From patchwork Sun Nov 5 09:34:17 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: =?utf-8?q?Uwe_Kleine-K=C3=B6nig?= X-Patchwork-Id: 13445674 Received: from lindbergh.monkeyblade.net (lindbergh.monkeyblade.net [23.128.96.19]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id B22E88C16 for ; Sun, 5 Nov 2023 09:34:44 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; dkim=none Received: from metis.whiteo.stw.pengutronix.de (metis.whiteo.stw.pengutronix.de [IPv6:2a0a:edc0:2:b01:1d::104]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 55CF8C0 for ; Sun, 5 Nov 2023 01:34:43 -0800 (PST) Received: from drehscheibe.grey.stw.pengutronix.de ([2a0a:edc0:0:c01:1d::a2]) by metis.whiteo.stw.pengutronix.de with esmtps (TLS1.3:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.92) (envelope-from ) id 1qzZWd-0000Pd-T4; Sun, 05 Nov 2023 10:34:35 +0100 Received: from [2a0a:edc0:0:900:1d::77] (helo=ptz.office.stw.pengutronix.de) by drehscheibe.grey.stw.pengutronix.de with esmtps (TLS1.3) tls TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384 (Exim 4.94.2) (envelope-from ) id 1qzZWb-006l7I-Te; Sun, 05 Nov 2023 10:34:33 +0100 Received: from ukl by ptz.office.stw.pengutronix.de with local (Exim 4.94.2) (envelope-from ) id 1qzZWb-00D8nf-KR; Sun, 05 Nov 2023 10:34:33 +0100 From: =?utf-8?q?Uwe_Kleine-K=C3=B6nig?= To: Vinod Koul Cc: Taichi Sugaya , Takao Orito , dmaengine@vger.kernel.org, linux-arm-kernel@lists.infradead.org, kernel@pengutronix.de Subject: [PATCH 1/4] dmaengine: milbeaut-hdmac: Convert to platform remove callback returning void Date: Sun, 5 Nov 2023 10:34:17 +0100 Message-ID: <20231105093415.3704633-7-u.kleine-koenig@pengutronix.de> X-Mailer: git-send-email 2.42.0 In-Reply-To: <20231105093415.3704633-6-u.kleine-koenig@pengutronix.de> References: <20231105093415.3704633-6-u.kleine-koenig@pengutronix.de> Precedence: bulk X-Mailing-List: dmaengine@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 X-Developer-Signature: v=1; a=openpgp-sha256; l=2642; i=u.kleine-koenig@pengutronix.de; h=from:subject; bh=iFA5SppBIUUrRTjlHh/H9b168YL1tbBN6Q5z7xRSG+A=; b=owEBbQGS/pANAwAKAY+A+1h9Ev5OAcsmYgBlR2GYE0b9wUxglKY+saD0HA5khqRUyH857dbZ5 Bg24LiORTKJATMEAAEKAB0WIQQ/gaxpOnoeWYmt/tOPgPtYfRL+TgUCZUdhmAAKCRCPgPtYfRL+ Tt96CACPBr/T5fmGsFg4nVxjGCv2TxQEa7qFpSvVlYTvnfx7FIGt6PIdltD6B3LEJDxvjq27B68 dnpAJsbhXY0ZcmaWX3TGcJrFpazAb1/ogMK6MEsglattVL7+RcLtzbr79qBADdwDaR3RkUUIZAh e8jXRrnxCtXBTL7xqv2rpdkIC2Lv5S8Wseu/SIynz3CdpTxcjzuL8SVVog/BDy+qeHz5mHLoPif jQeWAQiccAHnbnlY+FnerBcCnE65bKjT8iHTABdnb7h4C1jqW7WzT7FX5OObi7sdjqMyjUw/5Bg kizbOnbQZuha5RdvWCCcJeB8c1Teb5DKlG4DjATRUMokhgCp X-Developer-Key: i=u.kleine-koenig@pengutronix.de; a=openpgp; fpr=0D2511F322BFAB1C1580266BE2DCDD9132669BD6 X-SA-Exim-Connect-IP: 2a0a:edc0:0:c01:1d::a2 X-SA-Exim-Mail-From: ukl@pengutronix.de X-SA-Exim-Scanned: No (on metis.whiteo.stw.pengutronix.de); SAEximRunCond expanded to false X-PTX-Original-Recipient: dmaengine@vger.kernel.org The .remove() callback for a platform driver returns an int which makes many driver authors wrongly assume it's possible to do error handling by returning an error code. However the value returned is ignored (apart from emitting a warning) and this typically results in resource leaks. To improve here there is a quest to make the remove callback return void. In the first step of this quest all drivers are converted to .remove_new(), which already returns void. Eventually after all drivers are converted, .remove_new() will be renamed to .remove(). There is an error path that has the above mentioned problem. This patch only adds a more drastic error message. To properly fix it, dmaengine_terminate_sync() must be known to have succeeded (or that it's safe to not call it as other drivers seem to assume). Signed-off-by: Uwe Kleine-König --- drivers/dma/milbeaut-hdmac.c | 17 +++++++++++------ 1 file changed, 11 insertions(+), 6 deletions(-) diff --git a/drivers/dma/milbeaut-hdmac.c b/drivers/dma/milbeaut-hdmac.c index 1b0a95892627..7b41c670970a 100644 --- a/drivers/dma/milbeaut-hdmac.c +++ b/drivers/dma/milbeaut-hdmac.c @@ -531,7 +531,7 @@ static int milbeaut_hdmac_probe(struct platform_device *pdev) return ret; } -static int milbeaut_hdmac_remove(struct platform_device *pdev) +static void milbeaut_hdmac_remove(struct platform_device *pdev) { struct milbeaut_hdmac_device *mdev = platform_get_drvdata(pdev); struct dma_chan *chan; @@ -546,16 +546,21 @@ static int milbeaut_hdmac_remove(struct platform_device *pdev) */ list_for_each_entry(chan, &mdev->ddev.channels, device_node) { ret = dmaengine_terminate_sync(chan); - if (ret) - return ret; + if (ret) { + /* + * This results in resource leakage and maybe also + * use-after-free errors as e.g. *mdev is kfreed. + */ + dev_alert(&pdev->dev, "Failed to terminate channel %d (%pe)\n", + chan->chan_id, ERR_PTR(ret)); + return; + } milbeaut_hdmac_free_chan_resources(chan); } of_dma_controller_free(pdev->dev.of_node); dma_async_device_unregister(&mdev->ddev); clk_disable_unprepare(mdev->clk); - - return 0; } static const struct of_device_id milbeaut_hdmac_match[] = { @@ -566,7 +571,7 @@ MODULE_DEVICE_TABLE(of, milbeaut_hdmac_match); static struct platform_driver milbeaut_hdmac_driver = { .probe = milbeaut_hdmac_probe, - .remove = milbeaut_hdmac_remove, + .remove_new = milbeaut_hdmac_remove, .driver = { .name = "milbeaut-m10v-hdmac", .of_match_table = milbeaut_hdmac_match, From patchwork Sun Nov 5 09:34:18 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: =?utf-8?q?Uwe_Kleine-K=C3=B6nig?= X-Patchwork-Id: 13445675 Received: from lindbergh.monkeyblade.net (lindbergh.monkeyblade.net [23.128.96.19]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 1DA438F44 for ; Sun, 5 Nov 2023 09:34:45 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; dkim=none Received: from metis.whiteo.stw.pengutronix.de (metis.whiteo.stw.pengutronix.de [IPv6:2a0a:edc0:2:b01:1d::104]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 89106C5 for ; Sun, 5 Nov 2023 01:34:43 -0800 (PST) Received: from drehscheibe.grey.stw.pengutronix.de ([2a0a:edc0:0:c01:1d::a2]) by metis.whiteo.stw.pengutronix.de with esmtps (TLS1.3:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.92) (envelope-from ) id 1qzZWd-0000Pe-T3; Sun, 05 Nov 2023 10:34:35 +0100 Received: from [2a0a:edc0:0:900:1d::77] (helo=ptz.office.stw.pengutronix.de) by drehscheibe.grey.stw.pengutronix.de with esmtps (TLS1.3) tls TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384 (Exim 4.94.2) (envelope-from ) id 1qzZWc-006l7L-3u; Sun, 05 Nov 2023 10:34:34 +0100 Received: from ukl by ptz.office.stw.pengutronix.de with local (Exim 4.94.2) (envelope-from ) id 1qzZWb-00D8nj-Qv; Sun, 05 Nov 2023 10:34:33 +0100 From: =?utf-8?q?Uwe_Kleine-K=C3=B6nig?= To: Vinod Koul Cc: Taichi Sugaya , Takao Orito , dmaengine@vger.kernel.org, linux-arm-kernel@lists.infradead.org, kernel@pengutronix.de Subject: [PATCH 2/4] dmaengine: milbeaut-xdmac: Convert to platform remove callback returning void Date: Sun, 5 Nov 2023 10:34:18 +0100 Message-ID: <20231105093415.3704633-8-u.kleine-koenig@pengutronix.de> X-Mailer: git-send-email 2.42.0 In-Reply-To: <20231105093415.3704633-6-u.kleine-koenig@pengutronix.de> References: <20231105093415.3704633-6-u.kleine-koenig@pengutronix.de> Precedence: bulk X-Mailing-List: dmaengine@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 X-Developer-Signature: v=1; a=openpgp-sha256; l=2669; i=u.kleine-koenig@pengutronix.de; h=from:subject; bh=PbTnNrJDECIxqzCowQYTG7HFF0GLK6cy+LDhe3gdmRM=; b=owEBbQGS/pANAwAKAY+A+1h9Ev5OAcsmYgBlR2GZ83Sqgj3JZP8ncGZMRdJRTSgBLbsCK8wb9 D782Rkc+YiJATMEAAEKAB0WIQQ/gaxpOnoeWYmt/tOPgPtYfRL+TgUCZUdhmQAKCRCPgPtYfRL+ TrQ9B/0XEpv/vfeRJuDjJmweKk5xpOaMrcfLnGo85wUt+cIQWwBRcxfKUZaLtWvJuJez4wSDy+f hifDhyu/ARJ9H8TsLmngx0zOGyoufDSAsC4BJwcZkK53pNbdBtpM/9XIIA29oMlcpTTR6CNwTwH g4JwtQGbng7AyEyst/imEkMoSTlZ03bbDCLhZ1hOeev+fQI9IbAPIBznYJmMVnpnVxcpd0O4Q4X 4YK8KYdxk/VwcmcFtiwLXb04zJWZhen6iikra1KvDlx+njGFPtv4b3Bk2X4gB5o0jLNNee/ZdvC K8xbiQwD6gpIRfKHkMYszAO+7Y0dbKb6d/fxp7HbiR5gzxc1 X-Developer-Key: i=u.kleine-koenig@pengutronix.de; a=openpgp; fpr=0D2511F322BFAB1C1580266BE2DCDD9132669BD6 X-SA-Exim-Connect-IP: 2a0a:edc0:0:c01:1d::a2 X-SA-Exim-Mail-From: ukl@pengutronix.de X-SA-Exim-Scanned: No (on metis.whiteo.stw.pengutronix.de); SAEximRunCond expanded to false X-PTX-Original-Recipient: dmaengine@vger.kernel.org The .remove() callback for a platform driver returns an int which makes many driver authors wrongly assume it's possible to do error handling by returning an error code. However the value returned is ignored (apart from emitting a warning) and this typically results in resource leaks. To improve here there is a quest to make the remove callback return void. In the first step of this quest all drivers are converted to .remove_new(), which already returns void. Eventually after all drivers are converted, .remove_new() will be renamed to .remove(). There is an error path that has the above mentioned problem. This patch only adds a more drastic error message. To properly fix it, dmaengine_terminate_sync() must be known to have succeeded (or that it's safe to not call it as other drivers seem to assume). Signed-off-by: Uwe Kleine-König --- drivers/dma/milbeaut-xdmac.c | 17 +++++++++++------ 1 file changed, 11 insertions(+), 6 deletions(-) diff --git a/drivers/dma/milbeaut-xdmac.c b/drivers/dma/milbeaut-xdmac.c index d29d01e730aa..2cce529b448e 100644 --- a/drivers/dma/milbeaut-xdmac.c +++ b/drivers/dma/milbeaut-xdmac.c @@ -368,7 +368,7 @@ static int milbeaut_xdmac_probe(struct platform_device *pdev) return ret; } -static int milbeaut_xdmac_remove(struct platform_device *pdev) +static void milbeaut_xdmac_remove(struct platform_device *pdev) { struct milbeaut_xdmac_device *mdev = platform_get_drvdata(pdev); struct dma_chan *chan; @@ -383,8 +383,15 @@ static int milbeaut_xdmac_remove(struct platform_device *pdev) */ list_for_each_entry(chan, &mdev->ddev.channels, device_node) { ret = dmaengine_terminate_sync(chan); - if (ret) - return ret; + if (ret) { + /* + * This results in resource leakage and maybe also + * use-after-free errors as e.g. *mdev is kfreed. + */ + dev_alert(&pdev->dev, "Failed to terminate channel %d (%pe)\n", + chan->chan_id, ERR_PTR(ret)); + return; + } milbeaut_xdmac_free_chan_resources(chan); } @@ -392,8 +399,6 @@ static int milbeaut_xdmac_remove(struct platform_device *pdev) dma_async_device_unregister(&mdev->ddev); disable_xdmac(mdev); - - return 0; } static const struct of_device_id milbeaut_xdmac_match[] = { @@ -404,7 +409,7 @@ MODULE_DEVICE_TABLE(of, milbeaut_xdmac_match); static struct platform_driver milbeaut_xdmac_driver = { .probe = milbeaut_xdmac_probe, - .remove = milbeaut_xdmac_remove, + .remove_new = milbeaut_xdmac_remove, .driver = { .name = "milbeaut-m10v-xdmac", .of_match_table = milbeaut_xdmac_match, From patchwork Sun Nov 5 09:34:19 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: =?utf-8?q?Uwe_Kleine-K=C3=B6nig?= X-Patchwork-Id: 13445673 Received: from lindbergh.monkeyblade.net (lindbergh.monkeyblade.net [23.128.96.19]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 3407C8C13 for ; Sun, 5 Nov 2023 09:34:43 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; dkim=none Received: from metis.whiteo.stw.pengutronix.de (metis.whiteo.stw.pengutronix.de [IPv6:2a0a:edc0:2:b01:1d::104]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 8EF0BB7 for ; Sun, 5 Nov 2023 01:34:42 -0800 (PST) Received: from drehscheibe.grey.stw.pengutronix.de ([2a0a:edc0:0:c01:1d::a2]) by metis.whiteo.stw.pengutronix.de with esmtps (TLS1.3:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.92) (envelope-from ) id 1qzZWd-0000Pf-T9; Sun, 05 Nov 2023 10:34:35 +0100 Received: from [2a0a:edc0:0:900:1d::77] (helo=ptz.office.stw.pengutronix.de) by drehscheibe.grey.stw.pengutronix.de with esmtps (TLS1.3) tls TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384 (Exim 4.94.2) (envelope-from ) id 1qzZWc-006l7O-Ak; Sun, 05 Nov 2023 10:34:34 +0100 Received: from ukl by ptz.office.stw.pengutronix.de with local (Exim 4.94.2) (envelope-from ) id 1qzZWc-00D8nn-1o; Sun, 05 Nov 2023 10:34:34 +0100 From: =?utf-8?q?Uwe_Kleine-K=C3=B6nig?= To: Vinod Koul Cc: Kunihiko Hayashi , Masami Hiramatsu , dmaengine@vger.kernel.org, linux-arm-kernel@lists.infradead.org, kernel@pengutronix.de Subject: [PATCH 3/4] dmaengine: uniphier-mdmac: Convert to platform remove callback returning void Date: Sun, 5 Nov 2023 10:34:19 +0100 Message-ID: <20231105093415.3704633-9-u.kleine-koenig@pengutronix.de> X-Mailer: git-send-email 2.42.0 In-Reply-To: <20231105093415.3704633-6-u.kleine-koenig@pengutronix.de> References: <20231105093415.3704633-6-u.kleine-koenig@pengutronix.de> Precedence: bulk X-Mailing-List: dmaengine@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 X-Developer-Signature: v=1; a=openpgp-sha256; l=2640; i=u.kleine-koenig@pengutronix.de; h=from:subject; bh=KteA0sSG4oioIIBmCXZuIP2i8ZzuLBd1ds3YZc/Wpw0=; b=owEBbQGS/pANAwAKAY+A+1h9Ev5OAcsmYgBlR2GbzQegeQgG49nVuLEjV+AkhupybybvP492w XTJhuZ4chaJATMEAAEKAB0WIQQ/gaxpOnoeWYmt/tOPgPtYfRL+TgUCZUdhmwAKCRCPgPtYfRL+ TpE/CAC1VKU/vpTuUqO+Gq5U0eCAqLliLMByCCgKcNlEW7RJfqye/3E8kl9IZrLo4FhrRcDAuqX wJIlwLTkPVwvtFxcfVl+bu6b28zsv1052aqlS7OcMXl3be+Eu2LmG5c5fMvamZHoLVVjKve90pB 6vVmBX4VG8jDfHuB9oyEhud35+OYkI8eF51lq+wuyo80SCSeGDYW/eBOjQ91UHs0ALmQyuZvjpO QFBLlmPUE677TisT5Qc/rl9+ORNYUmv3ww2bQizxoryjgFj+bfbSlMRB7vtoDxHkmt4Ey3Xm1KI wn8ULnJNoFKrGQf/FmaHtLYpe+T43FVwZXvC21/u0tdxws4m X-Developer-Key: i=u.kleine-koenig@pengutronix.de; a=openpgp; fpr=0D2511F322BFAB1C1580266BE2DCDD9132669BD6 X-SA-Exim-Connect-IP: 2a0a:edc0:0:c01:1d::a2 X-SA-Exim-Mail-From: ukl@pengutronix.de X-SA-Exim-Scanned: No (on metis.whiteo.stw.pengutronix.de); SAEximRunCond expanded to false X-PTX-Original-Recipient: dmaengine@vger.kernel.org The .remove() callback for a platform driver returns an int which makes many driver authors wrongly assume it's possible to do error handling by returning an error code. However the value returned is ignored (apart from emitting a warning) and this typically results in resource leaks. To improve here there is a quest to make the remove callback return void. In the first step of this quest all drivers are converted to .remove_new(), which already returns void. Eventually after all drivers are converted, .remove_new() will be renamed to .remove(). There is an error path that has the above mentioned problem. This patch only adds a more drastic error message. To properly fix it, dmaengine_terminate_sync() must be known to have succeeded (or that it's safe to not call it as other drivers seem to assume). Signed-off-by: Uwe Kleine-König --- drivers/dma/uniphier-mdmac.c | 17 +++++++++++------ 1 file changed, 11 insertions(+), 6 deletions(-) diff --git a/drivers/dma/uniphier-mdmac.c b/drivers/dma/uniphier-mdmac.c index 618839df0748..ad7125f6e2ca 100644 --- a/drivers/dma/uniphier-mdmac.c +++ b/drivers/dma/uniphier-mdmac.c @@ -453,7 +453,7 @@ static int uniphier_mdmac_probe(struct platform_device *pdev) return ret; } -static int uniphier_mdmac_remove(struct platform_device *pdev) +static void uniphier_mdmac_remove(struct platform_device *pdev) { struct uniphier_mdmac_device *mdev = platform_get_drvdata(pdev); struct dma_chan *chan; @@ -468,16 +468,21 @@ static int uniphier_mdmac_remove(struct platform_device *pdev) */ list_for_each_entry(chan, &mdev->ddev.channels, device_node) { ret = dmaengine_terminate_sync(chan); - if (ret) - return ret; + if (ret) { + /* + * This results in resource leakage and maybe also + * use-after-free errors as e.g. *mdev is kfreed. + */ + dev_alert(&pdev->dev, "Failed to terminate channel %d (%pe)\n", + chan->chan_id, ERR_PTR(ret)); + return; + } uniphier_mdmac_free_chan_resources(chan); } of_dma_controller_free(pdev->dev.of_node); dma_async_device_unregister(&mdev->ddev); clk_disable_unprepare(mdev->clk); - - return 0; } static const struct of_device_id uniphier_mdmac_match[] = { @@ -488,7 +493,7 @@ MODULE_DEVICE_TABLE(of, uniphier_mdmac_match); static struct platform_driver uniphier_mdmac_driver = { .probe = uniphier_mdmac_probe, - .remove = uniphier_mdmac_remove, + .remove_new = uniphier_mdmac_remove, .driver = { .name = "uniphier-mio-dmac", .of_match_table = uniphier_mdmac_match, From patchwork Sun Nov 5 09:34:20 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: =?utf-8?q?Uwe_Kleine-K=C3=B6nig?= X-Patchwork-Id: 13445676 Received: from lindbergh.monkeyblade.net (lindbergh.monkeyblade.net [23.128.96.19]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 22F678F4B for ; Sun, 5 Nov 2023 09:34:45 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; dkim=none Received: from metis.whiteo.stw.pengutronix.de (metis.whiteo.stw.pengutronix.de [IPv6:2a0a:edc0:2:b01:1d::104]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id A89D8EE for ; Sun, 5 Nov 2023 01:34:43 -0800 (PST) Received: from drehscheibe.grey.stw.pengutronix.de ([2a0a:edc0:0:c01:1d::a2]) by metis.whiteo.stw.pengutronix.de with esmtps (TLS1.3:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.92) (envelope-from ) id 1qzZWd-0000Pg-T9; Sun, 05 Nov 2023 10:34:35 +0100 Received: from [2a0a:edc0:0:900:1d::77] (helo=ptz.office.stw.pengutronix.de) by drehscheibe.grey.stw.pengutronix.de with esmtps (TLS1.3) tls TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384 (Exim 4.94.2) (envelope-from ) id 1qzZWc-006l7R-Hl; Sun, 05 Nov 2023 10:34:34 +0100 Received: from ukl by ptz.office.stw.pengutronix.de with local (Exim 4.94.2) (envelope-from ) id 1qzZWc-00D8nr-8c; Sun, 05 Nov 2023 10:34:34 +0100 From: =?utf-8?q?Uwe_Kleine-K=C3=B6nig?= To: Vinod Koul Cc: Kunihiko Hayashi , Masami Hiramatsu , dmaengine@vger.kernel.org, linux-arm-kernel@lists.infradead.org, kernel@pengutronix.de Subject: [PATCH 4/4] dmaengine: uniphier-xdmac: Convert to platform remove callback returning void Date: Sun, 5 Nov 2023 10:34:20 +0100 Message-ID: <20231105093415.3704633-10-u.kleine-koenig@pengutronix.de> X-Mailer: git-send-email 2.42.0 In-Reply-To: <20231105093415.3704633-6-u.kleine-koenig@pengutronix.de> References: <20231105093415.3704633-6-u.kleine-koenig@pengutronix.de> Precedence: bulk X-Mailing-List: dmaengine@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 X-Developer-Signature: v=1; a=openpgp-sha256; l=2604; i=u.kleine-koenig@pengutronix.de; h=from:subject; bh=Wsv/SCBHHBOzEUfGKUTR96iDjLgTDVwgqqZx8UGatP8=; b=owEBbQGS/pANAwAKAY+A+1h9Ev5OAcsmYgBlR2Gc9/4L+wKj/bFGGeUlDQohLCMOow4pA/CO+ 7vLAdGsk1uJATMEAAEKAB0WIQQ/gaxpOnoeWYmt/tOPgPtYfRL+TgUCZUdhnAAKCRCPgPtYfRL+ TmXXCACx8B/yk11rm8+3h9qfsjq5+yW6EpK/1olvOKBg0lRSjNXwixyw8zlSPXzQQgGS8rUyNKl xkckt8x8XsfPi0qgDxJxxVrdIC/6y+chfwaerRjzvO+u3LIJaDtb9SuD9DlVO9hDGZZQJtro/R9 LHQ6AocPWjDbi8FWiLhvnURNNVqGQibcOXmPlY8glS+IM+/lr2HNrBR+g/mUD0gAyYe1jr/tw96 HIESy72063wNt8fqrmWgk+3g4kFFVgEAY0iL/CEiMNJStXw5ZpjkbqqYaOq5ysAWi+xMxfEPkqZ qm9bROZ3Lt//0qpfE0QQ70zIbOkFOI1/Yo/MzA+z3ZKwpHqS X-Developer-Key: i=u.kleine-koenig@pengutronix.de; a=openpgp; fpr=0D2511F322BFAB1C1580266BE2DCDD9132669BD6 X-SA-Exim-Connect-IP: 2a0a:edc0:0:c01:1d::a2 X-SA-Exim-Mail-From: ukl@pengutronix.de X-SA-Exim-Scanned: No (on metis.whiteo.stw.pengutronix.de); SAEximRunCond expanded to false X-PTX-Original-Recipient: dmaengine@vger.kernel.org The .remove() callback for a platform driver returns an int which makes many driver authors wrongly assume it's possible to do error handling by returning an error code. However the value returned is ignored (apart from emitting a warning) and this typically results in resource leaks. To improve here there is a quest to make the remove callback return void. In the first step of this quest all drivers are converted to .remove_new(), which already returns void. Eventually after all drivers are converted, .remove_new() will be renamed to .remove(). There is an error path that has the above mentioned problem. This patch only adds a more drastic error message. To properly fix it, dmaengine_terminate_sync() must be known to have succeeded (or that it's safe to not call it as other drivers seem to assume). Signed-off-by: Uwe Kleine-König --- drivers/dma/uniphier-xdmac.c | 17 +++++++++++------ 1 file changed, 11 insertions(+), 6 deletions(-) diff --git a/drivers/dma/uniphier-xdmac.c b/drivers/dma/uniphier-xdmac.c index 3a8ee2b173b5..3ce2dc2ad9de 100644 --- a/drivers/dma/uniphier-xdmac.c +++ b/drivers/dma/uniphier-xdmac.c @@ -563,7 +563,7 @@ static int uniphier_xdmac_probe(struct platform_device *pdev) return ret; } -static int uniphier_xdmac_remove(struct platform_device *pdev) +static void uniphier_xdmac_remove(struct platform_device *pdev) { struct uniphier_xdmac_device *xdev = platform_get_drvdata(pdev); struct dma_device *ddev = &xdev->ddev; @@ -579,15 +579,20 @@ static int uniphier_xdmac_remove(struct platform_device *pdev) */ list_for_each_entry(chan, &ddev->channels, device_node) { ret = dmaengine_terminate_sync(chan); - if (ret) - return ret; + if (ret) { + /* + * This results in resource leakage and maybe also + * use-after-free errors as e.g. *xdev is kfreed. + */ + dev_alert(&pdev->dev, "Failed to terminate channel %d (%pe)\n", + chan->chan_id, ERR_PTR(ret)); + return; + } uniphier_xdmac_free_chan_resources(chan); } of_dma_controller_free(pdev->dev.of_node); dma_async_device_unregister(ddev); - - return 0; } static const struct of_device_id uniphier_xdmac_match[] = { @@ -598,7 +603,7 @@ MODULE_DEVICE_TABLE(of, uniphier_xdmac_match); static struct platform_driver uniphier_xdmac_driver = { .probe = uniphier_xdmac_probe, - .remove = uniphier_xdmac_remove, + .remove_new = uniphier_xdmac_remove, .driver = { .name = "uniphier-xdmac", .of_match_table = uniphier_xdmac_match,