Re: [Qemu-devel] pseries machine updates

[David Gibson <david@gibson.dropbear.id.au>]

On Wed, Aug 10, 2011 at 05:24:59PM +0200, Alexander Graf wrote:
> On 08/10/2011 05:16 PM, Alexander Graf wrote:
> >On 08/04/2011 09:02 AM, David Gibson wrote:
> >>Hi Alex,
> >>
> >>Here's another batch of assorted updates for the pseries machine.
> >
> >Looks pretty nice. Please update patch 2/6 with the bug you found
> >and the whitespace problems. I'll put the others into my tree
> >already.
> 
> Sorry, patch 3 failed checkpatch. Please respin that one too.

Done.  Here you go.  I'll blame Ben again :)

>From 6d5418b479852edfe93c40b8ddb707dde5128e32 Mon Sep 17 00:00:00 2001
From: Ben Herrenschmidt <benh@kernel.crashing.org>
Date: Fri, 29 Jul 2011 12:06:33 +1000
Subject: [PATCH] pseries: Add real mode debugging hcalls

PAPR systems support several hypercalls intended for use in real mode
debugging tools.  These implement reads and writes to arbitrary guest
physical addresses.  This is useful for real mode software because it
allows access to IO addresses and memory outside the RMA without going
through the somewhat involved process of setting up the hash page table
and enabling translation.

We want these so that when we add real IO devices, the SLOF firmware can
boot from them without having to enter virtual mode.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: David Gibson <dwg@au1.ibm.com>
---
 hw/spapr_hcall.c |   73 ++++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 files changed, 73 insertions(+), 0 deletions(-)

diff --git a/hw/spapr_hcall.c b/hw/spapr_hcall.c
index f7ead04..05d6c54 100644
--- a/hw/spapr_hcall.c
+++ b/hw/spapr_hcall.c
@@ -449,6 +449,67 @@ static target_ulong h_rtas(CPUState *env, sPAPREnvironment *spapr,
                            nret, rtas_r3 + 12 + 4*nargs);
 }
 
+static target_ulong h_logical_load(CPUState *env, sPAPREnvironment *spapr,
+                                   target_ulong opcode, target_ulong *args)
+{
+    target_ulong size = args[0];
+    target_ulong addr = args[1];
+
+    switch (size) {
+    case 1:
+        args[0] = ldub_phys(addr);
+        return H_SUCCESS;
+    case 2:
+        args[0] = lduw_phys(addr);
+        return H_SUCCESS;
+    case 4:
+        args[0] = ldl_phys(addr);
+        return H_SUCCESS;
+    case 8:
+        args[0] = ldq_phys(addr);
+        return H_SUCCESS;
+    }
+    return H_PARAMETER;
+}
+
+static target_ulong h_logical_store(CPUState *env, sPAPREnvironment *spapr,
+                                    target_ulong opcode, target_ulong *args)
+{
+    target_ulong size = args[0];
+    target_ulong addr = args[1];
+    target_ulong val  = args[2];
+
+    switch (size) {
+    case 1:
+        stb_phys(addr, val);
+        return H_SUCCESS;
+    case 2:
+        stw_phys(addr, val);
+        return H_SUCCESS;
+    case 4:
+        stl_phys(addr, val);
+        return H_SUCCESS;
+    case 8:
+        stq_phys(addr, val);
+        return H_SUCCESS;
+    }
+    return H_PARAMETER;
+}
+
+static target_ulong h_logical_icbi(CPUState *env, sPAPREnvironment *spapr,
+                                   target_ulong opcode, target_ulong *args)
+{
+    /* Nothing to do on emulation, KVM will trap this in the kernel */
+    return H_SUCCESS;
+}
+
+static target_ulong h_logical_dcbf(CPUState *env, sPAPREnvironment *spapr,
+                                   target_ulong opcode, target_ulong *args)
+{
+    /* Nothing to do on emulation, KVM will trap this in the kernel */
+    return H_SUCCESS;
+}
+
 static spapr_hcall_fn papr_hypercall_table[(MAX_HCALL_OPCODE / 4) + 1];
 static spapr_hcall_fn kvmppc_hypercall_table[KVMPPC_HCALL_MAX - KVMPPC_HCALL_BASE + 1];
 
@@ -513,6 +574,18 @@ static void hypercall_init(void)
     spapr_register_hypercall(H_REGISTER_VPA, h_register_vpa);
     spapr_register_hypercall(H_CEDE, h_cede);
 
+    /* "debugger" hcalls (also used by SLOF). Note: We do -not- differenciate
+     * here between the "CI" and the "CACHE" variants, they will use whatever
+     * mapping attributes qemu is using. When using KVM, the kernel will
+     * enforce the attributes more strongly
+     */
+    spapr_register_hypercall(H_LOGICAL_CI_LOAD, h_logical_load);
+    spapr_register_hypercall(H_LOGICAL_CI_STORE, h_logical_store);
+    spapr_register_hypercall(H_LOGICAL_CACHE_LOAD, h_logical_load);
+    spapr_register_hypercall(H_LOGICAL_CACHE_STORE, h_logical_store);
+    spapr_register_hypercall(H_LOGICAL_ICBI, h_logical_icbi);
+    spapr_register_hypercall(H_LOGICAL_DCBF, h_logical_dcbf);
+
     /* qemu/KVM-PPC specific hcalls */
     spapr_register_hypercall(KVMPPC_H_RTAS, h_rtas);
 }
-- 
1.7.5.4



-- 
David Gibson			| I'll have my music baroque, and my code
david AT gibson.dropbear.id.au	| minimalist, thank you.  NOT _the_ _other_
				| _way_ _around_!
http://www.ozlabs.org/~dgibson