sergiotarxz/nuttx
sergiotarxz/nuttx
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

SAMA5 OHCI: Back out a change, the real root cause was a bug in the cache logic so the hack is no longer necessary

Browse Source
This commit is contained in:
Gregory Nutt 2013-09-22 07:53:51 -06:00
parent 9cb23c5ccb
commit 3ceac07042
2 changed files with 1 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
ChangeLog
View File

@ -5607,7 +5607,7 @@
descriptive task names when starting the EHCI and OHCI monitor tasks
(2013-9-20).
* arch/arm/src/sama5/sam_ohci.c: Fix a place where DMA-related data
needed to be flushed to data cache; Fix another where a virual address
needed to be flushed to data cache; Fix another where a virtual address
was being used in a register where a physical address was required
(2013-9-20).
* arch/arm/src/armv7-a/cp15_clean_dcache.S and cp15_flush_dcache.S:

8
arch/arm/src/sama5/sam_ohci.c
View File

@ -1337,14 +1337,6 @@ static int sam_enqueuetd(struct sam_rhport_s *rhport, struct sam_ed_s *ed,
{
cp15_clean_dcache((uintptr_t)buffer,
(uintptr_t)buffer + buflen);
/* REVISIT: This cache invalidation solves some transfer
* problems... but I don't understand why? Without this,
* the above clean seems to fail to flush all of the buffer.
*/
cp15_invalidate_dcache((uintptr_t)buffer,
(uintptr_t)buffer + buflen);
}
cp15_clean_dcache((uintptr_t)tdtail,