[PATCH] serial: mxc: Speed-up character transmission
Instead of waiting for empty FIFO condition before writing a character, wait for non-full FIFO condition.
This helps in saving several tens of milliseconds during boot (depending verbosity).
Signed-off-by: Loic Poulain loic.poulain@linaro.org --- drivers/serial/serial_mxc.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/drivers/serial/serial_mxc.c b/drivers/serial/serial_mxc.c index 82c0d84628..f8c49865be 100644 --- a/drivers/serial/serial_mxc.c +++ b/drivers/serial/serial_mxc.c @@ -223,11 +223,11 @@ static void mxc_serial_putc(const char c) if (c == '\n') serial_putc('\r');
- writel(c, &mxc_base->txd); - /* wait for transmitter to be ready */ - while (!(readl(&mxc_base->ts) & UTS_TXEMPTY)) + while (readl(&mxc_base->ts) & UTS_TXFULL) schedule(); + + writel(c, &mxc_base->txd); }
/* Test whether a character is in the RX buffer */ @@ -311,7 +311,7 @@ static int mxc_serial_putc(struct udevice *dev, const char ch) struct mxc_serial_plat *plat = dev_get_plat(dev); struct mxc_uart *const uart = plat->reg;
- if (!(readl(&uart->ts) & UTS_TXEMPTY)) + if (readl(&uart->ts) & UTS_TXFULL) return -EAGAIN;
writel(ch, &uart->txd);
Hi,
On Mon, 9 Jan 2023 15:59:25 +0100 Loic Poulain wrote:
Instead of waiting for empty FIFO condition before writing a character, wait for non-full FIFO condition.
This helps in saving several tens of milliseconds during boot (depending verbosity).
Signed-off-by: Loic Poulain loic.poulain@linaro.org
drivers/serial/serial_mxc.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/drivers/serial/serial_mxc.c b/drivers/serial/serial_mxc.c index 82c0d84628..f8c49865be 100644 --- a/drivers/serial/serial_mxc.c +++ b/drivers/serial/serial_mxc.c @@ -223,11 +223,11 @@ static void mxc_serial_putc(const char c) if (c == '\n') serial_putc('\r');
- writel(c, &mxc_base->txd);
- /* wait for transmitter to be ready */
- while (!(readl(&mxc_base->ts) & UTS_TXEMPTY))
- while (readl(&mxc_base->ts) & UTS_TXFULL) schedule();
- writel(c, &mxc_base->txd);
}
/* Test whether a character is in the RX buffer */ @@ -311,7 +311,7 @@ static int mxc_serial_putc(struct udevice *dev, const char ch) struct mxc_serial_plat *plat = dev_get_plat(dev); struct mxc_uart *const uart = plat->reg;
- if (!(readl(&uart->ts) & UTS_TXEMPTY))
if (readl(&uart->ts) & UTS_TXFULL) return -EAGAIN;
writel(ch, &uart->txd);
A previous attempt to do this in: |commit c7878a0483c59c48a730123bc0f4659fd40921bf |Author: Johannes Schneider johannes.schneider@leica-geosystems.com |Date: Tue Sep 6 14:15:04 2022 +0200 | | serial: mxc: have putc use the TXFIFO
has been reverted in: |commit fc1c1760de38823edbdc2cdd9606dff938a07f6e |Author: Fabio Estevam festevam@denx.de |Date: Tue Nov 8 08:39:33 2022 -0300 | | Revert "serial: mxc: have putc use the TXFIFO" | | This reverts commit c7878a0483c59c48a730123bc0f4659fd40921bf. | | Since commit c7878a0483c5 ("serial: mxc: have putc use the TXFIFO"), | serial console corruption can be seen when priting inside board_init(). | | Revert it to avoid the regression. | | Reported-by: Tim Harvey tharvey@gateworks.com | Signed-off-by: Fabio Estevam festevam@denx.de | Acked-by: Tim Harvey tharvey@gateworks.com
Lothar Waßmann
Hi Lothar,
On Tue, 10 Jan 2023 at 07:46, Lothar Waßmann LW@karo-electronics.de wrote:
A previous attempt to do this in: |commit c7878a0483c59c48a730123bc0f4659fd40921bf |Author: Johannes Schneider johannes.schneider@leica-geosystems.com |Date: Tue Sep 6 14:15:04 2022 +0200 | | serial: mxc: have putc use the TXFIFO
has been reverted in: |commit fc1c1760de38823edbdc2cdd9606dff938a07f6e |Author: Fabio Estevam festevam@denx.de |Date: Tue Nov 8 08:39:33 2022 -0300 | | Revert "serial: mxc: have putc use the TXFIFO" | | This reverts commit c7878a0483c59c48a730123bc0f4659fd40921bf. | | Since commit c7878a0483c5 ("serial: mxc: have putc use the TXFIFO"), | serial console corruption can be seen when priting inside board_init().
Thanks for highlighting this. Looked at it and reproduced some sort of corruption in board_init() as well, which seems due to reinitialization of the UART while TX is not complete. I'm going to follow up with a V2, including a fix for this.
Regards, Loic
participants (2)
-
Loic Poulain -
Lothar Waßmann