dw->max_rd_len * 4);
> > +
> > + /* Saving data structure reference */
> > + pci_set_drvdata(pdev, dw);
> > +
> > + /* Register misc device */
> > + err = misc_register(&dw->misc_dev);
> > + if (err)
> > + return err;
> > +
> > + return 0;
>
> How about:
> return misc_register(...);
I've reworked this part to include the kfree of the variable
dw->misc_dev.name
>
>
> > +}
> > +
> > +static void dw_xdata_pcie_remove(struct pci_dev *pdev)
> > +{
> > + struct dw_xdata *dw = pci_get_drvdata(pdev);
> > +
> > + if (dw) {
>
> How can this ever not be true? You never set this to NULL so this check
> is pointless.
It will be removed.
>
> > + dw_xdata_stop(dw);
> > + misc_deregister(&dw->misc_dev);
> > + }
> > +}
> > +
> > +static const struct pci_device_id dw_xdata_pcie_id_table[] = {
> > + { PCI_DEVICE_DATA(SYNOPSYS, EDDA, &snps_edda_data) },
>
> Why do you need a pointer to snps_edda_data here?
The structure snps_edda_data indicates the location of this IP block (BAR
and offset) for this particular endpoint.
It's very likely in the future to be more variants that for HW design
reasons might require this IP block to be on a different location.
-Gustavo
>
> thanks,
>
> greg k-h
Powered by blists - more mailing lists