From: Jae Hyun Yoo <jae.hyun.yoo@linux.intel.com>
To: Brendan Higgins <brendanhiggins@google.com>,
Wolfram Sang <wsa@the-dreams.de>,
Benjamin Herrenschmidt <benh@kernel.crashing.org>,
Joel Stanley <joel@jms.id.au>, Rob Herring <robh+dt@kernel.org>,
Mark Rutland <mark.rutland@arm.com>,
Andrew Jeffery <andrew@aj.id.au>, Tao Ren <taoren@fb.com>,
Cedric Le Goater <clg@kaod.org>
Cc: devicetree@vger.kernel.org,
Jae Hyun Yoo <jae.hyun.yoo@linux.intel.com>,
linux-aspeed@lists.ozlabs.org, openbmc@lists.ozlabs.org,
linux-i2c@vger.kernel.org, linux-arm-kernel@lists.infradead.org
Subject: [PATCH] i2c: aspeed: fix master pending state handling
Date: Wed, 9 Oct 2019 14:20:34 -0700 [thread overview]
Message-ID: <20191009212034.20325-1-jae.hyun.yoo@linux.intel.com> (raw)
In case of master pending state, it should not trigger a master
command, otherwise data could be corrupted because this H/W shares
the same data buffer for slave and master operations. It also means
that H/W command queue handling is unreliable because of the buffer
sharing issue. To fix this issue, it clears command queue if a
master command is queued in pending state to use S/W solution
instead of H/W command queue handling. Also, it refines restarting
mechanism of the pending master command.
Fixes: 2e57b7cebb98 ("i2c: aspeed: Add multi-master use case support")
Signed-off-by: Jae Hyun Yoo <jae.hyun.yoo@linux.intel.com>
---
drivers/i2c/busses/i2c-aspeed.c | 54 +++++++++++++++++++++------------
1 file changed, 34 insertions(+), 20 deletions(-)
diff --git a/drivers/i2c/busses/i2c-aspeed.c b/drivers/i2c/busses/i2c-aspeed.c
index fa66951b05d0..7b098ff5f5dd 100644
--- a/drivers/i2c/busses/i2c-aspeed.c
+++ b/drivers/i2c/busses/i2c-aspeed.c
@@ -108,6 +108,12 @@
#define ASPEED_I2CD_S_TX_CMD BIT(2)
#define ASPEED_I2CD_M_TX_CMD BIT(1)
#define ASPEED_I2CD_M_START_CMD BIT(0)
+#define ASPEED_I2CD_MASTER_CMDS_MASK \
+ (ASPEED_I2CD_M_STOP_CMD | \
+ ASPEED_I2CD_M_S_RX_CMD_LAST | \
+ ASPEED_I2CD_M_RX_CMD | \
+ ASPEED_I2CD_M_TX_CMD | \
+ ASPEED_I2CD_M_START_CMD)
/* 0x18 : I2CD Slave Device Address Register */
#define ASPEED_I2CD_DEV_ADDR_MASK GENMASK(6, 0)
@@ -336,18 +342,19 @@ static void aspeed_i2c_do_start(struct aspeed_i2c_bus *bus)
struct i2c_msg *msg = &bus->msgs[bus->msgs_index];
u8 slave_addr = i2c_8bit_addr_from_msg(msg);
- bus->master_state = ASPEED_I2C_MASTER_START;
-
#if IS_ENABLED(CONFIG_I2C_SLAVE)
/*
* If it's requested in the middle of a slave session, set the master
* state to 'pending' then H/W will continue handling this master
* command when the bus comes back to the idle state.
*/
- if (bus->slave_state != ASPEED_I2C_SLAVE_INACTIVE)
+ if (bus->slave_state != ASPEED_I2C_SLAVE_INACTIVE) {
bus->master_state = ASPEED_I2C_MASTER_PENDING;
+ return;
+ }
#endif /* CONFIG_I2C_SLAVE */
+ bus->master_state = ASPEED_I2C_MASTER_START;
bus->buf_index = 0;
if (msg->flags & I2C_M_RD) {
@@ -422,20 +429,6 @@ static u32 aspeed_i2c_master_irq(struct aspeed_i2c_bus *bus, u32 irq_status)
}
}
-#if IS_ENABLED(CONFIG_I2C_SLAVE)
- /*
- * A pending master command will be started by H/W when the bus comes
- * back to idle state after completing a slave operation so change the
- * master state from 'pending' to 'start' at here if slave is inactive.
- */
- if (bus->master_state == ASPEED_I2C_MASTER_PENDING) {
- if (bus->slave_state != ASPEED_I2C_SLAVE_INACTIVE)
- goto out_no_complete;
-
- bus->master_state = ASPEED_I2C_MASTER_START;
- }
-#endif /* CONFIG_I2C_SLAVE */
-
/* Master is not currently active, irq was for someone else. */
if (bus->master_state == ASPEED_I2C_MASTER_INACTIVE ||
bus->master_state == ASPEED_I2C_MASTER_PENDING)
@@ -462,11 +455,15 @@ static u32 aspeed_i2c_master_irq(struct aspeed_i2c_bus *bus, u32 irq_status)
#if IS_ENABLED(CONFIG_I2C_SLAVE)
/*
* If a peer master starts a xfer immediately after it queues a
- * master command, change its state to 'pending' then H/W will
- * continue the queued master xfer just after completing the
- * slave mode session.
+ * master command, clear the queued master command and change
+ * its state to 'pending'. To simplify handling of pending
+ * cases, it uses S/W solution instead of H/W command queue
+ * handling.
*/
if (unlikely(irq_status & ASPEED_I2CD_INTR_SLAVE_MATCH)) {
+ writel(readl(bus->base + ASPEED_I2C_CMD_REG) &
+ ~ASPEED_I2CD_MASTER_CMDS_MASK,
+ bus->base + ASPEED_I2C_CMD_REG);
bus->master_state = ASPEED_I2C_MASTER_PENDING;
dev_dbg(bus->dev,
"master goes pending due to a slave start\n");
@@ -629,6 +626,14 @@ static irqreturn_t aspeed_i2c_bus_irq(int irq, void *dev_id)
irq_handled |= aspeed_i2c_master_irq(bus,
irq_remaining);
}
+
+ /*
+ * Start a pending master command at here if a slave operation is
+ * completed.
+ */
+ if (bus->master_state == ASPEED_I2C_MASTER_PENDING &&
+ bus->slave_state == ASPEED_I2C_SLAVE_INACTIVE)
+ aspeed_i2c_do_start(bus);
#else
irq_handled = aspeed_i2c_master_irq(bus, irq_remaining);
#endif /* CONFIG_I2C_SLAVE */
@@ -691,6 +696,15 @@ static int aspeed_i2c_master_xfer(struct i2c_adapter *adap,
ASPEED_I2CD_BUS_BUSY_STS))
aspeed_i2c_recover_bus(bus);
+ /*
+ * If timed out and the state is still pending, drop the pending
+ * master command.
+ */
+ spin_lock_irqsave(&bus->lock, flags);
+ if (bus->master_state == ASPEED_I2C_MASTER_PENDING)
+ bus->master_state = ASPEED_I2C_MASTER_INACTIVE;
+ spin_unlock_irqrestore(&bus->lock, flags);
+
return -ETIMEDOUT;
}
--
2.23.0
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
next reply other threads:[~2019-10-09 21:20 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-10-09 21:20 Jae Hyun Yoo [this message]
2019-10-10 0:32 ` [PATCH] i2c: aspeed: fix master pending state handling Brendan Higgins
2019-10-10 1:26 ` Tao Ren
2019-10-10 6:37 ` Wolfram Sang
2019-10-10 6:48 ` Joel Stanley
2019-10-10 18:22 ` Jae Hyun Yoo
2019-10-10 21:20 ` Tao Ren
2019-10-10 22:04 ` Jae Hyun Yoo
2019-10-10 23:11 ` Tao Ren
2019-10-10 23:16 ` Jae Hyun Yoo
2019-10-10 23:24 ` Tao Ren
2019-10-10 23:29 ` Tao Ren
2019-10-10 23:52 ` Jae Hyun Yoo
2019-10-16 19:20 ` Brendan Higgins
2019-10-21 12:09 ` Wolfram Sang
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20191009212034.20325-1-jae.hyun.yoo@linux.intel.com \
--to=jae.hyun.yoo@linux.intel.com \
--cc=andrew@aj.id.au \
--cc=benh@kernel.crashing.org \
--cc=brendanhiggins@google.com \
--cc=clg@kaod.org \
--cc=devicetree@vger.kernel.org \
--cc=joel@jms.id.au \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-aspeed@lists.ozlabs.org \
--cc=linux-i2c@vger.kernel.org \
--cc=mark.rutland@arm.com \
--cc=openbmc@lists.ozlabs.org \
--cc=robh+dt@kernel.org \
--cc=taoren@fb.com \
--cc=wsa@the-dreams.de \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).