public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Alex Elder <elder@linaro.org>
To: davem@davemloft.net, kuba@kernel.org
Cc: evgreen@chromium.org, subashab@codeaurora.org,
	cpratapa@codeaurora.org, bjorn.andersson@linaro.org,
	netdev@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: [PATCH net-next 09/13] net: ipa: only enable generic command completion IRQ when needed
Date: Thu,  5 Nov 2020 12:14:03 -0600	[thread overview]
Message-ID: <20201105181407.8006-10-elder@linaro.org> (raw)
In-Reply-To: <20201105181407.8006-1-elder@linaro.org>

The completion of a generic EE GSI command is signaled by a global
interrupt of type GP_INT1.  The only other used type for a global
interrupt is a hardware error report.

First, disallow all global interrupt types in gsi_irq_setup().  We
want to know about hardware errors, so re-enable the interrupt type
in gsi_irq_enable(), to allow hardware errors to be reported.
Disable that interrupt type again in gsi_irq_disable().

We only issue generic EE commands one at a time, and there's no
reason to keep the completion interrupt enabled when no generic
EE command is pending.  We furthermore have no need to enable the
GP_INT2 or GP_INT3 interrupt types (which aren't used).

The change in gsi_irq_enable() makes GSI_CNTXT_GLOB_IRQ_ALL unused,
so get rid of it.  Have gsi_generic_command() enable the GP_INT1
interrupt type (in addition to the ERROR_INT type) only while a
generic command is pending.

Signed-off-by: Alex Elder <elder@linaro.org>
---
 drivers/net/ipa/gsi.c     | 35 +++++++++++++++++++++++++++--------
 drivers/net/ipa/gsi_reg.h |  1 -
 2 files changed, 27 insertions(+), 9 deletions(-)

diff --git a/drivers/net/ipa/gsi.c b/drivers/net/ipa/gsi.c
index 2c01a04e07b70..4ab1d89f642ea 100644
--- a/drivers/net/ipa/gsi.c
+++ b/drivers/net/ipa/gsi.c
@@ -257,6 +257,7 @@ static void gsi_irq_setup(struct gsi *gsi)
 
 	iowrite32(0, gsi->virt + GSI_CNTXT_SRC_CH_IRQ_MSK_OFFSET);
 	iowrite32(0, gsi->virt + GSI_CNTXT_SRC_EV_CH_IRQ_MSK_OFFSET);
+	iowrite32(0, gsi->virt + GSI_CNTXT_GLOB_IRQ_EN_OFFSET);
 }
 
 /* Turn off all GSI interrupts when we're all done */
@@ -289,14 +290,16 @@ static void gsi_irq_enable(struct gsi *gsi)
 {
 	u32 val;
 
+	/* Global interrupts include hardware error reports.  Enable
+	 * that so we can at least report the error should it occur.
+	 */
+	iowrite32(ERROR_INT_FMASK, gsi->virt + GSI_CNTXT_GLOB_IRQ_EN_OFFSET);
+	gsi->type_enabled_bitmap |= BIT(GSI_GLOB_EE);
+
 	/* Each IEOB interrupt is enabled (later) as needed by channels */
 	iowrite32(0, gsi->virt + GSI_CNTXT_SRC_IEOB_IRQ_MSK_OFFSET);
 	gsi->type_enabled_bitmap |= BIT(GSI_IEOB);
 
-	val = GSI_CNTXT_GLOB_IRQ_ALL;
-	iowrite32(val, gsi->virt + GSI_CNTXT_GLOB_IRQ_EN_OFFSET);
-	gsi->type_enabled_bitmap |= BIT(GSI_GLOB_EE);
-
 	/* We don't use inter-EE channel or event interrupts */
 
 	/* Never enable GSI_BREAK_POINT */
@@ -315,8 +318,8 @@ static void gsi_irq_disable(struct gsi *gsi)
 	gsi_irq_type_update(gsi);
 
 	iowrite32(0, gsi->virt + GSI_CNTXT_GSI_IRQ_EN_OFFSET);
-	iowrite32(0, gsi->virt + GSI_CNTXT_GLOB_IRQ_EN_OFFSET);
 	iowrite32(0, gsi->virt + GSI_CNTXT_SRC_IEOB_IRQ_MSK_OFFSET);
+	iowrite32(0, gsi->virt + GSI_CNTXT_GLOB_IRQ_EN_OFFSET);
 }
 
 /* Return the virtual address associated with a ring index */
@@ -1101,8 +1104,8 @@ static void gsi_isr_glob_err(struct gsi *gsi)
 	iowrite32(~0, gsi->virt + GSI_ERROR_LOG_CLR_OFFSET);
 
 	ee = u32_get_bits(val, ERR_EE_FMASK);
-	which = u32_get_bits(val, ERR_VIRT_IDX_FMASK);
 	type = u32_get_bits(val, ERR_TYPE_FMASK);
+	which = u32_get_bits(val, ERR_VIRT_IDX_FMASK);
 	code = u32_get_bits(val, ERR_CODE_FMASK);
 
 	if (type == GSI_ERR_TYPE_CHAN)
@@ -1606,8 +1609,19 @@ static int gsi_generic_command(struct gsi *gsi, u32 channel_id,
 			       enum gsi_generic_cmd_opcode opcode)
 {
 	struct completion *completion = &gsi->completion;
+	bool success;
 	u32 val;
 
+	/* The error global interrupt type is always enabled (until we
+	 * teardown), so we won't change that.  A generic EE command
+	 * completes with a GSI global interrupt of type GP_INT1.  We
+	 * only perform one generic command at a time (to allocate or
+	 * halt a modem channel) and only from this function.  So we
+	 * enable the GP_INT1 IRQ type here while we're expecting it.
+	 */
+	val = ERROR_INT_FMASK | GP_INT1_FMASK;
+	iowrite32(val, gsi->virt + GSI_CNTXT_GLOB_IRQ_EN_OFFSET);
+
 	/* First zero the result code field */
 	val = ioread32(gsi->virt + GSI_CNTXT_SCRATCH_0_OFFSET);
 	val &= ~GENERIC_EE_RESULT_FMASK;
@@ -1618,8 +1632,13 @@ static int gsi_generic_command(struct gsi *gsi, u32 channel_id,
 	val |= u32_encode_bits(channel_id, GENERIC_CHID_FMASK);
 	val |= u32_encode_bits(GSI_EE_MODEM, GENERIC_EE_FMASK);
 
-	if (gsi_command(gsi, GSI_GENERIC_CMD_OFFSET, val, completion))
-		return 0;	/* Success! */
+	success = gsi_command(gsi, GSI_GENERIC_CMD_OFFSET, val, completion);
+
+	/* Disable the GP_INT1 IRQ type again */
+	iowrite32(ERROR_INT_FMASK, gsi->virt + GSI_CNTXT_GLOB_IRQ_EN_OFFSET);
+
+	if (success)
+		return 0;
 
 	dev_err(gsi->dev, "GSI generic command %u to channel %u timed out\n",
 		opcode, channel_id);
diff --git a/drivers/net/ipa/gsi_reg.h b/drivers/net/ipa/gsi_reg.h
index 1dd81cf0b46a8..ae00aff1cfa50 100644
--- a/drivers/net/ipa/gsi_reg.h
+++ b/drivers/net/ipa/gsi_reg.h
@@ -335,7 +335,6 @@ enum gsi_irq_type_id {
 #define GP_INT1_FMASK			GENMASK(1, 1)
 #define GP_INT2_FMASK			GENMASK(2, 2)
 #define GP_INT3_FMASK			GENMASK(3, 3)
-#define GSI_CNTXT_GLOB_IRQ_ALL		GENMASK(3, 0)
 
 #define GSI_CNTXT_GSI_IRQ_STTS_OFFSET \
 			GSI_EE_N_CNTXT_GSI_IRQ_STTS_OFFSET(GSI_EE_AP)
-- 
2.20.1


  parent reply	other threads:[~2020-11-05 18:15 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-11-05 18:13 [PATCH net-next 00/13] net: ipa: constrain GSI interrupts Alex Elder
2020-11-05 18:13 ` [PATCH net-next 01/13] net: ipa: refer to IPA versions, not GSI Alex Elder
2020-11-05 18:13 ` [PATCH net-next 02/13] net: ipa: request GSI IRQ later Alex Elder
2020-11-05 18:13 ` [PATCH net-next 03/13] net: ipa: rename gsi->event_enable_bitmap Alex Elder
2020-11-05 18:13 ` [PATCH net-next 04/13] net: ipa: define GSI interrupt types with an enum Alex Elder
2020-11-05 18:13 ` [PATCH net-next 05/13] net: ipa: disable all GSI interrupt types initially Alex Elder
2020-11-05 18:14 ` [PATCH net-next 06/13] net: ipa: cache last-saved GSI IRQ enabled type Alex Elder
2020-11-05 18:14 ` [PATCH net-next 07/13] net: ipa: only enable GSI channel control IRQs when needed Alex Elder
2020-11-05 18:14 ` [PATCH net-next 08/13] net: ipa: only enable GSI event " Alex Elder
2020-11-05 18:14 ` Alex Elder [this message]
2020-11-05 18:14 ` [PATCH net-next 10/13] net: ipa: only enable GSI IEOB " Alex Elder
2020-11-05 18:14 ` [PATCH net-next 11/13] net: ipa: explicitly disallow inter-EE interrupts Alex Elder
2020-11-05 18:14 ` [PATCH net-next 12/13] net: ipa: only enable GSI general IRQs when needed Alex Elder
2020-11-05 18:14 ` [PATCH net-next 13/13] net: ipa: pass a value to gsi_irq_type_update() Alex Elder
2020-11-07 23:50 ` [PATCH net-next 00/13] net: ipa: constrain GSI interrupts patchwork-bot+netdevbpf

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=20201105181407.8006-10-elder@linaro.org \
    --to=elder@linaro.org \
    --cc=bjorn.andersson@linaro.org \
    --cc=cpratapa@codeaurora.org \
    --cc=davem@davemloft.net \
    --cc=evgreen@chromium.org \
    --cc=kuba@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=subashab@codeaurora.org \
    /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