xen/mm: Remove INVALID_{MFN,GFN}_INITIALIZER
PATCH v2 1/1 xen/mm: Remove INVALID_{MFN,GFN}_INITIALIZER
These existed to work around a bug in older GCC when using struct-casting for variable initialisation. However, our baseline toolchain is new enough to not suffer this bug.
Removing these resolves two MISRA Rule 9.2 violations which exist in release builds of Xen only, where "= { ... }" is disallowed for simple scalar initialisation.
The BUILD_BUG_ON() in xenmem_add_to_physmap() cannot stay as it is, because INVALID_GFN is not an Integer Constant Expression. Replace it BUILD_ERROR() which is the nearest available alternative.
No functional change.
---
CC: Anthony PERARD <[email protected]>
CC: Michal Orzel <[email protected]>
CC: Jan Beulich <[email protected]>
CC: Julien Grall <[email protected]>
CC: Roger Pau Monné <[email protected]>
CC: Stefano Stabellini <[email protected]>
CC: Volodymyr Babchuk <[email protected]>
CC: Bertrand Marquis <[email protected]>
CC: Oleksii Kurochko <[email protected]>
v2:
* Drop INVALID_{MFN,GFN}_INITIALIZER entirely.
For 4.22. staging-4.22 is currently blocked by this bug, following the switch from debug builds to release builds.
Passing pipeline:
https://gitlab.com/xen-project/hardware/xen-staging/-/pipelines/2638483933
--- xen/arch/arm/mmu/setup.c | 2 +- xen/arch/riscv/mm.c | 2 +- xen/arch/x86/efi/runtime.h | 2 +- xen/common/memory.c | 4 +++- xen/common/page_alloc.c | 2 +- xen/include/xen/mm-frame.h | 16 ++-------------- 6 files changed, 9 insertions(+), 19 deletions(-)
xen/arch/arm/mmu/setup.c +1 -1
@@ -59,7 +59,7 @@ DEFINE_BOOT_PAGE_TABLE(xen_fixmap); static DEFINE_PAGE_TABLES(xen_xenmap, XEN_NR_ENTRIES(2)); /* Limits of the Xen heap */ -mfn_t directmap_mfn_start __read_mostly = INVALID_MFN_INITIALIZER; +mfn_t directmap_mfn_start __read_mostly = INVALID_MFN; mfn_t directmap_mfn_end __read_mostly; vaddr_t directmap_virt_end __read_mostly; #ifdef CONFIG_ARM_64
xen/arch/riscv/mm.c +1 -1
@@ -480,7 +480,7 @@ static void __init setup_frametable_mappings(paddr_t ps, paddr_t pe)static void __init setup_directmap_mappings (unsigned long base_mfn,unsigned long nr_mfns) { -static mfn_t __initdata directmap_mfn_start =INVALID_MFN_INITIALIZER ; +static mfn_t __initdata directmap_mfn_start =INVALID_MFN ;mfn_t base_mfn_t =_mfn (base_mfn);unsigned long base_addr =mfn_to_maddr (base_mfn_t);
xen/arch/x86/efi/runtime.h +1 -1
@@ -7,7 +7,7 @@#include <asm/mc146818rtc.h> #ifndef COMPAT -mfn_t __read_mostly efi_l4_mfn =INVALID_MFN_INITIALIZER ; +mfn_t __read_mostly efi_l4_mfn =INVALID_MFN ;void efi_update_l4_pgtable (unsigned int l4idx,l4_pgentry_t l4e) {
xen/common/memory.c +3 -1
@@ -991,7 +991,9 @@ int xenmem_add_to_physmap(struct domain *d, struct xen_add_to_physmap *xatp,* guaranteeing that it won't fall in the middle of the * [xatp->gpfn, xatp->gpfn + xatp->size) range checked above. */ -BUILD_BUG_ON (INVALID_GFN_RAW +1 ); +if (gfn_x (INVALID_GFN ) +1 ) +BUILD_ERROR ("bad INVALID_GFN" ); +return -EOVERFLOW ; }
xen/common/page_alloc.c +1 -1
@@ -275,7 +275,7 @@ static PAGE_LIST_HEAD(page_broken_list);* When !CONFIG_NUMA first_valid_mfn is non-static, for use by respective * stubs. */ -STATIC_IF(CONFIG_NUMA) mfn_t first_valid_mfn = INVALID_MFN_INITIALIZER; +STATIC_IF(CONFIG_NUMA) mfn_t first_valid_mfn = INVALID_MFN;struct bootmem_region {unsigned long s, e;/* MFNs @s through @e-1 inclusive are free */
xen/include/xen/mm-frame.h +2 -14
@@ -6,13 +6,7 @@ TYPE_SAFE(unsigned long, mfn);#define PRI_mfn "05lx" -#define INVALID_MFN_RAW (~0UL) -#define INVALID_MFN _mfn(INVALID_MFN_RAW) -/* -* To be used for global variable initialization. This workaround a bug -* in GCC < 5.0. -*/ -#define INVALID_MFN_INITIALIZER { INVALID_MFN_RAW } +#define INVALID_MFN ((mfn_t){ ~0UL })#if 0 #define mfn_t/* Grep fodder: mfn_t, _mfn() and mfn_x() are defined above */
@@ -42,13 +36,7 @@ static inline bool mfn_eq(mfn_t x, mfn_t y) TYPE_SAFE(unsigned long, gfn);#define PRI_gfn "05lx" -#define INVALID_GFN_RAW (~0UL) -#define INVALID_GFN _gfn(INVALID_GFN_RAW) -/* -* To be used for global variable initialization. This workaround a bug -* in GCC < 5.0 https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64856 -*/ -#define INVALID_GFN_INITIALIZER { INVALID_GFN_RAW } +#define INVALID_GFN ((gfn_t){ ~0UL })#if 0 #define gfn_t/* Grep fodder: gfn_t, _gfn() and gfn_x() are defined above */
Signature · 1 line
2.39.5
Quoted 8 lines · Andrew Cooper
These existed to work around a bug in older GCC when using struct-casting for variable initialisation. However, our baseline toolchain is new enough to not suffer this bug.
Removing these resolves two MISRA Rule 9.2 violations which exist in release builds of Xen only, where "= { ... }" is disallowed for simple scalar initialisation.
Quoted 7 lines
The
BUILD_BUG_ON()inxenmem_add_to_physmap()cannot stay as it is, becauseINVALID_GFNis not an Integer Constant Expression. Replace itBUILD_ERROR()which is the nearest available alternative.No functional change.
Signed-off-by Andrew Cooper <[email protected]>
Quoted 29 lines · Andrew Cooper
These existed to work around a bug in older GCC when using struct-casting for variable initialisation. However, our baseline toolchain is new enough to not suffer this bug.
Removing these resolves two MISRA Rule 9.2 violations which exist in release builds of Xen only, where "= { ... }" is disallowed for simple scalar initialisation.
The
BUILD_BUG_ON()inxenmem_add_to_physmap()cannot stay as it is, becauseINVALID_GFNis not an Integer Constant Expression. Replace itBUILD_ERROR()which is the nearest available alternative.No functional change.
Signed-off-by Andrew Cooper <[email protected]>---
CC: Anthony PERARD <[email protected]>
CC: Michal Orzel <[email protected]>
CC: Jan Beulich <[email protected]>
CC: Julien Grall <[email protected]>
CC: Roger Pau Monné <[email protected]>
CC: Stefano Stabellini <[email protected]>
CC: Volodymyr Babchuk <[email protected]>
CC: Bertrand Marquis <[email protected]>
CC: Oleksii Kurochko <[email protected]>v2:
* Drop INVALID_{MFN,GFN}_INITIALIZER entirely.
Quoted 6 lines
For 4.22. staging-4.22 is currently blocked by this bug, following the switch from debug builds to release builds.
Passing pipeline:
https://gitlab.com/xen-project/hardware/xen-staging/-/pipelines/2638483933---
~ Oleksii
Quoted 8 lines · Andrew Cooper
These existed to work around a bug in older GCC when using struct-casting for variable initialisation. However, our baseline toolchain is new enough to not suffer this bug.
Removing these resolves two MISRA Rule 9.2 violations which exist in release builds of Xen only, where "= { ... }" is disallowed for simple scalar initialisation.
Quoted 7 lines
The
BUILD_BUG_ON()inxenmem_add_to_physmap()cannot stay as it is, becauseINVALID_GFNis not an Integer Constant Expression. Replace itBUILD_ERROR()which is the nearest available alternative.No functional change.
Signed-off-by Andrew Cooper <[email protected]>
~Michal