mm: memblock: do not enforce current limit for memblock_phys* family

Commit Message

Mike Rapoport Oct. 12, 2019, 9:31 p.m. UTC

From: Mike Rapoport <[e-mail]>

Until commit 92d12f9544b7 ("memblock: refactor internal allocation
functions") the maximal address for memblock allocations was forced to
memblock.current_limit only for the allocation functions returning virtual
address. The changes introduced by that commit moved the limit enforcement
into the allocation core and as a result the allocation functions returning
physical address also started to limit allocations to
memblock.current_limit.

This caused breakage of etnaviv GPU driver:

[    3.682347] etnaviv etnaviv: bound 130000.gpu (ops gpu_ops)
[    3.688669] etnaviv etnaviv: bound 134000.gpu (ops gpu_ops)
[    3.695099] etnaviv etnaviv: bound 2204000.gpu (ops gpu_ops)
[    3.700800] etnaviv-gpu 130000.gpu: model: GC2000, revision: 5108
[    3.723013] etnaviv-gpu 130000.gpu: command buffer outside valid
memory window
[    3.731308] etnaviv-gpu 134000.gpu: model: GC320, revision: 5007
[    3.752437] etnaviv-gpu 134000.gpu: command buffer outside valid
memory window
[    3.760583] etnaviv-gpu 2204000.gpu: model: GC355, revision: 1215
[    3.766766] etnaviv-gpu 2204000.gpu: Ignoring GPU with VG and FE2.0

Restore the behaviour of memblock_phys* family so that these functions will
not enforce memblock.current_limit.

Fixes: 92d12f9544b7 ("memblock: refactor internal allocation functions")
Reported-by: Adam Ford <[e-mail]>
Tested-by: Adam Ford <[e-mail]> #imx6q-logicpd
Signed-off-by: Mike Rapoport <[e-mail]>
---
 mm/memblock.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

Comments

Andrew Morton Oct. 14, 2019, 10:44 p.m. UTC | #1

On Sun, 13 Oct 2019 00:31:01 +0300 Mike Rapoport <[e-mail]> wrote:

> Until commit 92d12f9544b7 ("memblock: refactor internal allocation
> functions") the maximal address for memblock allocations was forced to
> memblock.current_limit only for the allocation functions returning virtual
> address. The changes introduced by that commit moved the limit enforcement
> into the allocation core and as a result the allocation functions returning
> physical address also started to limit allocations to
> memblock.current_limit.
> 
> This caused breakage of etnaviv GPU driver:
> 
> ...
>

So I'll add a cc:stable, yes?

Mike Rapoport Oct. 15, 2019, 6:42 a.m. UTC | #2

On October 15, 2019 12:44:23 AM GMT+02:00, Andrew Morton <[e-mail]> wrote:
>On Sun, 13 Oct 2019 00:31:01 +0300 Mike Rapoport <[e-mail]>
>wrote:
>
>> Until commit 92d12f9544b7 ("memblock: refactor internal allocation
>> functions") the maximal address for memblock allocations was forced
>to
>> memblock.current_limit only for the allocation functions returning
>virtual
>> address. The changes introduced by that commit moved the limit
>enforcement
>> into the allocation core and as a result the allocation functions
>returning
>> physical address also started to limit allocations to
>> memblock.current_limit.
>> 
>> This caused breakage of etnaviv GPU driver:
>> 
>> ...
>>
>
>So I'll add a cc:stable, yes?

Yeah, right. Somehow I've missed that...

Patch

diff --git a/mm/memblock.c b/mm/memblock.c
index 7d4f61a..c4b16ca 100644
--- a/mm/memblock.c
+++ b/mm/memblock.c
@@ -1356,9 +1356,6 @@  static phys_addr_t __init memblock_alloc_range_nid(phys_addr_t size,
 		align = SMP_CACHE_BYTES;
 	}
 
-	if (end > memblock.current_limit)
-		end = memblock.current_limit;
-
 again:
 	found = memblock_find_in_range_node(size, align, start, end, nid,
 					    flags);
@@ -1469,6 +1466,9 @@  static void * __init memblock_alloc_internal(
 	if (WARN_ON_ONCE(slab_is_available()))
 		return kzalloc_node(size, GFP_NOWAIT, nid);
 
+	if (max_addr > memblock.current_limit)
+		max_addr = memblock.current_limit;
+
 	alloc = memblock_alloc_range_nid(size, align, min_addr, max_addr, nid);
 
 	/* retry allocation without lower limit */