From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jack Steiner Date: Sat, 22 May 2004 12:11:59 +0000 Subject: Re: [PATCH] Reduce TLB flushing during process migration Message-Id: <20040522121159.GB19454@sgi.com> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: linux-ia64@vger.kernel.org On Wed, May 12, 2004 at 03:08:36PM -0700, Andrew Morton wrote: > David Mosberger wrote: > > > > Please remove it until a proper solution exists. > > I'll submit this on Friday, when bugalugs returns. > Here is an updated patch for flushing the TLB on SN2 after a process migration. The patch is based on the latest "//linux.bkbits.net/linux-2.5" bitkeeper tree. ------------------------------------- --- linuxbase/kernel/sched.c 2004-05-21 20:47:33.000000000 -0500 +++ linux/kernel/sched.c 2004-05-21 22:42:39.000000000 -0500 @@ -40,6 +40,7 @@ #include #include #include +#include #ifdef CONFIG_NUMA #define cpu_to_node_mask(cpu) node_to_cpumask(cpu_to_node(cpu)) @@ -1311,6 +1312,7 @@ wake_up_process(mt); put_task_struct(mt); wait_for_completion(&req.done); + tlb_migrate_finish(p->mm); return; } out: @@ -3317,6 +3319,7 @@ task_rq_unlock(rq, &flags); wake_up_process(rq->migration_thread); wait_for_completion(&req.done); + tlb_migrate_finish(p->mm); return 0; } out: --- linuxbase/arch/ia64/sn/kernel/sn2/sn2_smp.c 2004-05-21 20:47:20.000000000 -0500 +++ linux/arch/ia64/sn/kernel/sn2/sn2_smp.c 2004-05-21 20:48:25.000000000 -0500 @@ -27,6 +27,7 @@ #include #include #include +#include #include #include #include @@ -60,6 +61,13 @@ } +void +sn_tlb_migrate_finish(struct mm_struct *mm) +{ + if (mm = current->mm) + flush_tlb_mm(mm); +} + /** * sn2_global_tlb_purge - globally purge translation cache of virtual address range @@ -114,6 +122,13 @@ return; } + if (atomic_read(&mm->mm_users) = 1) { + flush_tlb_mm(mm); + preempt_enable(); + return; + } + + nix = 0; for (cnode=find_first_bit(&nodes_flushed, NR_NODES); cnode < NR_NODES; cnode=find_next_bit(&nodes_flushed, NR_NODES, ++cnode)) --- linuxbase/arch/ia64/kernel/machvec.c 2004-05-21 20:47:20.000000000 -0500 +++ linux/arch/ia64/kernel/machvec.c 2004-05-21 20:48:25.000000000 -0500 @@ -62,6 +62,12 @@ EXPORT_SYMBOL(machvec_timer_interrupt); void +machvec_tlb_migrate_finish (struct mm_struct *mm) +{ +} +EXPORT_SYMBOL(machvec_tlb_migrate_finish); + +void machvec_dma_sync_single (struct device *hwdev, dma_addr_t dma_handle, size_t size, int dir) { mb(); --- linuxbase/include/asm-generic/tlb.h 2004-05-21 20:47:28.000000000 -0500 +++ linux/include/asm-generic/tlb.h 2004-05-21 20:48:25.000000000 -0500 @@ -146,4 +146,6 @@ __pmd_free_tlb(tlb, pmdp); \ } while (0) +#define tlb_migrate_finish(mm) flush_tlb_mm(mm) + #endif /* _ASM_GENERIC__TLB_H */ --- linuxbase/include/asm-ia64/tlb.h 2004-05-21 20:47:29.000000000 -0500 +++ linux/include/asm-ia64/tlb.h 2004-05-21 20:48:25.000000000 -0500 @@ -44,6 +44,7 @@ #include #include #include +#include #ifdef CONFIG_SMP # define FREE_PTE_NR 2048 @@ -211,6 +212,8 @@ tlb->end_addr = address + PAGE_SIZE; } +#define tlb_migrate_finish(mm) platform_tlb_migrate_finish(mm) + #define tlb_start_vma(tlb, vma) do { } while (0) #define tlb_end_vma(tlb, vma) do { } while (0) --- linuxbase/include/asm-ia64/machvec.h 2004-05-21 20:47:29.000000000 -0500 +++ linux/include/asm-ia64/machvec.h 2004-05-21 20:48:25.000000000 -0500 @@ -19,6 +19,7 @@ struct scatterlist; struct irq_desc; struct page; +struct mm_struct; typedef void ia64_mv_setup_t (char **); typedef void ia64_mv_cpu_init_t (void); @@ -26,6 +27,7 @@ typedef void ia64_mv_send_ipi_t (int, int, int, int); typedef void ia64_mv_timer_interrupt_t (int, void *, struct pt_regs *); typedef void ia64_mv_global_tlb_purge_t (unsigned long, unsigned long, unsigned long); +typedef void ia64_mv_tlb_migrate_finish_t (struct mm_struct *); typedef struct irq_desc *ia64_mv_irq_desc (unsigned int); typedef u8 ia64_mv_irq_to_vector (u8); typedef unsigned int ia64_mv_local_vector_to_irq (u8 vector); @@ -72,6 +74,7 @@ extern void machvec_noop (void); extern void machvec_setup (char **); extern void machvec_timer_interrupt (int, void *, struct pt_regs *); +extern void machvec_tlb_migrate_finish (struct mm_struct *); extern void machvec_dma_sync_single (struct device *, dma_addr_t, size_t, int); extern void machvec_dma_sync_sg (struct device *, struct scatterlist *, int, int); @@ -95,6 +98,7 @@ # define platform_send_ipi ia64_mv.send_ipi # define platform_timer_interrupt ia64_mv.timer_interrupt # define platform_global_tlb_purge ia64_mv.global_tlb_purge +# define platform_tlb_migrate_finish ia64_mv.tlb_migrate_finish # define platform_dma_init ia64_mv.dma_init # define platform_dma_alloc_coherent ia64_mv.dma_alloc_coherent # define platform_dma_free_coherent ia64_mv.dma_free_coherent @@ -140,6 +144,7 @@ ia64_mv_send_ipi_t *send_ipi; ia64_mv_timer_interrupt_t *timer_interrupt; ia64_mv_global_tlb_purge_t *global_tlb_purge; + ia64_mv_tlb_migrate_finish_t *tlb_migrate_finish; ia64_mv_dma_init *dma_init; ia64_mv_dma_alloc_coherent *dma_alloc_coherent; ia64_mv_dma_free_coherent *dma_free_coherent; @@ -181,6 +186,7 @@ platform_send_ipi, \ platform_timer_interrupt, \ platform_global_tlb_purge, \ + platform_tlb_migrate_finish, \ platform_dma_init, \ platform_dma_alloc_coherent, \ platform_dma_free_coherent, \ @@ -260,6 +266,9 @@ #ifndef platform_global_tlb_purge # define platform_global_tlb_purge ia64_global_tlb_purge /* default to architected version */ #endif +#ifndef platform_tlb_migrate_finish +# define platform_tlb_migrate_finish machvec_tlb_migrate_finish +#endif #ifndef platform_dma_init # define platform_dma_init swiotlb_init #endif --- linuxbase/include/asm-ia64/machvec_sn2.h 2004-05-21 20:47:29.000000000 -0500 +++ linux/include/asm-ia64/machvec_sn2.h 2004-05-21 20:48:25.000000000 -0500 @@ -39,6 +39,7 @@ extern ia64_mv_send_ipi_t sn2_send_IPI; extern ia64_mv_timer_interrupt_t sn_timer_interrupt; extern ia64_mv_global_tlb_purge_t sn2_global_tlb_purge; +extern ia64_mv_tlb_migrate_finish_t sn_tlb_migrate_finish; extern ia64_mv_irq_desc sn_irq_desc; extern ia64_mv_irq_to_vector sn_irq_to_vector; extern ia64_mv_local_vector_to_irq sn_local_vector_to_irq; @@ -83,6 +84,7 @@ #define platform_send_ipi sn2_send_IPI #define platform_timer_interrupt sn_timer_interrupt #define platform_global_tlb_purge sn2_global_tlb_purge +#define platform_tlb_migrate_finish sn_tlb_migrate_finish #define platform_pci_fixup sn_pci_fixup #define platform_inb __sn_inb #define platform_inw __sn_inw -- Thanks Jack Steiner (steiner@sgi.com) 651-683-5302 Principal Engineer SGI - Silicon Graphics, Inc.