From: Yinghai Lu <yinghai@kernel.org>
To: Jesse Barnes <jbarnes@virtuousgeek.org>,
Benjamin Herrenschmidt <benh@kernel.crashing.org>,
Tony Luck <tony.luck@intel.com>,
David Miller <davem@davemloft.net>, x86 <x86@kernel.org>
Cc: Bjorn Helgaas <bhelgaas@google.com>,
Dominik Brodowski <linux@dominikbrodowski.net>,
linux-pci@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-arch@vger.kernel.org, Yinghai Lu <yinghai@kernel.org>
Subject: [PATCH 08/39] x86, PCI: add host bridge resource release for using _CRS
Date: Wed, 29 Feb 2012 15:07:07 -0800 [thread overview]
Message-ID: <1330556858-11768-9-git-send-email-yinghai@kernel.org> (raw)
In-Reply-To: <1330556858-11768-1-git-send-email-yinghai@kernel.org>
1. allocate pci_root_info instead of use local stack. we need to pass around info
for release fn.
2. add release_pci_root_info
3. set x86 own host bridge release fn, so will make sure root bridge
related resources get freed during root bus removal.
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
---
arch/x86/pci/acpi.c | 63 +++++++++++++++++++++++++++++++++++++++++---------
1 files changed, 51 insertions(+), 12 deletions(-)
diff --git a/arch/x86/pci/acpi.c b/arch/x86/pci/acpi.c
index 763f7bb..c7a230f 100644
--- a/arch/x86/pci/acpi.c
+++ b/arch/x86/pci/acpi.c
@@ -305,11 +305,40 @@ static void add_resources(struct pci_root_info *info,
}
}
-static void free_pci_root_info(struct pci_root_info *info)
+static void free_pci_root_info_res(struct pci_root_info *info)
{
kfree(info->name);
kfree(info->res);
- memset(info, 0, sizeof(struct pci_root_info));
+ info->res = NULL;
+ info->res_num = 0;
+}
+
+static void __release_pci_root_info(struct pci_root_info *info)
+{
+ int i;
+ struct resource *res;
+
+ for (i = 0; i < info->res_num; i++) {
+ res = &info->res[i];
+
+ if (!res->parent)
+ continue;
+
+ if (!(res->flags & (IORESOURCE_MEM | IORESOURCE_IO)))
+ continue;
+
+ release_resource(res);
+ }
+
+ free_pci_root_info_res(info);
+
+ kfree(info);
+}
+static void release_pci_root_info(struct pci_host_bridge *bridge)
+{
+ struct pci_root_info *info = bridge->release_data;
+
+ __release_pci_root_info(info);
}
static void
@@ -342,7 +371,7 @@ probe_pci_root_info(struct pci_root_info *info, struct acpi_device *device,
struct pci_bus * __devinit pci_acpi_scan_root(struct acpi_pci_root *root)
{
struct acpi_device *device = root->device;
- struct pci_root_info info;
+ struct pci_root_info *info = NULL;
int domain = root->segment;
int busnum = root->secondary.start;
LIST_HEAD(resources);
@@ -387,7 +416,13 @@ struct pci_bus * __devinit pci_acpi_scan_root(struct acpi_pci_root *root)
sd->domain = domain;
sd->node = node;
- memset(&info, 0, sizeof(struct pci_root_info));
+ info = kzalloc(sizeof(*info), GFP_KERNEL);
+ if (!info) {
+ kfree(sd);
+ printk(KERN_WARNING "pci_bus %04x:%02x: "
+ "ignored (out of memory)\n", domain, busnum);
+ return NULL;
+ }
/*
* Maybe the desired pci bus has been already scanned. In such case
* it is unnecessary to scan the pci bus with the given domain,busnum.
@@ -399,29 +434,33 @@ struct pci_bus * __devinit pci_acpi_scan_root(struct acpi_pci_root *root)
* be replaced by sd.
*/
memcpy(bus->sysdata, sd, sizeof(*sd));
+ kfree(info);
kfree(sd);
} else {
- probe_pci_root_info(&info, device, busnum, domain);
+ probe_pci_root_info(info, device, busnum, domain);
/*
* _CRS with no apertures is normal, so only fall back to
* defaults or native bridge info if we're ignoring _CRS.
*/
if (pci_use_crs)
- add_resources(&info, &resources);
+ add_resources(info, &resources);
else {
- free_pci_root_info(&info);
+ free_pci_root_info_res(info);
x86_pci_root_bus_resources(busnum, &resources);
}
+
bus = pci_create_root_bus(NULL, busnum, &pci_root_ops, sd,
&resources);
- if (bus)
+ if (bus) {
bus->subordinate = pci_scan_child_bus(bus);
- else
+ pci_set_host_bridge_release(
+ to_pci_host_bridge(bus->bridge),
+ release_pci_root_info, info);
+ } else {
pci_free_resource_list(&resources);
-
- if (!bus && pci_use_crs)
- free_pci_root_info(&info);
+ __release_pci_root_info(info);
+ }
}
/* After the PCI-E bus has been walked and all devices discovered,
--
1.7.7
next prev parent reply other threads:[~2012-02-29 23:08 UTC|newest]
Thread overview: 55+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-02-29 23:06 [PATCH 00/39] PCI: pci_host_bridge related cleanup and busn_alloc Yinghai Lu
2012-02-29 23:07 ` [PATCH 01/39] PCI: Separate host_bridge code out from probe.c Yinghai Lu
2012-02-29 23:07 ` [PATCH 02/39] x86, PCI: have own version for pcibios_bus_to_resource Yinghai Lu
2012-02-29 23:20 ` Bjorn Helgaas
2012-02-29 23:33 ` Yinghai Lu
2012-03-01 0:42 ` Bjorn Helgaas
2012-03-01 2:54 ` Yinghai Lu
2012-02-29 23:07 ` [PATCH 03/39] x86, PCI: Fix memleak with get_current_resources Yinghai Lu
2012-02-29 23:07 ` [PATCH 04/39] PCI: rename pci_host_bridge() to find_pci_root_bridge() Yinghai Lu
2012-02-29 23:07 ` [PATCH 05/39] PCI: add generic device into pci_host_bridge struct Yinghai Lu
2012-02-29 23:07 ` [PATCH 06/39] PCI: add host bridge release support Yinghai Lu
2012-02-29 23:07 ` [PATCH 07/39] x86, PCI: break down get_current_resource() Yinghai Lu
2012-02-29 23:07 ` Yinghai Lu [this message]
2012-02-29 23:07 ` [PATCH 09/39] x86, PCI: embed name acpi version pci_root_info struct Yinghai Lu
2012-02-29 23:07 ` [PATCH 10/39] x86, PCI: embed pci_sysdata into pci_root_info on acpi path Yinghai Lu
2012-02-29 23:07 ` [PATCH 11/39] x86, PCI: Allocating pci_root_info for not using _CRS path Yinghai Lu
2012-02-29 23:07 ` [PATCH 12/39] x86, PCI: Merge root info printing for nocrs path Yinghai Lu
2012-02-29 23:07 ` [PATCH 13/39] x86, PCI: add print all root info " Yinghai Lu
2012-02-29 23:07 ` [PATCH 14/39] x86, PCI: allocate temp range array in amd_bus pci_root_info probing Yinghai Lu
2012-02-29 23:07 ` [PATCH 15/39] x86, PCI: Merge pcibios_scan_root and pci_scan_bus_on_node Yinghai Lu
2012-02-29 23:07 ` [PATCH 16/39] PCI: skip busn resource at first Yinghai Lu
2012-02-29 23:07 ` [PATCH 17/39] x86, PCI: put busn resource in pci_root_info for acpi path Yinghai Lu
2012-02-29 23:07 ` [PATCH 18/39] PCI: default busn_resource Yinghai Lu
2012-02-29 23:07 ` [PATCH 19/39] x86, PCI: put busn resource in pci_root_info for no_crs path Yinghai Lu
2012-02-29 23:07 ` [PATCH 20/39] PCI: Add busn_res into struct pci_bus Yinghai Lu
2012-02-29 23:07 ` [PATCH 21/39] PCI: Add busn_res operation functions Yinghai Lu
2012-02-29 23:07 ` [PATCH 22/39] PCI: release busn when removing bus Yinghai Lu
2012-02-29 23:07 ` [PATCH 23/39] PCI: insert busn_res in pci_create_root_bus Yinghai Lu
2012-02-29 23:07 ` [PATCH 24/39] PCI: checking busn_res in pci_scan_root_bus Yinghai Lu
2012-02-29 23:07 ` [PATCH 25/39] PCI: add default res for pci_scan_bus Yinghai Lu
2012-02-29 23:07 ` [PATCH 26/39] PCI, ia64: Register busn_res for root buses Yinghai Lu
2012-02-29 23:07 ` Yinghai Lu
2012-02-29 23:07 ` [PATCH 27/39] PCI, sparc: " Yinghai Lu
2012-02-29 23:07 ` Yinghai Lu
2012-02-29 23:07 ` [PATCH 28/39] PCI, powerpc: " Yinghai Lu
2012-02-29 23:07 ` Yinghai Lu
2012-02-29 23:07 ` [PATCH 29/39] PCI, parisc: " Yinghai Lu
2012-02-29 23:07 ` [PATCH 30/39] PCI: Add pci_bus_extend/shrink_top() Yinghai Lu
2012-02-29 23:07 ` [PATCH 31/39] PCI: Probe safe range that we can use for unassigned bridge Yinghai Lu
2012-02-29 23:07 ` [PATCH 32/39] PCI: Strict checking of valid range for bridge Yinghai Lu
2012-02-29 23:07 ` [PATCH 33/39] PCI: Allocate bus range instead of use max blindly Yinghai Lu
2012-02-29 23:07 ` [PATCH 34/39] PCI: kill pci_fixup_parent_subordinate_busnr() Yinghai Lu
2012-02-29 23:07 ` [PATCH 35/39] PCI: Seperate child bus scanning to two passes overall Yinghai Lu
2012-02-29 23:07 ` [PATCH 36/39] pcmcia: remove workaround for fixing pci parent bus subordinate Yinghai Lu
2012-02-29 23:07 ` Yinghai Lu
2012-02-29 23:07 ` [PATCH 37/39] PCI: Double checking setting for bus register and bus struct Yinghai Lu
2012-02-29 23:07 ` [PATCH 38/39] PCI, pciehp: Remove not needed bus number range checking Yinghai Lu
2012-02-29 23:07 ` [PATCH 39/39] x86, PCI: kill busn in acpi pci_root_info Yinghai Lu
2012-02-29 23:32 ` Bjorn Helgaas
2012-02-29 23:37 ` Yinghai Lu
2012-02-29 23:51 ` Greg KH
2012-03-01 0:27 ` Jesse Barnes
2012-03-01 2:57 ` Yinghai Lu
2012-03-01 18:51 ` [PATCH 00/39] PCI: pci_host_bridge related cleanup and busn_alloc Myron Stowe
2012-03-01 19:03 ` Yinghai Lu
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1330556858-11768-9-git-send-email-yinghai@kernel.org \
--to=yinghai@kernel.org \
--cc=benh@kernel.crashing.org \
--cc=bhelgaas@google.com \
--cc=davem@davemloft.net \
--cc=jbarnes@virtuousgeek.org \
--cc=linux-arch@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-pci@vger.kernel.org \
--cc=linux@dominikbrodowski.net \
--cc=tony.luck@intel.com \
--cc=x86@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.