Kernel 4.7

[kilaueabart]

Hi Kilaueabart & Mechanic,
...
Could you please test the RC5?...

It failed on my X1000 twice.

Thank you for testing. Maybe you need to reformat the CF card.

Reformatted the card. (NO FUN!) That seems to have solved the problem, and the 30.7 MB version of RC5 now works with my Radeon HD5xxx.

[xeno74]

Reformatted the card. (NO FUN!) That seems to have solved the problem, and the 30.7 MB version of RC5 now works with my Radeon HD5xxx.

Thank you for reformating the CF card. Good to know that the RC5 works now.

[xeno74]

Could you please test the RC5 with a Radeon HD 7xxx or higher?

Download: vmlinux-4.7-rc5-AmigaONE_X1000.tar.gz

[xeno74]

Great news from Darren:

Hello Benjamin

On 27/06/2016, Benjamin Herrenschmidt wrote:
> Tell me more, when is that mapping done ? I'm changing things so that
> platform probe is called much later so that might have an impact.
>
> What consumes pci_io_base before it's been initialized ?

pas_pci_init() is the culprit. Following on from Aneesh's suggestion an
improved patch will follow shoertly.

Regards
Darren

Hello Aneesh

On 28/06/2016, Aneesh Kumar K.V wrote:
> Another option is to init it along with rest of the variables as done in
> hash__early_init_mmu(void)/radix__early_init_mmu(void)

*FACEPALM* Why didn't I think of that! I've made this change and seems to work
- obviously I can't test on a Radix system though, as I don't have access to
one.

Patch comming shortly

Regards
Darren

Commit:d6a9996e84ac4beb7713e9485f4563e100a9b03e (powerpc/mm:
vmalloc abstraction in preparation for radix) turned kernel memory
and IO addresses from #defined constants to variables initialised
at runtime.

On PA6T systems the setup_arch machine call initialises the onboard
PCI-e root-ports, and uses pci_io_base to do this, which is now before
its value has been set resulting in a panic right after 'booting
linux via __start()'

Move the pci_io_base initialisation to the same place as vmalloc
ranges are set (hash__early_init_mmu()/radix__early_init_mmu())

Reported-by: Christian Zigotzky <[email protected]>
Signed-off-by: Darren Stevens <[email protected]>

---

Tested on my AmigaOneX1000, I don't have access to a refence board system,
and our developer with one is on honeymoon.

I am hoping to follow this patch with others to reduce the size of the nemo
patch we apply, eventually I'd like to see the patch gone, but that is a
very big job.

Kind regards
Darren

pa6t-bootfix-v2.patch:

Code: Select all

diff --git a/arch/powerpc/include/asm/book3s/64/pgtable.h b/arch/powerpc/include/asm/book3s/64/pgtable.h
index 88a5eca..ab84c89 100644
--- a/arch/powerpc/include/asm/book3s/64/pgtable.h
+++ b/arch/powerpc/include/asm/book3s/64/pgtable.h
@@ -230,6 +230,7 @@ extern unsigned long __kernel_virt_size;
 #define KERN_VIRT_SIZE  __kernel_virt_size
 extern struct page *vmemmap;
 extern unsigned long ioremap_bot;
+extern unsigned long pci_io_base;
 #endif /* __ASSEMBLY__ */
 
 #include <asm/book3s/64/hash.h>
diff --git a/arch/powerpc/kernel/pci_64.c b/arch/powerpc/kernel/pci_64.c
index 3759df5..a5ae49a 100644
--- a/arch/powerpc/kernel/pci_64.c
+++ b/arch/powerpc/kernel/pci_64.c
@@ -47,7 +47,6 @@ static int __init pcibios_init(void)
 
 	printk(KERN_INFO "PCI: Probing PCI hardware\n");
 
-	pci_io_base = ISA_IO_BASE;
 	/* For now, override phys_mem_access_prot. If we need it,g
 	 * later, we may move that initialization to each ppc_md
 	 */
diff --git a/arch/powerpc/mm/hash_utils_64.c b/arch/powerpc/mm/hash_utils_64.c
index 5b22ba0..b5b5fe6 100644
--- a/arch/powerpc/mm/hash_utils_64.c
+++ b/arch/powerpc/mm/hash_utils_64.c
@@ -922,6 +922,8 @@ void __init hash__early_init_mmu(void)
 	vmemmap = (struct page *)H_VMEMMAP_BASE;
 	ioremap_bot = IOREMAP_BASE;
 
+	pci_io_base = ISA_IO_BASE;
+
 	/* Initialize the MMU Hash table and create the linear mapping
 	 * of memory. Has to be done before SLB initialization as this is
 	 * currently where the page size encoding is obtained.
diff --git a/arch/powerpc/mm/pgtable-radix.c b/arch/powerpc/mm/pgtable-radix.c
index e58707d..095fbfa 100644
--- a/arch/powerpc/mm/pgtable-radix.c
+++ b/arch/powerpc/mm/pgtable-radix.c
@@ -328,6 +328,9 @@ void __init radix__early_init_mmu(void)
 	__vmalloc_end = RADIX_VMALLOC_END;
 	vmemmap = (struct page *)RADIX_VMEMMAP_BASE;
 	ioremap_bot = IOREMAP_BASE;
+
+	pci_io_base = ISA_IO_BASE;
+
 	/*
 	 * For now radix also use the same frag size
 	 */

[xeno74]

Hi all,

I cross compiled the RC5-2 with Darren's latest patch (pa6t-bootfix-v2.patch) today.

Please test the RC5-2.

Download: vmlinux-4.7-rc5-2-AmigaONE_X1000.tar.gz

Thanks in advance,

Christian

[zzd10h]

Hi,
tested 4.7rc5 and 4.7rc5.2 under my x1000 with R7-250X.

Network is OK but with these 2 kernels, my monitor resolution is not detected (even after to have copied the /lib/modules files).

See Kernel4.7_R7250X.png attached.

To compare, see Kernel4.6_R7250X that works fine.

[xeno74]

Hi,
tested 4.7rc5 and 4.7rc5.2 under my x1000 with R7-250X.

Network is OK but with these 2 kernels, my monitor resolution is not detected (even after to have copied the /lib/modules files).

See Kernel4.7_R7250X.png attached.

To compare, see Kernel4.6_R7250X that works fine.

Thanks a million for testing!

I created the RC3 without the PowerPC updates. Maybe the RC3 set up the correct monitor resolution.
Could you please test the RC3 as well?
Maybe there is another issue in the PowerPC updates.

[Srtest]

Everything works here and boots fine except of course the graphics driver on SI GCN 1.0 card.

[xeno74]

Everything works here and boots fine except of course the graphics driver on SI GCN 1.0 card.

Srtest,

Thanks a lot for testing! Which kernel did you test? Did the kernels set up the correct monitor resolution?

Cheers,

Christian

[Srtest]

I used the first 2 rc5 kernels with both boot patches and both worked without a noticeable difference.

Just a reminder: there is a possibility of the next kernel having amdgpu for GCN 1.0 without being enabled by default. Maybe you will be able to enable it for us so we can test if it fixes anything.