2008-08-19 19:18:19

by Eduardo Habkost

[permalink] [raw]
Subject: [PATCH] KVM_GUEST: use the right parameter type on kvm_release_pt() (was Re: i386 allmodconfig in linux-next)

On Tue, Aug 19, 2008 at 04:41:41PM +0300, Avi Kivity wrote:
> Andrew Morton wrote:
>> arch/x86/kernel/kvm.c: In function 'paravirt_ops_setup':
>> arch/x86/kernel/kvm.c:233: warning: assignment from incompatible pointer type
>> arch/x86/kernel/kvm.c:234: warning: assignment from incompatible pointer type
>> arch/x86/kernel/kvm.c:235: warning: assignment from incompatible pointer type
>>
>
> Eduardo, this is your
>
>> commit fa24f8e67c189b8b904c8a34f12e6dae9c14dbba
>> Author: Eduardo Habkost <[email protected]>
>> Date: Wed Jul 30 18:32:27 2008 -0300
>>
>> x86, paravirt_ops: use unsigned long instead of u32 for alloc_p*()
>> pfn args
>
> Please compile-test (and run-test?) paravirt_ops changes with
> CONFIG_KVM_GUEST and CONFIG_KVM_CLOCK enabled. Thanks.
>

Oops, sorry. I've grepped for 'alloc_p*' and didn't notice KVM was using
release_p*. I've checked for binary code changes with all paravirt guest
options enabled, but missed the compiler warnings. Fix below.

---
From: Eduardo Habkost <[email protected]>
Subject: [PATCH] KVM_GUEST: Use the right parameter type on kvm_release_pt()

This fixes the following compiler warnings, that were introduced by me
on commit fa24f8e67c189b8b904c8a34f12e6dae9c14dbba.

arch/x86/kernel/kvm.c: In function 'paravirt_ops_setup':
arch/x86/kernel/kvm.c:233: warning: assignment from incompatible pointer type
arch/x86/kernel/kvm.c:234: warning: assignment from incompatible pointer type
arch/x86/kernel/kvm.c:235: warning: assignment from incompatible pointer type

Signed-off-by: Eduardo Habkost <[email protected]>
---
arch/x86/kernel/kvm.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c
index 8b7a3cf..478bca9 100644
--- a/arch/x86/kernel/kvm.c
+++ b/arch/x86/kernel/kvm.c
@@ -178,7 +178,7 @@ static void kvm_flush_tlb(void)
kvm_deferred_mmu_op(&ftlb, sizeof ftlb);
}

-static void kvm_release_pt(u32 pfn)
+static void kvm_release_pt(unsigned long pfn)
{
struct kvm_mmu_op_release_pt rpt = {
.header.op = KVM_MMU_OP_RELEASE_PT,
--
1.5.5.GIT

--
Eduardo


2008-08-21 13:50:57

by Avi Kivity

[permalink] [raw]
Subject: Re: [PATCH] KVM_GUEST: use the right parameter type on kvm_release_pt() (was Re: i386 allmodconfig in linux-next)

Eduardo Habkost wrote:
> From: Eduardo Habkost <[email protected]>
> Subject: [PATCH] KVM_GUEST: Use the right parameter type on kvm_release_pt()
>
> This fixes the following compiler warnings, that were introduced by me
> on commit fa24f8e67c189b8b904c8a34f12e6dae9c14dbba.
>
> arch/x86/kernel/kvm.c: In function 'paravirt_ops_setup':
> arch/x86/kernel/kvm.c:233: warning: assignment from incompatible pointer type
> arch/x86/kernel/kvm.c:234: warning: assignment from incompatible pointer type
> arch/x86/kernel/kvm.c:235: warning: assignment from incompatible pointer type
>

Ingo? Can you queue/fold this into the original patch?

> Signed-off-by: Eduardo Habkost <[email protected]>
> ---
> arch/x86/kernel/kvm.c | 2 +-
> 1 files changed, 1 insertions(+), 1 deletions(-)
>
> diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c
> index 8b7a3cf..478bca9 100644
> --- a/arch/x86/kernel/kvm.c
> +++ b/arch/x86/kernel/kvm.c
> @@ -178,7 +178,7 @@ static void kvm_flush_tlb(void)
> kvm_deferred_mmu_op(&ftlb, sizeof ftlb);
> }
>
> -static void kvm_release_pt(u32 pfn)
> +static void kvm_release_pt(unsigned long pfn)
> {
> struct kvm_mmu_op_release_pt rpt = {
> .header.op = KVM_MMU_OP_RELEASE_PT,
>


--
error compiling committee.c: too many arguments to function

2008-08-22 03:36:50

by Ingo Molnar

[permalink] [raw]
Subject: Re: [PATCH] KVM_GUEST: use the right parameter type on kvm_release_pt() (was Re: i386 allmodconfig in linux-next)


* Avi Kivity <[email protected]> wrote:

> Eduardo Habkost wrote:
>> From: Eduardo Habkost <[email protected]>
>> Subject: [PATCH] KVM_GUEST: Use the right parameter type on kvm_release_pt()
>>
>> This fixes the following compiler warnings, that were introduced by me
>> on commit fa24f8e67c189b8b904c8a34f12e6dae9c14dbba.
>>
>> arch/x86/kernel/kvm.c: In function 'paravirt_ops_setup':
>> arch/x86/kernel/kvm.c:233: warning: assignment from incompatible pointer type
>> arch/x86/kernel/kvm.c:234: warning: assignment from incompatible pointer type
>> arch/x86/kernel/kvm.c:235: warning: assignment from incompatible pointer type
>>
>
> Ingo? Can you queue/fold this into the original patch?

done - updated v2 combo patch below, queued in tip/x86/paravirt.

Ingo

-------------------->
>From f86399396ce7a4f4069828b7dceac5aa5113dfb5 Mon Sep 17 00:00:00 2001
From: Eduardo Habkost <[email protected]>
Date: Wed, 30 Jul 2008 18:32:27 -0300
Subject: [PATCH] x86, paravirt_ops: use unsigned long instead of u32 for alloc_p*() pfn args

This patch changes the pfn args from 'u32' to 'unsigned long'
on alloc_p*() functions on paravirt_ops, and the corresponding
implementations for Xen and VMI. The prototypes for CONFIG_PARAVIRT=n
are already using unsigned long, so paravirt.h now matches the prototypes
on asm-x86/pgalloc.h.

It shouldn't result in any changes on generated code on 32-bit, with
or without CONFIG_PARAVIRT. On both cases, 'codiff -f' didn't show any
change after applying this patch.

On 64-bit, there are (expected) binary changes only when CONFIG_PARAVIRT
is enabled, as the patch is really supposed to change the size of the
pfn args.

[ v2: KVM_GUEST: use the right parameter type on kvm_release_pt() ]

Signed-off-by: Eduardo Habkost <[email protected]>
Acked-by: Jeremy Fitzhardinge <[email protected]>
Acked-by: Zachary Amsden <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
---
arch/x86/kernel/kvm.c | 2 +-
arch/x86/kernel/vmi_32.c | 10 +++++-----
arch/x86/xen/enlighten.c | 20 ++++++++++----------
include/asm-x86/paravirt.h | 30 +++++++++++++++---------------
4 files changed, 31 insertions(+), 31 deletions(-)

diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c
index 8b7a3cf..478bca9 100644
--- a/arch/x86/kernel/kvm.c
+++ b/arch/x86/kernel/kvm.c
@@ -178,7 +178,7 @@ static void kvm_flush_tlb(void)
kvm_deferred_mmu_op(&ftlb, sizeof ftlb);
}

-static void kvm_release_pt(u32 pfn)
+static void kvm_release_pt(unsigned long pfn)
{
struct kvm_mmu_op_release_pt rpt = {
.header.op = KVM_MMU_OP_RELEASE_PT,
diff --git a/arch/x86/kernel/vmi_32.c b/arch/x86/kernel/vmi_32.c
index 0a1b1a9..340b036 100644
--- a/arch/x86/kernel/vmi_32.c
+++ b/arch/x86/kernel/vmi_32.c
@@ -392,13 +392,13 @@ static void *vmi_kmap_atomic_pte(struct page *page, enum km_type type)
}
#endif

-static void vmi_allocate_pte(struct mm_struct *mm, u32 pfn)
+static void vmi_allocate_pte(struct mm_struct *mm, unsigned long pfn)
{
vmi_set_page_type(pfn, VMI_PAGE_L1);
vmi_ops.allocate_page(pfn, VMI_PAGE_L1, 0, 0, 0);
}

-static void vmi_allocate_pmd(struct mm_struct *mm, u32 pfn)
+static void vmi_allocate_pmd(struct mm_struct *mm, unsigned long pfn)
{
/*
* This call comes in very early, before mem_map is setup.
@@ -409,20 +409,20 @@ static void vmi_allocate_pmd(struct mm_struct *mm, u32 pfn)
vmi_ops.allocate_page(pfn, VMI_PAGE_L2, 0, 0, 0);
}

-static void vmi_allocate_pmd_clone(u32 pfn, u32 clonepfn, u32 start, u32 count)
+static void vmi_allocate_pmd_clone(unsigned long pfn, unsigned long clonepfn, unsigned long start, unsigned long count)
{
vmi_set_page_type(pfn, VMI_PAGE_L2 | VMI_PAGE_CLONE);
vmi_check_page_type(clonepfn, VMI_PAGE_L2);
vmi_ops.allocate_page(pfn, VMI_PAGE_L2 | VMI_PAGE_CLONE, clonepfn, start, count);
}

-static void vmi_release_pte(u32 pfn)
+static void vmi_release_pte(unsigned long pfn)
{
vmi_ops.release_page(pfn, VMI_PAGE_L1);
vmi_set_page_type(pfn, VMI_PAGE_NORMAL);
}

-static void vmi_release_pmd(u32 pfn)
+static void vmi_release_pmd(unsigned long pfn)
{
vmi_ops.release_page(pfn, VMI_PAGE_L2);
vmi_set_page_type(pfn, VMI_PAGE_NORMAL);
diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c
index 9ff6e3c..db970bd 100644
--- a/arch/x86/xen/enlighten.c
+++ b/arch/x86/xen/enlighten.c
@@ -812,7 +812,7 @@ static int xen_write_msr_safe(unsigned int msr, unsigned low, unsigned high)

/* Early in boot, while setting up the initial pagetable, assume
everything is pinned. */
-static __init void xen_alloc_pte_init(struct mm_struct *mm, u32 pfn)
+static __init void xen_alloc_pte_init(struct mm_struct *mm, unsigned long pfn)
{
#ifdef CONFIG_FLATMEM
BUG_ON(mem_map); /* should only be used early */
@@ -822,7 +822,7 @@ static __init void xen_alloc_pte_init(struct mm_struct *mm, u32 pfn)

/* Early release_pte assumes that all pts are pinned, since there's
only init_mm and anything attached to that is pinned. */
-static void xen_release_pte_init(u32 pfn)
+static void xen_release_pte_init(unsigned long pfn)
{
make_lowmem_page_readwrite(__va(PFN_PHYS(pfn)));
}
@@ -838,7 +838,7 @@ static void pin_pagetable_pfn(unsigned cmd, unsigned long pfn)

/* This needs to make sure the new pte page is pinned iff its being
attached to a pinned pagetable. */
-static void xen_alloc_ptpage(struct mm_struct *mm, u32 pfn, unsigned level)
+static void xen_alloc_ptpage(struct mm_struct *mm, unsigned long pfn, unsigned level)
{
struct page *page = pfn_to_page(pfn);

@@ -856,12 +856,12 @@ static void xen_alloc_ptpage(struct mm_struct *mm, u32 pfn, unsigned level)
}
}

-static void xen_alloc_pte(struct mm_struct *mm, u32 pfn)
+static void xen_alloc_pte(struct mm_struct *mm, unsigned long pfn)
{
xen_alloc_ptpage(mm, pfn, PT_PTE);
}

-static void xen_alloc_pmd(struct mm_struct *mm, u32 pfn)
+static void xen_alloc_pmd(struct mm_struct *mm, unsigned long pfn)
{
xen_alloc_ptpage(mm, pfn, PT_PMD);
}
@@ -909,7 +909,7 @@ static void xen_pgd_free(struct mm_struct *mm, pgd_t *pgd)
}

/* This should never happen until we're OK to use struct page */
-static void xen_release_ptpage(u32 pfn, unsigned level)
+static void xen_release_ptpage(unsigned long pfn, unsigned level)
{
struct page *page = pfn_to_page(pfn);

@@ -923,23 +923,23 @@ static void xen_release_ptpage(u32 pfn, unsigned level)
}
}

-static void xen_release_pte(u32 pfn)
+static void xen_release_pte(unsigned long pfn)
{
xen_release_ptpage(pfn, PT_PTE);
}

-static void xen_release_pmd(u32 pfn)
+static void xen_release_pmd(unsigned long pfn)
{
xen_release_ptpage(pfn, PT_PMD);
}

#if PAGETABLE_LEVELS == 4
-static void xen_alloc_pud(struct mm_struct *mm, u32 pfn)
+static void xen_alloc_pud(struct mm_struct *mm, unsigned long pfn)
{
xen_alloc_ptpage(mm, pfn, PT_PUD);
}

-static void xen_release_pud(u32 pfn)
+static void xen_release_pud(unsigned long pfn)
{
xen_release_ptpage(pfn, PT_PUD);
}
diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h
index fbbde93..497aea0 100644
--- a/include/asm-x86/paravirt.h
+++ b/include/asm-x86/paravirt.h
@@ -257,13 +257,13 @@ struct pv_mmu_ops {
* Hooks for allocating/releasing pagetable pages when they're
* attached to a pagetable
*/
- void (*alloc_pte)(struct mm_struct *mm, u32 pfn);
- void (*alloc_pmd)(struct mm_struct *mm, u32 pfn);
- void (*alloc_pmd_clone)(u32 pfn, u32 clonepfn, u32 start, u32 count);
- void (*alloc_pud)(struct mm_struct *mm, u32 pfn);
- void (*release_pte)(u32 pfn);
- void (*release_pmd)(u32 pfn);
- void (*release_pud)(u32 pfn);
+ void (*alloc_pte)(struct mm_struct *mm, unsigned long pfn);
+ void (*alloc_pmd)(struct mm_struct *mm, unsigned long pfn);
+ void (*alloc_pmd_clone)(unsigned long pfn, unsigned long clonepfn, unsigned long start, unsigned long count);
+ void (*alloc_pud)(struct mm_struct *mm, unsigned long pfn);
+ void (*release_pte)(unsigned long pfn);
+ void (*release_pmd)(unsigned long pfn);
+ void (*release_pud)(unsigned long pfn);

/* Pagetable manipulation functions */
void (*set_pte)(pte_t *ptep, pte_t pteval);
@@ -993,35 +993,35 @@ static inline void paravirt_pgd_free(struct mm_struct *mm, pgd_t *pgd)
PVOP_VCALL2(pv_mmu_ops.pgd_free, mm, pgd);
}

-static inline void paravirt_alloc_pte(struct mm_struct *mm, unsigned pfn)
+static inline void paravirt_alloc_pte(struct mm_struct *mm, unsigned long pfn)
{
PVOP_VCALL2(pv_mmu_ops.alloc_pte, mm, pfn);
}
-static inline void paravirt_release_pte(unsigned pfn)
+static inline void paravirt_release_pte(unsigned long pfn)
{
PVOP_VCALL1(pv_mmu_ops.release_pte, pfn);
}

-static inline void paravirt_alloc_pmd(struct mm_struct *mm, unsigned pfn)
+static inline void paravirt_alloc_pmd(struct mm_struct *mm, unsigned long pfn)
{
PVOP_VCALL2(pv_mmu_ops.alloc_pmd, mm, pfn);
}

-static inline void paravirt_alloc_pmd_clone(unsigned pfn, unsigned clonepfn,
- unsigned start, unsigned count)
+static inline void paravirt_alloc_pmd_clone(unsigned long pfn, unsigned long clonepfn,
+ unsigned long start, unsigned long count)
{
PVOP_VCALL4(pv_mmu_ops.alloc_pmd_clone, pfn, clonepfn, start, count);
}
-static inline void paravirt_release_pmd(unsigned pfn)
+static inline void paravirt_release_pmd(unsigned long pfn)
{
PVOP_VCALL1(pv_mmu_ops.release_pmd, pfn);
}

-static inline void paravirt_alloc_pud(struct mm_struct *mm, unsigned pfn)
+static inline void paravirt_alloc_pud(struct mm_struct *mm, unsigned long pfn)
{
PVOP_VCALL2(pv_mmu_ops.alloc_pud, mm, pfn);
}
-static inline void paravirt_release_pud(unsigned pfn)
+static inline void paravirt_release_pud(unsigned long pfn)
{
PVOP_VCALL1(pv_mmu_ops.release_pud, pfn);
}