linux-kernel - Re: [PATCH v3 2/3] openrisc: Add pci bus support

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date:   Fri, 29 Jul 2022 16:54:52 +0900
From:   Stafford Horne <shorne@...il.com>
To:     Guenter Roeck <linux@...ck-us.net>
Cc:     LKML <linux-kernel@...r.kernel.org>,
        Bjorn Helgaas <helgaas@...nel.org>,
        Arnd Bergmann <arnd@...db.de>, Jonas Bonn <jonas@...thpole.se>,
        Stefan Kristiansson <stefan.kristiansson@...nalahti.fi>,
        Palmer Dabbelt <palmer@...osinc.com>,
        Peter Zijlstra <peterz@...radead.org>,
        openrisc@...ts.librecores.org
Subject: Re: [PATCH v3 2/3] openrisc: Add pci bus support

On Thu, Jul 28, 2022 at 11:10:37PM -0700, Guenter Roeck wrote:
> On 7/28/22 22:50, Stafford Horne wrote:
> > On Thu, Jul 28, 2022 at 08:37:28PM -0700, Guenter Roeck wrote:
> > > On Mon, Jul 25, 2022 at 11:07:36AM +0900, Stafford Horne wrote:
> > > > This patch adds required definitions to allow for PCI buses on OpenRISC.
> > > > This is being tested on the OpenRISC QEMU virt platform which is in
> > > > development.
> > > > 
> > > > OpenRISC does not have IO ports so we keep the definition of
> > > > IO_SPACE_LIMIT and PIO_RESERVED to be 0.
> > > > 
> > > > Note, since commit 66bcd06099bb ("parport_pc: Also enable driver for PCI
> > > > systems") all platforms that support PCI also need to support parallel
> > > > port.  We add a generic header to support compiling parallel port
> > > > drivers, though they generally will not work as they require IO ports.
> > > > 
> > > > Signed-off-by: Stafford Horne <shorne@...il.com>
> > > 
> > > This patch results in
> > > 
> > > Building openrisc:allmodconfig ... failed
> > > --------------
> > > Error log:
> > > drivers/video/fbdev/riva/fbdev.c: In function 'rivafb_probe':
> > > drivers/video/fbdev/riva/fbdev.c:2062:42: error:
> > > 	passing argument 1 of 'iounmap' discards 'volatile' qualifier from pointer target type
> > > 
> > > drivers/video/fbdev/nvidia/nvidia.c: In function 'nvidiafb_probe':
> > > drivers/video/fbdev/nvidia/nvidia.c:1414:20: error:
> > > 	passing argument 1 of 'iounmap' discards 'volatile' qualifier from pointer target type
> > > 
> > > drivers/scsi/aic7xxx/aic7xxx_osm.c: In function 'ahc_platform_free':
> > > drivers/scsi/aic7xxx/aic7xxx_osm.c:1231:41: error:
> > > 	passing argument 1 of 'iounmap' discards 'volatile' qualifier from pointer target type
> > > 
> > > ... and so on.
> > > 
> > > Prior to this patch, the code was not enabled because it depends on PCI.
> > 
> > Hi Guenter,
> > 
> > Thanks for reporting this.
> > 
> > It's interesting, I don't get this on the openrisc/for-next branch.
> > 
> 
> Hmm, weird. I see it all over the place. Complete log is at
> https://kerneltests.org/builders/next-openrisc-next/builds/1880/steps/buildcommand/logs/stdio
> if you are interested.
> 
> > BTW, do you turn off WERROR on the allmodconfig config?  I get many warnings
> > such as the below, but I haven't looked into it much yet:
> > 
> 
> No, I don't. Disabling it would defeat its purpose.
> 
> >      fs/exec.c: In function 'shift_arg_pages':
> >      fs/exec.c:687:27: error: 'tlb' is used uninitialized [-Werror=uninitialized]
> >        687 |         struct mmu_gather tlb;
> > 	  |                           ^~~
> > 
> 
> I don't see that in next-20220728. I tried with gcc-11.2 and 11.3.
> Which compiler do you use ?

I am using gcc 12.0.1 with next-20220728.  That might exaplain it, I am doing
compiler development at the same time so I always end up with the latest and
greatest warnings.

$ or1k-linux-gcc -v
  ..
  gcc version 12.0.1 20220210 (experimental) (GCC)

I can see the issue now:

    drivers/video/fbdev/riva/fbdev.c:2062:42: warning: passing argument 1 of 'iounmap' discards 'volatile' qualifier from pointer target type [-Wdiscarded-qualifiers]
     2062 |                 iounmap(default_par->riva.PRAMIN);
	  |                         ~~~~~~~~~~~~~~~~~^~~~~~~

Just adding volatile does seem to fix this, I will do some more testing and
create a formal patch.

--

diff --git a/arch/openrisc/include/asm/io.h b/arch/openrisc/include/asm/io.h
index 625ac6ad1205..ee6043a03173 100644
--- a/arch/openrisc/include/asm/io.h
+++ b/arch/openrisc/include/asm/io.h
@@ -31,7 +31,7 @@
 void __iomem *ioremap(phys_addr_t offset, unsigned long size);
 
 #define iounmap iounmap
-extern void iounmap(void __iomem *addr);
+extern void iounmap(volatile void __iomem *addr);
 
 #include <asm-generic/io.h>
 
diff --git a/arch/openrisc/mm/ioremap.c b/arch/openrisc/mm/ioremap.c
index daae13a76743..8ec0dafecf25 100644
--- a/arch/openrisc/mm/ioremap.c
+++ b/arch/openrisc/mm/ioremap.c
@@ -77,7 +77,7 @@ void __iomem *__ref ioremap(phys_addr_t addr, unsigned long size)
 }
 EXPORT_SYMBOL(ioremap);
 
-void iounmap(void __iomem *addr)
+void iounmap(volatile void __iomem *addr)
 {
        /* If the page is from the fixmap pool then we just clear out
         * the fixmap mapping.