Thread (8 messages) flat view 8 messages, 3 authors, 2013-07-16

Re: [PATCH] powerpc/fsl-booke: Work around erratum A-006958

From: Aneesh Kumar K.V <hidden>
Date: 2013-07-15 06:03:32

Scott Wood [off-list ref] writes:


quoted hunk ↗ jump to hunk
diff --git a/arch/powerpc/platforms/Kconfig.cputype b/arch/powerpc/platforms/Kconfig.cputype
index ae0aaea..7f0e2e5 100644
--- a/arch/powerpc/platforms/Kconfig.cputype
+++ b/arch/powerpc/platforms/Kconfig.cputype
@@ -165,6 +165,17 @@ config PPC_E500MC
 	  such as e5500/e6500), and must be disabled for running on
 	  e500v1 or e500v2.

+config FSL_ERRATUM_A_006958
+	bool
+	depends on PPC_E500MC && PPC64
+	default y
+	help
+	  Workaround for erratum A-006958, which says that 64-bit
+	  timebase reads are not atomic.  The workaround is to fall back
+	  to the 32-bit method of reading timebase.  Note that timebase
+	  is readable by userspace, so any non-vdso userspace accesses
+	  will need to have the workaround applied separately.
+
 config PPC_FPU
 	bool
 	default y if PPC64

I am completely new to this area, so ignore if it is silly. But how do
we expect to select this config ? Should that happen via .config ?
That seems strange, because without this change some of the configs
will surely be broken/buggy right ?. I was expecting it not be a config
entry, primarily because I haven't seem something similar on other
archs. And if there is an erratum against a cpu release, we should by
default apply this when we know we are running on those cpus right ?
Which implies this should not be a config option ?

-aneesh
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help