* [PATCH] spi: s3c64xx: requests spi-dma channel only during data transfer [not found] <CGME20220524140211epcas5p1b9b61a0ea112beb2327848c37d3ba000@epcas5p1.samsung.com> @ 2022-05-24 14:01 ` Adithya K V 0 siblings, 0 replies; 4+ messages in thread From: Adithya K V @ 2022-05-24 14:01 UTC (permalink / raw) To: linux-spi, linux-arm-kernel Cc: linux-samsung-soc, linux-kernel, krzysztof.kozlowski, andi, broonie, alim.akhtar, pankaj.dubey, Adithya K V Current s3c64xx SPI driver acquires DMA channel during driver probe and holds on it even when channels are not used (no DMA transfer). This is a problem especially when all the DMA channels are exhausted (as other IPs on the same DMA controller also acquires DMA channel) and if a new IP/Device requests for a DMA channel (on the same DMA controller), it won’t get DMA channel allocated. The said issue can be avoided if s3c64xx driver request and release DMA channel before and after data transfer. Let’s modify the driver to request and release DMA channel before and after DMA mode data transfer. Signed-off-by: Adithya K V <adithya.kv@samsung.com> --- drivers/spi/spi-s3c64xx.c | 56 ++++++++++++++++++++++----------------- 1 file changed, 32 insertions(+), 24 deletions(-) diff --git a/drivers/spi/spi-s3c64xx.c b/drivers/spi/spi-s3c64xx.c index c26440e9058d..82558e37c735 100644 --- a/drivers/spi/spi-s3c64xx.c +++ b/drivers/spi/spi-s3c64xx.c @@ -350,6 +350,23 @@ static int s3c64xx_spi_prepare_transfer(struct spi_master *spi) if (is_polling(sdd)) return 0; + /* Requests DMA channels */ + sdd->rx_dma.ch = dma_request_chan(&sdd->pdev->dev, "rx"); + if (IS_ERR(sdd->rx_dma.ch)) { + dev_err(&sdd->pdev->dev, "Failed to get RX DMA channel\n"); + sdd->rx_dma.ch = 0; + return 0; + } + + sdd->tx_dma.ch = dma_request_chan(&sdd->pdev->dev, "tx"); + if (IS_ERR(sdd->tx_dma.ch)) { + dev_err(&sdd->pdev->dev, "Failed to get TX DMA hannel\n"); + dma_release_channel(sdd->rx_dma.ch); + sdd->tx_dma.ch = 0; + sdd->rx_dma.ch = 0; + return 0; + } + spi->dma_rx = sdd->rx_dma.ch; spi->dma_tx = sdd->tx_dma.ch; @@ -362,7 +379,12 @@ static bool s3c64xx_spi_can_dma(struct spi_master *master, { struct s3c64xx_spi_driver_data *sdd = spi_master_get_devdata(master); - return xfer->len > (FIFO_LVL_MASK(sdd) >> 1) + 1; + if (sdd->rx_dma.ch && sdd->tx_dma.ch) { + return xfer->len > (FIFO_LVL_MASK(sdd) >> 1) + 1; + } else { + return 0; + } + } static int s3c64xx_enable_datapath(struct s3c64xx_spi_driver_data *sdd, @@ -697,7 +719,7 @@ static int s3c64xx_spi_transfer_one(struct spi_master *master, sdd->rx_dma.ch && sdd->tx_dma.ch) { use_dma = 1; - } else if (is_polling(sdd) && xfer->len > fifo_len) { + } else if (xfer->len > fifo_len) { tx_buf = xfer->tx_buf; rx_buf = xfer->rx_buf; origin_len = xfer->len; @@ -782,6 +804,14 @@ static int s3c64xx_spi_transfer_one(struct spi_master *master, xfer->len = origin_len; } + /* Releases DMA channels after data transfer is completed */ + if (sdd->rx_dma.ch && sdd->tx_dma.ch) { + dma_release_channel(sdd->rx_dma.ch); + dma_release_channel(sdd->tx_dma.ch); + sdd->rx_dma.ch = 0; + sdd->tx_dma.ch = 0; + } + return status; } @@ -1167,22 +1197,6 @@ static int s3c64xx_spi_probe(struct platform_device *pdev) } } - if (!is_polling(sdd)) { - /* Acquire DMA channels */ - sdd->rx_dma.ch = dma_request_chan(&pdev->dev, "rx"); - if (IS_ERR(sdd->rx_dma.ch)) { - dev_err(&pdev->dev, "Failed to get RX DMA channel\n"); - ret = PTR_ERR(sdd->rx_dma.ch); - goto err_disable_io_clk; - } - sdd->tx_dma.ch = dma_request_chan(&pdev->dev, "tx"); - if (IS_ERR(sdd->tx_dma.ch)) { - dev_err(&pdev->dev, "Failed to get TX DMA channel\n"); - ret = PTR_ERR(sdd->tx_dma.ch); - goto err_release_rx_dma; - } - } - pm_runtime_set_autosuspend_delay(&pdev->dev, AUTOSUSPEND_TIMEOUT); pm_runtime_use_autosuspend(&pdev->dev); pm_runtime_set_active(&pdev->dev); @@ -1228,12 +1242,6 @@ static int s3c64xx_spi_probe(struct platform_device *pdev) pm_runtime_disable(&pdev->dev); pm_runtime_set_suspended(&pdev->dev); - if (!is_polling(sdd)) - dma_release_channel(sdd->tx_dma.ch); -err_release_rx_dma: - if (!is_polling(sdd)) - dma_release_channel(sdd->rx_dma.ch); -err_disable_io_clk: clk_disable_unprepare(sdd->ioclk); err_disable_src_clk: clk_disable_unprepare(sdd->src_clk); base-commit: 09ce5091ff971cdbfd67ad84dc561ea27f10d67a -- 2.17.1 ^ permalink raw reply related [flat|nested] 4+ messages in thread
* [PATCH] spi: s3c64xx: requests spi-dma channel only during data transfer @ 2022-05-24 14:01 ` Adithya K V 0 siblings, 0 replies; 4+ messages in thread From: Adithya K V @ 2022-05-24 14:01 UTC (permalink / raw) To: linux-spi, linux-arm-kernel Cc: linux-samsung-soc, linux-kernel, krzysztof.kozlowski, andi, broonie, alim.akhtar, pankaj.dubey, Adithya K V Current s3c64xx SPI driver acquires DMA channel during driver probe and holds on it even when channels are not used (no DMA transfer). This is a problem especially when all the DMA channels are exhausted (as other IPs on the same DMA controller also acquires DMA channel) and if a new IP/Device requests for a DMA channel (on the same DMA controller), it won’t get DMA channel allocated. The said issue can be avoided if s3c64xx driver request and release DMA channel before and after data transfer. Let’s modify the driver to request and release DMA channel before and after DMA mode data transfer. Signed-off-by: Adithya K V <adithya.kv@samsung.com> --- drivers/spi/spi-s3c64xx.c | 56 ++++++++++++++++++++++----------------- 1 file changed, 32 insertions(+), 24 deletions(-) diff --git a/drivers/spi/spi-s3c64xx.c b/drivers/spi/spi-s3c64xx.c index c26440e9058d..82558e37c735 100644 --- a/drivers/spi/spi-s3c64xx.c +++ b/drivers/spi/spi-s3c64xx.c @@ -350,6 +350,23 @@ static int s3c64xx_spi_prepare_transfer(struct spi_master *spi) if (is_polling(sdd)) return 0; + /* Requests DMA channels */ + sdd->rx_dma.ch = dma_request_chan(&sdd->pdev->dev, "rx"); + if (IS_ERR(sdd->rx_dma.ch)) { + dev_err(&sdd->pdev->dev, "Failed to get RX DMA channel\n"); + sdd->rx_dma.ch = 0; + return 0; + } + + sdd->tx_dma.ch = dma_request_chan(&sdd->pdev->dev, "tx"); + if (IS_ERR(sdd->tx_dma.ch)) { + dev_err(&sdd->pdev->dev, "Failed to get TX DMA hannel\n"); + dma_release_channel(sdd->rx_dma.ch); + sdd->tx_dma.ch = 0; + sdd->rx_dma.ch = 0; + return 0; + } + spi->dma_rx = sdd->rx_dma.ch; spi->dma_tx = sdd->tx_dma.ch; @@ -362,7 +379,12 @@ static bool s3c64xx_spi_can_dma(struct spi_master *master, { struct s3c64xx_spi_driver_data *sdd = spi_master_get_devdata(master); - return xfer->len > (FIFO_LVL_MASK(sdd) >> 1) + 1; + if (sdd->rx_dma.ch && sdd->tx_dma.ch) { + return xfer->len > (FIFO_LVL_MASK(sdd) >> 1) + 1; + } else { + return 0; + } + } static int s3c64xx_enable_datapath(struct s3c64xx_spi_driver_data *sdd, @@ -697,7 +719,7 @@ static int s3c64xx_spi_transfer_one(struct spi_master *master, sdd->rx_dma.ch && sdd->tx_dma.ch) { use_dma = 1; - } else if (is_polling(sdd) && xfer->len > fifo_len) { + } else if (xfer->len > fifo_len) { tx_buf = xfer->tx_buf; rx_buf = xfer->rx_buf; origin_len = xfer->len; @@ -782,6 +804,14 @@ static int s3c64xx_spi_transfer_one(struct spi_master *master, xfer->len = origin_len; } + /* Releases DMA channels after data transfer is completed */ + if (sdd->rx_dma.ch && sdd->tx_dma.ch) { + dma_release_channel(sdd->rx_dma.ch); + dma_release_channel(sdd->tx_dma.ch); + sdd->rx_dma.ch = 0; + sdd->tx_dma.ch = 0; + } + return status; } @@ -1167,22 +1197,6 @@ static int s3c64xx_spi_probe(struct platform_device *pdev) } } - if (!is_polling(sdd)) { - /* Acquire DMA channels */ - sdd->rx_dma.ch = dma_request_chan(&pdev->dev, "rx"); - if (IS_ERR(sdd->rx_dma.ch)) { - dev_err(&pdev->dev, "Failed to get RX DMA channel\n"); - ret = PTR_ERR(sdd->rx_dma.ch); - goto err_disable_io_clk; - } - sdd->tx_dma.ch = dma_request_chan(&pdev->dev, "tx"); - if (IS_ERR(sdd->tx_dma.ch)) { - dev_err(&pdev->dev, "Failed to get TX DMA channel\n"); - ret = PTR_ERR(sdd->tx_dma.ch); - goto err_release_rx_dma; - } - } - pm_runtime_set_autosuspend_delay(&pdev->dev, AUTOSUSPEND_TIMEOUT); pm_runtime_use_autosuspend(&pdev->dev); pm_runtime_set_active(&pdev->dev); @@ -1228,12 +1242,6 @@ static int s3c64xx_spi_probe(struct platform_device *pdev) pm_runtime_disable(&pdev->dev); pm_runtime_set_suspended(&pdev->dev); - if (!is_polling(sdd)) - dma_release_channel(sdd->tx_dma.ch); -err_release_rx_dma: - if (!is_polling(sdd)) - dma_release_channel(sdd->rx_dma.ch); -err_disable_io_clk: clk_disable_unprepare(sdd->ioclk); err_disable_src_clk: clk_disable_unprepare(sdd->src_clk); base-commit: 09ce5091ff971cdbfd67ad84dc561ea27f10d67a -- 2.17.1 _______________________________________________ linux-arm-kernel mailing list linux-arm-kernel@lists.infradead.org http://lists.infradead.org/mailman/listinfo/linux-arm-kernel ^ permalink raw reply related [flat|nested] 4+ messages in thread
* Re: [PATCH] spi: s3c64xx: requests spi-dma channel only during data transfer 2022-05-24 14:01 ` Adithya K V @ 2022-06-07 10:46 ` Mark Brown -1 siblings, 0 replies; 4+ messages in thread From: Mark Brown @ 2022-06-07 10:46 UTC (permalink / raw) To: linux-arm-kernel, adithya.kv, linux-spi Cc: alim.akhtar, linux-kernel, andi, krzysztof.kozlowski, linux-samsung-soc, pankaj.dubey On Tue, 24 May 2022 19:31:32 +0530, Adithya K V wrote: > Current s3c64xx SPI driver acquires DMA channel during driver > probe and holds on it even when channels are not used > (no DMA transfer). This is a problem especially when all the > DMA channels are exhausted (as other IPs on the same DMA > controller also acquires DMA channel) and if a new IP/Device > requests for a DMA channel (on the same DMA controller), it won’t > get DMA channel allocated. > The said issue can be avoided if s3c64xx driver request and > release DMA channel before and after data transfer. Let’s modify > the driver to request and release DMA channel before and after > DMA mode data transfer. > > [...] Applied to https://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi.git for-next Thanks! [1/1] spi: s3c64xx: requests spi-dma channel only during data transfer commit: f52b03c707444c5a3d1a0b9c5724f93ddc3c588e All being well this means that it will be integrated into the linux-next tree (usually sometime in the next 24 hours) and sent to Linus during the next merge window (or sooner if it is a bug fix), however if problems are discovered then the patch may be dropped or reverted. You may get further e-mails resulting from automated or manual testing and review of the tree, please engage with people reporting problems and send followup patches addressing any issues that are reported if needed. If any updates are required or you are submitting further changes they should be sent as incremental updates against current git, existing patches will not be replaced. Please add any relevant lists and maintainers to the CCs when replying to this mail. Thanks, Mark ^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH] spi: s3c64xx: requests spi-dma channel only during data transfer @ 2022-06-07 10:46 ` Mark Brown 0 siblings, 0 replies; 4+ messages in thread From: Mark Brown @ 2022-06-07 10:46 UTC (permalink / raw) To: linux-arm-kernel, adithya.kv, linux-spi Cc: alim.akhtar, linux-kernel, andi, krzysztof.kozlowski, linux-samsung-soc, pankaj.dubey On Tue, 24 May 2022 19:31:32 +0530, Adithya K V wrote: > Current s3c64xx SPI driver acquires DMA channel during driver > probe and holds on it even when channels are not used > (no DMA transfer). This is a problem especially when all the > DMA channels are exhausted (as other IPs on the same DMA > controller also acquires DMA channel) and if a new IP/Device > requests for a DMA channel (on the same DMA controller), it won’t > get DMA channel allocated. > The said issue can be avoided if s3c64xx driver request and > release DMA channel before and after data transfer. Let’s modify > the driver to request and release DMA channel before and after > DMA mode data transfer. > > [...] Applied to https://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi.git for-next Thanks! [1/1] spi: s3c64xx: requests spi-dma channel only during data transfer commit: f52b03c707444c5a3d1a0b9c5724f93ddc3c588e All being well this means that it will be integrated into the linux-next tree (usually sometime in the next 24 hours) and sent to Linus during the next merge window (or sooner if it is a bug fix), however if problems are discovered then the patch may be dropped or reverted. You may get further e-mails resulting from automated or manual testing and review of the tree, please engage with people reporting problems and send followup patches addressing any issues that are reported if needed. If any updates are required or you are submitting further changes they should be sent as incremental updates against current git, existing patches will not be replaced. Please add any relevant lists and maintainers to the CCs when replying to this mail. Thanks, Mark _______________________________________________ linux-arm-kernel mailing list linux-arm-kernel@lists.infradead.org http://lists.infradead.org/mailman/listinfo/linux-arm-kernel ^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2022-06-07 10:50 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
[not found] <CGME20220524140211epcas5p1b9b61a0ea112beb2327848c37d3ba000@epcas5p1.samsung.com>
2022-05-24 14:01 ` [PATCH] spi: s3c64xx: requests spi-dma channel only during data transfer Adithya K V
2022-05-24 14:01 ` Adithya K V
2022-06-07 10:46 ` Mark Brown
2022-06-07 10:46 ` Mark Brown
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.