Re: [PATCH v1 2/2] hwrng: npcm: add NPCM RNG driver

From: Daniel Thompson
Date: Tue Sep 10 2019 - 07:29:30 EST


On Tue, Sep 10, 2019 at 01:52:35PM +0300, Tomer Maimon wrote:
> Hi Daniel,
>
> Thanks for your prompt reply,
>
>
>
> On Mon, 9 Sep 2019 at 18:10, Daniel Thompson <daniel.thompson@xxxxxxxxxx>
> wrote:
>
> > On Mon, Sep 09, 2019 at 05:31:30PM +0300, Tomer Maimon wrote:
> > > Hi Daniel,
> > >
> > > appreciate your comments and sorry for the late reply
> > >
> > > On Thu, 29 Aug 2019 at 13:47, Daniel Thompson <
> > daniel.thompson@xxxxxxxxxx>
> > > wrote:
> > >
> > > > On Wed, Aug 28, 2019 at 07:26:17PM +0300, Tomer Maimon wrote:
> > > > > Add Nuvoton NPCM BMC Random Number Generator(RNG) driver.
> > > > >
> > > > > Signed-off-by: Tomer Maimon <tmaimon77@xxxxxxxxx>
> > > > > ---
> > > > > drivers/char/hw_random/Kconfig | 13 ++
> > > > > drivers/char/hw_random/Makefile | 1 +
> > > > > drivers/char/hw_random/npcm-rng.c | 207
> > ++++++++++++++++++++++++++++++
> > > > > 3 files changed, 221 insertions(+)
> > > > > create mode 100644 drivers/char/hw_random/npcm-rng.c
> > > > >
> > > > > diff --git a/drivers/char/hw_random/npcm-rng.c
> > > > b/drivers/char/hw_random/npcm-rng.c
> > > > > new file mode 100644
> > > > > index 000000000000..5b4b1b6cb362
> > > > > --- /dev/null
> > > > > +++ b/drivers/char/hw_random/npcm-rng.c
> > > > > @@ -0,0 +1,207 @@
> > > > > +// SPDX-License-Identifier: GPL-2.0
> > > > > +// Copyright (c) 2019 Nuvoton Technology corporation.
> > > > > +
> > > > > +#include <linux/kernel.h>
> > > > > +#include <linux/module.h>
> > > > > +#include <linux/io.h>
> > > > > +#include <linux/iopoll.h>
> > > > > +#include <linux/init.h>
> > > > > +#include <linux/random.h>
> > > > > +#include <linux/err.h>
> > > > > +#include <linux/platform_device.h>
> > > > > +#include <linux/hw_random.h>
> > > > > +#include <linux/delay.h>
> > > > > +#include <linux/of_irq.h>
> > > > > +#include <linux/pm_runtime.h>
> > > > > +
> > > > > +#define NPCM_RNGCS_REG 0x00 /* Control and status
> > > > register */
> > > > > +#define NPCM_RNGD_REG 0x04 /* Data register */
> > > > > +#define NPCM_RNGMODE_REG 0x08 /* Mode register */
> > > > > +
> > > > > +#define NPCM_RNG_CLK_SET_25MHZ GENMASK(4, 3) /* 20-25 MHz */
> > > > > +#define NPCM_RNG_DATA_VALID BIT(1)
> > > > > +#define NPCM_RNG_ENABLE BIT(0)
> > > > > +#define NPCM_RNG_M1ROSEL BIT(1)
> > > > > +
> > > > > +#define NPCM_RNG_TIMEOUT_POLL 20
> > > >
> > > > Might be better to define this in real-world units (such as
> > > > milliseconds) since the timeout is effectively the longest time the
> > > > hardware can take to generate 4 bytes.
> > > >
> > > > > +
> > > > > +#define to_npcm_rng(p) container_of(p, struct npcm_rng, rng)
> > > > > +
> > > > > +struct npcm_rng {
> > > > > + void __iomem *base;
> > > > > + struct hwrng rng;
> > > > > +};
> > > > > +
> > > > > +static int npcm_rng_init(struct hwrng *rng)
> > > > > +{
> > > > > + struct npcm_rng *priv = to_npcm_rng(rng);
> > > > > + u32 val;
> > > > > +
> > > > > + val = readl(priv->base + NPCM_RNGCS_REG);
> > > > > + val |= NPCM_RNG_ENABLE;
> > > > > + writel(val, priv->base + NPCM_RNGCS_REG);
> > > > > +
> > > > > + return 0;
> > > > > +}
> > > > > +
> > > > > +static void npcm_rng_cleanup(struct hwrng *rng)
> > > > > +{
> > > > > + struct npcm_rng *priv = to_npcm_rng(rng);
> > > > > + u32 val;
> > > > > +
> > > > > + val = readl(priv->base + NPCM_RNGCS_REG);
> > > > > + val &= ~NPCM_RNG_ENABLE;
> > > > > + writel(val, priv->base + NPCM_RNGCS_REG);
> > > > > +}
> > > > > +
> > > > > +static bool npcm_rng_wait_ready(struct hwrng *rng, bool wait)
> > > > > +{
> > > > > + struct npcm_rng *priv = to_npcm_rng(rng);
> > > > > + int timeout_cnt = 0;
> > > > > + int ready;
> > > > > +
> > > > > + ready = readl(priv->base + NPCM_RNGCS_REG) &
> > NPCM_RNG_DATA_VALID;
> > > > > + while ((ready == 0) && (timeout_cnt < NPCM_RNG_TIMEOUT_POLL)) {
> > > > > + usleep_range(500, 1000);
> > > > > + ready = readl(priv->base + NPCM_RNGCS_REG) &
> > > > > + NPCM_RNG_DATA_VALID;
> > > > > + timeout_cnt++;
> > > > > + }
> > > > > +
> > > > > + return !!ready;
> > > > > +}
> > > >
> > > > This looks like an open-coded version of readl_poll_timeout()... better
> > > > to use the library function.
> > > >
> > > > Also the sleep looks a bit long to me. What is the generation rate of
> > > > the peripheral? Most RNG drivers have short intervals between data
> > > > generation so they use delays rather than sleeps (a.k.a.
> > > > readl_poll_timeout_atomic() ).
> > >
> > > the HWRNG generate byte of random data in a few milliseconds so it is
> > > better to use the sleep command.
> >
> > That's fine, just use readl_poll_timeout() then.
> >
> >
> > > > > +
> > > > > +static int npcm_rng_read(struct hwrng *rng, void *buf, size_t max,
> > bool
> > > > wait)
> > > > > +{
> > > > > + struct npcm_rng *priv = to_npcm_rng(rng);
> > > > > + int retval = 0;
> > > > > +
> > > > > + pm_runtime_get_sync((struct device *)priv->rng.priv);
> > > > > +
> > > > > + while (max >= sizeof(u32)) {
> > > > > + if (!npcm_rng_wait_ready(rng, wait))
> > > > > + break;
> > > >
> > > > The code as currently written does not honour the wait parameter (e.g.
> > > > it sleeps even when wait is false).
> > > >
> > > >
> > > > > +
> > > > > + *(u32 *)buf = readl(priv->base + NPCM_RNGD_REG);
> > > > > + retval += sizeof(u32);
> > > > > + buf += sizeof(u32);
> > > > > + max -= sizeof(u32);
> > > > > + }
> > > > > +
> > > > > + pm_runtime_mark_last_busy((struct device *)priv->rng.priv);
> > > > > + pm_runtime_put_sync_autosuspend((struct device
> > *)priv->rng.priv);
> > > > > +
> > > > > + return retval || !wait ? retval : -EIO;
> > > > > +}
> > > > > +
> > > > > +static int npcm_rng_probe(struct platform_device *pdev)
> > > > > +{
> > > > > + struct npcm_rng *priv;
> > > > > + struct resource *res;
> > > > > + u32 quality;
> > > > > + int ret;
> > > > > +
> > > > > + priv = devm_kzalloc(&pdev->dev, sizeof(*priv), GFP_KERNEL);
> > > > > + if (!priv)
> > > > > + return -ENOMEM;
> > > > > +
> > > > > + res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
> > > > > + priv->base = devm_ioremap_resource(&pdev->dev, res);
> > > > > + if (IS_ERR(priv->base))
> > > > > + return PTR_ERR(priv->base);
> > > > > +
> > > > > + priv->rng.name = pdev->name;
> > > > > +#ifndef CONFIG_PM
> > > > > + priv->rng.init = npcm_rng_init;
> > > > > + priv->rng.cleanup = npcm_rng_cleanup;
> > > > > +#endif
> > > > > + priv->rng.read = npcm_rng_read;
> > > > > + priv->rng.priv = (unsigned long)&pdev->dev;
> > > > > + if (of_property_read_u32(pdev->dev.of_node, "quality",
> > &quality))
> > > > > + priv->rng.quality = 1000;
> > > > > + else
> > > > > + priv->rng.quality = quality;
> > > > > +
> > > > > + writel(NPCM_RNG_M1ROSEL, priv->base + NPCM_RNGMODE_REG);
> > > > > +#ifndef CONFIG_PM
> > > > > + writel(NPCM_RNG_CLK_SET_25MHZ, priv->base + NPCM_RNGCS_REG);
> > > > > +#else
> > > > > + writel(NPCM_RNG_CLK_SET_25MHZ | NPCM_RNG_ENABLE,
> > > > > + priv->base + NPCM_RNGCS_REG);
> > > > > +#endif
> > > >
> > > > If this initialization was moved to npcm_rng_init() then there would be
> > > > no need for the additional ifdefing. It would also get rid of the
> > > > (potentially slow) readl calls on the PM wakeup path.
> > > >
> > >
> > > But when the Kernel have PM configuration than the priv->rng.init is not
> > > set and
> > > *add_early_randomness* function is called. for the *add_early_randomness*
> > > success
> > > the hwrng need to enabled in the probe.
> >
> > Sorry but I don't understand this reply.
> >
> > When CONFIG_PM is enabled then the probe function does not currently set
> > NPCM_RNG_ENABLE; instead is relies on npcm_rng_init() being called by
> >
>
> Sorry maybe I miss understood, but when the CONFIG_PM enabled so the
> NPCM_RNG_ENABLE sets (the code use ifndef and not ifdef)
> *#ifndef CONFIG_PM*
> writel(NPCM_RNG_CLK_SET_25MHZ, priv->base + NPCM_RNGCS_REG);
> #else (*CONFIG_PM enabled*)
> writel(NPCM_RNG_CLK_SET_25MHZ | NPCM_RNG_ENABLE,
> priv->base + NPCM_RNGCS_REG);
> #endif
>
> And the hwrng needed to be enabled to run *add_early_randomness *function
> successfully.
>
> If the hwrng driver will relay on PM logic to enable the hwrng will be
> disable when *add_early_randomness *function is called.
>
> the PM logic (as part of pm_runtime_get_sync() ).

I ended up reading my mail out of order and replied to the v2 patch.

The question is *why* the driver doesn't resume properly when adding
early randomness! I think it is because the hwrng_register() is being
called before PM runtime is enabled.


Daniel.