2022-11-25 13:31:12

by Maxime Chevallier

Subject: [PATCH net-next 1/3] net: pcs: altera-tse: use read_poll_timeout to wait for reset

Software resets on the TSE PCS don't clear registers, but rather reset
all internal state machines regarding AN, comma detection and
encoding/decoding. Use read_poll_timeout to wait for the reset to clear
instead of manually polling the register.

Signed-off-by: Maxime Chevallier <[email protected]>
---
drivers/net/pcs/pcs-altera-tse.c | 11 +++--------
1 file changed, 3 insertions(+), 8 deletions(-)

diff --git a/drivers/net/pcs/pcs-altera-tse.c b/drivers/net/pcs/pcs-altera-tse.c
index 97a7cabff962..e86cadc391e8 100644
--- a/drivers/net/pcs/pcs-altera-tse.c
+++ b/drivers/net/pcs/pcs-altera-tse.c
@@ -60,7 +60,6 @@ static void tse_pcs_write(struct altera_tse_pcs *tse_pcs, int regnum,

static int tse_pcs_reset(struct altera_tse_pcs *tse_pcs)
{
- int i = 0;
u16 bmcr;

/* Reset PCS block */
@@ -68,13 +67,9 @@ static int tse_pcs_reset(struct altera_tse_pcs *tse_pcs)
bmcr |= BMCR_RESET;
tse_pcs_write(tse_pcs, MII_BMCR, bmcr);

- for (i = 0; i < SGMII_PCS_SW_RESET_TIMEOUT; i++) {
- if (!(tse_pcs_read(tse_pcs, MII_BMCR) & BMCR_RESET))
- return 0;
- udelay(1);
- }
-
- return -ETIMEDOUT;
+ return read_poll_timeout(tse_pcs_read, bmcr, (bmcr & BMCR_RESET),
+ 10, SGMII_PCS_SW_RESET_TIMEOUT, 1,
+ tse_pcs, MII_BMCR);
}

static int alt_tse_pcs_validate(struct phylink_pcs *pcs,
--
2.38.1

2022-11-30 04:42:55

by Jakub Kicinski

Subject: Re: [PATCH net-next 1/3] net: pcs: altera-tse: use read_poll_timeout to wait for reset

On Fri, 25 Nov 2022 14:17:59 +0100 Maxime Chevallier wrote:
> - for (i = 0; i < SGMII_PCS_SW_RESET_TIMEOUT; i++) {
> - if (!(tse_pcs_read(tse_pcs, MII_BMCR) & BMCR_RESET))
> - return 0;
> - udelay(1);
> - }
> -
> - return -ETIMEDOUT;
> + return read_poll_timeout(tse_pcs_read, bmcr, (bmcr & BMCR_RESET),
> + 10, SGMII_PCS_SW_RESET_TIMEOUT, 1,
> + tse_pcs, MII_BMCR);

You say "no functional change intended" in the cover letter but you
switch from udelay to usleep and change timeouts here. I presume this
is intentional but should be mentioned in the commit message, I think.