* [PATCH v3] media: saa7164: add missing ioremap error handling @ 2026-03-12 13:32 Wang Jun 2026-03-12 15:28 ` Markus Elfring 0 siblings, 1 reply; 12+ messages in thread From: Wang Jun @ 2026-03-12 13:32 UTC (permalink / raw) To: Mauro Carvalho Chehab, Wang Jun, linux-media, linux-kernel Cc: gszhai, 25125332, 25125283, 23120469 Add checks for ioremap return values in saa7164_dev_setup(). If ioremap for BAR0 or BAR2 fails, release the already allocated PCI memory regions, remove the device from the global list, decrement the device count, and return -ENODEV. This prevents potential null pointer dereferences and ensures proper cleanup on memory mapping failures. Signed-off-by: Wang Jun <1742789905@qq.com> --- drivers/media/pci/saa7164/saa7164-core.c | 29 ++++++++++++++++++++++++ 1 file changed, 29 insertions(+) diff --git a/drivers/media/pci/saa7164/saa7164-core.c b/drivers/media/pci/saa7164/saa7164-core.c index a8a004f28ca0..dc68d7cac7cf 100644 --- a/drivers/media/pci/saa7164/saa7164-core.c +++ b/drivers/media/pci/saa7164/saa7164-core.c @@ -998,9 +998,21 @@ static int saa7164_dev_setup(struct saa7164_dev *dev) /* PCI/e allocations */ dev->lmmio = ioremap(pci_resource_start(dev->pci, 0), pci_resource_len(dev->pci, 0)); + if (!dev->lmmio) { + dev_err(&dev->pci->dev, + "failed to remap MMIO memory @ 0x%llx\n", + (u64)pci_resource_start(dev->pci, 0)); + goto err_ioremap; + } dev->lmmio2 = ioremap(pci_resource_start(dev->pci, 2), pci_resource_len(dev->pci, 2)); + if (!dev->lmmio2) { + dev_err(&dev->pci->dev, + "failed to remap MMIO memory @ 0x%llx\n", + (u64)pci_resource_start(dev->pci, 2)); + goto err_ioremap2; + } dev->bmmio = (u8 __iomem *)dev->lmmio; dev->bmmio2 = (u8 __iomem *)dev->lmmio2; @@ -1019,6 +1031,23 @@ static int saa7164_dev_setup(struct saa7164_dev *dev) saa7164_pci_quirks(dev); return 0; + +err_ioremap2: + iounmap(dev->lmmio); +err_ioremap: + /* Release the PCI memory regions allocated in get_resources() */ + release_mem_region(pci_resource_start(dev->pci, 0), + pci_resource_len(dev->pci, 0)); + release_mem_region(pci_resource_start(dev->pci, 2), + pci_resource_len(dev->pci, 2)); + + /* Remove from device list and decrement count */ + mutex_lock(&devlist); + list_del(&dev->devlist); + mutex_unlock(&devlist); + saa7164_devcount--; + + return -ENODEV; } static void saa7164_dev_unregister(struct saa7164_dev *dev) -- 2.43.0 ^ permalink raw reply related [flat|nested] 12+ messages in thread
* Re: [PATCH v3] media: saa7164: add missing ioremap error handling 2026-03-12 13:32 [PATCH v3] media: saa7164: add missing ioremap error handling Wang Jun @ 2026-03-12 15:28 ` Markus Elfring 2026-03-13 8:08 ` [PATCH v4 0/1] media: saa7164: add ioremap return checks and cleanups Wang Jun ` (9 more replies) 0 siblings, 10 replies; 12+ messages in thread From: Markus Elfring @ 2026-03-12 15:28 UTC (permalink / raw) To: Wang Jun, linux-media, Mauro Carvalho Chehab Cc: LKML, kernel-janitors, gszhai, 23120469, 25125332, 25125283 > Add checks for ioremap return values in saa7164_dev_setup(). If > ioremap for BAR0 or BAR2 fails, release the already allocated PCI > memory regions, remove the device from the global list, decrement > the device count, and return -ENODEV. See also once more: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/Documentation/process/submitting-patches.rst?h=v7.0-rc3#n659 > This prevents potential null pointer dereferences and ensures proper > cleanup on memory mapping failures. How do you think about to add any tags (like “Fixes” and “Cc”) accordingly? https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/Documentation/process/submitting-patches.rst?h=v7.0-rc3#n145 … > +++ b/drivers/media/pci/saa7164/saa7164-core.c > @@ -998,9 +998,21 @@ static int saa7164_dev_setup(struct saa7164_dev *dev) > /* PCI/e allocations */ > dev->lmmio = ioremap(pci_resource_start(dev->pci, 0), > pci_resource_len(dev->pci, 0)); > + if (!dev->lmmio) { > + dev_err(&dev->pci->dev, > + "failed to remap MMIO memory @ 0x%llx\n", > + (u64)pci_resource_start(dev->pci, 0)); > + goto err_ioremap; > + } > > dev->lmmio2 = ioremap(pci_resource_start(dev->pci, 2), > pci_resource_len(dev->pci, 2)); … Would you like to avoid duplicate source code here? … > @@ -1019,6 +1031,23 @@ static int saa7164_dev_setup(struct saa7164_dev *dev) > saa7164_pci_quirks(dev); > > return 0; … > + /* Remove from device list and decrement count */ > + mutex_lock(&devlist); > + list_del(&dev->devlist); > + mutex_unlock(&devlist); > + saa7164_devcount--; … Will development interests grow to apply a call like “scoped_guard(mutex, &devlist)”? https://elixir.bootlin.com/linux/v7.0-rc3/source/include/linux/mutex.h#L253 Regards, Markus ^ permalink raw reply [flat|nested] 12+ messages in thread
* [PATCH v4 0/1] media: saa7164: add ioremap return checks and cleanups 2026-03-12 15:28 ` Markus Elfring @ 2026-03-13 8:08 ` Wang Jun [not found] ` <20260313080845.169663-1-1742789905@qq.com> ` (8 subsequent siblings) 9 siblings, 0 replies; 12+ messages in thread From: Wang Jun @ 2026-03-13 8:08 UTC (permalink / raw) To: Mauro Carvalho Chehab, Wang Jun, linux-media, linux-kernel Cc: gszhai, 25125332, 25125283, 23120469 Hi Markus, Thank you for your thorough review of v3. This v4 addresses all your comments: - Added `Cc: stable@vger.kernel.org` to ensure the fix reaches stable kernels. - Refactored the duplicate ioremap error handling into a helper function to improve code clarity and reduce duplication. - Replaced manual mutex lock/unlock with `scoped_guard()` for better safety and readability. Regarding the `Fixes` tag: I've added a tag pointing to the commit that originally introduced the driver. At that time, the file was located at drivers/media/video/saa7164/saa7164-core.c; the `Fixes` tag correctly identifies the commit where the issue first appeared, regardless of the later file move. The patch adds missing error checks for two ioremap calls in saa7164_dev_setup(). If either mapping fails, the function now properly releases previously allocated PCI resources, removes the device from the global list, and returns -ENODEV. This prevents potential null pointer dereferences and ensures proper cleanup on failure. Please review the updated patch. Thanks! Wang Jun (1): media: saa7164: add ioremap return checks and cleanups drivers/media/pci/saa7164/saa7164-core.c | 41 +++++++++++++++++++++--- 1 file changed, 37 insertions(+), 4 deletions(-) -- 2.43.0 ^ permalink raw reply [flat|nested] 12+ messages in thread
[parent not found: <20260313080845.169663-1-1742789905@qq.com>]
* [PATCH v4] media: saa7164: add ioremap return checks and cleanups [not found] ` <20260313080845.169663-1-1742789905@qq.com> @ 2026-03-13 8:08 ` Wang Jun 0 siblings, 0 replies; 12+ messages in thread From: Wang Jun @ 2026-03-13 8:08 UTC (permalink / raw) To: Mauro Carvalho Chehab, Wang Jun, linux-media, linux-kernel Cc: gszhai, 25125332, 25125283, 23120469, stable Add checks for ioremap return values in saa7164_dev_setup(). If ioremap for BAR0 or BAR2 fails, release the already allocated PCI memory regions, remove the device from the global list, decrement the device count, and return -ENODEV. This prevents potential null pointer dereferences and ensures proper cleanup on memory mapping failures. Fixes: 443c1228d505 ("V4L/DVB (12923): SAA7164: Add support for the NXP SAA7164 silicon") Cc: stable@vger.kernel.org Signed-off-by: Wang Jun <1742789905@qq.com> --- drivers/media/pci/saa7164/saa7164-core.c | 47 ++++++++++++++++++------ 1 file changed, 35 insertions(+), 12 deletions(-) diff --git a/drivers/media/pci/saa7164/saa7164-core.c b/drivers/media/pci/saa7164/saa7164-core.c index 74406d5ea0a5..47b0594042c6 100644 --- a/drivers/media/pci/saa7164/saa7164-core.c +++ b/drivers/media/pci/saa7164/saa7164-core.c @@ -888,6 +888,15 @@ static int get_resources(struct saa7164_dev *dev) return -EBUSY; } +static void release_resources(struct saa7164_dev *dev) +{ + release_mem_region(pci_resource_start(dev->pci, 0), + pci_resource_len(dev->pci, 0)); + + release_mem_region(pci_resource_start(dev->pci, 2), + pci_resource_len(dev->pci, 2)); +} + static int saa7164_port_init(struct saa7164_dev *dev, int portnr) { struct saa7164_port *port = NULL; @@ -947,9 +956,9 @@ static int saa7164_dev_setup(struct saa7164_dev *dev) snprintf(dev->name, sizeof(dev->name), "saa7164[%d]", dev->nr); - mutex_lock(&devlist); - list_add_tail(&dev->devlist, &saa7164_devlist); - mutex_unlock(&devlist); + scoped_guard(mutex, &devlist) { + list_add_tail(&dev->devlist, &saa7164_devlist); + } /* board config */ dev->board = UNSET; @@ -996,11 +1005,17 @@ static int saa7164_dev_setup(struct saa7164_dev *dev) } /* PCI/e allocations */ - dev->lmmio = ioremap(pci_resource_start(dev->pci, 0), - pci_resource_len(dev->pci, 0)); + dev->immio = pci_ioremap_bar(dev->pci_dev, 0); + if (!dev->immio) { + dev_err(&dev->pci_dev->dev, "Failed to remap MMIO BAR 0\n"); + goto err_ioremap_bar0; + } - dev->lmmio2 = ioremap(pci_resource_start(dev->pci, 2), - pci_resource_len(dev->pci, 2)); + dev->lmmio = pci_ioremap_bar(dev->pci_dev, 2); + if (!dev->lmmio) { + dev_err(&dev->pci_dev->dev, "Failed to remap MMIO BAR 2\n"); + goto err_ioremap_bar2; + } dev->bmmio = (u8 __iomem *)dev->lmmio; dev->bmmio2 = (u8 __iomem *)dev->lmmio2; @@ -1019,17 +1034,25 @@ static int saa7164_dev_setup(struct saa7164_dev *dev) saa7164_pci_quirks(dev); return 0; + +err_ioremap_bar2: + iounmap(dev->lmmio); +err_ioremap_bar0: + release_resources(dev); + + scoped_guard(mutex, &devlist) { + list_del(&dev->devlist); + } + saa7164_devcount--; + + return -ENODEV; } static void saa7164_dev_unregister(struct saa7164_dev *dev) { dprintk(1, "%s()\n", __func__); - release_mem_region(pci_resource_start(dev->pci, 0), - pci_resource_len(dev->pci, 0)); - - release_mem_region(pci_resource_start(dev->pci, 2), - pci_resource_len(dev->pci, 2)); + release_resources(dev); if (!atomic_dec_and_test(&dev->refcount)) return; -- 2.43.0 ^ permalink raw reply related [flat|nested] 12+ messages in thread
* [PATCH v5 0/1] media: saa7164: add ioremap return checks and cleanups 2026-03-12 15:28 ` Markus Elfring 2026-03-13 8:08 ` [PATCH v4 0/1] media: saa7164: add ioremap return checks and cleanups Wang Jun [not found] ` <20260313080845.169663-1-1742789905@qq.com> @ 2026-03-16 0:48 ` Wang Jun [not found] ` <20260316004854.1711-1-1742789905@qq.com> ` (6 subsequent siblings) 9 siblings, 0 replies; 12+ messages in thread From: Wang Jun @ 2026-03-16 0:48 UTC (permalink / raw) To: Mauro Carvalho Chehab, Wang Jun, linux-media, linux-kernel Cc: gszhai, 25125332, 25125283, 23120469 Hi Markus, Thank you for your thorough review of v3. This v5 addresses the issues reported by the Media CI robot after v4 was submitted: - Fixed build errors caused by incorrect struct member names (`immio` → `lmmio`, `pci_dev` → `pci`). - Adjusted code alignment to satisfy checkpatch.pl (two CHECK warnings about open parenthesis alignment). This v4 addresses all your comments: - Added `Cc: stable@vger.kernel.org` to ensure the fix reaches stable kernels. - Refactored the duplicate ioremap error handling into a helper function to improve code clarity and reduce duplication. - Replaced manual mutex lock/unlock with `scoped_guard()` for better safety and readability. Regarding the `Fixes` tag: I've added a tag pointing to the commit that originally introduced the driver. At that time, the file was located at drivers/media/video/saa7164/saa7164-core.c; the `Fixes` tag correctly identifies the commit where the issue first appeared, regardless of the later file move. The patch adds missing error checks for two ioremap calls in saa7164_dev_setup(). If either mapping fails, the function now properly releases previously allocated PCI resources, removes the device from the global list, and returns -ENODEV. This prevents potential null pointer dereferences and ensures proper cleanup on failure. Please review the updated patch. Thanks! Wang Jun (1): media: saa7164: add ioremap return checks and cleanups drivers/media/pci/saa7164/saa7164-core.c | 47 ++++++++++++++++++------ 1 file changed, 35 insertions(+), 12 deletions(-) -- 2.43.0 ^ permalink raw reply [flat|nested] 12+ messages in thread
[parent not found: <20260316004854.1711-1-1742789905@qq.com>]
* [PATCH v5 1/1] media: saa7164: add ioremap return checks and cleanups [not found] ` <20260316004854.1711-1-1742789905@qq.com> @ 2026-03-16 0:48 ` Wang Jun 0 siblings, 0 replies; 12+ messages in thread From: Wang Jun @ 2026-03-16 0:48 UTC (permalink / raw) To: Mauro Carvalho Chehab, Wang Jun, linux-media, linux-kernel Cc: gszhai, 25125332, 25125283, 23120469, stable Add checks for ioremap return values in saa7164_dev_setup(). If ioremap for BAR0 or BAR2 fails, release the already allocated PCI memory regions, remove the device from the global list, decrement the device count, and return -ENODEV. This prevents potential null pointer dereferences and ensures proper cleanup on memory mapping failures. Fixes: 443c1228d505 ("V4L/DVB (12923): SAA7164: Add support for the NXP SAA7164 silicon") Cc: stable@vger.kernel.org Signed-off-by: Wang Jun <1742789905@qq.com> --- drivers/media/pci/saa7164/saa7164-core.c | 47 ++++++++++++++++++------ 1 file changed, 35 insertions(+), 12 deletions(-) diff --git a/drivers/media/pci/saa7164/saa7164-core.c b/drivers/media/pci/saa7164/saa7164-core.c index 74406d5ea0a5..b410e24c403c 100644 --- a/drivers/media/pci/saa7164/saa7164-core.c +++ b/drivers/media/pci/saa7164/saa7164-core.c @@ -888,6 +888,15 @@ static int get_resources(struct saa7164_dev *dev) return -EBUSY; } +static void release_resources(struct saa7164_dev *dev) +{ + release_mem_region(pci_resource_start(dev->pci, 0), + pci_resource_len(dev->pci, 0)); + + release_mem_region(pci_resource_start(dev->pci, 2), + pci_resource_len(dev->pci, 2)); +} + static int saa7164_port_init(struct saa7164_dev *dev, int portnr) { struct saa7164_port *port = NULL; @@ -947,9 +956,9 @@ static int saa7164_dev_setup(struct saa7164_dev *dev) snprintf(dev->name, sizeof(dev->name), "saa7164[%d]", dev->nr); - mutex_lock(&devlist); - list_add_tail(&dev->devlist, &saa7164_devlist); - mutex_unlock(&devlist); + scoped_guard(mutex, &devlist) { + list_add_tail(&dev->devlist, &saa7164_devlist); + } /* board config */ dev->board = UNSET; @@ -996,11 +1005,17 @@ static int saa7164_dev_setup(struct saa7164_dev *dev) } /* PCI/e allocations */ - dev->lmmio = ioremap(pci_resource_start(dev->pci, 0), - pci_resource_len(dev->pci, 0)); + dev->lmmio = pci_ioremap_bar(dev->pci_dev, 0); + if (!dev->lmmio) { + dev_err(&dev->pci->dev, "Failed to remap MMIO BAR 0\n"); + goto err_ioremap_bar0; + } - dev->lmmio2 = ioremap(pci_resource_start(dev->pci, 2), - pci_resource_len(dev->pci, 2)); + dev->lmmio = pci_ioremap_bar(dev->pci_dev, 2); + if (!dev->lmmio) { + dev_err(&dev->pci->dev, "Failed to remap MMIO BAR 2\n"); + goto err_ioremap_bar2; + } dev->bmmio = (u8 __iomem *)dev->lmmio; dev->bmmio2 = (u8 __iomem *)dev->lmmio2; @@ -1019,17 +1034,25 @@ static int saa7164_dev_setup(struct saa7164_dev *dev) saa7164_pci_quirks(dev); return 0; + +err_ioremap_bar2: + iounmap(dev->lmmio); +err_ioremap_bar0: + release_resources(dev); + + scoped_guard(mutex, &devlist) { + list_del(&dev->devlist); + } + saa7164_devcount--; + + return -ENODEV; } static void saa7164_dev_unregister(struct saa7164_dev *dev) { dprintk(1, "%s()\n", __func__); - release_mem_region(pci_resource_start(dev->pci, 0), - pci_resource_len(dev->pci, 0)); - - release_mem_region(pci_resource_start(dev->pci, 2), - pci_resource_len(dev->pci, 2)); + release_resources(dev); if (!atomic_dec_and_test(&dev->refcount)) return; -- 2.43.0 ^ permalink raw reply related [flat|nested] 12+ messages in thread
* [PATCH v5 0/1] media: saa7164: add ioremap return checks and cleanups 2026-03-12 15:28 ` Markus Elfring ` (3 preceding siblings ...) [not found] ` <20260316004854.1711-1-1742789905@qq.com> @ 2026-03-16 6:09 ` Wang Jun [not found] ` <20260316060940.9659-1-1742789905@qq.com> ` (4 subsequent siblings) 9 siblings, 0 replies; 12+ messages in thread From: Wang Jun @ 2026-03-16 6:09 UTC (permalink / raw) To: Mauro Carvalho Chehab, Wang Jun, linux-media, linux-kernel Cc: gszhai, 25125332, 25125283, 23120469 Hi Markus, Thank you for your thorough review of v3. This v5 addresses the issues reported by the Media CI robot after v4 was submitted: - Fixed build errors caused by incorrect struct member names (`immio` → `lmmio`, `pci_dev` → `pci`). - Adjusted code alignment to satisfy checkpatch.pl (two CHECK warnings about open parenthesis alignment). This v4 addresses all your comments: - Added `Cc: stable@vger.kernel.org` to ensure the fix reaches stable kernels. - Refactored the duplicate ioremap error handling into a helper function to improve code clarity and reduce duplication. - Replaced manual mutex lock/unlock with `scoped_guard()` for better safety and readability. Regarding the `Fixes` tag: I've added a tag pointing to the commit that originally introduced the driver. At that time, the file was located at drivers/media/video/saa7164/saa7164-core.c; the `Fixes` tag correctly identifies the commit where the issue first appeared, regardless of the later file move. The patch adds missing error checks for two ioremap calls in saa7164_dev_setup(). If either mapping fails, the function now properly releases previously allocated PCI resources, removes the device from the global list, and returns -ENODEV. This prevents potential null pointer dereferences and ensures proper cleanup on failure. Please review the updated patch. Thanks! Wang Jun (1): media: saa7164: add ioremap return checks and cleanups drivers/media/pci/saa7164/saa7164-core.c | 47 ++++++++++++++++++------ 1 file changed, 35 insertions(+), 12 deletions(-) -- 2.43.0 ^ permalink raw reply [flat|nested] 12+ messages in thread
[parent not found: <20260316060940.9659-1-1742789905@qq.com>]
* [PATCH v5] media: saa7164: add ioremap return checks and cleanups [not found] ` <20260316060940.9659-1-1742789905@qq.com> @ 2026-03-16 6:09 ` Wang Jun 0 siblings, 0 replies; 12+ messages in thread From: Wang Jun @ 2026-03-16 6:09 UTC (permalink / raw) To: Mauro Carvalho Chehab, Wang Jun, linux-media, linux-kernel Cc: gszhai, 25125332, 25125283, 23120469, stable Add checks for ioremap return values in saa7164_dev_setup(). If ioremap for BAR0 or BAR2 fails, release the already allocated PCI memory regions, remove the device from the global list, decrement the device count, and return -ENODEV. This prevents potential null pointer dereferences and ensures proper cleanup on memory mapping failures. Fixes: 443c1228d505 ("V4L/DVB (12923): SAA7164: Add support for the NXP SAA7164 silicon") Cc: stable@vger.kernel.org Signed-off-by: Wang Jun <1742789905@qq.com> --- drivers/media/pci/saa7164/saa7164-core.c | 47 ++++++++++++++++++------ 1 file changed, 35 insertions(+), 12 deletions(-) diff --git a/drivers/media/pci/saa7164/saa7164-core.c b/drivers/media/pci/saa7164/saa7164-core.c index 74406d5ea0a5..5e9e85c6ddb4 100644 --- a/drivers/media/pci/saa7164/saa7164-core.c +++ b/drivers/media/pci/saa7164/saa7164-core.c @@ -888,6 +888,15 @@ static int get_resources(struct saa7164_dev *dev) return -EBUSY; } +static void release_resources(struct saa7164_dev *dev) +{ + release_mem_region(pci_resource_start(dev->pci, 0), + pci_resource_len(dev->pci, 0)); + + release_mem_region(pci_resource_start(dev->pci, 2), + pci_resource_len(dev->pci, 2)); +} + static int saa7164_port_init(struct saa7164_dev *dev, int portnr) { struct saa7164_port *port = NULL; @@ -947,9 +956,9 @@ static int saa7164_dev_setup(struct saa7164_dev *dev) snprintf(dev->name, sizeof(dev->name), "saa7164[%d]", dev->nr); - mutex_lock(&devlist); - list_add_tail(&dev->devlist, &saa7164_devlist); - mutex_unlock(&devlist); + scoped_guard(mutex, &devlist) { + list_add_tail(&dev->devlist, &saa7164_devlist); + } /* board config */ dev->board = UNSET; @@ -996,11 +1005,17 @@ static int saa7164_dev_setup(struct saa7164_dev *dev) } /* PCI/e allocations */ - dev->lmmio = ioremap(pci_resource_start(dev->pci, 0), - pci_resource_len(dev->pci, 0)); + dev->lmmio = pci_ioremap_bar(dev->pci, 0); + if (!dev->lmmio) { + dev_err(&dev->pci->dev, "Failed to remap MMIO BAR 0\n"); + goto err_ioremap_bar0; + } - dev->lmmio2 = ioremap(pci_resource_start(dev->pci, 2), - pci_resource_len(dev->pci, 2)); + dev->lmmio = pci_ioremap_bar(dev->pci, 2); + if (!dev->lmmio) { + dev_err(&dev->pci->dev, "Failed to remap MMIO BAR 2\n"); + goto err_ioremap_bar2; + } dev->bmmio = (u8 __iomem *)dev->lmmio; dev->bmmio2 = (u8 __iomem *)dev->lmmio2; @@ -1019,17 +1034,25 @@ static int saa7164_dev_setup(struct saa7164_dev *dev) saa7164_pci_quirks(dev); return 0; + +err_ioremap_bar2: + iounmap(dev->lmmio); +err_ioremap_bar0: + release_resources(dev); + + scoped_guard(mutex, &devlist) { + list_del(&dev->devlist); + } + saa7164_devcount--; + + return -ENODEV; } static void saa7164_dev_unregister(struct saa7164_dev *dev) { dprintk(1, "%s()\n", __func__); - release_mem_region(pci_resource_start(dev->pci, 0), - pci_resource_len(dev->pci, 0)); - - release_mem_region(pci_resource_start(dev->pci, 2), - pci_resource_len(dev->pci, 2)); + release_resources(dev); if (!atomic_dec_and_test(&dev->refcount)) return; -- 2.43.0 ^ permalink raw reply related [flat|nested] 12+ messages in thread
* [PATCH v6 0/1] media: saa7164: add ioremap return checks and cleanups 2026-03-12 15:28 ` Markus Elfring ` (5 preceding siblings ...) [not found] ` <20260316060940.9659-1-1742789905@qq.com> @ 2026-03-16 9:16 ` Wang Jun [not found] ` <20260316091649.14827-1-1742789905@qq.com> ` (2 subsequent siblings) 9 siblings, 0 replies; 12+ messages in thread From: Wang Jun @ 2026-03-16 9:16 UTC (permalink / raw) To: Mauro Carvalho Chehab, Wang Jun, linux-media, linux-kernel Cc: gszhai, 25125332, 25125283, 23120469 Hi Markus, This v4 addressed all your comments: - Added `Cc: stable@vger.kernel.org` to ensure the fix reaches stable kernels. - Refactored the duplicate ioremap error handling into a helper function to improve code clarity and reduce duplication. - Replaced manual mutex lock/unlock with `scoped_guard()` for better safety and readability. After submitting v4 (which became v5 in the patchwork series due to a version bump), the Media CI robot detected two issues: - Build errors caused by incorrect struct member names (`immio` → `lmmio`, `pci_dev` → `pci`). - Two checkpatch `CHECK` warnings about alignment of function call arguments (open parenthesis alignment). This **v6** fixes those issues: - Corrected the member names to match the actual struct definitions. - Adjusted the code alignment in the `release_resources()` helper to satisfy `checkpatch.pl --strict`. Regarding the `Fixes` tag: I've added a tag pointing to the commit that originally introduced the driver. At that time, the file was located at drivers/media/video/saa7164/saa7164-core.c; the `Fixes` tag correctly identifies the commit where the issue first appeared, regardless of the later file move. The patch adds missing error checks for two ioremap calls in saa7164_dev_setup(). If either mapping fails, the function now properly releases previously allocated PCI resources, removes the device from the global list, and returns -ENODEV. This prevents potential null pointer dereferences and ensures proper cleanup on failure. Please review the updated patch. Thanks! Wang Jun (1): media: saa7164: add ioremap return checks and cleanups drivers/media/pci/saa7164/saa7164-core.c | 47 ++++++++++++++++++------ 1 file changed, 35 insertions(+), 12 deletions(-) -- 2.43.0 ^ permalink raw reply [flat|nested] 12+ messages in thread
[parent not found: <20260316091649.14827-1-1742789905@qq.com>]
* [PATCH v6 1/1] media: saa7164: add ioremap return checks and cleanups [not found] ` <20260316091649.14827-1-1742789905@qq.com> @ 2026-03-16 9:16 ` Wang Jun 0 siblings, 0 replies; 12+ messages in thread From: Wang Jun @ 2026-03-16 9:16 UTC (permalink / raw) To: Mauro Carvalho Chehab, Wang Jun, linux-media, linux-kernel Cc: gszhai, 25125332, 25125283, 23120469, stable Add checks for ioremap return values in saa7164_dev_setup(). If ioremap for BAR0 or BAR2 fails, release the already allocated PCI memory regions, remove the device from the global list, decrement the device count, and return -ENODEV. This prevents potential null pointer dereferences and ensures proper cleanup on memory mapping failures. Fixes: 443c1228d505 ("V4L/DVB (12923): SAA7164: Add support for the NXP SAA7164 silicon") Cc: stable@vger.kernel.org Signed-off-by: Wang Jun <1742789905@qq.com> --- drivers/media/pci/saa7164/saa7164-core.c | 47 ++++++++++++++++++------ 1 file changed, 35 insertions(+), 12 deletions(-) diff --git a/drivers/media/pci/saa7164/saa7164-core.c b/drivers/media/pci/saa7164/saa7164-core.c index 74406d5ea0a5..5aea5ab34c3f 100644 --- a/drivers/media/pci/saa7164/saa7164-core.c +++ b/drivers/media/pci/saa7164/saa7164-core.c @@ -888,6 +888,15 @@ static int get_resources(struct saa7164_dev *dev) return -EBUSY; } +static void release_resources(struct saa7164_dev *dev) +{ + release_mem_region(pci_resource_start(dev->pci, 0), + pci_resource_len(dev->pci, 0)); + + release_mem_region(pci_resource_start(dev->pci, 2), + pci_resource_len(dev->pci, 2)); +} + static int saa7164_port_init(struct saa7164_dev *dev, int portnr) { struct saa7164_port *port = NULL; @@ -947,9 +956,9 @@ static int saa7164_dev_setup(struct saa7164_dev *dev) snprintf(dev->name, sizeof(dev->name), "saa7164[%d]", dev->nr); - mutex_lock(&devlist); - list_add_tail(&dev->devlist, &saa7164_devlist); - mutex_unlock(&devlist); + scoped_guard(mutex, &devlist) { + list_add_tail(&dev->devlist, &saa7164_devlist); + } /* board config */ dev->board = UNSET; @@ -996,11 +1005,17 @@ static int saa7164_dev_setup(struct saa7164_dev *dev) } /* PCI/e allocations */ - dev->lmmio = ioremap(pci_resource_start(dev->pci, 0), - pci_resource_len(dev->pci, 0)); + dev->lmmio = pci_ioremap_bar(dev->pci, 0); + if (!dev->lmmio) { + dev_err(&dev->pci->dev, "Failed to remap MMIO BAR 0\n"); + goto err_ioremap_bar0; + } - dev->lmmio2 = ioremap(pci_resource_start(dev->pci, 2), - pci_resource_len(dev->pci, 2)); + dev->lmmio = pci_ioremap_bar(dev->pci, 2); + if (!dev->lmmio) { + dev_err(&dev->pci->dev, "Failed to remap MMIO BAR 2\n"); + goto err_ioremap_bar2; + } dev->bmmio = (u8 __iomem *)dev->lmmio; dev->bmmio2 = (u8 __iomem *)dev->lmmio2; @@ -1019,17 +1034,25 @@ static int saa7164_dev_setup(struct saa7164_dev *dev) saa7164_pci_quirks(dev); return 0; + +err_ioremap_bar2: + iounmap(dev->lmmio); +err_ioremap_bar0: + release_resources(dev); + + scoped_guard(mutex, &devlist) { + list_del(&dev->devlist); + } + saa7164_devcount--; + + return -ENODEV; } static void saa7164_dev_unregister(struct saa7164_dev *dev) { dprintk(1, "%s()\n", __func__); - release_mem_region(pci_resource_start(dev->pci, 0), - pci_resource_len(dev->pci, 0)); - - release_mem_region(pci_resource_start(dev->pci, 2), - pci_resource_len(dev->pci, 2)); + release_resources(dev); if (!atomic_dec_and_test(&dev->refcount)) return; -- 2.43.0 ^ permalink raw reply related [flat|nested] 12+ messages in thread
* [PATCH v7 0/1] media: saa7164: add ioremap return checks and cleanups 2026-03-12 15:28 ` Markus Elfring ` (7 preceding siblings ...) [not found] ` <20260316091649.14827-1-1742789905@qq.com> @ 2026-03-16 12:24 ` Wang Jun [not found] ` <20260316122403.23337-1-1742789905@qq.com> 9 siblings, 0 replies; 12+ messages in thread From: Wang Jun @ 2026-03-16 12:24 UTC (permalink / raw) To: Mauro Carvalho Chehab, Wang Jun, linux-media, linux-kernel Cc: gszhai, 25125332, 25125283, 23120469 Hi Markus, This v4 addressed all your comments: - Added `Cc: stable@vger.kernel.org` to ensure the fix reaches stable kernels. - Refactored the duplicate ioremap error handling into a helper function to improve code clarity and reduce duplication. - Replaced manual mutex lock/unlock with `scoped_guard()` for better safety and readability. After submitting v4 (which became v5 in the patchwork series due to a version bump), the Media CI robot detected two issues: - Build errors caused by incorrect struct member names (`immio` → `lmmio`, `pci_dev` → `pci`). - Two checkpatch `CHECK` warnings about alignment of function call arguments (open parenthesis alignment). This v6 fixes those issues: - Corrected the member names to match the actual struct definitions. - Adjusted the code alignment in the `release_resources()` helper to satisfy `checkpatch.pl --strict`. This v7 addresses two additional issues that were identified in the v6 submission: Corrected a remaining instance where dev->lmmio was used instead of the correct dev->lmmio2 in the foo_bar() function. Regarding the `Fixes` tag: I've added a tag pointing to the commit that originally introduced the driver. At that time, the file was located at drivers/media/video/saa7164/saa7164-core.c; the `Fixes` tag correctly identifies the commit where the issue first appeared, regardless of the later file move. The patch adds missing error checks for two ioremap calls in saa7164_dev_setup(). If either mapping fails, the function now properly releases previously allocated PCI resources, removes the device from the global list, and returns -ENODEV. This prevents potential null pointer dereferences and ensures proper cleanup on failure. Please review the updated patch. Thanks! Wang Jun (1): media: saa7164: add ioremap return checks and cleanups drivers/media/pci/saa7164/saa7164-core.c | 47 ++++++++++++++++++------ 1 file changed, 35 insertions(+), 12 deletions(-) -- 2.43.0 ^ permalink raw reply [flat|nested] 12+ messages in thread
[parent not found: <20260316122403.23337-1-1742789905@qq.com>]
* [PATCH v7 1/1] media: saa7164: add ioremap return checks and cleanups [not found] ` <20260316122403.23337-1-1742789905@qq.com> @ 2026-03-16 12:24 ` Wang Jun 0 siblings, 0 replies; 12+ messages in thread From: Wang Jun @ 2026-03-16 12:24 UTC (permalink / raw) To: Mauro Carvalho Chehab, Wang Jun, linux-media, linux-kernel Cc: gszhai, 25125332, 25125283, 23120469, stable Add checks for ioremap return values in saa7164_dev_setup(). If ioremap for BAR0 or BAR2 fails, release the already allocated PCI memory regions, remove the device from the global list, decrement the device count, and return -ENODEV. This prevents potential null pointer dereferences and ensures proper cleanup on memory mapping failures. Fixes: 443c1228d505 ("V4L/DVB (12923): SAA7164: Add support for the NXP SAA7164 silicon") Cc: stable@vger.kernel.org Signed-off-by: Wang Jun <1742789905@qq.com> --- drivers/media/pci/saa7164/saa7164-core.c | 47 ++++++++++++++++++------ 1 file changed, 35 insertions(+), 12 deletions(-) diff --git a/drivers/media/pci/saa7164/saa7164-core.c b/drivers/media/pci/saa7164/saa7164-core.c index 74406d5ea0a5..e8037ac1db73 100644 --- a/drivers/media/pci/saa7164/saa7164-core.c +++ b/drivers/media/pci/saa7164/saa7164-core.c @@ -888,6 +888,15 @@ static int get_resources(struct saa7164_dev *dev) return -EBUSY; } +static void release_resources(struct saa7164_dev *dev) +{ + release_mem_region(pci_resource_start(dev->pci, 0), + pci_resource_len(dev->pci, 0)); + + release_mem_region(pci_resource_start(dev->pci, 2), + pci_resource_len(dev->pci, 2)); +} + static int saa7164_port_init(struct saa7164_dev *dev, int portnr) { struct saa7164_port *port = NULL; @@ -947,9 +956,9 @@ static int saa7164_dev_setup(struct saa7164_dev *dev) snprintf(dev->name, sizeof(dev->name), "saa7164[%d]", dev->nr); - mutex_lock(&devlist); - list_add_tail(&dev->devlist, &saa7164_devlist); - mutex_unlock(&devlist); + scoped_guard(mutex, &devlist) { + list_add_tail(&dev->devlist, &saa7164_devlist); + } /* board config */ dev->board = UNSET; @@ -996,11 +1005,17 @@ static int saa7164_dev_setup(struct saa7164_dev *dev) } /* PCI/e allocations */ - dev->lmmio = ioremap(pci_resource_start(dev->pci, 0), - pci_resource_len(dev->pci, 0)); + dev->lmmio = pci_ioremap_bar(dev->pci, 0); + if (!dev->lmmio) { + dev_err(&dev->pci->dev, "Failed to remap MMIO BAR 0\n"); + goto err_ioremap_bar0; + } - dev->lmmio2 = ioremap(pci_resource_start(dev->pci, 2), - pci_resource_len(dev->pci, 2)); + dev->lmmio2 = pci_ioremap_bar(dev->pci, 2); + if (!dev->lmmio2) { + dev_err(&dev->pci->dev, "Failed to remap MMIO BAR 2\n"); + goto err_ioremap_bar2; + } dev->bmmio = (u8 __iomem *)dev->lmmio; dev->bmmio2 = (u8 __iomem *)dev->lmmio2; @@ -1019,17 +1034,25 @@ static int saa7164_dev_setup(struct saa7164_dev *dev) saa7164_pci_quirks(dev); return 0; + +err_ioremap_bar2: + iounmap(dev->lmmio); +err_ioremap_bar0: + release_resources(dev); + + scoped_guard(mutex, &devlist) { + list_del(&dev->devlist); + } + saa7164_devcount--; + + return -ENODEV; } static void saa7164_dev_unregister(struct saa7164_dev *dev) { dprintk(1, "%s()\n", __func__); - release_mem_region(pci_resource_start(dev->pci, 0), - pci_resource_len(dev->pci, 0)); - - release_mem_region(pci_resource_start(dev->pci, 2), - pci_resource_len(dev->pci, 2)); + release_resources(dev); if (!atomic_dec_and_test(&dev->refcount)) return; -- 2.43.0 ^ permalink raw reply related [flat|nested] 12+ messages in thread
end of thread, other threads:[~2026-03-16 12:25 UTC | newest]
Thread overview: 12+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-03-12 13:32 [PATCH v3] media: saa7164: add missing ioremap error handling Wang Jun
2026-03-12 15:28 ` Markus Elfring
2026-03-13 8:08 ` [PATCH v4 0/1] media: saa7164: add ioremap return checks and cleanups Wang Jun
[not found] ` <20260313080845.169663-1-1742789905@qq.com>
2026-03-13 8:08 ` [PATCH v4] " Wang Jun
2026-03-16 0:48 ` [PATCH v5 0/1] " Wang Jun
[not found] ` <20260316004854.1711-1-1742789905@qq.com>
2026-03-16 0:48 ` [PATCH v5 1/1] " Wang Jun
2026-03-16 6:09 ` [PATCH v5 0/1] " Wang Jun
[not found] ` <20260316060940.9659-1-1742789905@qq.com>
2026-03-16 6:09 ` [PATCH v5] " Wang Jun
2026-03-16 9:16 ` [PATCH v6 0/1] " Wang Jun
[not found] ` <20260316091649.14827-1-1742789905@qq.com>
2026-03-16 9:16 ` [PATCH v6 1/1] " Wang Jun
2026-03-16 12:24 ` [PATCH v7 0/1] " Wang Jun
[not found] ` <20260316122403.23337-1-1742789905@qq.com>
2026-03-16 12:24 ` [PATCH v7 1/1] " Wang Jun
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox