Thread (7 messages) flat view 7 messages, 3 authors, 2014-01-07

Regression (ARM) arch/arm/mm/init.c doesn't build without CONFIG_ARM_PATCH_PHYS_VIRT.

From: Russell King - ARM Linux <hidden>
Date: 2014-01-07 17:55:19
Also in: lkml

On Tue, Jan 07, 2014 at 12:45:42PM -0500, Santosh Shilimkar wrote:
On Monday 06 January 2014 08:11 PM, Russell King - ARM Linux wrote:
quoted
/--------------------------------------------------------------
| > > Right, so, with Assabet, which has CONFIG_DMA_ZONE=y and
| > > CONFIG_ARM_PATCH_PHYS_VIRT=y:
| > > 
| > > $ make O=../build/assabet arch/arm/mm/init.i
| > > 
| > > gives:
| > >   arm_dma_limit = __pv_phys_offset + arm_dma_zone_size - 1;
| > > 
| > > with or without Santosh's patch.
\--------------------------------------------------------------
quoted
May be I missed your point but I ended up creating the patch because
the CMA reservation was failing on Keystone since the arm_dma_limit
wasn't right since it wasn't considering the actual __pv_phys_offset.

Isn't that an issue ?
See the above.  Your patch has _no_ effect what so ever when
CONFIG_ARM_PATCH_PHYS_VIRT=y - which you have on the Keystone, right?

If you don't believe me, ask make to produce arch/arm/mm/init.i, which
is the output from the preprocessor, comparing the resulting generated
file both with and without your patch applied.
Looks like you are right. I did two fixes to have right arm_dma_limit in
below order.

1. 787b0d5 {ARM: 7908/1: mm: Fix the arm_dma_limit calculation}
2. 7c92732 {ARM: 7909/1: mm: Call setup_dma_zone() post early_paging_init()}

But with 2 alone the issue gets fixed since with ARM_PATCH_PHYS_VIRT and
below pre-processor, the PHYS_OFFSET also gets an updated value. I never
realised that 1 becomes redundant after second patch while creating them. 

#define PHYS_OFFSET __pv_phys_offset

So indeed, 787b0d5{ARM: 7908/1: mm: Fix the arm_dma_limit calculation}
won't be needed anymore and can be reverted. Sorry it took some time
for me to reach to your conclusion.
Okay, reverted.

-- 
FTTC broadband for 0.8mile line: 5.8Mbps down 500kbps up.  Estimation
in database were 13.1 to 19Mbit for a good line, about 7.5+ for a bad.
Estimate before purchase was "up to 13.2Mbit".
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help