Patchwork [V3,3/4] PCI: Add support for non-BAR ROMs

login
register
mail settings
Submitter Matthew Garrett
Date Aug. 23, 2012, 4:36 p.m.
Message ID <1345739803-21017-4-git-send-email-mjg@redhat.com>
Download mbox | patch
Permalink /patch/179683/
State Accepted
Headers show

Comments

Matthew Garrett - Aug. 23, 2012, 4:36 p.m.
Platforms may provide their own mechanisms for obtaining ROMs. Add support
for using data provided by the platform in that case.

Signed-off-by: Matthew Garrett <mjg@redhat.com>
---
 drivers/pci/rom.c   | 11 +++++++++--
 include/linux/pci.h |  2 ++
 2 files changed, 11 insertions(+), 2 deletions(-)
Don Dutile - Sept. 5, 2012, 2:18 a.m.
On 08/23/2012 12:36 PM, Matthew Garrett wrote:
> Platforms may provide their own mechanisms for obtaining ROMs. Add support
> for using data provided by the platform in that case.
>
> Signed-off-by: Matthew Garrett<mjg@redhat.com>
> ---
>   drivers/pci/rom.c   | 11 +++++++++--
>   include/linux/pci.h |  2 ++
>   2 files changed, 11 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/pci/rom.c b/drivers/pci/rom.c
> index 48ebdb2..46026e4 100644
> --- a/drivers/pci/rom.c
> +++ b/drivers/pci/rom.c
> @@ -118,11 +118,17 @@ void __iomem *pci_map_rom(struct pci_dev *pdev, size_t *size)
>   	void __iomem *rom;
>
>   	/*
> +	 * Some devices may provide ROMs via a source other than the BAR
> +	 */
> +	if (pdev->rom&&  pdev->romlen) {
> +		*size = pdev->romlen;
> +		return phys_to_virt((phys_addr_t)pdev->rom);
                        ^^^^^
                          ioremap_nocache() ? ... or is caching rom ok?

> +	/*
>   	 * IORESOURCE_ROM_SHADOW set on x86, x86_64 and IA64 supports legacy
>   	 * memory map if the VGA enable bit of the Bridge Control register is
>   	 * set for embedded VGA.
>   	 */
> -	if (res->flags&  IORESOURCE_ROM_SHADOW) {
> +	} else if (res->flags&  IORESOURCE_ROM_SHADOW) {
>   		/* primary video rom always starts here */
>   		start = (loff_t)0xC0000;
>   		*size = 0x20000; /* cover C000:0 through E000:0 */
> @@ -219,7 +225,8 @@ void pci_unmap_rom(struct pci_dev *pdev, void __iomem *rom)
>   	if (res->flags&  (IORESOURCE_ROM_COPY | IORESOURCE_ROM_BIOS_COPY))
>   		return;
>
> -	iounmap(rom);
> +	if (!pdev->rom || !pdev->romlen)
> +		iounmap(rom);
>
>   	/* Disable again before continuing, leave enabled if pci=rom */
>   	if (!(res->flags&  (IORESOURCE_ROM_ENABLE | IORESOURCE_ROM_SHADOW)))
> diff --git a/include/linux/pci.h b/include/linux/pci.h
> index 6a2625c..2668bb9 100644
> --- a/include/linux/pci.h
> +++ b/include/linux/pci.h
> @@ -355,6 +355,8 @@ struct pci_dev {
>   	};
>   	struct pci_ats	*ats;	/* Address Translation Service */
>   #endif
> +	void *rom; /* Physical pointer to ROM if it's not from the BAR */
> +	size_t romlen; /* Length of ROM if it's not from the BAR */
>   };
>
>   static inline struct pci_dev *pci_physfn(struct pci_dev *dev)

--
To unsubscribe from this list: send the line "unsubscribe linux-pci" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Matthew Garrett - Sept. 5, 2012, 2:29 a.m.
On Tue, Sep 04, 2012 at 10:18:48PM -0400, Don Dutile wrote:
> >  	/*
> >+	 * Some devices may provide ROMs via a source other than the BAR
> >+	 */
> >+	if (pdev->rom&&  pdev->romlen) {
> >+		*size = pdev->romlen;
> >+		return phys_to_virt((phys_addr_t)pdev->rom);
>                        ^^^^^
>                          ioremap_nocache() ? ... or is caching rom ok?

If it's appearing through this pathway then it's not actually in ROM, 
the platform has pulled it out of somewhere else.
Alan Cox - Sept. 5, 2012, 12:46 p.m.
On Wed, 5 Sep 2012 03:29:32 +0100
Matthew Garrett <mjg@redhat.com> wrote:

> On Tue, Sep 04, 2012 at 10:18:48PM -0400, Don Dutile wrote:
> > >  	/*
> > >+	 * Some devices may provide ROMs via a source other than the BAR
> > >+	 */
> > >+	if (pdev->rom&&  pdev->romlen) {
> > >+		*size = pdev->romlen;
> > >+		return phys_to_virt((phys_addr_t)pdev->rom);
> >                        ^^^^^
> >                          ioremap_nocache() ? ... or is caching rom ok?
> 
> If it's appearing through this pathway then it's not actually in ROM, 
> the platform has pulled it out of somewhere else.

If that somewhere else is on the PCI bus then it should be a bus not a
virt translation surely ?
--
To unsubscribe from this list: send the line "unsubscribe linux-pci" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Matthew Garrett - Sept. 5, 2012, 1:20 p.m.
On Wed, Sep 05, 2012 at 01:46:21PM +0100, Alan Cox wrote:
> On Wed, 5 Sep 2012 03:29:32 +0100
> Matthew Garrett <mjg@redhat.com> wrote:
> 
> > On Tue, Sep 04, 2012 at 10:18:48PM -0400, Don Dutile wrote:
> > > >  	/*
> > > >+	 * Some devices may provide ROMs via a source other than the BAR
> > > >+	 */
> > > >+	if (pdev->rom&&  pdev->romlen) {
> > > >+		*size = pdev->romlen;
> > > >+		return phys_to_virt((phys_addr_t)pdev->rom);
> > >                        ^^^^^
> > >                          ioremap_nocache() ? ... or is caching rom ok?
> > 
> > If it's appearing through this pathway then it's not actually in ROM, 
> > the platform has pulled it out of somewhere else.
> 
> If that somewhere else is on the PCI bus then it should be a bus not a
> virt translation surely ?

We've no good way of knowing what the firmware's giving us - we copy it 
to RAM in the EFI init process, so by the time we're here it certainly 
shouldn't be on the PCI bus.
Alan Cox - Sept. 5, 2012, 1:30 p.m.
On Wed, 5 Sep 2012 14:20:07 +0100
Matthew Garrett <mjg@redhat.com> wrote:

> On Wed, Sep 05, 2012 at 01:46:21PM +0100, Alan Cox wrote:
> > On Wed, 5 Sep 2012 03:29:32 +0100
> > Matthew Garrett <mjg@redhat.com> wrote:
> > 
> > > On Tue, Sep 04, 2012 at 10:18:48PM -0400, Don Dutile wrote:
> > > > >  	/*
> > > > >+	 * Some devices may provide ROMs via a source other than the BAR
> > > > >+	 */
> > > > >+	if (pdev->rom&&  pdev->romlen) {
> > > > >+		*size = pdev->romlen;
> > > > >+		return phys_to_virt((phys_addr_t)pdev->rom);
> > > >                        ^^^^^
> > > >                          ioremap_nocache() ? ... or is caching rom ok?
> > > 
> > > If it's appearing through this pathway then it's not actually in ROM, 
> > > the platform has pulled it out of somewhere else.
> > 
> > If that somewhere else is on the PCI bus then it should be a bus not a
> > virt translation surely ?
> 
> We've no good way of knowing what the firmware's giving us - we copy it 
> to RAM in the EFI init process, so by the time we're here it certainly 
> shouldn't be on the PCI bus.

In which case how do you know that physical address you have been given
actually has a mapping in kernel virtual space. It seems like it ought to
be getting an ioremap in that case ?

phys_to_virt is not valid for arbitary addresses.

Alan


--
To unsubscribe from this list: send the line "unsubscribe linux-pci" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Matthew Garrett - Sept. 5, 2012, 1:44 p.m.
On Wed, Sep 05, 2012 at 02:30:10PM +0100, Alan Cox wrote:
> On Wed, 5 Sep 2012 14:20:07 +0100
> Matthew Garrett <mjg@redhat.com> wrote:
> > We've no good way of knowing what the firmware's giving us - we copy it 
> > to RAM in the EFI init process, so by the time we're here it certainly 
> > shouldn't be on the PCI bus.
> 
> In which case how do you know that physical address you have been given
> actually has a mapping in kernel virtual space. It seems like it ought to
> be getting an ioremap in that case ?
> 
> phys_to_virt is not valid for arbitary addresses.

The pages are allocated as EFI_LOADER_DATA, and that gets mapped as 
E820_RAM.

Patch

diff --git a/drivers/pci/rom.c b/drivers/pci/rom.c
index 48ebdb2..46026e4 100644
--- a/drivers/pci/rom.c
+++ b/drivers/pci/rom.c
@@ -118,11 +118,17 @@  void __iomem *pci_map_rom(struct pci_dev *pdev, size_t *size)
 	void __iomem *rom;
 
 	/*
+	 * Some devices may provide ROMs via a source other than the BAR
+	 */
+	if (pdev->rom && pdev->romlen) {
+		*size = pdev->romlen;
+		return phys_to_virt((phys_addr_t)pdev->rom);
+	/*
 	 * IORESOURCE_ROM_SHADOW set on x86, x86_64 and IA64 supports legacy
 	 * memory map if the VGA enable bit of the Bridge Control register is
 	 * set for embedded VGA.
 	 */
-	if (res->flags & IORESOURCE_ROM_SHADOW) {
+	} else if (res->flags & IORESOURCE_ROM_SHADOW) {
 		/* primary video rom always starts here */
 		start = (loff_t)0xC0000;
 		*size = 0x20000; /* cover C000:0 through E000:0 */
@@ -219,7 +225,8 @@  void pci_unmap_rom(struct pci_dev *pdev, void __iomem *rom)
 	if (res->flags & (IORESOURCE_ROM_COPY | IORESOURCE_ROM_BIOS_COPY))
 		return;
 
-	iounmap(rom);
+	if (!pdev->rom || !pdev->romlen)
+		iounmap(rom);
 
 	/* Disable again before continuing, leave enabled if pci=rom */
 	if (!(res->flags & (IORESOURCE_ROM_ENABLE | IORESOURCE_ROM_SHADOW)))
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 6a2625c..2668bb9 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -355,6 +355,8 @@  struct pci_dev {
 	};
 	struct pci_ats	*ats;	/* Address Translation Service */
 #endif
+	void *rom; /* Physical pointer to ROM if it's not from the BAR */
+	size_t romlen; /* Length of ROM if it's not from the BAR */
 };
 
 static inline struct pci_dev *pci_physfn(struct pci_dev *dev)