On 04/08/2015 04:29 PM, Alexander Graf wrote:



Am 08.04.2015 um 03:45 schrieb Alexey Kardashevskiy <a...@ozlabs.ru>:

On 04/08/2015 06:43 AM, Alexander Graf wrote:
On 04/02/2015 08:26 AM, Alexey Kardashevskiy wrote:
Since 8dfa3a5e "target-ppc: Add "compat" CPU option" (which was a part of
client-architecture-support patchset) there is a "compat" option
of "-cpu". However it was not documented at all and this is what this
patch is trying to do.

Signed-off-by: Alexey Kardashevskiy <a...@ozlabs.ru>
---
  qemu-options.hx | 11 ++++++++++-
  1 file changed, 10 insertions(+), 1 deletion(-)

diff --git a/qemu-options.hx b/qemu-options.hx
index c513352..bd15214 100644
--- a/qemu-options.hx
+++ b/qemu-options.hx
@@ -84,11 +84,20 @@ HXCOMM Deprecated by -machine
  DEF("M", HAS_ARG, QEMU_OPTION_M, "", QEMU_ARCH_ALL)
  DEF("cpu", HAS_ARG, QEMU_OPTION_cpu,
-    "-cpu cpu        select CPU ('-cpu help' for list)\n", QEMU_ARCH_ALL)
+    "-cpu cpu,[compat=MODE]\n"
+    "                select CPU ('-cpu help' for list)\n"
+    "                compat= processor compatibility mode (PPC64 only)\n",
+        QEMU_ARCH_ALL)

Shouldn't there be a way to show this only for the ppc64 target?


I do not really know, I did not dig deep. I just saw "Intel IOMMU (VT-d)" in 
"qemu-system-ppc64 --help" output and concluded there is no easy way.

Please dig again :).


Ok. I did. I can do as below. QEMU shows different things on "-help" on PPC and x86_64. But qemu-doc.html has 2 entries for "-cpu" and it does not mention architecture per "-cpu" instance there (unless I keep the original "(PPC64 only)" which I find rather ugly):

===
   -cpu model
Select CPU model (-cpu help for list and additional feature selection)

   -cpu model
Select CPU model (-cpu help for list and additional feature selection)

        compat=power6|power7|power8
Enables architecture compatibility mode. "power6" enables PowerISA 2.05, "power7" - 2.06, "power8" - 2.07.

====


For the sake of a nice looking html, I would stick to the original patch but since you seem to know this stuff better, any help is very appreciated :)




diff --git a/qemu-options.hx b/qemu-options.hx
index c513352..428359a 100644
--- a/qemu-options.hx
+++ b/qemu-options.hx
@@ -84,13 +84,30 @@ HXCOMM Deprecated by -machine
 DEF("M", HAS_ARG, QEMU_OPTION_M, "", QEMU_ARCH_ALL)

 DEF("cpu", HAS_ARG, QEMU_OPTION_cpu,
-    "-cpu cpu        select CPU ('-cpu help' for list)\n", QEMU_ARCH_ALL)
+    "-cpu cpu        select CPU ('-cpu help' for list)\n",
+            QEMU_ARCH_ALL & ~QEMU_ARCH_PPC)
 STEXI
 @item -cpu @var{model}
 @findex -cpu
 Select CPU model (@code{-cpu help} for list and additional feature selection)
 ETEXI

+DEF("cpu", HAS_ARG, QEMU_OPTION_cpu_ppc,
+    "-cpu cpu,[compat=MODE]\n"
+    "                select CPU ('-cpu help' for list)\n"
+    "                compat= processor compatibility mode\n",
+        QEMU_ARCH_PPC)
+STEXI
+@item -cpu @var{model}
+@findex -cpu
+Select CPU model (@code{-cpu help} for list and additional feature selection)
+@table @option
+@item compat=power6|power7|power8
+Enables architecture compatibility mode. "power6" enables PowerISA 2.05,
+"power7" - 2.06, "power8" - 2.07.
+@end table
+ETEXI
+
 DEF("smp", HAS_ARG, QEMU_OPTION_smp,
"-smp [cpus=]n[,maxcpus=cpus][,cores=cores][,threads=threads][,sockets=sockets]\n"
     "                set the number of CPUs to 'n' [default=1]\n"



--
Alexey

Reply via email to