search for: __meminit

...- iounmap(addr); + if ((void *)round_down((unsigned long)addr, LARGE_PAGE_SIZE) != temp_mappings[0].address) + printk("early_iounmap: bad address %p\n", addr); + set_pmd(temp_mappings[0].pmd, __pmd(0)); + set_pmd(temp_mappings[1].pmd, __pmd(0)); + __flush_tlb(); } +#endif static void __meminit phys_pmd_init(pmd_t *pmd, unsigned long address, unsigned long end) @@ -648,9 +670,9 @@ void __init extend_init_mapping(unsigned } } -static void __init find_early_table_space(unsigned long end) +static unsigned long __init find_early_table_space(unsigned long end) { - unsigned long puds, pm...

...it should be done as kernel boot option (on by default to not break things using this interface now). If it be useful for maintainers of mm/memory_hotplug.c and drivers/base/memory.c code then I could do that. Currently original arch/x86/Kconfig version is restored. > >+/* we are OK calling __meminit stuff here - we have CONFIG_MEMORY_HOTPLUG > >*/ > >+static int __ref xen_add_memory(int nid, u64 start, u64 size) > > Could this be __meminit too then? Good question. I looked throught the code and could not find any simple explanation why mm/memory_hotplug.c authors used __ref...

Hi all, This is a new release of bio-cgroup which provides an IO tracking mechanism. The patches can be applied to the kernel 2.6.30-rc1 and you can also download them from the following site. http://people.valinux.co.jp/~ryov/bio-cgroup/ What's bio-cgroup all about? ============================ With this feature, you can determine the owners of any type of I/Os. This makes

Hi all, This is a new release of bio-cgroup which provides an IO tracking mechanism. The patches can be applied to the kernel 2.6.30-rc1 and you can also download them from the following site. http://people.valinux.co.jp/~ryov/bio-cgroup/ What's bio-cgroup all about? ============================ With this feature, you can determine the owners of any type of I/Os. This makes

Hi all, This is a new release of bio-cgroup which provides an IO tracking mechanism. The patches can be applied to the kernel 2.6.30-rc1 and you can also download them from the following site. http://people.valinux.co.jp/~ryov/bio-cgroup/ What's bio-cgroup all about? ============================ With this feature, you can determine the owners of any type of I/Os. This makes

Hi all, This is a new release of blkio-cgroup which provides an IO tracking mechanism. You can also download this series of patches from http://people.valinux.co.jp/~ryov/blkio-cgroup/ Changes from the previous release ================================= - bio-cgroup renamed to blkio-cgroup. - Use part of page_cgroup->flags to store the blkio-cgroup ID. This code is taken from Andrea's

Hi all, This is a new release of blkio-cgroup which provides an IO tracking mechanism. You can also download this series of patches from http://people.valinux.co.jp/~ryov/blkio-cgroup/ Changes from the previous release ================================= - bio-cgroup renamed to blkio-cgroup. - Use part of page_cgroup->flags to store the blkio-cgroup ID. This code is taken from Andrea's

Hi all, This is a new release of blkio-cgroup which provides an IO tracking mechanism. You can also download this series of patches from http://people.valinux.co.jp/~ryov/blkio-cgroup/ Changes from the previous release ================================= - bio-cgroup renamed to blkio-cgroup. - Use part of page_cgroup->flags to store the blkio-cgroup ID. This code is taken from Andrea's

Hi all, This is a new release of blkio-cgroup v11. Changes: - The function cgroup_get_from_page() is added which determines to which cgroup a given page belongs. This function is introduced from Vivek's io-controller patch. Thanks Vivek. - Fix a type mismatch of the refcount of io_context. The refcount has been changed to atomic_long_t. This patch can be applied to 2.6.31-rc7. The

Hi all, This is a new release of blkio-cgroup v11. Changes: - The function cgroup_get_from_page() is added which determines to which cgroup a given page belongs. This function is introduced from Vivek's io-controller patch. Thanks Vivek. - Fix a type mismatch of the refcount of io_context. The refcount has been changed to atomic_long_t. This patch can be applied to 2.6.31-rc7. The

Hi all, This is a new release of blkio-cgroup v11. Changes: - The function cgroup_get_from_page() is added which determines to which cgroup a given page belongs. This function is introduced from Vivek's io-controller patch. Thanks Vivek. - Fix a type mismatch of the refcount of io_context. The refcount has been changed to atomic_long_t. This patch can be applied to 2.6.31-rc7. The

Hi all, This is a new release of blkio-cgroup v10. This release reduces IO tracking overhead and fixes an issue that could cause a deadlock since lock_page_cgroup() is no longer used. Thank you KAMEZAWA-san for your suggestions and pointing out the issue. This patch can be applied to 2.6.31-rc3-mmotm0716 and 2.6.31-rc4. The list of the patches: [PATCH 1/7] blkio-cgroup-v10: Introduction

Hi all, This is a new release of blkio-cgroup v10. This release reduces IO tracking overhead and fixes an issue that could cause a deadlock since lock_page_cgroup() is no longer used. Thank you KAMEZAWA-san for your suggestions and pointing out the issue. This patch can be applied to 2.6.31-rc3-mmotm0716 and 2.6.31-rc4. The list of the patches: [PATCH 1/7] blkio-cgroup-v10: Introduction

Hi all, This is a new release of blkio-cgroup v10. This release reduces IO tracking overhead and fixes an issue that could cause a deadlock since lock_page_cgroup() is no longer used. Thank you KAMEZAWA-san for your suggestions and pointing out the issue. This patch can be applied to 2.6.31-rc3-mmotm0716 and 2.6.31-rc4. The list of the patches: [PATCH 1/7] blkio-cgroup-v10: Introduction

These patches are not ready to be merged because I was unable to measure a performance improvement. I'm publishing them so they are archived in case someone picks up this work again in the future. The goal of these patches is to allocate virtqueues and driver state from the device's NUMA node for optimal memory access latency. Only guests with a vNUMA topology and virtio devices spread

These patches are not ready to be merged because I was unable to measure a performance improvement. I'm publishing them so they are archived in case someone picks up this work again in the future. The goal of these patches is to allocate virtqueues and driver state from the device's NUMA node for optimal memory access latency. Only guests with a vNUMA topology and virtio devices spread

...0)) #define BALLOON_CLASS_NAME "xen_memory" @@ -199,6 +197,196 @@ static inline unsigned long current_target(void) { return balloon_stats.target_pages; } + +static inline u64 is_memory_resource_reserved(void) +{ + return balloon_stats.hotplug_start_paddr; +} + +/* we are OK calling __meminit stuff here - we have CONFIG_MEMORY_HOTPLUG */ +static int __ref xen_add_memory(int nid, u64 start, u64 size) +{ + pg_data_t *pgdat = NULL; + int new_pgdat = 0, ret; + + lock_system_sleep(); + + if (!node_online(nid)) { + pgdat = hotadd_new_pgdat(nid, start); + ret = -ENOMEM; + if (!pgdat) + go...

..."=r" (address)); + address = read_cr2(); info.si_code = SEGV_MAPERR; Index: clean-start/arch/x86_64/mm/init.c =================================================================== --- clean-start.orig/arch/x86_64/mm/init.c +++ clean-start/arch/x86_64/mm/init.c @@ -384,7 +384,7 @@ void __meminit init_memory_mapping(unsig } if (!after_bootmem) - asm volatile("movq %%cr4,%0" : "=r" (mmu_cr4_features)); + mmu_cr4_features = read_cr4(); __flush_tlb_all(); } @@ -398,7 +398,7 @@ void __cpuinit zap_low_mappings(int cpu) * For AP's, zap the low identity ma...

..."=r" (address)); + address = read_cr2(); info.si_code = SEGV_MAPERR; Index: clean-start/arch/x86_64/mm/init.c =================================================================== --- clean-start.orig/arch/x86_64/mm/init.c +++ clean-start/arch/x86_64/mm/init.c @@ -384,7 +384,7 @@ void __meminit init_memory_mapping(unsig } if (!after_bootmem) - asm volatile("movq %%cr4,%0" : "=r" (mmu_cr4_features)); + mmu_cr4_features = read_cr4(); __flush_tlb_all(); } @@ -398,7 +398,7 @@ void __cpuinit zap_low_mappings(int cpu) * For AP's, zap the low identity ma...

Add a new field, VIRTIO_BALLOON_S_AVAIL, to virtio_balloon memory statistics protocol, corresponding to 'Available' in /proc/meminfo. It indicates to the hypervisor how big the balloon can be inflated without pushing the guest system to swap. This metric would be very useful in VM orchestration software to improve memory management of different VMs under overcommit. Signed-off-by: Igor