diff mbox series

[v3,6/6] gpio: aspeed: Add the flush write to ensure the write complete.

Message ID 20240913074325.239390-7-billy_tsai@aspeedtech.com (mailing list archive)
State New, archived
Headers show
Series Add Aspeed G7 gpio support | expand

Commit Message

Billy Tsai Sept. 13, 2024, 7:43 a.m. UTC
Performing a dummy read ensures that the register write operation is fully
completed, mitigating any potential bus delays that could otherwise impact
the frequency of bitbang usage. E.g., if the JTAG application uses GPIO to
control the JTAG pins (TCK, TMS, TDI, TDO, and TRST), and the application
sets the TCK clock to 1 MHz, the GPIO’s high/low transitions will rely on
a delay function to ensure the clock frequency does not exceed 1 MHz.
However, this can lead to rapid toggling of the GPIO because the write
operation is POSTed and does not wait for a bus acknowledgment.

Signed-off-by: Billy Tsai <billy_tsai@aspeedtech.com>
---
 drivers/gpio/gpio-aspeed.c | 2 ++
 1 file changed, 2 insertions(+)

Comments

Andrew Jeffery Sept. 16, 2024, 3:03 a.m. UTC | #1
On Fri, 2024-09-13 at 15:43 +0800, Billy Tsai wrote:
> Performing a dummy read ensures that the register write operation is fully
> completed, mitigating any potential bus delays that could otherwise impact
> the frequency of bitbang usage. E.g., if the JTAG application uses GPIO to
> control the JTAG pins (TCK, TMS, TDI, TDO, and TRST), and the application
> sets the TCK clock to 1 MHz, the GPIO’s high/low transitions will rely on
> a delay function to ensure the clock frequency does not exceed 1 MHz.
> However, this can lead to rapid toggling of the GPIO because the write
> operation is POSTed and does not wait for a bus acknowledgment.
> 
> Signed-off-by: Billy Tsai <billy_tsai@aspeedtech.com>

... are you aware of any other driver concerns of a similar nature wrt
the architecture of the SoCs?
Billy Tsai Sept. 19, 2024, 6:07 a.m. UTC | #2
> > Performing a dummy read ensures that the register write operation is fully
> > completed, mitigating any potential bus delays that could otherwise impact
> > the frequency of bitbang usage. E.g., if the JTAG application uses GPIO to
> > control the JTAG pins (TCK, TMS, TDI, TDO, and TRST), and the application
> > sets the TCK clock to 1 MHz, the GPIO’s high/low transitions will rely on
> > a delay function to ensure the clock frequency does not exceed 1 MHz.
> > However, this can lead to rapid toggling of the GPIO because the write
> > operation is POSTed and does not wait for a bus acknowledgment.
> >
> > Signed-off-by: Billy Tsai <billy_tsai@aspeedtech.com>

> ... are you aware of any other driver concerns of a similar nature wrt
> the architecture of the SoCs?

No, we are only aware of this issue with the GPIO controller, which affects
the output pin behavior immediately after register write.
diff mbox series

Patch

diff --git a/drivers/gpio/gpio-aspeed.c b/drivers/gpio/gpio-aspeed.c
index d322c03481a8..d1e61ee48da2 100644
--- a/drivers/gpio/gpio-aspeed.c
+++ b/drivers/gpio/gpio-aspeed.c
@@ -406,6 +406,8 @@  static void __aspeed_gpio_set(struct gpio_chip *gc, unsigned int offset,
 	gpio->dcache[GPIO_BANK(offset)] = reg;
 
 	gpio->config->llops->reg_bits_set(gpio, offset, reg_val, val);
+	// flush write
+	gpio->config->llops->reg_bits_read(gpio, offset, reg_val);
 }
 
 static void aspeed_gpio_set(struct gpio_chip *gc, unsigned int offset,