Commit 064cc44e authored by Sonic Zhang's avatar Sonic Zhang Committed by Mike Frysinger

Blackfin: SMP: kgdb: flush core internal write buffer before flushinv

KGDB single step in SMP kernel may hang forever in flushinv without a
CSYNC ahead.  This is because the core internal write buffers need to
be flushed before invalidating the data cache to make sure the insn
fetch is not out of sync.
Signed-off-by: default avatarSonic Zhang <sonic.zhang@analog.com>
Signed-off-by: default avatarMike Frysinger <vapier@gentoo.org>
parent e54b6730
......@@ -49,6 +49,7 @@ ENTRY(_get_core_lock)
jump .Lretry_corelock
.Ldone_corelock:
p0 = r1;
/* flush core internal write buffer before invalidate dcache */
CSYNC(r2);
flushinv[p0];
SSYNC(r2);
......@@ -685,6 +686,8 @@ ENTRY(___raw_atomic_test_asm)
r1 = -L1_CACHE_BYTES;
r1 = r0 & r1;
p0 = r1;
/* flush core internal write buffer before invalidate dcache */
CSYNC(r2);
flushinv[p0];
SSYNC(r2);
r0 = [p1];
......@@ -907,6 +910,8 @@ ENTRY(___raw_uncached_fetch_asm)
r1 = -L1_CACHE_BYTES;
r1 = r0 & r1;
p0 = r1;
/* flush core internal write buffer before invalidate dcache */
CSYNC(r2);
flushinv[p0];
SSYNC(r2);
r0 = [p1];
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment