From nobody Tue Dec 16 11:42:59 2025 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.129.124]) (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 B2CB5175B2 for ; Mon, 15 Jan 2024 14:47:17 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=redhat.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=redhat.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=redhat.com header.i=@redhat.com header.b="gLwHSP41" DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1705330036; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=wbybY9iQcnkB73xM1yKkpppr/vxHnl4OLedIb79V9wo=; b=gLwHSP41U//rWmV8owp25U3YaQNVgJFw9UbMi7yzgBlfcINPn9bMcpIHJufLiBuVHfJHQf 8N3bY16osyAem6UkPWakw8legmBZdBW3nfuOTT1NEjfE5xWL8+uNISZ11qtlVZ7aWLYQnu MDwZMkaQnwZ32+o01bH2flLVcs+GFYo= Received: from mail-qv1-f69.google.com (mail-qv1-f69.google.com [209.85.219.69]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.3, cipher=TLS_AES_256_GCM_SHA384) id us-mta-631-yiXFP30TNT28dBfq-SI91A-1; Mon, 15 Jan 2024 09:47:15 -0500 X-MC-Unique: yiXFP30TNT28dBfq-SI91A-1 Received: by mail-qv1-f69.google.com with SMTP id 6a1803df08f44-67f826cc03dso24819606d6.1 for ; Mon, 15 Jan 2024 06:47:15 -0800 (PST) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1705330034; x=1705934834; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=wbybY9iQcnkB73xM1yKkpppr/vxHnl4OLedIb79V9wo=; b=V/+8bgIcqXDLtrL5bklgRQaVpW3rroXaigl1BwnUdsXbb8WnMqIkN713COqkEOqspW +wHQzIUvIoPc1it+JaZLDe56ubvvAD1LmjBU9AKkw3tGi7FD/yteIcnGO7xqkSVJk1SJ mLmbNfMnuaMTA+ZTaBFnogZ1TJ6ujEzf8qbEGFHIYVbNp9/tsMu18xrqprwnqtbPLDVB xZDGoDlmJ91vDlYf+/FMiazfAH9m9ETUPGRdpvdlLZ8d8+EiT7Q4UE2LvxJjKx3DXnF8 AuCvYmxMiwvIhA4ALuQ6fg//5UyTZrGqhgwfODcDcmDPS2zHM9i02N/4R3fQGKI+fdhb 2H6A== X-Gm-Message-State: AOJu0YwwIsG6v2p0YsaPhfFZA7X3unBz3/v4nqqoOJjKKa54EkiRtj1l s5jMKsLlMZywOfglv+8SDEcvRj3esJs04LMdP/o1I14yXVR2qFiRTxFXbiEnv0ZJAaTSS38drAz D9gPFU6zqSvd2Blf5KYyTt54VkKprl52S X-Received: by 2002:a05:6214:1c45:b0:681:554a:a94b with SMTP id if5-20020a0562141c4500b00681554aa94bmr6344627qvb.4.1705330034304; Mon, 15 Jan 2024 06:47:14 -0800 (PST) X-Google-Smtp-Source: AGHT+IFp3HxThPTri3utY9IJvRlcM6dRy9DG1DBfIgbFThv8dyL1vLRK4mkJab5Ju7PxXxoqgvW/fA== X-Received: by 2002:a05:6214:1c45:b0:681:554a:a94b with SMTP id if5-20020a0562141c4500b00681554aa94bmr6344602qvb.4.1705330033916; Mon, 15 Jan 2024 06:47:13 -0800 (PST) Received: from pstanner-thinkpadt14sgen1.muc.redhat.com (nat-pool-muc-t.redhat.com. [149.14.88.26]) by smtp.gmail.com with ESMTPSA id ne13-20020a056214424d00b006815cf9a644sm1020720qvb.55.2024.01.15.06.47.11 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Mon, 15 Jan 2024 06:47:13 -0800 (PST) From: Philipp Stanner To: Jonathan Corbet , Hans de Goede , Maarten Lankhorst , Maxime Ripard , Thomas Zimmermann , David Airlie , Daniel Vetter , Bjorn Helgaas , Philipp Stanner , Sam Ravnborg , dakr@redhat.com Cc: linux-doc@vger.kernel.org, linux-kernel@vger.kernel.org, dri-devel@lists.freedesktop.org, linux-pci@vger.kernel.org Subject: [PATCH 01/10] pci: add new set of devres functions Date: Mon, 15 Jan 2024 15:46:12 +0100 Message-ID: <20240115144655.32046-3-pstanner@redhat.com> X-Mailer: git-send-email 2.43.0 In-Reply-To: <20240115144655.32046-2-pstanner@redhat.com> References: <20240115144655.32046-2-pstanner@redhat.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" PCI's devres API is not extensible to ranged mappings and has bug-provoking features. Improve that by providing better alternatives. When the original devres API for PCI was implemented, priority was given to the creation of a set of "pural functions" such as pcim_request_regions(). These functions have bit masks as parameters to specify which BARs shall get mapped. Most users, however, only use those to mapp 1-3 BARs. A complete set of "singular functions" does not exist. As functions mapping / requesting multiple BARs at once have (almost) no mechanism in C to return the resources to the caller of the plural function, the devres API utilizes the iomap-table administrated by the function pcim_iomap_table(). The entire PCI devres implementation was strongly tied to that table which only allows for mapping whole, complete BARs, as the BAR's index is used as table index. Consequently, it's not possible to, e.g., have a pcim_iomap_range() function with that mechanism. An additional problem is that pci-devres has been ipmlemented in a sort of "hybrid-mode": Some unmanaged functions have managed counterparts (e.g.: pci_iomap() <-> pcim_iomap()), making their managed nature obvious to the programmer. However, the region-request functions in pci.c, prefixed with pci_, behave either managed or unmanaged, depending on whether pci_enable_device() or pcim_enable_device() has been called in advance. This hybrid API is confusing and should be more cleanly separated by providing always-managed functions prefixed with pcim_. Thus, the existing devres API is not desirable because: a) The vast majority of the users of the plural functions only ever sets a single bit in the bit mask, consequently making them singular functions anyways. b) There is no mechanism to request / iomap only part of a BAR. c) The iomap-table mechanism is over-engineered, complicated and can by definition not perform bounds checks, thus, provoking memory faults: pcim_iomap_table(pdev)[42] d) region-request functions being sometimes managed and sometimes not is bug-provoking. Implement a set of singular pcim_ functions that use devres directly and bypass the legacy iomap table mechanism. Add devres.c to driver-api documentation. Signed-off-by: Philipp Stanner --- Documentation/driver-api/pci/pci.rst | 3 + drivers/pci/devres.c | 481 ++++++++++++++++++++++++++- include/linux/pci.h | 11 + 3 files changed, 490 insertions(+), 5 deletions(-) diff --git a/Documentation/driver-api/pci/pci.rst b/Documentation/driver-ap= i/pci/pci.rst index 4843cfad4f60..92b11775344e 100644 --- a/Documentation/driver-api/pci/pci.rst +++ b/Documentation/driver-api/pci/pci.rst @@ -4,6 +4,9 @@ PCI Support Library .. kernel-doc:: drivers/pci/pci.c :export: =20 +.. kernel-doc:: drivers/pci/devres.c + :export: + .. kernel-doc:: drivers/pci/pci-driver.c :export: =20 diff --git a/drivers/pci/devres.c b/drivers/pci/devres.c index 4bd1e125bca1..cc8c1501eb13 100644 --- a/drivers/pci/devres.c +++ b/drivers/pci/devres.c @@ -8,10 +8,223 @@ */ #define PCIM_IOMAP_MAX PCI_STD_NUM_BARS =20 +/* + * Legacy struct storing addresses to whole mapped bars. + */ struct pcim_iomap_devres { void __iomem *table[PCIM_IOMAP_MAX]; }; =20 +enum pcim_addr_devres_type { + /* Default initializer. */ + PCIM_ADDR_DEVRES_TYPE_INVALID, + + /* A region spaning an entire bar. */ + PCIM_ADDR_DEVRES_TYPE_REGION, + + /* A region spaning an entire bar, and a mapping for that whole bar. */ + PCIM_ADDR_DEVRES_TYPE_REGION_MAPPING, + + /* + * A mapping within a bar, either spaning the whole bar or just a range. + * Without a requested region. + */ + PCIM_ADDR_DEVRES_TYPE_MAPPING, + + /* A ranged region within a bar, with a mapping spaning that range. */ + PCIM_ADDR_DEVRES_TYPE_REGION_RANGE_MAPPING +}; + +/* + * This struct envelopes IO or MEM addresses, that means mappings and regi= on + * requests, because those are very frequently requested and released toge= ther. + */ +struct pcim_addr_devres { + enum pcim_addr_devres_type type; + void __iomem *baseaddr; + unsigned long offset; + unsigned long len; + short bar; +}; + +static inline void pcim_addr_devres_clear(struct pcim_addr_devres *res) +{ + res->type =3D PCIM_ADDR_DEVRES_TYPE_INVALID; + res->bar =3D -1; + res->baseaddr =3D NULL; + res->offset =3D 0; + res->len =3D 0; +} + +/* + * The following functions, __pcim_*_region*, exist as counterparts to the + * versions from pci.c - which, unfortunately, can be in "hybrid mode", i.= e., + * sometimes managed, sometimes not. + * + * To separate the APIs cleanly, we define our own, simplified versions he= re. + */ + +/** + * __pcim_request_region_range - Request a ranged region + * @pdev: PCI device the region belongs to + * @bar: The BAR the region is within + * @offset: offset from the BAR's start address + * @maxlen: length in bytes, beginning at @offset + * @name: name associated with the request + * @exclusive: whether the mapping shall be exclusively for kernelspace + * + * Returns: 0 on success, a negative error code on failure. + * + * Request a ranged region within a device's PCI BAR. This function perfor= ms + * sanity checks on the input. + */ +static int __pcim_request_region_range(struct pci_dev *pdev, int bar, + unsigned long offset, unsigned long maxlen, + const char *name, int exclusive) +{ + resource_size_t start =3D pci_resource_start(pdev, bar); + resource_size_t len =3D pci_resource_len(pdev, bar); + unsigned long flags =3D pci_resource_flags(pdev, bar); + + if (start =3D=3D 0 || len =3D=3D 0) /* that's an unused BAR. */ + return 0; + if (len <=3D offset) + return -EINVAL; + + start +=3D offset; + len -=3D offset; + + if (len > maxlen && maxlen !=3D 0) + len =3D maxlen; + + if (flags & IORESOURCE_IO) { + if (!request_region(start, len, name)) + return -EBUSY; + } else if (flags & IORESOURCE_MEM) { + if (!__request_mem_region(start, len, name, exclusive)) + return -EBUSY; + } else { + /* That's not a device we can request anything on. */ + return -ENODEV; + } + + return 0; +} + +static void __pcim_release_region_range(struct pci_dev *pdev, int bar, + unsigned long offset, unsigned long maxlen) +{ + resource_size_t start =3D pci_resource_start(pdev, bar); + resource_size_t len =3D pci_resource_len(pdev, bar); + unsigned long flags =3D pci_resource_flags(pdev, bar); + + if (len <=3D offset || start =3D=3D 0) + return; + + if (len =3D=3D 0 || maxlen =3D=3D 0) /* This an unused BAR. Do nothing. */ + return; + + start +=3D offset; + len -=3D offset; + + if (len > maxlen) + len =3D maxlen; + + if (flags & IORESOURCE_IO) + release_region(start, len); + else if (flags & IORESOURCE_MEM) + release_mem_region(start, len); +} + +static int __pcim_request_region(struct pci_dev *pdev, int bar, + const char *name, int exclusive) +{ + const unsigned long offset =3D 0; + const unsigned long len =3D pci_resource_len(pdev, bar); + + return __pcim_request_region_range(pdev, bar, offset, len, name, exclusiv= e); +} + +static void __pcim_release_region(struct pci_dev *pdev, int bar) +{ + const unsigned long offset =3D 0; + const unsigned long len =3D pci_resource_len(pdev, bar); + + __pcim_release_region_range(pdev, bar, offset, len); +} + +static void pcim_addr_resource_release(struct device *dev, void *resource_= raw) +{ + struct pci_dev *pdev =3D to_pci_dev(dev); + struct pcim_addr_devres *res =3D resource_raw; + + switch (res->type) { + case PCIM_ADDR_DEVRES_TYPE_REGION: + __pcim_release_region(pdev, res->bar); + break; + case PCIM_ADDR_DEVRES_TYPE_REGION_MAPPING: + pci_iounmap(pdev, res->baseaddr); + __pcim_release_region(pdev, res->bar); + break; + case PCIM_ADDR_DEVRES_TYPE_MAPPING: + pci_iounmap(pdev, res->baseaddr); + break; + case PCIM_ADDR_DEVRES_TYPE_REGION_RANGE_MAPPING: + pci_iounmap(pdev, res->baseaddr); + __pcim_release_region_range(pdev, res->bar, res->offset, res->len); + break; + default: + break; + } +} + +static struct pcim_addr_devres *pcim_addr_devres_alloc(struct pci_dev *pde= v) +{ + struct pcim_addr_devres *res; + + res =3D devres_alloc_node(pcim_addr_resource_release, sizeof(*res), + GFP_KERNEL, dev_to_node(&pdev->dev)); + if (res) + pcim_addr_devres_clear(res); + return res; +} + +/* Just for consistency and readability. */ +static inline void pcim_addr_devres_free(struct pcim_addr_devres *res) +{ + devres_free(res); +} + +/* + * Used to identify a resource in devres_*() functions. + */ +static int pcim_addr_resources_match(struct device *dev, void *a_raw, void= *b_raw) +{ + struct pcim_addr_devres *a, *b; + + a =3D a_raw; + b =3D b_raw; + + (void)dev; /* unused. */ + + if (a->type !=3D b->type) + return 0; + + switch (a->type) { + case PCIM_ADDR_DEVRES_TYPE_REGION: + case PCIM_ADDR_DEVRES_TYPE_REGION_MAPPING: + return a->bar =3D=3D b->bar; + case PCIM_ADDR_DEVRES_TYPE_MAPPING: + return a->baseaddr =3D=3D b->baseaddr; + case PCIM_ADDR_DEVRES_TYPE_REGION_RANGE_MAPPING: + return a->bar =3D=3D b->bar && + a->offset =3D=3D b->offset && a->len =3D=3D b->len; + default: + break; + } + + return 0; +} =20 static void devm_pci_unmap_iospace(struct device *dev, void *ptr) { @@ -92,8 +305,8 @@ EXPORT_SYMBOL(devm_pci_remap_cfgspace); * * All operations are managed and will be undone on driver detach. * - * Returns a pointer to the remapped memory or an ERR_PTR() encoded error = code - * on failure. Usage example:: + * Returns a pointer to the remapped memory or an IOMEM_ERR_PTR() encoded = error + * code on failure. Usage example:: * * res =3D platform_get_resource(pdev, IORESOURCE_MEM, 0); * base =3D devm_pci_remap_cfg_resource(&pdev->dev, res); @@ -341,15 +554,80 @@ void pcim_iounmap(struct pci_dev *pdev, void __iomem = *addr) tbl[i] =3D NULL; return; } - WARN_ON(1); } EXPORT_SYMBOL(pcim_iounmap); =20 +/** + * pcim_iomap_region - Request and iomap a PCI BAR + * @pdev: PCI device to map IO resources for + * @bar: Index of a BAR to map + * @name: Name associated with the request + * + * Returns __iomem pointer on success, an IOMEM_ERR_PTR on failure. + * + * Mapping and region will get automatically released on driver detach. If + * desired, release manually only with pcim_iounmap_region(). + */ +void __iomem *pcim_iomap_region(struct pci_dev *pdev, int bar, const char = *name) +{ + int ret =3D 0; + struct pcim_addr_devres *res; + + res =3D pcim_addr_devres_alloc(pdev); + if (!res) + return IOMEM_ERR_PTR(-ENOMEM); + + res->type =3D PCIM_ADDR_DEVRES_TYPE_REGION_MAPPING; + res->bar =3D bar; + + ret =3D __pcim_request_region(pdev, bar, name, 0); + if (ret !=3D 0) + goto err_region; + + res->baseaddr =3D pci_iomap(pdev, bar, 0); + if (!res->baseaddr) { + ret =3D -EINVAL; + goto err_iomap; + } + + devres_add(&pdev->dev, res); + return res->baseaddr; + +err_iomap: + __pcim_release_region(pdev, bar); +err_region: + pcim_addr_devres_free(res); + + return IOMEM_ERR_PTR(ret); +} +EXPORT_SYMBOL(pcim_iomap_region); + +/** + * pcim_iounmap_region - Unmap and release a PCI BAR + * @pdev: PCI device to operate on + * @bar: Index of BAR to unmap and release + * + * Unmap a BAR and release its region manually. Only pass BARs that were + * previously mapped by pcim_iomap_region(). + */ +void pcim_iounmap_region(struct pci_dev *pdev, int bar) +{ + struct pcim_addr_devres res_searched; + + pcim_addr_devres_clear(&res_searched); + res_searched.type =3D PCIM_ADDR_DEVRES_TYPE_REGION_MAPPING; + res_searched.bar =3D bar; + + devres_release(&pdev->dev, pcim_addr_resource_release, + pcim_addr_resources_match, &res_searched); +} +EXPORT_SYMBOL(pcim_iounmap_region); + /** * pcim_iomap_regions - Request and iomap PCI BARs * @pdev: PCI device to map IO resources for * @mask: Mask of BARs to request and iomap - * @name: Name used when requesting regions + * @name: Name associated with the requests * * Request and iomap regions specified by @mask. */ @@ -402,7 +680,7 @@ EXPORT_SYMBOL(pcim_iomap_regions); * pcim_iomap_regions_request_all - Request all BARs and iomap specified o= nes * @pdev: PCI device to map IO resources for * @mask: Mask of BARs to iomap - * @name: Name used when requesting regions + * @name: Name associated with the requests * * Request all PCI BARs and iomap regions specified by @mask. */ @@ -448,3 +726,196 @@ void pcim_iounmap_regions(struct pci_dev *pdev, int m= ask) } } EXPORT_SYMBOL(pcim_iounmap_regions); + +static int _pcim_request_region(struct pci_dev *pdev, int bar, const char = *name, + int request_flags) +{ + int ret =3D 0; + struct pcim_addr_devres *res; + + res =3D pcim_addr_devres_alloc(pdev); + if (!res) + return -ENOMEM; + res->type =3D PCIM_ADDR_DEVRES_TYPE_REGION; + res->bar =3D bar; + + ret =3D __pcim_request_region(pdev, bar, name, request_flags); + if (ret !=3D 0) { + pcim_addr_devres_free(res); + return ret; + } + + devres_add(&pdev->dev, res); + return 0; +} + +/** + * pcim_request_region - Request a PCI BAR + * @pdev: PCI device to requestion region for + * @bar: Index of BAR to request + * @name: Name associated with the request + * + * Returns: 0 on success, a negative error code on failure. + * + * Request region specified by @bar. + * + * The region will automatically be released on driver detach. If desired, + * release manually only with pcim_release_region(). + */ +int pcim_request_region(struct pci_dev *pdev, int bar, const char *name) +{ + return _pcim_request_region(pdev, bar, name, 0); +} +EXPORT_SYMBOL(pcim_request_region); + +/** + * pcim_release_region - Release a PCI BAR + * @pdev: PCI device to operate on + * @bar: Index of BAR to release + * + * Release a region manually that was previously requested by + * pcim_request_region(). + */ +void pcim_release_region(struct pci_dev *pdev, int bar) +{ + struct pcim_addr_devres res_searched; + + pcim_addr_devres_clear(&res_searched); + res_searched.type =3D PCIM_ADDR_DEVRES_TYPE_REGION; + res_searched.bar =3D bar; + + devres_release(&pdev->dev, pcim_addr_resource_release, + pcim_addr_resources_match, &res_searched); +} +EXPORT_SYMBOL(pcim_release_region); + +/** + * pcim_iomap_range - Create a ranged __iomap mapping within a PCI BAR + * @pdev: PCI device to map IO resources for + * @bar: Index of the BAR + * @offset: Offset from the begin of the BAR + * @len: Length in bytes for the mapping + * + * Returns: __iomem pointer on success, an IOMEM_ERR_PTR on failure. + * + * Creates a new IO-Mapping within the specified @bar, ranging from @offse= t to + * @offset + @len. + * + * The mapping will automatically get unmapped on driver detach. If desire= d, + * release manually only with pcim_iounmap(). + */ +void __iomem *pcim_iomap_range(struct pci_dev *pdev, int bar, + unsigned long offset, unsigned long len) +{ + void __iomem *mapping; + struct pcim_addr_devres *res; + + res =3D pcim_addr_devres_alloc(pdev); + if (!res) + return IOMEM_ERR_PTR(-ENOMEM); + + mapping =3D pci_iomap_range(pdev, bar, offset, len); + if (!mapping) { + pcim_addr_devres_free(res); + return IOMEM_ERR_PTR(-EINVAL); + } + + res->type =3D PCIM_ADDR_DEVRES_TYPE_MAPPING; + res->baseaddr =3D mapping; + + /* + * Ranged mappings don't get added to the legacy-table, since the table + * only ever keeps track of whole BARs. + */ + + devres_add(&pdev->dev, res); + return mapping; +} +EXPORT_SYMBOL(pcim_iomap_range); + +/** + * pcim_iomap_region_range - Request and map a range within a PCI BAR + * @pdev: PCI device to map IO resources for + * @bar: Index of BAR to request within + * @offset: Offset from the begin of the BAR + * @len: Length in bytes for the mapping + * @name: Name associated with the request + * + * Returns: __iomem pointer on success, an IOMEM_ERR_PTR on failure. + * + * Request region with a range specified by @offset and @len within @bar a= nd + * iomap it. + * + * The region will automatically be released and the mapping be unmapped on + * driver detach. If desired, release manually only with + * pcim_iounmap_region_range(). + * + * You probably should only use this function if you explicitly do not wan= t to + * request the entire BAR. For most use-cases, combining pcim_request_regi= on() + * and pcim_iomap_range() should be sufficient. + */ +void __iomem *pcim_iomap_region_range(struct pci_dev *pdev, int bar, + unsigned long offset, unsigned long len, const char *name) +{ + int ret =3D 0; + struct pcim_addr_devres *res; + + res =3D pcim_addr_devres_alloc(pdev); + if (!res) + return IOMEM_ERR_PTR(-ENOMEM); + + res->type =3D PCIM_ADDR_DEVRES_TYPE_REGION_RANGE_MAPPING; + res->bar =3D bar; + res->offset =3D offset; + res->len =3D len; + + ret =3D __pcim_request_region_range(pdev, bar, offset, len, name, 0); + if (ret !=3D 0) + goto err_region; + + res->baseaddr =3D pci_iomap_range(pdev, bar, offset, len); + if (!res->baseaddr) { + ret =3D -EINVAL; + goto err_iomap; + } + + devres_add(&pdev->dev, res); + return res->baseaddr; + +err_iomap: + __pcim_release_region_range(pdev, bar, offset, len); +err_region: + pcim_addr_devres_free(res); + + return IOMEM_ERR_PTR(ret); +} +EXPORT_SYMBOL(pcim_iomap_region_range); + +/** + * pcim_iounmap_region_range - Unmap and release a range within a PCI BAR + * @pdev: PCI device to operate on + * @bar: Index of BAR containing the range + * @offset: Offset from the begin of the BAR + * @len: Length in bytes for the mapping + * + * Unmaps and releases a memory area within the specified PCI BAR. + * + * This function may not be used to free only part of a range. Only use th= is + * function with the exact parameters you previously used successfully in + * pcim_iomap_region_range(). + */ +void pcim_iounmap_region_range(struct pci_dev *pdev, int bar, + unsigned long offset, unsigned long len) +{ + struct pcim_addr_devres res_searched; + pcim_addr_devres_clear(&res_searched); + + res_searched.type =3D PCIM_ADDR_DEVRES_TYPE_REGION_RANGE_MAPPING; + res_searched.bar =3D bar; + res_searched.offset =3D offset; + res_searched.len =3D len; + + devres_release(&pdev->dev, pcim_addr_resource_release, + pcim_addr_resources_match, &res_searched); +} +EXPORT_SYMBOL(pcim_iounmap_region_range); diff --git a/include/linux/pci.h b/include/linux/pci.h index 58a4c976c39b..1b45a4888703 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -2314,10 +2314,21 @@ static inline void pci_fixup_device(enum pci_fixup_= pass pass, void __iomem *pcim_iomap(struct pci_dev *pdev, int bar, unsigned long maxl= en); void pcim_iounmap(struct pci_dev *pdev, void __iomem *addr); void __iomem * const *pcim_iomap_table(struct pci_dev *pdev); +void __iomem *pcim_iomap_region(struct pci_dev *pdev, int bar, const char = *name); +void pcim_iounmap_region(struct pci_dev *pdev, int bar); int pcim_iomap_regions(struct pci_dev *pdev, int mask, const char *name); int pcim_iomap_regions_request_all(struct pci_dev *pdev, int mask, const char *name); void pcim_iounmap_regions(struct pci_dev *pdev, int mask); +int pcim_request_region(struct pci_dev *pdev, int bar, const char *res_nam= e); +void pcim_release_region(struct pci_dev *pdev, int bar); +void __iomem *pcim_iomap_range(struct pci_dev *pdev, int bar, + unsigned long offset, unsigned long len); +void __iomem *pcim_iomap_region_range(struct pci_dev *pdev, int bar, + unsigned long offset, unsigned long len, + const char *res_name); +void pcim_iounmap_region_range(struct pci_dev *pdev, int bar, + unsigned long offset, unsigned long len); =20 extern int pci_pci_problems; #define PCIPCI_FAIL 1 /* No PCI PCI DMA */ --=20 2.43.0 From nobody Tue Dec 16 11:42:59 2025 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.133.124]) (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 02C1D17998 for ; Mon, 15 Jan 2024 14:47:24 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=redhat.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=redhat.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=redhat.com header.i=@redhat.com header.b="fXJupGQP" DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1705330044; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=O/4sb7EtZnZXDAXsZ1TnbdCzfHG1Mac4zhgSYFspvX4=; b=fXJupGQPWdntqND6HUaT6E9hWEoyk+qmbk1pmjsSLL+Rj0EOBMd3zMyqtAmIfOw5klies4 c7zLOzHtlVyKwSimuX+/LYqLcWf2WQ7d4Nfuj+QaFUwcYlTUV/eymuuANgbDa3p7iP+hdb JYrony9qXQEpsaqLIJkKex6Ekeul0w4= Received: from mail-yw1-f199.google.com (mail-yw1-f199.google.com [209.85.128.199]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.3, cipher=TLS_AES_256_GCM_SHA384) id us-mta-576-WbnvMcgJMG-JTIwxSc3cTw-1; Mon, 15 Jan 2024 09:47:17 -0500 X-MC-Unique: WbnvMcgJMG-JTIwxSc3cTw-1 Received: by mail-yw1-f199.google.com with SMTP id 00721157ae682-5fb837a22a2so5953317b3.0 for ; Mon, 15 Jan 2024 06:47:17 -0800 (PST) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1705330037; x=1705934837; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=O/4sb7EtZnZXDAXsZ1TnbdCzfHG1Mac4zhgSYFspvX4=; b=DDsOIN6UotPSkPiqHTBb0JVzouy/2oH9dnvgtGLc8Hr+cZmvhLh+8jXhW1U8RVzarE Rq7Kyb2z4EGOzskKSI4qmYi5ror2SNsUa97qsCu31m9cfW1EgmsRTwEv+yV5kglzHmx1 ecl/kdylNBKdGRZDFStpZa6nHot8P/PlEAfTrS3HYMXSQ0T/CJ/1TX+SnyYiZ5y6xKoT YB7LQVaQFNqY5etY1kpZt5bZvbshWBJwg6gL5l3xMTjfzyctIjve1I8wnomQTYVUWFEL Pr84+EoVaAcdX/+NgBxUyG6MyAfXhINQnD3wy6iE8twS75U3L0OSLub7SJY+lbJmOW8/ 3QSw== X-Gm-Message-State: AOJu0YyUQkMqjT8cwomrDBbutHTkF3xApjL9RMRp08KUxH91qK0VKgHP XgeoDDO7F1tWffqQv6j/jb9365ykJEbumdtnvS1cT0oPGyumdLFeJGNA6lT9bx0ssWJKiOwChSf rbxVUKDjiD9N2E45kHRh75+DIgvVJboIf X-Received: by 2002:a81:48d1:0:b0:5ed:95b2:ca48 with SMTP id v200-20020a8148d1000000b005ed95b2ca48mr4529241ywa.5.1705330036594; Mon, 15 Jan 2024 06:47:16 -0800 (PST) X-Google-Smtp-Source: AGHT+IEFikiSFust9uxkTlR/9heFIFjf1A0XIIogDr1MtKcQAB8Xk4Knp3bRyPE3d8Lwwogh0DxVXg== X-Received: by 2002:a81:48d1:0:b0:5ed:95b2:ca48 with SMTP id v200-20020a8148d1000000b005ed95b2ca48mr4529224ywa.5.1705330036207; Mon, 15 Jan 2024 06:47:16 -0800 (PST) Received: from pstanner-thinkpadt14sgen1.muc.redhat.com (nat-pool-muc-t.redhat.com. [149.14.88.26]) by smtp.gmail.com with ESMTPSA id ne13-20020a056214424d00b006815cf9a644sm1020720qvb.55.2024.01.15.06.47.14 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Mon, 15 Jan 2024 06:47:15 -0800 (PST) From: Philipp Stanner To: Jonathan Corbet , Hans de Goede , Maarten Lankhorst , Maxime Ripard , Thomas Zimmermann , David Airlie , Daniel Vetter , Bjorn Helgaas , Philipp Stanner , Sam Ravnborg , dakr@redhat.com Cc: linux-doc@vger.kernel.org, linux-kernel@vger.kernel.org, dri-devel@lists.freedesktop.org, linux-pci@vger.kernel.org Subject: [PATCH 02/10] pci: deprecate iomap-table functions Date: Mon, 15 Jan 2024 15:46:13 +0100 Message-ID: <20240115144655.32046-4-pstanner@redhat.com> X-Mailer: git-send-email 2.43.0 In-Reply-To: <20240115144655.32046-2-pstanner@redhat.com> References: <20240115144655.32046-2-pstanner@redhat.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" The old plural devres functions tie PCI's devres implementation to the iomap-table mechanism which unfortunately is not extensible. As the purlal functions are almost never used with more than one bit set in their bit-mask, deprecating them and encouraging users to use the new singular functions instead is reasonable. Furthermore, to make the implementation more consistent and extensible, the plural functions should use the singular functions. Add new wrapper to request / release all BARs. Make the plural functions call into the singular functions. Mark the plural functions as deprecated. Remove as much of the iomap-table-mechanism as possible. Add comments describing the path towards a cleaned-up API. Signed-off-by: Philipp Stanner --- drivers/pci/devres.c | 369 +++++++++++++++++++++++++++++++++---------- drivers/pci/pci.c | 20 +++ drivers/pci/pci.h | 5 + include/linux/pci.h | 2 + 4 files changed, 313 insertions(+), 83 deletions(-) diff --git a/drivers/pci/devres.c b/drivers/pci/devres.c index cc8c1501eb13..e221919ebbe2 100644 --- a/drivers/pci/devres.c +++ b/drivers/pci/devres.c @@ -4,15 +4,43 @@ #include "pci.h" =20 /* - * PCI iomap devres + * On the state of PCI's devres implementation: + * + * The older devres API for PCI has two significant problems: + * + * 1. It is very strongly tied to the statically allocated mapping table in + * struct pcim_iomap_devres below. This is mostly solved in the sense o= f the + * pcim_ functions in this file providing things like ranged mapping by + * bypassing this table, wheras the functions that were present in the = old + * API still enter the mapping addresses into the table for users of th= e old + * API. + * 2. The region-request-functions in pci.c do become managed IF the devic= e has + * been enabled with pcim_enable_device() instead of pci_enable_device(= ). + * This resulted in the API becoming inconsistent: Some functions have = an + * obviously managed counter-part (e.g., pci_iomap() <-> pcim_iomap()), + * whereas some don't and are never managed, while others don't and are + * _sometimes_ managed (e.g. pci_request_region()). + * Consequently, in the new API, region requests performed by the pcim_ + * functions are automatically cleaned up through the devres callback + * pcim_addr_resource_release(), while requests performed by + * pcim_enable_device() + pci_*region*() are automatically cleaned up + * through the for-loop in pcim_release(). + * + * TODO 1: + * Remove the legacy table entirely once all calls to pcim_iomap_table() in + * the kernel have been removed. + * + * TODO 2: + * Port everyone calling pcim_enable_device() + pci_*region*() to using the + * pcim_ functions. Then, remove all devres functionality from pci_*region= *() + * functions and remove the associated cleanups described above in point #= 2. */ -#define PCIM_IOMAP_MAX PCI_STD_NUM_BARS =20 /* * Legacy struct storing addresses to whole mapped bars. */ struct pcim_iomap_devres { - void __iomem *table[PCIM_IOMAP_MAX]; + void __iomem *table[PCI_STD_NUM_BARS]; }; =20 enum pcim_addr_devres_type { @@ -381,6 +409,16 @@ static void pcim_release(struct device *gendev, void *= res) struct pci_devres *this =3D res; int i; =20 + /* + * This is legacy code. + * All regions requested by a pcim_ function do get released through + * pcim_addr_resource_release(). Thanks to the hybrid nature of the pci_ + * region-request functions, this for-loop has to release the regions + * if they have been requested by such a function. + * + * TODO: Remove this once all users of pcim_enable_device() PLUS + * pci-region-request-functions have been ported to pcim_ functions. + */ for (i =3D 0; i < DEVICE_COUNT_RESOURCE; i++) if (this->region_mask & (1 << i)) pci_release_region(dev, i); @@ -469,17 +507,16 @@ EXPORT_SYMBOL(pcim_pin_device); =20 static void pcim_iomap_release(struct device *gendev, void *res) { - struct pci_dev *dev =3D to_pci_dev(gendev); - struct pcim_iomap_devres *this =3D res; - int i; - - for (i =3D 0; i < PCIM_IOMAP_MAX; i++) - if (this->table[i]) - pci_iounmap(dev, this->table[i]); + /* + * Do nothing. This is legacy code. + * + * Cleanup of the mappings is now done directly through the callbacks + * registered when creating them. + */ } =20 /** - * pcim_iomap_table - access iomap allocation table + * pcim_iomap_table - access iomap allocation table (DEPRECATED) * @pdev: PCI device to access iomap table for * * Access iomap allocation table for @dev. If iomap table doesn't @@ -490,6 +527,11 @@ static void pcim_iomap_release(struct device *gendev, = void *res) * This function might sleep when the table is first allocated but can * be safely called without context and guaranteed to succeed once * allocated. + * + * This function is DEPRECATED. Do not use it in new code. + * Instead, obtain a mapping's address directly from one of the pcim_* map= ping + * functions. For example: + * void __iomem *mappy =3D pcim_iomap(pdev, barnr, length); */ void __iomem * const *pcim_iomap_table(struct pci_dev *pdev) { @@ -508,27 +550,109 @@ void __iomem * const *pcim_iomap_table(struct pci_de= v *pdev) } EXPORT_SYMBOL(pcim_iomap_table); =20 +/* + * Fill the legacy mapping-table, so that drivers using the old API + * can still get a BAR's mapping address through pcim_iomap_table(). + */ +static int pcim_add_mapping_to_legacy_table(struct pci_dev *pdev, + void __iomem *mapping, short bar) +{ + void __iomem **legacy_iomap_table; + + if (bar >=3D PCI_STD_NUM_BARS) + return -EINVAL; + + legacy_iomap_table =3D (void __iomem **)pcim_iomap_table(pdev); + if (!legacy_iomap_table) + return -ENOMEM; + + /* The legacy mechanism doesn't allow for duplicate mappings. */ + WARN_ON(legacy_iomap_table[bar]); + + legacy_iomap_table[bar] =3D mapping; + + return 0; +} + +/* + * Removes a mapping. The table only contains whole-bar-mappings, so this = will + * never interfere with ranged mappings. + */ +static void pcim_remove_mapping_from_legacy_table(struct pci_dev *pdev, + void __iomem *addr) +{ + short bar; + void __iomem **legacy_iomap_table; + + legacy_iomap_table =3D (void __iomem **)pcim_iomap_table(pdev); + if (!legacy_iomap_table) + return; + + for (bar =3D 0; bar < PCI_STD_NUM_BARS; bar++) { + if (legacy_iomap_table[bar] =3D=3D addr) { + legacy_iomap_table[bar] =3D NULL; + return; + } + } +} + +/* + * The same as pcim_remove_mapping_from_legacy_table(), but identifies the + * mapping by its BAR index. + */ +static void pcim_remove_bar_from_legacy_table(struct pci_dev *pdev, short = bar) +{ + void __iomem **legacy_iomap_table; + + if (bar >=3D PCI_STD_NUM_BARS) + return; + + legacy_iomap_table =3D (void __iomem **)pcim_iomap_table(pdev); + if (!legacy_iomap_table) + return; + + legacy_iomap_table[bar] =3D NULL; +} + /** * pcim_iomap - Managed pcim_iomap() * @pdev: PCI device to iomap for * @bar: BAR to iomap * @maxlen: Maximum length of iomap * - * Managed pci_iomap(). Map is automatically unmapped on driver - * detach. + * Returns __iomem pointer on success, NULL on failure. + * + * Managed pci_iomap(). Map is automatically unmapped on driver detach. If + * desired, unmap manually only with pcim_iounmap(). + * + * This SHOULD only be used once per BAR. */ void __iomem *pcim_iomap(struct pci_dev *pdev, int bar, unsigned long maxl= en) { - void __iomem **tbl; - - BUG_ON(bar >=3D PCIM_IOMAP_MAX); + void __iomem *mapping; + struct pcim_addr_devres *res; =20 - tbl =3D (void __iomem **)pcim_iomap_table(pdev); - if (!tbl || tbl[bar]) /* duplicate mappings not allowed */ + res =3D pcim_addr_devres_alloc(pdev); + if (!res) return NULL; + res->type =3D PCIM_ADDR_DEVRES_TYPE_MAPPING; =20 - tbl[bar] =3D pci_iomap(pdev, bar, maxlen); - return tbl[bar]; + mapping =3D pci_iomap(pdev, bar, maxlen); + if (!mapping) + goto err_iomap; + res->baseaddr =3D mapping; + + if (pcim_add_mapping_to_legacy_table(pdev, mapping, bar) !=3D 0) + goto err_table; + + devres_add(&pdev->dev, res); + return mapping; + +err_table: + pci_iounmap(pdev, mapping); +err_iomap: + pcim_addr_devres_free(res); + return NULL; } EXPORT_SYMBOL(pcim_iomap); =20 @@ -537,23 +661,24 @@ EXPORT_SYMBOL(pcim_iomap); * @pdev: PCI device to iounmap for * @addr: Address to unmap * - * Managed pci_iounmap(). @addr must have been mapped using pcim_iomap(). + * Managed pci_iounmap(). @addr must have been mapped using pcim_iomap() or + * pcim_iomap_range(). */ void pcim_iounmap(struct pci_dev *pdev, void __iomem *addr) { - void __iomem **tbl; - int i; + struct pcim_addr_devres res_searched; =20 - pci_iounmap(pdev, addr); + pcim_addr_devres_clear(&res_searched); + res_searched.type =3D PCIM_ADDR_DEVRES_TYPE_MAPPING; + res_searched.baseaddr =3D addr; =20 - tbl =3D (void __iomem **)pcim_iomap_table(pdev); - BUG_ON(!tbl); + if (devres_release(&pdev->dev, pcim_addr_resource_release, + pcim_addr_resources_match, &res_searched) !=3D 0) { + /* Doesn't exist. User passed nonsense. */ + return; + } =20 - for (i =3D 0; i < PCIM_IOMAP_MAX; i++) - if (tbl[i] =3D=3D addr) { - tbl[i] =3D NULL; - return; - } + pcim_remove_mapping_from_legacy_table(pdev, addr); } EXPORT_SYMBOL(pcim_iounmap); =20 @@ -623,106 +748,184 @@ void pcim_iounmap_region(struct pci_dev *pdev, int = bar) } EXPORT_SYMBOL(pcim_iounmap_region); =20 +static inline bool mask_contains_bar(int mask, int bar) +{ + return mask & (1 << bar); +} + /** - * pcim_iomap_regions - Request and iomap PCI BARs + * pcim_iomap_regions - Request and iomap PCI BARs (DEPRECATED) * @pdev: PCI device to map IO resources for * @mask: Mask of BARs to request and iomap * @name: Name associated with the requests * + * Returns 0 on success, negative error code on failure. + * * Request and iomap regions specified by @mask. + * + * This function is DEPRECATED. Don't use it in new code. + * Use pcim_iomap_region() instead. */ int pcim_iomap_regions(struct pci_dev *pdev, int mask, const char *name) { - void __iomem * const *iomap; - int i, rc; + int ret =3D -1; + short bar; + void __iomem *mapping; =20 - iomap =3D pcim_iomap_table(pdev); - if (!iomap) - return -ENOMEM; + for (bar =3D 0; bar < DEVICE_COUNT_RESOURCE; bar++) { + if (!mask_contains_bar(mask, bar)) + continue; =20 - for (i =3D 0; i < DEVICE_COUNT_RESOURCE; i++) { - unsigned long len; + mapping =3D pcim_iomap_region(pdev, bar, name); + if (IS_ERR(mapping)) { + ret =3D PTR_ERR(mapping); + goto err; + } + ret =3D pcim_add_mapping_to_legacy_table(pdev, mapping, bar); + if (ret !=3D 0) + goto err; + } =20 - if (!(mask & (1 << i))) - continue; + return 0; =20 - rc =3D -EINVAL; - len =3D pci_resource_len(pdev, i); - if (!len) - goto err_inval; + err: + while (--bar >=3D 0) { + pcim_iounmap_region(pdev, bar); + pcim_remove_bar_from_legacy_table(pdev, bar); + } =20 - rc =3D pci_request_region(pdev, i, name); - if (rc) - goto err_inval; + return ret; +} +EXPORT_SYMBOL(pcim_iomap_regions); =20 - rc =3D -ENOMEM; - if (!pcim_iomap(pdev, i, 0)) - goto err_region; +/** + * pcim_release_all_regions - Release all regions of a PCI-device + * @pdev: the PCI device + * + * Returns 0 on success, negative error code on failure. + * + * Will release all regions previously requested through pcim_request_regi= on() + * or pcim_request_all_regions(). + * + * Can be called from any context, i.e., not necessarily as a counterpart = to + * pcim_request_all_regions(). + */ +void pcim_release_all_regions(struct pci_dev *pdev) +{ + short bar; + + for (bar =3D 0; bar < PCI_STD_NUM_BARS; bar++) + pcim_release_region(pdev, bar); +} +EXPORT_SYMBOL(pcim_release_all_regions); + +/** + * pcim_request_all_regions - Request all regions + * @pdev: PCI device to map IO resources for + * @name: name associated with the request + * + * Requested regions will automatically be released at driver detach. If d= esired, + * release individual regions with pcim_release_region() or all of them at= once + * with pcim_release_all_regions(). + */ +int pcim_request_all_regions(struct pci_dev *pdev, const char *name) +{ + int ret; + short bar; + + for (bar =3D 0; bar < PCI_STD_NUM_BARS; bar++) { + ret =3D pcim_request_region(pdev, bar, name); + if (ret !=3D 0) + goto err; } =20 return 0; =20 - err_region: - pci_release_region(pdev, i); - err_inval: - while (--i >=3D 0) { - if (!(mask & (1 << i))) - continue; - pcim_iounmap(pdev, iomap[i]); - pci_release_region(pdev, i); - } +err: + pcim_release_all_regions(pdev); =20 - return rc; + return ret; } -EXPORT_SYMBOL(pcim_iomap_regions); +EXPORT_SYMBOL(pcim_request_all_regions); =20 /** - * pcim_iomap_regions_request_all - Request all BARs and iomap specified o= nes + * pcim_iomap_regions_request_all - Request all BARs and iomap specified o= nes (DEPRECATED) * @pdev: PCI device to map IO resources for * @mask: Mask of BARs to iomap * @name: Name associated with the requests * + * Returns 0 on success, negative error code on failure. + * * Request all PCI BARs and iomap regions specified by @mask. + * + * To release these resources manually, call pcim_release_region() for the + * regions and pcim_iounmap() for the mappings. + * + * This function is DEPRECATED. Don't use it in new code. + * Use pcim_request_all_regions() + pcim_iomap*() instead. */ int pcim_iomap_regions_request_all(struct pci_dev *pdev, int mask, const char *name) { - int request_mask =3D ((1 << 6) - 1) & ~mask; - int rc; + short bar; + int ret =3D -1; + void __iomem **legacy_iomap_table; =20 - rc =3D pci_request_selected_regions(pdev, request_mask, name); - if (rc) - return rc; + ret =3D pcim_request_all_regions(pdev, name); + if (ret !=3D 0) + return ret; =20 - rc =3D pcim_iomap_regions(pdev, mask, name); - if (rc) - pci_release_selected_regions(pdev, request_mask); - return rc; + for (bar =3D 0; bar < PCI_STD_NUM_BARS; bar++) { + if (!mask_contains_bar(mask, bar)) + continue; + if (!pcim_iomap(pdev, bar, 0)) + goto err; + } + + return 0; + +err: + /* + * Here it gets tricky: pcim_iomap() above has most likely + * failed because it got an OOM when trying to create the + * legacy-table. + * We check here if that has happened. If not, pcim_iomap() + * must have failed because of EINVAL. + */ + legacy_iomap_table =3D (void __iomem **)pcim_iomap_table(pdev); + if (!legacy_iomap_table) + ret =3D -ENOMEM; + else + ret =3D -EINVAL; + + while (--bar >=3D 0) + pcim_iounmap(pdev, legacy_iomap_table[bar]); + + pcim_release_all_regions(pdev); + + return ret; } EXPORT_SYMBOL(pcim_iomap_regions_request_all); =20 /** - * pcim_iounmap_regions - Unmap and release PCI BARs + * pcim_iounmap_regions - Unmap and release PCI BARs (DEPRECATED) * @pdev: PCI device to map IO resources for * @mask: Mask of BARs to unmap and release * * Unmap and release regions specified by @mask. + * + * This function is DEPRECATED. Don't use it in new code. */ void pcim_iounmap_regions(struct pci_dev *pdev, int mask) { - void __iomem * const *iomap; - int i; - - iomap =3D pcim_iomap_table(pdev); - if (!iomap) - return; + short bar; =20 - for (i =3D 0; i < PCIM_IOMAP_MAX; i++) { - if (!(mask & (1 << i))) + for (bar =3D 0; bar < PCI_STD_NUM_BARS; bar++) { + if (!mask_contains_bar(mask, bar)) continue; =20 - pcim_iounmap(pdev, iomap[i]); - pci_release_region(pdev, i); + pcim_iounmap_region(pdev, bar); + pcim_remove_bar_from_legacy_table(pdev, bar); } } EXPORT_SYMBOL(pcim_iounmap_regions); diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index bf5f9c3a1908..2899df77a285 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -3884,6 +3884,16 @@ void pci_release_region(struct pci_dev *pdev, int ba= r) release_mem_region(pci_resource_start(pdev, bar), pci_resource_len(pdev, bar)); =20 + /* + * This devres utility makes this function sometimes managed + * (when pcim_enable_device() has been called before). + * This is bad because it conflicts with the pcim_ functions being + * exclusively responsible for managed pci. Its "sometimes yes, sometimes + * no" nature can cause bugs. + * + * TODO: Remove this once all users that use pcim_enable_device() PLUS + * a region request function have been ported to using pcim_ functions. + */ dr =3D find_pci_dr(pdev); if (dr) dr->region_mask &=3D ~(1 << bar); @@ -3928,6 +3938,16 @@ static int __pci_request_region(struct pci_dev *pdev= , int bar, goto err_out; } =20 + /* + * This devres utility makes this function sometimes managed + * (when pcim_enable_device() has been called before). + * This is bad because it conflicts with the pcim_ functions being + * exclusively responsible for managed pci. Its "sometimes yes, sometimes + * no" nature can cause bugs. + * + * TODO: Remove this once all users that use pcim_enable_device() PLUS + * a region request function have been ported to using pcim_ functions. + */ dr =3D find_pci_dr(pdev); if (dr) dr->region_mask |=3D 1 << bar; diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h index de6d61d1e07f..4d4bcc2d850f 100644 --- a/drivers/pci/pci.h +++ b/drivers/pci/pci.h @@ -814,6 +814,11 @@ struct pci_devres { unsigned int orig_intx:1; unsigned int restore_intx:1; unsigned int mwi:1; + + /* + * TODO: remove the region_mask once everyone calling + * pcim_enable_device() + pci_*region*() is ported to pcim_ functions. + */ u32 region_mask; }; =20 diff --git a/include/linux/pci.h b/include/linux/pci.h index 1b45a4888703..3dcd8a5e10b5 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -2322,6 +2322,8 @@ int pcim_iomap_regions_request_all(struct pci_dev *pd= ev, int mask, void pcim_iounmap_regions(struct pci_dev *pdev, int mask); int pcim_request_region(struct pci_dev *pdev, int bar, const char *res_nam= e); void pcim_release_region(struct pci_dev *pdev, int bar); +void pcim_release_all_regions(struct pci_dev *pdev); +int pcim_request_all_regions(struct pci_dev *pdev, const char *name); void __iomem *pcim_iomap_range(struct pci_dev *pdev, int bar, unsigned long offset, unsigned long len); void __iomem *pcim_iomap_region_range(struct pci_dev *pdev, int bar, --=20 2.43.0 From nobody Tue Dec 16 11:42:59 2025 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.133.124]) (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 BAE091775E for ; Mon, 15 Jan 2024 14:47:22 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=redhat.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=redhat.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=redhat.com header.i=@redhat.com header.b="DjOmKabw" DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1705330041; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=ppwvkSEuixS2SG0TbgxgzQKgr6fgJxktZuxIJo/ERYU=; b=DjOmKabwUK1SeXmk+bEQa5KxBBqlQjNouFrXq2LrEoUVDlYnfFfwibnoc0jCPF9GXWWZr6 v8oTHH0vek/QMrK1m7PWKybiFnM7MJZCZinkIWPKlazBeV91l2f5mAMZ3wAwikLavsKtXB Nztyahy3PJQKxhlpxi3uikoPAaK7yCw= Received: from mail-qv1-f71.google.com (mail-qv1-f71.google.com [209.85.219.71]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.3, cipher=TLS_AES_256_GCM_SHA384) id us-mta-640--sBCLrDlMIKeA8PiYoUQbw-1; Mon, 15 Jan 2024 09:47:19 -0500 X-MC-Unique: -sBCLrDlMIKeA8PiYoUQbw-1 Received: by mail-qv1-f71.google.com with SMTP id 6a1803df08f44-67f4860c970so20144706d6.0 for ; Mon, 15 Jan 2024 06:47:19 -0800 (PST) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1705330039; x=1705934839; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=ppwvkSEuixS2SG0TbgxgzQKgr6fgJxktZuxIJo/ERYU=; b=Fa+dhGQtEhiUcyLA9iJPAZxvweGK2Y8uf769OAqrs+gAfQp51mb/1YOEvv+HP7tImc QAXm60WBrxBpQ68nUtCYzj3o07HMA0+/TD7YKznjgYLhe5Gc/nh1FdGMcwfyFG1stAPg egPcmfm4ueCbKItxus5oBfxwk6V4fsi3is5rgYavb9JD3CCjG0epn3dN16Rek613ndOq E8gGGyOlvCK2niwcjnpYbKjz3u8fDo1Ioc4SRBQZd3J5OnBqQ/lcA269hLcWA89Zg3uo 5f6sFwrboWbpXqw6VA3DYMKPAaFw3sHNCfJ89el/gg1i165nxyuupsrTVMYuvFGucl43 Bfow== X-Gm-Message-State: AOJu0YybO9oDtJhS7COHUeTwmJ5TJotNBAiJG57n0yESetIYLFNq4RcP IQm6DPFiPATmJEjps2RoE+4++xUcDNEgiGgMH4ZL7mVP6jlODYkRrtVNYxvFUkXjLMxplCX7wJ7 Bmvj2NDbm/LtX4eC1T9oEDK2r+g8/BJR5 X-Received: by 2002:a05:6214:4109:b0:681:3016:2ca6 with SMTP id kc9-20020a056214410900b0068130162ca6mr10705068qvb.6.1705330038839; Mon, 15 Jan 2024 06:47:18 -0800 (PST) X-Google-Smtp-Source: AGHT+IHUSOs/arWl9oiDpq0rWblAC6qIoaPYTdyyYZ8nxT4UMvv5M5B4KQKyYzih3gp3DMQa57ou9A== X-Received: by 2002:a05:6214:4109:b0:681:3016:2ca6 with SMTP id kc9-20020a056214410900b0068130162ca6mr10705033qvb.6.1705330038467; Mon, 15 Jan 2024 06:47:18 -0800 (PST) Received: from pstanner-thinkpadt14sgen1.muc.redhat.com (nat-pool-muc-t.redhat.com. [149.14.88.26]) by smtp.gmail.com with ESMTPSA id ne13-20020a056214424d00b006815cf9a644sm1020720qvb.55.2024.01.15.06.47.16 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Mon, 15 Jan 2024 06:47:18 -0800 (PST) From: Philipp Stanner To: Jonathan Corbet , Hans de Goede , Maarten Lankhorst , Maxime Ripard , Thomas Zimmermann , David Airlie , Daniel Vetter , Bjorn Helgaas , Philipp Stanner , Sam Ravnborg , dakr@redhat.com Cc: linux-doc@vger.kernel.org, linux-kernel@vger.kernel.org, dri-devel@lists.freedesktop.org, linux-pci@vger.kernel.org Subject: [PATCH 03/10] pci: warn users about complicated devres nature Date: Mon, 15 Jan 2024 15:46:14 +0100 Message-ID: <20240115144655.32046-5-pstanner@redhat.com> X-Mailer: git-send-email 2.43.0 In-Reply-To: <20240115144655.32046-2-pstanner@redhat.com> References: <20240115144655.32046-2-pstanner@redhat.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" The PCI region-request functions become managed functions when pcim_enable_device() has been called previously instead of pci_enable_device(). This has already caused bugs by confusing users, who came to believe that all pci functions, such as pci_iomap(), suddenly are managed that way. This is not the case. Add comments to the relevant functions' docstrings that warn users about this behavior. Signed-off-by: Philipp Stanner --- drivers/pci/iomap.c | 18 ++++++++++++++ drivers/pci/pci.c | 60 ++++++++++++++++++++++++++++++++++++++++++++- 2 files changed, 77 insertions(+), 1 deletion(-) diff --git a/drivers/pci/iomap.c b/drivers/pci/iomap.c index b7faf22ec8f5..0b7fcc2f7a49 100644 --- a/drivers/pci/iomap.c +++ b/drivers/pci/iomap.c @@ -23,6 +23,11 @@ * * @maxlen specifies the maximum length to map. If you want to get access = to * the complete BAR from offset to the end, pass %0 here. + * + * NOTE: + * This function is never managed, even if you initialized with + * pcim_enable_device(). + * If you need automatic cleanup, use pcim_iomap_range(). * */ void __iomem *pci_iomap_range(struct pci_dev *dev, int bar, @@ -63,6 +68,10 @@ EXPORT_SYMBOL(pci_iomap_range); * * @maxlen specifies the maximum length to map. If you want to get access = to * the complete BAR from offset to the end, pass %0 here. + * + * NOTE: + * This function is never managed, even if you initialized with + * pcim_enable_device(). * */ void __iomem *pci_iomap_wc_range(struct pci_dev *dev, int bar, @@ -106,6 +115,11 @@ EXPORT_SYMBOL_GPL(pci_iomap_wc_range); * * @maxlen specifies the maximum length to map. If you want to get access = to * the complete BAR without checking for its length first, pass %0 here. + * + * NOTE: + * This function is never managed, even if you initialized with + * pcim_enable_device(). + * If you need automatic cleanup, use pcim_iomap(). * */ void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long maxlen) { @@ -127,6 +141,10 @@ EXPORT_SYMBOL(pci_iomap); * * @maxlen specifies the maximum length to map. If you want to get access = to * the complete BAR without checking for its length first, pass %0 here. + * + * NOTE: + * This function is never managed, even if you initialized with + * pcim_enable_device(). * */ void __iomem *pci_iomap_wc(struct pci_dev *dev, int bar, unsigned long max= len) { diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index 2899df77a285..7ca860acf351 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -3907,6 +3907,8 @@ EXPORT_SYMBOL(pci_release_region); * @res_name: Name to be associated with resource. * @exclusive: whether the region access is exclusive or not * + * Returns: 0 on success, negative error code on failure. + * * Mark the PCI region associated with PCI device @pdev BAR @bar as * being reserved by owner @res_name. Do not access any * address inside the PCI regions unless this call returns @@ -3918,6 +3920,13 @@ EXPORT_SYMBOL(pci_release_region); * * Returns 0 on success, or %EBUSY on error. A warning * message is also printed on failure. + * + * NOTE: + * This is a "hybrid" function: Its normally unmanaged, but becomes managed + * when pcim_enable_device() has been called in advance. + * This hybrid feature is DEPRECATED! If you need to implement a new pci + * function that does automatic cleanup, write a new pcim_* function that = uses + * devres directly. */ static int __pci_request_region(struct pci_dev *pdev, int bar, const char *res_name, int exclusive) @@ -3966,6 +3975,8 @@ static int __pci_request_region(struct pci_dev *pdev,= int bar, * @bar: BAR to be reserved * @res_name: Name to be associated with resource * + * Returns: 0 on success, negative error code on failure. + * * Mark the PCI region associated with PCI device @pdev BAR @bar as * being reserved by owner @res_name. Do not access any * address inside the PCI regions unless this call returns @@ -3973,6 +3984,12 @@ static int __pci_request_region(struct pci_dev *pdev= , int bar, * * Returns 0 on success, or %EBUSY on error. A warning * message is also printed on failure. + * + * NOTE: + * This is a "hybrid" function: Its normally unmanaged, but becomes managed + * when pcim_enable_device() has been called in advance. + * This hybrid feature is DEPRECATED! If you want managed cleanup, use the + * pcim_* functions instead. */ int pci_request_region(struct pci_dev *pdev, int bar, const char *res_name) { @@ -3998,6 +4015,13 @@ void pci_release_selected_regions(struct pci_dev *pd= ev, int bars) } EXPORT_SYMBOL(pci_release_selected_regions); =20 +/* + * NOTE: + * This is a "hybrid" function: Its normally unmanaged, but becomes managed + * when pcim_enable_device() has been called in advance. + * This hybrid feature is DEPRECATED! If you want managed cleanup, use the + * pcim_* functions instead. + */ static int __pci_request_selected_regions(struct pci_dev *pdev, int bars, const char *res_name, int excl) { @@ -4023,6 +4047,14 @@ static int __pci_request_selected_regions(struct pci= _dev *pdev, int bars, * @pdev: PCI device whose resources are to be reserved * @bars: Bitmask of BARs to be requested * @res_name: Name to be associated with resource + * + * Returns: 0 on success, negative error code on failure. + * + * NOTE: + * This is a "hybrid" function: Its normally unmanaged, but becomes managed + * when pcim_enable_device() has been called in advance. + * This hybrid feature is DEPRECATED! If you want managed cleanup, use the + * pcim_* functions instead. */ int pci_request_selected_regions(struct pci_dev *pdev, int bars, const char *res_name) @@ -4031,6 +4063,20 @@ int pci_request_selected_regions(struct pci_dev *pde= v, int bars, } EXPORT_SYMBOL(pci_request_selected_regions); =20 +/** + * pci_request_selected_regions_exclusive - Request regions exclusively + * @pdev: PCI device to request regions from + * @bars: bit mask of bars to request + * @res_name: name to be associated with the requests + * + * Returns: 0 on success, negative error code on failure. + * + * NOTE: + * This is a "hybrid" function: Its normally unmanaged, but becomes managed + * when pcim_enable_device() has been called in advance. + * This hybrid feature is DEPRECATED! If you want managed cleanup, use the + * pcim_* functions instead. + */ int pci_request_selected_regions_exclusive(struct pci_dev *pdev, int bars, const char *res_name) { @@ -4048,7 +4094,6 @@ EXPORT_SYMBOL(pci_request_selected_regions_exclusive); * successful call to pci_request_regions(). Call this function only * after all use of the PCI regions has ceased. */ - void pci_release_regions(struct pci_dev *pdev) { pci_release_selected_regions(pdev, (1 << PCI_STD_NUM_BARS) - 1); @@ -4080,6 +4125,8 @@ EXPORT_SYMBOL(pci_request_regions); * @pdev: PCI device whose resources are to be reserved * @res_name: Name to be associated with resource. * + * Returns: 0 on success, negative error code on failure. + * * Mark all PCI regions associated with PCI device @pdev as being reserved * by owner @res_name. Do not access any address inside the PCI regions * unless this call returns successfully. @@ -4089,6 +4136,12 @@ EXPORT_SYMBOL(pci_request_regions); * * Returns 0 on success, or %EBUSY on error. A warning message is also * printed on failure. + * + * NOTE: + * This is a "hybrid" function: Its normally unmanaged, but becomes managed + * when pcim_enable_device() has been called in advance. + * This hybrid feature is DEPRECATED! If you want managed cleanup, use the + * pcim_* functions instead. */ int pci_request_regions_exclusive(struct pci_dev *pdev, const char *res_na= me) { @@ -4420,6 +4473,11 @@ void pci_disable_parity(struct pci_dev *dev) * @enable: boolean: whether to enable or disable PCI INTx * * Enables/disables PCI INTx for device @pdev + * + * NOTE: + * This is a "hybrid" function: Its normally unmanaged, but becomes managed + * when pcim_enable_device() has been called in advance. + * This hybrid feature is DEPRECATED! */ void pci_intx(struct pci_dev *pdev, int enable) { --=20 2.43.0 From nobody Tue Dec 16 11:42:59 2025 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.133.124]) (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 B86E81799F for ; Mon, 15 Jan 2024 14:47:25 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=redhat.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=redhat.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=redhat.com header.i=@redhat.com header.b="SV7SESBy" DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1705330044; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=GuPy6o0xdtswTFzGt//PWEkxlEqwBN5Ui7YfUNE5A+E=; b=SV7SESByT91/P8alxXhjdOmMPktimB418QKhoVC3H4zRIo7RIZ5Ubv4Z0WNMRN1IRC3V+z xG4NveWgKebKrN/KpmLjR59PsBh2gaIFgH2K3DgDOoyP4NhOP3Ib4+GBzfAjDXsXWrBJr+ W8t/GQ87c5te2kglQNGoC4XyLXM7HCg= Received: from mail-ua1-f70.google.com (mail-ua1-f70.google.com [209.85.222.70]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.3, cipher=TLS_AES_256_GCM_SHA384) id us-mta-662-MwaY5k1tNu-PaaaSI2QuRA-1; Mon, 15 Jan 2024 09:47:21 -0500 X-MC-Unique: MwaY5k1tNu-PaaaSI2QuRA-1 Received: by mail-ua1-f70.google.com with SMTP id a1e0cc1a2514c-7cdaffe6b9aso377456241.0 for ; Mon, 15 Jan 2024 06:47:21 -0800 (PST) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1705330041; x=1705934841; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=GuPy6o0xdtswTFzGt//PWEkxlEqwBN5Ui7YfUNE5A+E=; b=FqSPvc6KqJgTrURorjBfXpeA98dp92vvhnznjuFUl6eDwxpJm5IJf2SnYoL4qDBF3M 6HITTMpbnyOgTV5BfMXQuiSr51RkCFcZ67lIxGxm7SOmggbyCDmjjMdZJ3SrX4LnpYcm FpX53ZCwoiqQMSOM2plcIJ/m97ZF7+LkI5dJK+hi/wN2/a183y+Mlks6vrG9M5GkWvqd rRa4/oIwZ2HqFqqsVYXEoN3UstCWgMsw4JKMxSzgWmM5D3nAX4cbQdcIQQ3+6csINy2b hDGPjIFnvNPUDiG7vXSONiM7B1ij0L5ZZ4QM+vjMhnO+CB5NJ60SqQ1l91DElJPt/xOA dZtg== X-Gm-Message-State: AOJu0YyuIQo5CSPrV1ngvtaisFJKxOJbqWZM4RftHpV/HJ0cHMO5YqHl kxgvZk5/Hdq0fl1NuvBA7cgvF5dxvYR6Z0dLRh2TtC5GOgt9Avc/7oTq8RViO5dw4Ki1MnjRbAF JDwEHbBlCKPJxg7yNJ+X5sIS/SrR+N5Tr X-Received: by 2002:a05:6122:200a:b0:4b6:e3fa:7599 with SMTP id l10-20020a056122200a00b004b6e3fa7599mr4678900vkd.0.1705330041139; Mon, 15 Jan 2024 06:47:21 -0800 (PST) X-Google-Smtp-Source: AGHT+IHCsY0ZuDYFJr87bladWt7s6eCEJzNh1RKNCl6gfgHXUBWkBCjqEMu+F8KwMd/Grf0wk26qBg== X-Received: by 2002:a05:6122:200a:b0:4b6:e3fa:7599 with SMTP id l10-20020a056122200a00b004b6e3fa7599mr4678878vkd.0.1705330040810; Mon, 15 Jan 2024 06:47:20 -0800 (PST) Received: from pstanner-thinkpadt14sgen1.muc.redhat.com (nat-pool-muc-t.redhat.com. [149.14.88.26]) by smtp.gmail.com with ESMTPSA id ne13-20020a056214424d00b006815cf9a644sm1020720qvb.55.2024.01.15.06.47.18 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Mon, 15 Jan 2024 06:47:20 -0800 (PST) From: Philipp Stanner To: Jonathan Corbet , Hans de Goede , Maarten Lankhorst , Maxime Ripard , Thomas Zimmermann , David Airlie , Daniel Vetter , Bjorn Helgaas , Philipp Stanner , Sam Ravnborg , dakr@redhat.com Cc: linux-doc@vger.kernel.org, linux-kernel@vger.kernel.org, dri-devel@lists.freedesktop.org, linux-pci@vger.kernel.org Subject: [PATCH 04/10] pci: devres: make devres region requests consistent Date: Mon, 15 Jan 2024 15:46:15 +0100 Message-ID: <20240115144655.32046-6-pstanner@redhat.com> X-Mailer: git-send-email 2.43.0 In-Reply-To: <20240115144655.32046-2-pstanner@redhat.com> References: <20240115144655.32046-2-pstanner@redhat.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" Now that pure managed region request functions are available, the implementation of the hybrid-functions which are only sometimes managed can be made more consistent and readable by wrapping those always-managed functions. Implement a new pcim_ function for exclusively requested regions. Have the pci_request / release functions call their pcim_ counterparts. Remove the now surplus region_mask from the devres struct. Signed-off-by: Philipp Stanner --- drivers/pci/devres.c | 49 ++++++++++++++++++++++--------------------- drivers/pci/pci.c | 50 +++++++++++++++----------------------------- drivers/pci/pci.h | 6 ------ include/linux/pci.h | 1 + 4 files changed, 43 insertions(+), 63 deletions(-) diff --git a/drivers/pci/devres.c b/drivers/pci/devres.c index e221919ebbe2..1e5cf950775d 100644 --- a/drivers/pci/devres.c +++ b/drivers/pci/devres.c @@ -22,18 +22,15 @@ * _sometimes_ managed (e.g. pci_request_region()). * Consequently, in the new API, region requests performed by the pcim_ * functions are automatically cleaned up through the devres callback - * pcim_addr_resource_release(), while requests performed by - * pcim_enable_device() + pci_*region*() are automatically cleaned up - * through the for-loop in pcim_release(). + * pcim_addr_resource_release(). + * Users utilizing pcim_enable_device() + pci_*region*() are redirected= in + * pci.c to the managed functions here in this file. This isn't exactly + * perfect, but the only alternative way would be to port ALL drivers u= sing + * said combination to pcim_ functions. * - * TODO 1: + * TODO: * Remove the legacy table entirely once all calls to pcim_iomap_table() in * the kernel have been removed. - * - * TODO 2: - * Port everyone calling pcim_enable_device() + pci_*region*() to using the - * pcim_ functions. Then, remove all devres functionality from pci_*region= *() - * functions and remove the associated cleanups described above in point #= 2. */ =20 /* @@ -407,21 +404,6 @@ static void pcim_release(struct device *gendev, void *= res) { struct pci_dev *dev =3D to_pci_dev(gendev); struct pci_devres *this =3D res; - int i; - - /* - * This is legacy code. - * All regions requested by a pcim_ function do get released through - * pcim_addr_resource_release(). Thanks to the hybrid nature of the pci_ - * region-request functions, this for-loop has to release the regions - * if they have been requested by such a function. - * - * TODO: Remove this once all users of pcim_enable_device() PLUS - * pci-region-request-functions have been ported to pcim_ functions. - */ - for (i =3D 0; i < DEVICE_COUNT_RESOURCE; i++) - if (this->region_mask & (1 << i)) - pci_release_region(dev, i); =20 if (this->mwi) pci_clear_mwi(dev); @@ -971,6 +953,25 @@ int pcim_request_region(struct pci_dev *pdev, int bar,= const char *name) } EXPORT_SYMBOL(pcim_request_region); =20 +/** + * pcim_request_region_exclusive - Request a PCI BAR exclusively + * @pdev: PCI device to requestion region for + * @bar: Index of BAR to request + * @name: Name associated with the request + * + * Returns: 0 on success, a negative error code on failure. + * + * Request region specified by @bar exclusively. + * + * The region will automatically be released on driver detach. If desired, + * release manually only with pcim_release_region(). + */ +int pcim_request_region_exclusive(struct pci_dev *pdev, int bar, const cha= r *name) +{ + return _pcim_request_region(pdev, bar, name, IORESOURCE_EXCLUSIVE); +} +EXPORT_SYMBOL(pcim_request_region_exclusive); + /** * pcim_release_region - Release a PCI BAR * @pdev: PCI device to operate on diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index 7ca860acf351..66639dd754c7 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -3873,7 +3873,15 @@ EXPORT_SYMBOL_GPL(pci_common_swizzle); */ void pci_release_region(struct pci_dev *pdev, int bar) { - struct pci_devres *dr; + /* + * This is done for backwards compatibility, because the old pci-devres + * API had a mode in which the function became managed if it had been + * enabled with pcim_enable_device() instead of pci_enable_device(). + */ + if (pci_is_managed(pdev)) { + pcim_release_region(pdev, bar); + return; + } =20 if (pci_resource_len(pdev, bar) =3D=3D 0) return; @@ -3883,20 +3891,6 @@ void pci_release_region(struct pci_dev *pdev, int ba= r) else if (pci_resource_flags(pdev, bar) & IORESOURCE_MEM) release_mem_region(pci_resource_start(pdev, bar), pci_resource_len(pdev, bar)); - - /* - * This devres utility makes this function sometimes managed - * (when pcim_enable_device() has been called before). - * This is bad because it conflicts with the pcim_ functions being - * exclusively responsible for managed pci. Its "sometimes yes, sometimes - * no" nature can cause bugs. - * - * TODO: Remove this once all users that use pcim_enable_device() PLUS - * a region request function have been ported to using pcim_ functions. - */ - dr =3D find_pci_dr(pdev); - if (dr) - dr->region_mask &=3D ~(1 << bar); } EXPORT_SYMBOL(pci_release_region); =20 @@ -3924,14 +3918,18 @@ EXPORT_SYMBOL(pci_release_region); * NOTE: * This is a "hybrid" function: Its normally unmanaged, but becomes managed * when pcim_enable_device() has been called in advance. - * This hybrid feature is DEPRECATED! If you need to implement a new pci - * function that does automatic cleanup, write a new pcim_* function that = uses - * devres directly. + * This hybrid feature is DEPRECATED! If you want managed cleanup, use the + * pcim_* functions instead. */ static int __pci_request_region(struct pci_dev *pdev, int bar, const char *res_name, int exclusive) { - struct pci_devres *dr; + if (pci_is_managed(pdev)) { + if (exclusive =3D=3D IORESOURCE_EXCLUSIVE) + return pcim_request_region_exclusive(pdev, bar, res_name); + else + return pcim_request_region(pdev, bar, res_name); + } =20 if (pci_resource_len(pdev, bar) =3D=3D 0) return 0; @@ -3947,20 +3945,6 @@ static int __pci_request_region(struct pci_dev *pdev= , int bar, goto err_out; } =20 - /* - * This devres utility makes this function sometimes managed - * (when pcim_enable_device() has been called before). - * This is bad because it conflicts with the pcim_ functions being - * exclusively responsible for managed pci. Its "sometimes yes, sometimes - * no" nature can cause bugs. - * - * TODO: Remove this once all users that use pcim_enable_device() PLUS - * a region request function have been ported to using pcim_ functions. - */ - dr =3D find_pci_dr(pdev); - if (dr) - dr->region_mask |=3D 1 << bar; - return 0; =20 err_out: diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h index 4d4bcc2d850f..7c2181677760 100644 --- a/drivers/pci/pci.h +++ b/drivers/pci/pci.h @@ -814,12 +814,6 @@ struct pci_devres { unsigned int orig_intx:1; unsigned int restore_intx:1; unsigned int mwi:1; - - /* - * TODO: remove the region_mask once everyone calling - * pcim_enable_device() + pci_*region*() is ported to pcim_ functions. - */ - u32 region_mask; }; =20 struct pci_devres *find_pci_dr(struct pci_dev *pdev); diff --git a/include/linux/pci.h b/include/linux/pci.h index 3dcd8a5e10b5..641ee30f7d2d 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -2321,6 +2321,7 @@ int pcim_iomap_regions_request_all(struct pci_dev *pd= ev, int mask, const char *name); void pcim_iounmap_regions(struct pci_dev *pdev, int mask); int pcim_request_region(struct pci_dev *pdev, int bar, const char *res_nam= e); +int pcim_request_region_exclusive(struct pci_dev *pdev, int bar, const cha= r *name); void pcim_release_region(struct pci_dev *pdev, int bar); void pcim_release_all_regions(struct pci_dev *pdev); int pcim_request_all_regions(struct pci_dev *pdev, const char *name); --=20 2.43.0 From nobody Tue Dec 16 11:42:59 2025 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.133.124]) (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 0F860179A5 for ; Mon, 15 Jan 2024 14:47:25 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=redhat.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=redhat.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=redhat.com header.i=@redhat.com header.b="C0HHSfn0" DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1705330045; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=RO0mZDJz5TK909+4LedW3qKsNhM3MJg2yInHziL/J8s=; b=C0HHSfn0qZzHu/IAqsZ53H2L3xXeNpzULGhL2STwLlECph0/nohD/e/tN1QnCM5gnl+tYC QWURZVZEpS5SR9UICD2C07LJjikEbecsEFxvuId1OmdoPMI97ofhXR8ROJQ0gu7uOUWyFe 9jJJL/ZYEWuzgBIKanbmNlPDHedB6Iw= Received: from mail-ua1-f72.google.com (mail-ua1-f72.google.com [209.85.222.72]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.3, cipher=TLS_AES_256_GCM_SHA384) id us-mta-680-9-iHSTusNGitpIQMXQploQ-1; Mon, 15 Jan 2024 09:47:24 -0500 X-MC-Unique: 9-iHSTusNGitpIQMXQploQ-1 Received: by mail-ua1-f72.google.com with SMTP id a1e0cc1a2514c-7cdaffe6b9aso377467241.0 for ; Mon, 15 Jan 2024 06:47:24 -0800 (PST) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1705330043; x=1705934843; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=RO0mZDJz5TK909+4LedW3qKsNhM3MJg2yInHziL/J8s=; b=X9AFvPh0x1COb4Pwp8et/I+2yS6HV+lTnTtJcjtNQ7UEXvPNWEhCYl+uEAYEOI9KqT mZhNeFhcLRPy4hKckqvAh3yc8MCNtr5DgyiPMozTCMVcbgv/mJhnNlYqAxLlCyxKNgdq xJM2kQfRglhgg0qw5Lb/NBCuaoGtSJWze5clbjuto6gFVIVlaxKFfIpWSxWrSrmXLR72 c4DA4zPGAyfDZ2+WuxoFptyhHTvfoyNZPAuPyQCqZSuKtEKsIVlsZuNo6kgNE5SajnGg +3vQW0sBcJzUYZTTKBEcinCCWs6HgovkfCk5aZ4yBpvOSNoIpK27bP2uIexqjlMY9w47 TVzw== X-Gm-Message-State: AOJu0Yw5kXEoiqYxxhebZ/elA2tQYmMYqhH8BNtCfJUBhsqbePylfz1k 4qBTvPYlco5tpk0QaaL7pD+RIF1GtDPKj+5YrakSnvl3e5C/Nvv9dSoUjoqHdiqcrjOBu/r2b3Z R8y0CbiL2Y3LHMeBE+RvLOGnJwYVGPy0a X-Received: by 2002:a05:6122:200a:b0:4b6:e3fa:7599 with SMTP id l10-20020a056122200a00b004b6e3fa7599mr4678997vkd.0.1705330043473; Mon, 15 Jan 2024 06:47:23 -0800 (PST) X-Google-Smtp-Source: AGHT+IH6fwAzY9p/B8Ojoob7vJlPit+LMYVgGiOaxw4BacA7x0sKclRH2EeCFfpGqDBHtgs3IUs2RQ== X-Received: by 2002:a05:6122:200a:b0:4b6:e3fa:7599 with SMTP id l10-20020a056122200a00b004b6e3fa7599mr4678977vkd.0.1705330043173; Mon, 15 Jan 2024 06:47:23 -0800 (PST) Received: from pstanner-thinkpadt14sgen1.muc.redhat.com (nat-pool-muc-t.redhat.com. [149.14.88.26]) by smtp.gmail.com with ESMTPSA id ne13-20020a056214424d00b006815cf9a644sm1020720qvb.55.2024.01.15.06.47.21 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Mon, 15 Jan 2024 06:47:22 -0800 (PST) From: Philipp Stanner To: Jonathan Corbet , Hans de Goede , Maarten Lankhorst , Maxime Ripard , Thomas Zimmermann , David Airlie , Daniel Vetter , Bjorn Helgaas , Philipp Stanner , Sam Ravnborg , dakr@redhat.com Cc: linux-doc@vger.kernel.org, linux-kernel@vger.kernel.org, dri-devel@lists.freedesktop.org, linux-pci@vger.kernel.org Subject: [PATCH 05/10] pci: move enabled status bit to pci_dev struct Date: Mon, 15 Jan 2024 15:46:16 +0100 Message-ID: <20240115144655.32046-7-pstanner@redhat.com> X-Mailer: git-send-email 2.43.0 In-Reply-To: <20240115144655.32046-2-pstanner@redhat.com> References: <20240115144655.32046-2-pstanner@redhat.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" The bit describing whether the PCI device is currently enabled is stored in the devres-struct. Besides this struct being subject of a cleanup process, the device-struct is in general the right place to store this information, since it is not devres-specific. Move the 'enabled' boolean bit to struct pci_dev. Signed-off-by: Philipp Stanner --- drivers/pci/devres.c | 11 ++++------- drivers/pci/pci.c | 17 ++++++++++------- drivers/pci/pci.h | 1 - include/linux/pci.h | 1 + 4 files changed, 15 insertions(+), 15 deletions(-) diff --git a/drivers/pci/devres.c b/drivers/pci/devres.c index 1e5cf950775d..bf957f0bc5ac 100644 --- a/drivers/pci/devres.c +++ b/drivers/pci/devres.c @@ -411,7 +411,7 @@ static void pcim_release(struct device *gendev, void *r= es) if (this->restore_intx) pci_intx(dev, this->orig_intx); =20 - if (this->enabled && !this->pinned) + if (!this->pinned) pci_disable_device(dev); } =20 @@ -456,14 +456,11 @@ int pcim_enable_device(struct pci_dev *pdev) dr =3D get_pci_dr(pdev); if (unlikely(!dr)) return -ENOMEM; - if (dr->enabled) - return 0; =20 rc =3D pci_enable_device(pdev); - if (!rc) { + if (!rc) pdev->is_managed =3D 1; - dr->enabled =3D 1; - } + return rc; } EXPORT_SYMBOL(pcim_enable_device); @@ -481,7 +478,7 @@ void pcim_pin_device(struct pci_dev *pdev) struct pci_devres *dr; =20 dr =3D find_pci_dr(pdev); - WARN_ON(!dr || !dr->enabled); + WARN_ON(!dr || !pdev->enabled); if (dr) dr->pinned =3D 1; } diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index 66639dd754c7..0fb9e7bb0e43 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -1953,6 +1953,9 @@ static int do_pci_enable_device(struct pci_dev *dev, = int bars) u16 cmd; u8 pin; =20 + if (dev->enabled) + return 0; + err =3D pci_set_power_state(dev, PCI_D0); if (err < 0 && err !=3D -EIO) return err; @@ -1967,7 +1970,7 @@ static int do_pci_enable_device(struct pci_dev *dev, = int bars) pci_fixup_device(pci_fixup_enable, dev); =20 if (dev->msi_enabled || dev->msix_enabled) - return 0; + goto success_out; =20 pci_read_config_byte(dev, PCI_INTERRUPT_PIN, &pin); if (pin) { @@ -1977,6 +1980,8 @@ static int do_pci_enable_device(struct pci_dev *dev, = int bars) cmd & ~PCI_COMMAND_INTX_DISABLE); } =20 +success_out: + dev->enabled =3D true; return 0; } =20 @@ -2146,6 +2151,9 @@ static void do_pci_disable_device(struct pci_dev *dev) { u16 pci_command; =20 + if (!dev->enabled) + return; + pci_read_config_word(dev, PCI_COMMAND, &pci_command); if (pci_command & PCI_COMMAND_MASTER) { pci_command &=3D ~PCI_COMMAND_MASTER; @@ -2153,6 +2161,7 @@ static void do_pci_disable_device(struct pci_dev *dev) } =20 pcibios_disable_device(dev); + dev->enabled =3D false; } =20 /** @@ -2180,12 +2189,6 @@ void pci_disable_enabled_device(struct pci_dev *dev) */ void pci_disable_device(struct pci_dev *dev) { - struct pci_devres *dr; - - dr =3D find_pci_dr(dev); - if (dr) - dr->enabled =3D 0; - dev_WARN_ONCE(&dev->dev, atomic_read(&dev->enable_cnt) <=3D 0, "disabling already-disabled device"); =20 diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h index 7c2181677760..dbb76a3fb0e4 100644 --- a/drivers/pci/pci.h +++ b/drivers/pci/pci.h @@ -809,7 +809,6 @@ static inline pci_power_t mid_pci_get_power_state(struc= t pci_dev *pdev) * when a device is enabled using managed PCI device enable interface. */ struct pci_devres { - unsigned int enabled:1; unsigned int pinned:1; unsigned int orig_intx:1; unsigned int restore_intx:1; diff --git a/include/linux/pci.h b/include/linux/pci.h index 641ee30f7d2d..a356bdcc14cc 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -367,6 +367,7 @@ struct pci_dev { this is D0-D3, D0 being fully functional, and D3 being off. */ u8 pm_cap; /* PM capability offset */ + unsigned int enabled:1; /* Whether this dev is enabled */ unsigned int imm_ready:1; /* Supports Immediate Readiness */ unsigned int pme_support:5; /* Bitmask of states from which PME# can be generated */ --=20 2.43.0 From nobody Tue Dec 16 11:42:59 2025 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.129.124]) (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 181CE17BC7 for ; Mon, 15 Jan 2024 14:47:28 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=redhat.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=redhat.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=redhat.com header.i=@redhat.com header.b="BD+zOudQ" DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1705330048; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=k3vu21hkZlZsapFSJtp8nhKKQ/3OZTkpZfYIme9+49k=; b=BD+zOudQkPf/CYbLETcnHrpIm1cUoXYFUOO0K4nZiU8n/wHtXXFmiYUQ8e0eJ7CZAwvYhj FJt/pq46Q3Ioil40c0MRFS0A6MkGDOxHEHRMeAClUf0d41M/CUbUnmsoc+sacGgLsx2NS1 059A8ROOW+4lKQJK1eNdwcr2vCbwsHQ= Received: from mail-vk1-f197.google.com (mail-vk1-f197.google.com [209.85.221.197]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.3, cipher=TLS_AES_256_GCM_SHA384) id us-mta-303-DiF4Cu-INhiTmwYDeM7lVw-1; Mon, 15 Jan 2024 09:47:26 -0500 X-MC-Unique: DiF4Cu-INhiTmwYDeM7lVw-1 Received: by mail-vk1-f197.google.com with SMTP id 71dfb90a1353d-4b7bdc844fcso135153e0c.0 for ; Mon, 15 Jan 2024 06:47:26 -0800 (PST) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1705330046; x=1705934846; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=k3vu21hkZlZsapFSJtp8nhKKQ/3OZTkpZfYIme9+49k=; b=WQ11PYcA1fq4S8fu5h10ekqn0bgPbYDl97kPbXHH+ysKq5GnZWAqOSTj2P0a/ZFA4t qEro/2jw1sBwqKCWH4wZCVyCAv5wOtapMXYQcdlC4Wlf4FfLMAeXOMoCo5CapvidvX8W eIrwJFaBMkvDpgHz22gnnwnm1Li+RvYdOpfmvqX6xgz3gtomthzgRFE8N15Rk2BuinAg 45ihnR7AS9MXF2Q9Fln6ilwTHi6EPWxmRSwkbxbNW+DvTh+AZgTsUCxmYZMvV6Hkcugc Qzdx/yTlrg8UfnNGj1v9q/yPI+U6s6bJRttMfpBueF1Wy7710z1PqBctKQssNzJDUytx S/DQ== X-Gm-Message-State: AOJu0YwOEKEPOv9ocCkP96Hh1C83HPL0t5RAwhtlaWs2qD61Rcfn/U5q qA4eMYKQN9IrCAAzO+M7KdF2OjmYwWaNMnzDSiHSa16EafVNgVpxyOhNeywKwu6kjpjUFHc+7Py UI5zoYq35BXkms7vXxRwI1n0kfzVIEaG/ X-Received: by 2002:a05:6122:2886:b0:4b7:32ac:f4e5 with SMTP id fl6-20020a056122288600b004b732acf4e5mr4682343vkb.1.1705330046110; Mon, 15 Jan 2024 06:47:26 -0800 (PST) X-Google-Smtp-Source: AGHT+IEl3g0muu7xD/Tj+Zr9TyY8gOWhOJboojbdmamx1smFfYvzduqw18AZ9+hpq2S3HXA+drCMKA== X-Received: by 2002:a05:6122:2886:b0:4b7:32ac:f4e5 with SMTP id fl6-20020a056122288600b004b732acf4e5mr4682321vkb.1.1705330045850; Mon, 15 Jan 2024 06:47:25 -0800 (PST) Received: from pstanner-thinkpadt14sgen1.muc.redhat.com (nat-pool-muc-t.redhat.com. [149.14.88.26]) by smtp.gmail.com with ESMTPSA id ne13-20020a056214424d00b006815cf9a644sm1020720qvb.55.2024.01.15.06.47.23 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Mon, 15 Jan 2024 06:47:25 -0800 (PST) From: Philipp Stanner To: Jonathan Corbet , Hans de Goede , Maarten Lankhorst , Maxime Ripard , Thomas Zimmermann , David Airlie , Daniel Vetter , Bjorn Helgaas , Philipp Stanner , Sam Ravnborg , dakr@redhat.com Cc: linux-doc@vger.kernel.org, linux-kernel@vger.kernel.org, dri-devel@lists.freedesktop.org, linux-pci@vger.kernel.org Subject: [PATCH 06/10] pci: move pinned status bit to pci_dev struct Date: Mon, 15 Jan 2024 15:46:17 +0100 Message-ID: <20240115144655.32046-8-pstanner@redhat.com> X-Mailer: git-send-email 2.43.0 In-Reply-To: <20240115144655.32046-2-pstanner@redhat.com> References: <20240115144655.32046-2-pstanner@redhat.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" The bit describing whether the PCI device is currently pinned is stored in the PCI devres struct. To clean up and simplify the pci-devres API, it's better if this information is stored in the pci_dev struct, because it allows for checking that device's pinned-status directly through the device struct. This will later permit simplifying pcim_enable_device(). Move the 'pinned' boolean bit to struct pci_dev. Signed-off-by: Philipp Stanner --- drivers/pci/devres.c | 14 ++++---------- drivers/pci/pci.h | 1 - include/linux/pci.h | 1 + 3 files changed, 5 insertions(+), 11 deletions(-) diff --git a/drivers/pci/devres.c b/drivers/pci/devres.c index bf957f0bc5ac..de8cf6f87dd7 100644 --- a/drivers/pci/devres.c +++ b/drivers/pci/devres.c @@ -411,7 +411,7 @@ static void pcim_release(struct device *gendev, void *r= es) if (this->restore_intx) pci_intx(dev, this->orig_intx); =20 - if (!this->pinned) + if (!dev->pinned) pci_disable_device(dev); } =20 @@ -469,18 +469,12 @@ EXPORT_SYMBOL(pcim_enable_device); * pcim_pin_device - Pin managed PCI device * @pdev: PCI device to pin * - * Pin managed PCI device @pdev. Pinned device won't be disabled on - * driver detach. @pdev must have been enabled with - * pcim_enable_device(). + * Pin managed PCI device @pdev. Pinned device won't be disabled on driver + * detach. @pdev must have been enabled with pcim_enable_device(). */ void pcim_pin_device(struct pci_dev *pdev) { - struct pci_devres *dr; - - dr =3D find_pci_dr(pdev); - WARN_ON(!dr || !pdev->enabled); - if (dr) - dr->pinned =3D 1; + pdev->pinned =3D true; } EXPORT_SYMBOL(pcim_pin_device); =20 diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h index dbb76a3fb0e4..3d9908a69ebf 100644 --- a/drivers/pci/pci.h +++ b/drivers/pci/pci.h @@ -809,7 +809,6 @@ static inline pci_power_t mid_pci_get_power_state(struc= t pci_dev *pdev) * when a device is enabled using managed PCI device enable interface. */ struct pci_devres { - unsigned int pinned:1; unsigned int orig_intx:1; unsigned int restore_intx:1; unsigned int mwi:1; diff --git a/include/linux/pci.h b/include/linux/pci.h index a356bdcc14cc..2f6f44991003 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -368,6 +368,7 @@ struct pci_dev { functional, and D3 being off. */ u8 pm_cap; /* PM capability offset */ unsigned int enabled:1; /* Whether this dev is enabled */ + unsigned int pinned:1; /* Whether this dev is pinned */ unsigned int imm_ready:1; /* Supports Immediate Readiness */ unsigned int pme_support:5; /* Bitmask of states from which PME# can be generated */ --=20 2.43.0 From nobody Tue Dec 16 11:42:59 2025 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.129.124]) (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 92C4F18029 for ; Mon, 15 Jan 2024 14:47:31 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=redhat.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=redhat.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=redhat.com header.i=@redhat.com header.b="MY9KF10d" DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1705330050; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=+s18P47JH053YQ7YJMWBY/f4LBPoNabsZFa7u4jw7Zc=; b=MY9KF10dJJriNnA+ntXlNTyECtuyfTx/iWDaWF94NhZBICXx1jE+Y0rLPMUKemMUmUyrff cTtJab0jF3hOPl8yUnylVIwPFkTbNqOTRA4PveHaJDtbcEUGzZoEFIKCW4Xt12rwks6O0X ckr+aClP7z21CR7afbA6rS8/q8OCFUM= Received: from mail-qv1-f72.google.com (mail-qv1-f72.google.com [209.85.219.72]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.3, cipher=TLS_AES_256_GCM_SHA384) id us-mta-85-CojAguqANSmZ28E_EP5WWQ-1; Mon, 15 Jan 2024 09:47:29 -0500 X-MC-Unique: CojAguqANSmZ28E_EP5WWQ-1 Received: by mail-qv1-f72.google.com with SMTP id 6a1803df08f44-6801dd788faso26142136d6.0 for ; Mon, 15 Jan 2024 06:47:29 -0800 (PST) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1705330048; x=1705934848; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=+s18P47JH053YQ7YJMWBY/f4LBPoNabsZFa7u4jw7Zc=; b=NPcaqbJZAXEu421uqPmzt4QrcWtX21BOFEEUhowikJTX9bO+T9kY5KrHxTLUwG+M5+ T6nrHr3aZ1qEJoAhwbyNYmwQzgmLtDYUcLAEquB/qItL1Yiib+nc9K0XGYyKxsblrFpZ YJxXrteX45n3SNpvYDYZjnAGysXudL2twzN1X+eEBjGqEdawj32fK7QxrwUBqBOJbqBt Lmn0XFAQUQT6fuAclO078XP0DZvdpr1xDtFAg1deanA/i9SGqgdmysujBvNJy31BnP27 5V1josV+3P/W8dHgOEzUf/iZHl8MJAqPdtp8pHbxJoY1L+wczXCJTihs3EsHeZxCej72 1GPQ== X-Gm-Message-State: AOJu0YxYAkHNRUpFsTOhkt44YbLLWVjCUYAo5L0/Pm3tXHFexeWJY3J4 wS835VbZqTEszHM9OxmJKBMlRm+9hz+qthV6/wKLbymb5db62mmAZvpHJxaEyryYm0KL+1/2Mcm mqPfoxx5uhEUCdB5LkK7mNV5GFu5NkU3q X-Received: by 2002:a05:6214:20ca:b0:681:555c:8e64 with SMTP id 10-20020a05621420ca00b00681555c8e64mr6594574qve.6.1705330048591; Mon, 15 Jan 2024 06:47:28 -0800 (PST) X-Google-Smtp-Source: AGHT+IHIFNFH3t9YBVxCpLVDR4bI9FQ3w+V6Y/t4RZxhNK4CjYtj3KjB51ezdoAU+KVMDwPukDm1tQ== X-Received: by 2002:a05:6214:20ca:b0:681:555c:8e64 with SMTP id 10-20020a05621420ca00b00681555c8e64mr6594555qve.6.1705330048349; Mon, 15 Jan 2024 06:47:28 -0800 (PST) Received: from pstanner-thinkpadt14sgen1.muc.redhat.com (nat-pool-muc-t.redhat.com. [149.14.88.26]) by smtp.gmail.com with ESMTPSA id ne13-20020a056214424d00b006815cf9a644sm1020720qvb.55.2024.01.15.06.47.26 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Mon, 15 Jan 2024 06:47:28 -0800 (PST) From: Philipp Stanner To: Jonathan Corbet , Hans de Goede , Maarten Lankhorst , Maxime Ripard , Thomas Zimmermann , David Airlie , Daniel Vetter , Bjorn Helgaas , Philipp Stanner , Sam Ravnborg , dakr@redhat.com Cc: linux-doc@vger.kernel.org, linux-kernel@vger.kernel.org, dri-devel@lists.freedesktop.org, linux-pci@vger.kernel.org Subject: [PATCH 07/10] pci: devres: give mwi its own callback Date: Mon, 15 Jan 2024 15:46:18 +0100 Message-ID: <20240115144655.32046-9-pstanner@redhat.com> X-Mailer: git-send-email 2.43.0 In-Reply-To: <20240115144655.32046-2-pstanner@redhat.com> References: <20240115144655.32046-2-pstanner@redhat.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" managing mwi with devres can easily be done with its own callback, without the necessity to store any state about it in a device-related struct. Remove the mwi state from the devres-struct. Make the devres-mwi functions set a separate devres-callback. Signed-off-by: Philipp Stanner --- drivers/pci/devres.c | 23 +++++++++++------------ drivers/pci/pci.h | 1 - 2 files changed, 11 insertions(+), 13 deletions(-) diff --git a/drivers/pci/devres.c b/drivers/pci/devres.c index de8cf6f87dd7..911c2037d9fd 100644 --- a/drivers/pci/devres.c +++ b/drivers/pci/devres.c @@ -378,24 +378,26 @@ void __iomem *devm_pci_remap_cfg_resource(struct devi= ce *dev, } EXPORT_SYMBOL(devm_pci_remap_cfg_resource); =20 +static void __pcim_clear_mwi(void *pdev_raw) +{ + struct pci_dev *pdev =3D pdev_raw; + + pci_clear_mwi(pdev); +} + /** * pcim_set_mwi - a device-managed pci_set_mwi() - * @dev: the PCI device for which MWI is enabled + * @pdev: the PCI device for which MWI is enabled * * Managed pci_set_mwi(). * * RETURNS: An appropriate -ERRNO error value on error, or zero for succes= s. */ -int pcim_set_mwi(struct pci_dev *dev) +int pcim_set_mwi(struct pci_dev *pdev) { - struct pci_devres *dr; - - dr =3D find_pci_dr(dev); - if (!dr) - return -ENOMEM; + devm_add_action(&pdev->dev, __pcim_clear_mwi, pdev); =20 - dr->mwi =3D 1; - return pci_set_mwi(dev); + return pci_set_mwi(pdev); } EXPORT_SYMBOL(pcim_set_mwi); =20 @@ -405,9 +407,6 @@ static void pcim_release(struct device *gendev, void *r= es) struct pci_dev *dev =3D to_pci_dev(gendev); struct pci_devres *this =3D res; =20 - if (this->mwi) - pci_clear_mwi(dev); - if (this->restore_intx) pci_intx(dev, this->orig_intx); =20 diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h index 3d9908a69ebf..667bfdd61d5e 100644 --- a/drivers/pci/pci.h +++ b/drivers/pci/pci.h @@ -811,7 +811,6 @@ static inline pci_power_t mid_pci_get_power_state(struc= t pci_dev *pdev) struct pci_devres { unsigned int orig_intx:1; unsigned int restore_intx:1; - unsigned int mwi:1; }; =20 struct pci_devres *find_pci_dr(struct pci_dev *pdev); --=20 2.43.0 From nobody Tue Dec 16 11:42:59 2025 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.129.124]) (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 29086182B5 for ; Mon, 15 Jan 2024 14:47:34 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=redhat.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=redhat.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=redhat.com header.i=@redhat.com header.b="G3MeuJj5" DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1705330053; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=NKQuejYaTfOy+ygLhxh3ezuOqu6KilFVvxO7msTQ6Q0=; b=G3MeuJj5ekiHdioLT1bKZtcShERGVuH4pcCLr82mzY9q4Me+aIQUjL2e3YuXwp3MMchP1w nJ7VdiVsKlsN/B9T9qDtx1m1N2AZNAwmKQ4L85dCHStwkMOxJ9eiP9E3W9v2myeZ5dxFvU zxaC5RXOhWDjwt1kgweaZu28HbXI3sw= Received: from mail-qv1-f72.google.com (mail-qv1-f72.google.com [209.85.219.72]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.3, cipher=TLS_AES_256_GCM_SHA384) id us-mta-371-DSKNP-RJNySRBlvuwo1TCQ-1; Mon, 15 Jan 2024 09:47:32 -0500 X-MC-Unique: DSKNP-RJNySRBlvuwo1TCQ-1 Received: by mail-qv1-f72.google.com with SMTP id 6a1803df08f44-67f826cc03dso24820026d6.1 for ; Mon, 15 Jan 2024 06:47:32 -0800 (PST) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1705330051; x=1705934851; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=NKQuejYaTfOy+ygLhxh3ezuOqu6KilFVvxO7msTQ6Q0=; b=KjXJY7nP9mQYblgKkBvOkf0VnyY7lXHsps2PIR35xLduKKlMmI095ANW+3y9YIiIhk i+ophGGV8rRtrdtc0WqGGeWy6x3WKyNFSufv3aKUjfcbc1W7Pu4HejU60OHFzh6d7sLw 3sTVca6wO0Hl6HEhk7pndKEyQ86yZHN/5lt8SQjB24LjJ9qAYdcA8yHePWpB6uD0DYpy OKmXzp27qJkoRt0tbLbE2f2ZcixflD3Jf80uMBe2AQaynYJFBw4RJJKzijR16ldlJxN/ 5XxCQQMb75BdZU22oMVaAOzdG2xY4Z1Rjvy6M3XojxynVGa3pNWqAR2Zxm+KOVKw0kEm ZQ4w== X-Gm-Message-State: AOJu0Yz+5xvy/M1RPbemwd/hQkpZ4vhPPSUGws4EI7C6VnFI7LBxXMmY cEmjndJE4VCML6QiSRJIwNknS3O2JzVA6LcPjT/UbteXNKbN3379RaHjTNW29Gso90dLbd9lqYp f6eL1TlZHcGfA8kus9qXoGFqoWZ2OxfaU X-Received: by 2002:a05:6214:2409:b0:680:fee0:b9e9 with SMTP id fv9-20020a056214240900b00680fee0b9e9mr10832203qvb.0.1705330051543; Mon, 15 Jan 2024 06:47:31 -0800 (PST) X-Google-Smtp-Source: AGHT+IHKb3jA3FUB/hnqVjWXw40a5VCI+1scQGITF6TP15jP4nIN+0I4b0P9xvnyisVbpi9btL+QGg== X-Received: by 2002:a05:6214:2409:b0:680:fee0:b9e9 with SMTP id fv9-20020a056214240900b00680fee0b9e9mr10832176qvb.0.1705330051170; Mon, 15 Jan 2024 06:47:31 -0800 (PST) Received: from pstanner-thinkpadt14sgen1.muc.redhat.com (nat-pool-muc-t.redhat.com. [149.14.88.26]) by smtp.gmail.com with ESMTPSA id ne13-20020a056214424d00b006815cf9a644sm1020720qvb.55.2024.01.15.06.47.28 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Mon, 15 Jan 2024 06:47:30 -0800 (PST) From: Philipp Stanner To: Jonathan Corbet , Hans de Goede , Maarten Lankhorst , Maxime Ripard , Thomas Zimmermann , David Airlie , Daniel Vetter , Bjorn Helgaas , Philipp Stanner , Sam Ravnborg , dakr@redhat.com Cc: linux-doc@vger.kernel.org, linux-kernel@vger.kernel.org, dri-devel@lists.freedesktop.org, linux-pci@vger.kernel.org Subject: [PATCH 08/10] pci: devres: give pci(m)_intx its own callback Date: Mon, 15 Jan 2024 15:46:19 +0100 Message-ID: <20240115144655.32046-10-pstanner@redhat.com> X-Mailer: git-send-email 2.43.0 In-Reply-To: <20240115144655.32046-2-pstanner@redhat.com> References: <20240115144655.32046-2-pstanner@redhat.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" pci_intx() is one of the functions that have "hybrid mode" (i.e., sometimes managed, sometimes not). Providing a separate pcim_intx() function with its own device resource and cleanup callback allows for removing further large parts of the legacy pci-devres implementation. As in the region-request-functions, pci_intx() has to call into its managed counterpart for backwards compatibility. Implement pcim_intx() with its own device resource. Make pci_intx() call pcim_intx() in the managed case. Remove the legacy devres struct from pci.h. Signed-off-by: Philipp Stanner --- drivers/pci/devres.c | 77 +++++++++++++++++++++++++++++++++++--------- drivers/pci/pci.c | 23 +++++++------ drivers/pci/pci.h | 20 ------------ include/linux/pci.h | 1 + 4 files changed, 75 insertions(+), 46 deletions(-) diff --git a/drivers/pci/devres.c b/drivers/pci/devres.c index 911c2037d9fd..7c4edcaeb8fe 100644 --- a/drivers/pci/devres.c +++ b/drivers/pci/devres.c @@ -40,6 +40,11 @@ struct pcim_iomap_devres { void __iomem *table[PCI_STD_NUM_BARS]; }; =20 +/* Used to restore the old intx state on driver detach. */ +struct pcim_intx_devres { + int orig_intx; +}; + enum pcim_addr_devres_type { /* Default initializer. */ PCIM_ADDR_DEVRES_TYPE_INVALID, @@ -401,31 +406,71 @@ int pcim_set_mwi(struct pci_dev *pdev) } EXPORT_SYMBOL(pcim_set_mwi); =20 +static void pcim_intx_restore(struct device *dev, void *data) +{ + struct pci_dev *pdev =3D to_pci_dev(dev); + struct pcim_intx_devres *res =3D data; =20 -static void pcim_release(struct device *gendev, void *res) + pci_intx(pdev, res->orig_intx); +} + +static struct pcim_intx_devres *get_or_create_intx_devres(struct device *d= ev) { - struct pci_dev *dev =3D to_pci_dev(gendev); - struct pci_devres *this =3D res; + struct pcim_intx_devres *res; =20 - if (this->restore_intx) - pci_intx(dev, this->orig_intx); + res =3D devres_find(dev, pcim_intx_restore, NULL, NULL); + if (res) + return res; =20 - if (!dev->pinned) - pci_disable_device(dev); + res =3D devres_alloc(pcim_intx_restore, sizeof(*res), GFP_KERNEL); + if (res) + devres_add(dev, res); + + return res; } =20 -/* - * TODO: - * Once the last four callers in pci.c are ported, this function here need= s to - * be made static again. +/** + * pcim_intx - managed pci_intx() + * @pdev: the PCI device to operate on + * @enable: boolean: whether to enable or disable PCI INTx + * + * Returns: 0 on success, -ENOMEM on error. + * + * Enables/disables PCI INTx for device @pdev. + * Restores the original state on driver detach. */ -struct pci_devres *find_pci_dr(struct pci_dev *pdev) +int pcim_intx(struct pci_dev *pdev, int enable) { - if (pci_is_managed(pdev)) - return devres_find(&pdev->dev, pcim_release, NULL, NULL); - return NULL; + u16 pci_command, new; + struct pcim_intx_devres *res; + + res =3D get_or_create_intx_devres(&pdev->dev); + if (!res) + return -ENOMEM; + + res->orig_intx =3D !enable; + + pci_read_config_word(pdev, PCI_COMMAND, &pci_command); + + if (enable) + new =3D pci_command & ~PCI_COMMAND_INTX_DISABLE; + else + new =3D pci_command | PCI_COMMAND_INTX_DISABLE; + + if (new !=3D pci_command) + pci_write_config_word(pdev, PCI_COMMAND, new); + + return 0; +} +EXPORT_SYMBOL_GPL(pcim_intx); + +static void pcim_release(struct device *gendev, void *res) +{ + struct pci_dev *dev =3D to_pci_dev(gendev); + + if (!dev->pinned) + pci_disable_device(dev); } -EXPORT_SYMBOL(find_pci_dr); =20 static struct pci_devres *get_pci_dr(struct pci_dev *pdev) { diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index 0fb9e7bb0e43..a03ad3f17540 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -4465,11 +4465,23 @@ void pci_disable_parity(struct pci_dev *dev) * This is a "hybrid" function: Its normally unmanaged, but becomes managed * when pcim_enable_device() has been called in advance. * This hybrid feature is DEPRECATED! + * Use pcim_intx() if you need a managed version. */ void pci_intx(struct pci_dev *pdev, int enable) { u16 pci_command, new; =20 + /* + * This is done for backwards compatibility, because the old pci-devres + * API had a mode in which this function became managed if the dev had + * been enabled with pcim_enable_device() instead of pci_enable_device(). + */ + if (pci_is_managed(pdev)) { + if (pcim_intx(pdev, enable) !=3D 0) + WARN_ON_ONCE(1); + return; + } + pci_read_config_word(pdev, PCI_COMMAND, &pci_command); =20 if (enable) @@ -4477,17 +4489,8 @@ void pci_intx(struct pci_dev *pdev, int enable) else new =3D pci_command | PCI_COMMAND_INTX_DISABLE; =20 - if (new !=3D pci_command) { - struct pci_devres *dr; - + if (new !=3D pci_command) pci_write_config_word(pdev, PCI_COMMAND, new); - - dr =3D find_pci_dr(pdev); - if (dr && !dr->restore_intx) { - dr->restore_intx =3D 1; - dr->orig_intx =3D !enable; - } - } } EXPORT_SYMBOL_GPL(pci_intx); =20 diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h index 667bfdd61d5e..f43873049d52 100644 --- a/drivers/pci/pci.h +++ b/drivers/pci/pci.h @@ -795,26 +795,6 @@ static inline pci_power_t mid_pci_get_power_state(stru= ct pci_dev *pdev) } #endif =20 -/* - * TODO: - * The following two components wouldn't need to be here if they weren't u= sed at - * four last places in pci.c - * Port or move these functions to devres.c and then remove the - * pci_devres-components from this header file here. - */ -/* - * Managed PCI resources. This manages device on/off, INTx/MSI/MSI-X - * on/off and BAR regions. pci_dev itself records MSI/MSI-X status, so - * there's no need to track it separately. pci_devres is initialized - * when a device is enabled using managed PCI device enable interface. - */ -struct pci_devres { - unsigned int orig_intx:1; - unsigned int restore_intx:1; -}; - -struct pci_devres *find_pci_dr(struct pci_dev *pdev); - /* * Config Address for PCI Configuration Mechanism #1 * diff --git a/include/linux/pci.h b/include/linux/pci.h index 2f6f44991003..721657cb59d8 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -1354,6 +1354,7 @@ int pci_try_set_mwi(struct pci_dev *dev); void pci_clear_mwi(struct pci_dev *dev); void pci_disable_parity(struct pci_dev *dev); void pci_intx(struct pci_dev *dev, int enable); +int pcim_intx(struct pci_dev *dev, int enable); bool pci_check_and_mask_intx(struct pci_dev *dev); bool pci_check_and_unmask_intx(struct pci_dev *dev); int pci_wait_for_pending(struct pci_dev *dev, int pos, u16 mask); --=20 2.43.0 From nobody Tue Dec 16 11:42:59 2025 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.129.124]) (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 8443718054 for ; Mon, 15 Jan 2024 14:47:36 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=redhat.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=redhat.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=redhat.com header.i=@redhat.com header.b="ZKQ3noT3" DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1705330055; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=6EAnPeJipP0+07a3Rzv2/tb0sjf4e9bqcclnBtzQ1Hw=; b=ZKQ3noT3lMoVquehcRddhaUCwGV6ujheO9KpWFHab+hmC6hI3flsI6pCuKys4shfIyBBqN baBenEE2mvEKS4NbrYPik3cvmQ2Kf8oIQENcet3PGysWAZOK1jdGPA9Uw6se1EqK1wG2w0 AeJ9ZD4eVynshdmzT7EPi84UsGog8HA= Received: from mail-yw1-f199.google.com (mail-yw1-f199.google.com [209.85.128.199]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.3, cipher=TLS_AES_256_GCM_SHA384) id us-mta-119-1vqw7SufOvyTCfaorb1p3w-1; Mon, 15 Jan 2024 09:47:34 -0500 X-MC-Unique: 1vqw7SufOvyTCfaorb1p3w-1 Received: by mail-yw1-f199.google.com with SMTP id 00721157ae682-5fbfa11838cso5004567b3.1 for ; Mon, 15 Jan 2024 06:47:34 -0800 (PST) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1705330054; x=1705934854; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=6EAnPeJipP0+07a3Rzv2/tb0sjf4e9bqcclnBtzQ1Hw=; b=gWB8l5xwervuu3qJG/lYwXiZSbAK0w3VnHjYpRKheNR8yPQ0/QmgpdJ44+UyYRKb+f Ef5e+LC+ROr4Im37WvNkxDWbkLQ3Jd2GZrEQrfzKRV+jeaWrE5Q8V9LLbXz/yUHeklK+ Y0hbSm408T14QBqHNJ7gImhnO7twQN6GoOpuYUzqu708mD3IRGmpnQPzg9umBBuZu9Qr A2uj66zSLTrkVteSKcVm82vS6ryNSNFVMyrWiCRNl5zzKmNhTdP42DZE9xgN1wJRaHyt wE3VSZOP8yPDzOEdZYVKro2Dp8eFMtdl7YVL/coUW5If+1TKkhm3mHEAGcj80oH+D7O0 8MHw== X-Gm-Message-State: AOJu0YzWsyluPl5TDhDcaVWiAVCJfHJBoM1O2IvCb1zAhDuWj7WlPxvC 8W2NNABqGHM6BxS+vQnazWiqcx385YghNAYu29QO9k2ygF+5mxoN2G8XTzd1+N6ZjEn9dkfd2/3 OrvJ4n3aTTlnlEwfEwDVs2t1+VxquQcKC X-Received: by 2002:a81:9201:0:b0:5af:778e:d53b with SMTP id j1-20020a819201000000b005af778ed53bmr4653018ywg.0.1705330053950; Mon, 15 Jan 2024 06:47:33 -0800 (PST) X-Google-Smtp-Source: AGHT+IFcCQGz1Mr7VFhpZST+41Zk2azRO3t8k7Tu1u+nJkXT6NHZsv59rkNJWBrwhybUyX7sFdqIfA== X-Received: by 2002:a81:9201:0:b0:5af:778e:d53b with SMTP id j1-20020a819201000000b005af778ed53bmr4653002ywg.0.1705330053647; Mon, 15 Jan 2024 06:47:33 -0800 (PST) Received: from pstanner-thinkpadt14sgen1.muc.redhat.com (nat-pool-muc-t.redhat.com. [149.14.88.26]) by smtp.gmail.com with ESMTPSA id ne13-20020a056214424d00b006815cf9a644sm1020720qvb.55.2024.01.15.06.47.31 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Mon, 15 Jan 2024 06:47:33 -0800 (PST) From: Philipp Stanner To: Jonathan Corbet , Hans de Goede , Maarten Lankhorst , Maxime Ripard , Thomas Zimmermann , David Airlie , Daniel Vetter , Bjorn Helgaas , Philipp Stanner , Sam Ravnborg , dakr@redhat.com Cc: linux-doc@vger.kernel.org, linux-kernel@vger.kernel.org, dri-devel@lists.freedesktop.org, linux-pci@vger.kernel.org Subject: [PATCH 09/10] pci: devres: remove legacy pcim_release() Date: Mon, 15 Jan 2024 15:46:20 +0100 Message-ID: <20240115144655.32046-11-pstanner@redhat.com> X-Mailer: git-send-email 2.43.0 In-Reply-To: <20240115144655.32046-2-pstanner@redhat.com> References: <20240115144655.32046-2-pstanner@redhat.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" Thanks to preceding cleanup steps, pcim_release() is now not needed anymore and can be replaced by pcim_disable_device(), which is the exact counterpart to pcim_enable_device(). This permits removing further parts of the old devres API. Replace pcim_release() with pcim_disable_device(). Remove the now surplus get_dr() function. Signed-off-by: Philipp Stanner --- drivers/pci/devres.c | 38 ++++++++------------------------------ 1 file changed, 8 insertions(+), 30 deletions(-) diff --git a/drivers/pci/devres.c b/drivers/pci/devres.c index 7c4edcaeb8fe..87bc62be21eb 100644 --- a/drivers/pci/devres.c +++ b/drivers/pci/devres.c @@ -464,48 +464,26 @@ int pcim_intx(struct pci_dev *pdev, int enable) } EXPORT_SYMBOL_GPL(pcim_intx); =20 -static void pcim_release(struct device *gendev, void *res) +static void pcim_disable_device(void *pdev_raw) { - struct pci_dev *dev =3D to_pci_dev(gendev); - - if (!dev->pinned) - pci_disable_device(dev); -} - -static struct pci_devres *get_pci_dr(struct pci_dev *pdev) -{ - struct pci_devres *dr, *new_dr; - - dr =3D devres_find(&pdev->dev, pcim_release, NULL, NULL); - if (dr) - return dr; + struct pci_dev *pdev =3D pdev_raw; =20 - new_dr =3D devres_alloc(pcim_release, sizeof(*new_dr), GFP_KERNEL); - if (!new_dr) - return NULL; - return devres_get(&pdev->dev, new_dr, NULL, NULL); + if (!pdev->pinned) + pci_disable_device(pdev); } =20 /** * pcim_enable_device - Managed pci_enable_device() * @pdev: PCI device to be initialized * - * Managed pci_enable_device(). + * Managed pci_enable_device(). Device will automatically be disabled on + * driver detach. */ int pcim_enable_device(struct pci_dev *pdev) { - struct pci_devres *dr; - int rc; - - dr =3D get_pci_dr(pdev); - if (unlikely(!dr)) - return -ENOMEM; - - rc =3D pci_enable_device(pdev); - if (!rc) - pdev->is_managed =3D 1; + devm_add_action(&pdev->dev, pcim_disable_device, pdev); =20 - return rc; + return pci_enable_device(pdev); } EXPORT_SYMBOL(pcim_enable_device); =20 --=20 2.43.0 From nobody Tue Dec 16 11:42:59 2025 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.129.124]) (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 377A618638 for ; Mon, 15 Jan 2024 14:47:39 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=redhat.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=redhat.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=redhat.com header.i=@redhat.com header.b="OvsBgFDH" DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1705330058; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=jIzHvpzkkigfxw83x1LWeciN350K4+qfOeyYElcXkdQ=; b=OvsBgFDHBCmsVrBMYQEvJrJ4nR3+j1Kze2Z282eczrlZuNoULV4+QgGwChLoHlCSQMx3lI abH4ZXhk4evOY8XQ6PjHb9GlUefEmjZp9IFYvIHE0qVVgIGVPusNXmRlVvtJ9+VeBtAKFQ zZyFDPKa1lB/u1dMbU6eIxUBPSV+cRI= Received: from mail-oo1-f69.google.com (mail-oo1-f69.google.com [209.85.161.69]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.3, cipher=TLS_AES_256_GCM_SHA384) id us-mta-413-v8PhBFXBPvG--hrAP6Y7YQ-1; Mon, 15 Jan 2024 09:47:37 -0500 X-MC-Unique: v8PhBFXBPvG--hrAP6Y7YQ-1 Received: by mail-oo1-f69.google.com with SMTP id 006d021491bc7-594cafcc66aso1838907eaf.0 for ; Mon, 15 Jan 2024 06:47:37 -0800 (PST) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1705330056; x=1705934856; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=jIzHvpzkkigfxw83x1LWeciN350K4+qfOeyYElcXkdQ=; b=wVhMXjVgXEotnDe/16BzO/s+oyhj/RruhzWXdEvYTKm1v8J6HXdSkRoS8kZuqunve9 QA0fQhlwRPyoKjBxHjZ4rvTWmef7/9m1TFq5jj2wW/5l6xbvUI2ybxfb9kvksyirdabW C4PsdxwEUn47xFVTLzROCfhQ1VvgV2KbY4LuP2hO8kK1xvFTOc33J/yjBISsOK4UBC0/ h27m7k2EVaflP1CXaL6t26aVpuYhUnXYWm39IPZVeXwrkhWfEZgbnMlV0EB/MTiOAfxf VpTy97CJY2uOQB/PsI/wqjtkRL8ThEbIk5A4b+cEsXym9xfDq+pBMClhiXuyYDvxkrXT VmRg== X-Gm-Message-State: AOJu0Yz5dB5uwwfDujY3ZfPyk0X47XFkegGU1JKThhJDxC8evH2RlJqC 1lezmAQo3GkXOUecCVSF8NcR0eeQns80Z64tNG6EjsqxKviHISmlQlWUjWHLyP1Ob1011nqHUPW 7ySe1OTA4HfV7DpMHSYDgAjzZupRRk92t X-Received: by 2002:a4a:dd8f:0:b0:598:b2d7:2499 with SMTP id h15-20020a4add8f000000b00598b2d72499mr7707342oov.0.1705330056429; Mon, 15 Jan 2024 06:47:36 -0800 (PST) X-Google-Smtp-Source: AGHT+IHi7+XsSK011xhs3AvyU5Z7vMLQIc2V6iqOiH+OrjwaSbXs/XiSSr4Ft9JY9kxXHF9woE1zRQ== X-Received: by 2002:a4a:dd8f:0:b0:598:b2d7:2499 with SMTP id h15-20020a4add8f000000b00598b2d72499mr7707328oov.0.1705330056145; Mon, 15 Jan 2024 06:47:36 -0800 (PST) Received: from pstanner-thinkpadt14sgen1.muc.redhat.com (nat-pool-muc-t.redhat.com. [149.14.88.26]) by smtp.gmail.com with ESMTPSA id ne13-20020a056214424d00b006815cf9a644sm1020720qvb.55.2024.01.15.06.47.33 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Mon, 15 Jan 2024 06:47:35 -0800 (PST) From: Philipp Stanner To: Jonathan Corbet , Hans de Goede , Maarten Lankhorst , Maxime Ripard , Thomas Zimmermann , David Airlie , Daniel Vetter , Bjorn Helgaas , Philipp Stanner , Sam Ravnborg , dakr@redhat.com Cc: linux-doc@vger.kernel.org, linux-kernel@vger.kernel.org, dri-devel@lists.freedesktop.org, linux-pci@vger.kernel.org, stable@kernel.vger.org Subject: [PATCH 10/10] drm/vboxvideo: fix mapping leaks Date: Mon, 15 Jan 2024 15:46:21 +0100 Message-ID: <20240115144655.32046-12-pstanner@redhat.com> X-Mailer: git-send-email 2.43.0 In-Reply-To: <20240115144655.32046-2-pstanner@redhat.com> References: <20240115144655.32046-2-pstanner@redhat.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" When the managed PCI-API was introduced to this driver, it was falsly assumed that initializing the device with pcim_enable_device() instead of pci_enable_device() will make all PCI functions managed. This is wrong and was caused by the quite confusing devres API for PCI: The function pci_iomap_range() is never managed. Replace pci_iomap_range() with the actually managed function pcim_iomap_range(). Additionally, add a call to pcim_request_region() to ensure exclusive access to BAR 0. CC: # v5.10+ Fixes: 8558de401b5f ("drm/vboxvideo: use managed pci functions") Signed-off-by: Philipp Stanner --- drivers/gpu/drm/vboxvideo/vbox_main.c | 24 +++++++++++++----------- 1 file changed, 13 insertions(+), 11 deletions(-) diff --git a/drivers/gpu/drm/vboxvideo/vbox_main.c b/drivers/gpu/drm/vboxvi= deo/vbox_main.c index 42c2d8a99509..7f686a0190e6 100644 --- a/drivers/gpu/drm/vboxvideo/vbox_main.c +++ b/drivers/gpu/drm/vboxvideo/vbox_main.c @@ -42,12 +42,11 @@ static int vbox_accel_init(struct vbox_private *vbox) /* Take a command buffer for each screen from the end of usable VRAM. */ vbox->available_vram_size -=3D vbox->num_crtcs * VBVA_MIN_BUFFER_SIZE; =20 - vbox->vbva_buffers =3D pci_iomap_range(pdev, 0, - vbox->available_vram_size, - vbox->num_crtcs * - VBVA_MIN_BUFFER_SIZE); - if (!vbox->vbva_buffers) - return -ENOMEM; + vbox->vbva_buffers =3D pcim_iomap_range( + pdev, 0, vbox->available_vram_size, + vbox->num_crtcs * VBVA_MIN_BUFFER_SIZE); + if (IS_ERR(vbox->vbva_buffers)) + return PTR_ERR(vbox->vbva_buffers); =20 for (i =3D 0; i < vbox->num_crtcs; ++i) { vbva_setup_buffer_context(&vbox->vbva_info[i], @@ -115,12 +114,15 @@ int vbox_hw_init(struct vbox_private *vbox) =20 DRM_INFO("VRAM %08x\n", vbox->full_vram_size); =20 + ret =3D pcim_request_region(pdev, 0, "vboxvideo"); + if (ret) + return ret; + /* Map guest-heap at end of vram */ - vbox->guest_heap =3D - pci_iomap_range(pdev, 0, GUEST_HEAP_OFFSET(vbox), - GUEST_HEAP_SIZE); - if (!vbox->guest_heap) - return -ENOMEM; + vbox->guest_heap =3D pcim_iomap_range(pdev, 0, + GUEST_HEAP_OFFSET(vbox), GUEST_HEAP_SIZE); + if (IS_ERR(vbox->guest_heap)) + return PTR_ERR(vbox->guest_heap); =20 /* Create guest-heap mem-pool use 2^4 =3D 16 byte chunks */ vbox->guest_pool =3D devm_gen_pool_create(vbox->ddev.dev, 4, -1, --=20 2.43.0