| Message ID | 20090106202245.GB17714@uranus.ravnborg.org |
|---|---|
| State | Accepted |
| Delegated to: | David Miller |
| Headers | show |
From: Sam Ravnborg <sam@ravnborg.org> Date: Tue, 6 Jan 2009 21:22:45 +0100 > The sparc64 allmodconfig build broke due to enabling of the > branch_tracer that does some very clever things with > all if conditions. This caused my gcc 3.4.5 to be so confused that > it emitted two warnings: > > arch/sparc/mm/init_64.c: In function `update_mmu_cache': > arch/sparc/mm/init_64.c:271: warning: 'pg_flags' might be used uninitialized in this function > arch/sparc/mm/init_64.c:272: warning: 'page' might be used uninitialized in this function > > And with -Werror this broke the build. > > Refactor code so it: > 1) becomes more readable > 2) no longer emit a warning with the branch_tracer enabled > > The refactoring uses a small helper function (flush_dcache()). > > Signed-off-by: Sam Ravnborg <sam@ravnborg.org> > Cc: Steven Rostedt <rostedt@goodmis.org> > --- > > The diff is not readable so you need to apply it to > see the end result. > > Please review extra carefully to check that the transformation of: > > (page = pfn_to_page(pfn), page_mapping(page)) > > is correct. Looks correct, applied. Thanks Sam. -- To unsubscribe from this list: send the line "unsubscribe sparclinux" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
diff --git a/arch/sparc/mm/init_64.c b/arch/sparc/mm/init_64.c index c04b111..6afb1a1 100644 --- a/arch/sparc/mm/init_64.c +++ b/arch/sparc/mm/init_64.c @@ -259,21 +259,16 @@ static inline void tsb_insert(struct tsb *ent, unsigned long tag, unsigned long unsigned long _PAGE_ALL_SZ_BITS __read_mostly; unsigned long _PAGE_SZBITS __read_mostly; -void update_mmu_cache(struct vm_area_struct *vma, unsigned long address, pte_t pte) +static void flush_dcache(unsigned long pfn) { - struct mm_struct *mm; - struct tsb *tsb; - unsigned long tag, flags; - unsigned long tsb_index, tsb_hash_shift; + struct page *page; - if (tlb_type != hypervisor) { - unsigned long pfn = pte_pfn(pte); + page = pfn_to_page(pfn); + if (page && page_mapping(page)) { unsigned long pg_flags; - struct page *page; - if (pfn_valid(pfn) && - (page = pfn_to_page(pfn), page_mapping(page)) && - ((pg_flags = page->flags) & (1UL << PG_dcache_dirty))) { + pg_flags = page->flags; + if (pg_flags & (1UL << PG_dcache_dirty)) { int cpu = ((pg_flags >> PG_dcache_cpu_shift) & PG_dcache_cpu_mask); int this_cpu = get_cpu(); @@ -291,6 +286,21 @@ void update_mmu_cache(struct vm_area_struct *vma, unsigned long address, pte_t p put_cpu(); } } +} + +void update_mmu_cache(struct vm_area_struct *vma, unsigned long address, pte_t pte) +{ + struct mm_struct *mm; + struct tsb *tsb; + unsigned long tag, flags; + unsigned long tsb_index, tsb_hash_shift; + + if (tlb_type != hypervisor) { + unsigned long pfn = pte_pfn(pte); + + if (pfn_valid(pfn)) + flush_dcache(pfn); + } mm = vma->vm_mm;
The sparc64 allmodconfig build broke due to enabling of the branch_tracer that does some very clever things with all if conditions. This caused my gcc 3.4.5 to be so confused that it emitted two warnings: arch/sparc/mm/init_64.c: In function `update_mmu_cache': arch/sparc/mm/init_64.c:271: warning: 'pg_flags' might be used uninitialized in this function arch/sparc/mm/init_64.c:272: warning: 'page' might be used uninitialized in this function And with -Werror this broke the build. Refactor code so it: 1) becomes more readable 2) no longer emit a warning with the branch_tracer enabled The refactoring uses a small helper function (flush_dcache()). Signed-off-by: Sam Ravnborg <sam@ravnborg.org> Cc: Steven Rostedt <rostedt@goodmis.org> --- The diff is not readable so you need to apply it to see the end result. Please review extra carefully to check that the transformation of: (page = pfn_to_page(pfn), page_mapping(page)) is correct. Thanks, Sam -- To unsubscribe from this list: send the line "unsubscribe sparclinux" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html