linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Oded Gabbay <oded.gabbay@gmail.com>
To: linux-kernel@vger.kernel.org, SW_Drivers@habana.ai
Cc: Ofir Bitton <obitton@habana.ai>
Subject: [PATCH 2/4] habanalabs: set clock gating according to mask
Date: Sun,  9 Aug 2020 14:53:35 +0300	[thread overview]
Message-ID: <20200809115337.23402-2-oded.gabbay@gmail.com> (raw)
In-Reply-To: <20200809115337.23402-1-oded.gabbay@gmail.com>

From: Ofir Bitton <obitton@habana.ai>

Once clock gating is set we enable clock gating according to mask,
we should also disable clock gating according to relevant bits.

Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
---
 drivers/misc/habanalabs/gaudi/gaudi.c | 44 +++++++++++++--------------
 1 file changed, 21 insertions(+), 23 deletions(-)

diff --git a/drivers/misc/habanalabs/gaudi/gaudi.c b/drivers/misc/habanalabs/gaudi/gaudi.c
index 08124873483f..7e0f9f64ffcb 100644
--- a/drivers/misc/habanalabs/gaudi/gaudi.c
+++ b/drivers/misc/habanalabs/gaudi/gaudi.c
@@ -2508,6 +2508,7 @@ static void gaudi_set_clock_gating(struct hl_device *hdev)
 {
 	struct gaudi_device *gaudi = hdev->asic_specific;
 	u32 qman_offset;
+	bool enable;
 	int i;
 
 	/* In case we are during debug session, don't enable the clock gate
@@ -2517,46 +2518,43 @@ static void gaudi_set_clock_gating(struct hl_device *hdev)
 		return;
 
 	for (i = GAUDI_PCI_DMA_1, qman_offset = 0 ; i < GAUDI_HBM_DMA_1 ; i++) {
-		if (!(hdev->clock_gating_mask &
-					(BIT_ULL(gaudi_dma_assignment[i]))))
-			continue;
+		enable = !!(hdev->clock_gating_mask &
+				(BIT_ULL(gaudi_dma_assignment[i])));
 
 		qman_offset = gaudi_dma_assignment[i] * DMA_QMAN_OFFSET;
-		WREG32(mmDMA0_QM_CGM_CFG1 + qman_offset, QMAN_CGM1_PWR_GATE_EN);
+		WREG32(mmDMA0_QM_CGM_CFG1 + qman_offset,
+				enable ? QMAN_CGM1_PWR_GATE_EN : 0);
 		WREG32(mmDMA0_QM_CGM_CFG + qman_offset,
-				QMAN_UPPER_CP_CGM_PWR_GATE_EN);
+				enable ? QMAN_UPPER_CP_CGM_PWR_GATE_EN : 0);
 	}
 
 	for (i = GAUDI_HBM_DMA_1 ; i < GAUDI_DMA_MAX ; i++) {
-		if (!(hdev->clock_gating_mask &
-					(BIT_ULL(gaudi_dma_assignment[i]))))
-			continue;
+		enable = !!(hdev->clock_gating_mask &
+				(BIT_ULL(gaudi_dma_assignment[i])));
 
 		qman_offset = gaudi_dma_assignment[i] * DMA_QMAN_OFFSET;
-		WREG32(mmDMA0_QM_CGM_CFG1 + qman_offset, QMAN_CGM1_PWR_GATE_EN);
+		WREG32(mmDMA0_QM_CGM_CFG1 + qman_offset,
+				enable ? QMAN_CGM1_PWR_GATE_EN : 0);
 		WREG32(mmDMA0_QM_CGM_CFG + qman_offset,
-				QMAN_COMMON_CP_CGM_PWR_GATE_EN);
+				enable ? QMAN_COMMON_CP_CGM_PWR_GATE_EN : 0);
 	}
 
-	if (hdev->clock_gating_mask & (BIT_ULL(GAUDI_ENGINE_ID_MME_0))) {
-		WREG32(mmMME0_QM_CGM_CFG1, QMAN_CGM1_PWR_GATE_EN);
-		WREG32(mmMME0_QM_CGM_CFG, QMAN_COMMON_CP_CGM_PWR_GATE_EN);
-	}
+	enable = !!(hdev->clock_gating_mask & (BIT_ULL(GAUDI_ENGINE_ID_MME_0)));
+	WREG32(mmMME0_QM_CGM_CFG1, enable ? QMAN_CGM1_PWR_GATE_EN : 0);
+	WREG32(mmMME0_QM_CGM_CFG, enable ? QMAN_COMMON_CP_CGM_PWR_GATE_EN : 0);
 
-	if (hdev->clock_gating_mask & (BIT_ULL(GAUDI_ENGINE_ID_MME_2))) {
-		WREG32(mmMME2_QM_CGM_CFG1, QMAN_CGM1_PWR_GATE_EN);
-		WREG32(mmMME2_QM_CGM_CFG, QMAN_COMMON_CP_CGM_PWR_GATE_EN);
-	}
+	enable = !!(hdev->clock_gating_mask & (BIT_ULL(GAUDI_ENGINE_ID_MME_2)));
+	WREG32(mmMME2_QM_CGM_CFG1, enable ? QMAN_CGM1_PWR_GATE_EN : 0);
+	WREG32(mmMME2_QM_CGM_CFG, enable ? QMAN_COMMON_CP_CGM_PWR_GATE_EN : 0);
 
 	for (i = 0, qman_offset = 0 ; i < TPC_NUMBER_OF_ENGINES ; i++) {
-		if (!(hdev->clock_gating_mask &
-					(BIT_ULL(GAUDI_ENGINE_ID_TPC_0 + i))))
-			continue;
+		enable = !!(hdev->clock_gating_mask &
+				(BIT_ULL(GAUDI_ENGINE_ID_TPC_0 + i)));
 
 		WREG32(mmTPC0_QM_CGM_CFG1 + qman_offset,
-				QMAN_CGM1_PWR_GATE_EN);
+				enable ? QMAN_CGM1_PWR_GATE_EN : 0);
 		WREG32(mmTPC0_QM_CGM_CFG + qman_offset,
-				QMAN_COMMON_CP_CGM_PWR_GATE_EN);
+				enable ? QMAN_COMMON_CP_CGM_PWR_GATE_EN : 0);
 
 		qman_offset += TPC_QMAN_OFFSET;
 	}
-- 
2.17.1


  reply	other threads:[~2020-08-09 11:53 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-08-09 11:53 [PATCH 1/4] habanalabs: verify user input in cs_ioctl_signal_wait Oded Gabbay
2020-08-09 11:53 ` Oded Gabbay [this message]
2020-08-09 11:53 ` [PATCH 3/4] habanalabs: proper handling of alloc size in coresight Oded Gabbay
2020-08-09 11:53 ` [PATCH 4/4] habanalabs: set max power according to card type Oded Gabbay
2020-08-09 14:52   ` kernel test robot

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=20200809115337.23402-2-oded.gabbay@gmail.com \
    --to=oded.gabbay@gmail.com \
    --cc=SW_Drivers@habana.ai \
    --cc=linux-kernel@vger.kernel.org \
    --cc=obitton@habana.ai \
    /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).