Sign Up
Log In
Log In
or
Sign Up
Places
All Projects
Status Monitor
Collapse sidebar
SUSE:SLE-15-SP3:Update
xen.22515
libxc.sr.superpage.patch
Overview
Repositories
Revisions
Requests
Users
Attributes
Meta
File libxc.sr.superpage.patch of Package xen.22515
tools/libxc: use superpages during restore of HVM guest bsc#1035231 - migration of HVM domU does not use superpages on destination dom0 bsc#1055695 - XEN: 11SP4 and 12SP3 HVM guests can not be restored During creating of a HVM domU meminit_hvm() tries to map superpages. After save/restore or migration this mapping is lost, everything is allocated in single pages. This causes a performance degradition after migration. Add neccessary code to preallocate a superpage for the chunk of pfns that is received. In case a pfn was not populated on the sending side it must be freed on the receiving side to avoid over-allocation. The existing code for x86_pv is moved unmodified into its own file. --- a/tools/libxc/xc_dom_x86.c +++ b/tools/libxc/xc_dom_x86.c @@ -45,11 +45,6 @@ #define SUPERPAGE_BATCH_SIZE 512 -#define SUPERPAGE_2MB_SHIFT 9 -#define SUPERPAGE_2MB_NR_PFNS (1UL << SUPERPAGE_2MB_SHIFT) -#define SUPERPAGE_1GB_SHIFT 18 -#define SUPERPAGE_1GB_NR_PFNS (1UL << SUPERPAGE_1GB_SHIFT) - #define X86_CR0_PE 0x01 #define X86_CR0_ET 0x10 --- a/tools/libxc/xc_private.h +++ b/tools/libxc/xc_private.h @@ -71,6 +71,11 @@ struct iovec { #define DECLARE_FLASK_OP struct xen_flask_op op #define DECLARE_PLATFORM_OP struct xen_platform_op platform_op +#define SUPERPAGE_2MB_SHIFT 9 +#define SUPERPAGE_2MB_NR_PFNS (1UL << SUPERPAGE_2MB_SHIFT) +#define SUPERPAGE_1GB_SHIFT 18 +#define SUPERPAGE_1GB_NR_PFNS (1UL << SUPERPAGE_1GB_SHIFT) + #undef PAGE_SHIFT #undef PAGE_SIZE #undef PAGE_MASK --- a/tools/libxc/xc_sr_common.c +++ b/tools/libxc/xc_sr_common.c @@ -164,6 +164,47 @@ static void __attribute__((unused)) buil } /* + * Expand the tracking structures as needed. + * To avoid realloc()ing too excessively, the size increased to the nearest power + * of two large enough to contain the required number of bits. + */ +bool _xc_sr_bitmap_resize(struct xc_sr_bitmap *bm, unsigned long bits) +{ + if (bits > bm->bits) + { + size_t new_max; + size_t old_sz, new_sz; + void *p; + + /* Round up to the nearest power of two larger than bit, less 1. */ + new_max = bits; + new_max |= new_max >> 1; + new_max |= new_max >> 2; + new_max |= new_max >> 4; + new_max |= new_max >> 8; + new_max |= new_max >> 16; +#ifdef __x86_64__ + new_max |= new_max >> 32; +#endif + + old_sz = bitmap_size(bm->bits + 1); + new_sz = bitmap_size(new_max + 1); + p = realloc(bm->p, new_sz); + if (!p) + return false; + + if (bm->p) + memset(p + old_sz, 0, new_sz - old_sz); + else + memset(p, 0, new_sz); + + bm->p = p; + bm->bits = new_max; + } + return true; +} + +/* * Local variables: * mode: C * c-file-style: "BSD" --- a/tools/libxc/xc_sr_common.h +++ b/tools/libxc/xc_sr_common.h @@ -19,6 +19,12 @@ const char *rec_type_to_str(uint32_t typ struct xc_sr_context; struct xc_sr_record; +struct xc_sr_bitmap +{ + void *p; + unsigned long bits; +}; + /** * Save operations. To be implemented for each type of guest, for use by the * common save algorithm. @@ -142,6 +148,16 @@ struct xc_sr_restore_ops int (*setup)(struct xc_sr_context *ctx); /** + * Populate PFNs + * + * Given a set of pfns, obtain memory from Xen to fill the physmap for the + * unpopulated subset. + */ + int (*populate_pfns)(struct xc_sr_context *ctx, unsigned count, + const xen_pfn_t *original_pfns, const uint32_t *types); + + + /** * Process an individual record from the stream. The caller shall take * care of processing common records (e.g. END, PAGE_DATA). * @@ -288,6 +304,8 @@ struct xc_sr_context int send_back_fd; unsigned long p2m_size; + unsigned long max_pages; + unsigned long tot_pages; xc_hypercall_buffer_t dirty_bitmap_hbuf; /* From Image Header. */ @@ -325,8 +343,7 @@ struct xc_sr_context uint32_t xenstore_domid, console_domid; /* Bitmap of currently populated PFNs during restore. */ - unsigned long *populated_pfns; - xen_pfn_t max_populated_pfn; + struct xc_sr_bitmap populated_pfns; /* Sender has invoked verify mode on the stream. */ bool verify; @@ -426,6 +443,14 @@ struct xc_sr_context { /* HVM context blob. */ struct xc_sr_blob context; + + /* Bitmap of currently allocated PFNs during restore. */ + struct xc_sr_bitmap attempted_1g; + struct xc_sr_bitmap attempted_2m; + struct xc_sr_bitmap allocated_pfns; + xen_pfn_t prev_populated_pfn; + xen_pfn_t iteration_tracker_pfn; + unsigned long iteration; } restore; }; } hvm; @@ -440,6 +465,78 @@ extern struct xc_sr_save_ops save_ops_x8 extern struct xc_sr_restore_ops restore_ops_x86_pv; extern struct xc_sr_restore_ops restore_ops_x86_hvm; +extern bool _xc_sr_bitmap_resize(struct xc_sr_bitmap *bm, unsigned long bits); + +static inline bool xc_sr_bitmap_resize(struct xc_sr_bitmap *bm, unsigned long bits) +{ + if (bits > bm->bits) + return _xc_sr_bitmap_resize(bm, bits); + return true; +} + +static inline void xc_sr_bitmap_free(struct xc_sr_bitmap *bm) +{ + free(bm->p); + bm->p = NULL; +} + +static inline bool xc_sr_set_bit(unsigned long bit, struct xc_sr_bitmap *bm) +{ + if (!xc_sr_bitmap_resize(bm, bit)) + return false; + + set_bit(bit, bm->p); + return true; +} + +static inline bool xc_sr_set_long_bit(unsigned long base_bit, struct xc_sr_bitmap *bm) +{ + if (!xc_sr_bitmap_resize(bm, base_bit + BITS_PER_LONG)) + return false; + + set_bit_long(base_bit, bm->p); + return true; +} + +static inline bool xc_sr_test_bit(unsigned long bit, struct xc_sr_bitmap *bm) +{ + if (bit > bm->bits) + return false; + return !!test_bit(bit, bm->p); +} + +static inline bool xc_sr_test_and_clear_bit(unsigned long bit, struct xc_sr_bitmap *bm) +{ + if (bit > bm->bits) + return false; + return !!test_and_clear_bit(bit, bm->p); +} + +static inline bool xc_sr_test_and_set_bit(unsigned long bit, struct xc_sr_bitmap *bm) +{ + if (bit > bm->bits) + return false; + return !!test_and_set_bit(bit, bm->p); +} + +static inline bool pfn_is_populated(struct xc_sr_context *ctx, xen_pfn_t pfn) +{ + return xc_sr_test_bit(pfn, &ctx->restore.populated_pfns); +} + +static inline int pfn_set_populated(struct xc_sr_context *ctx, xen_pfn_t pfn) +{ + xc_interface *xch = ctx->xch; + + if ( !xc_sr_set_bit(pfn, &ctx->restore.populated_pfns) ) + { + ERROR("Failed to realloc populated_pfns bitmap"); + errno = ENOMEM; + return -1; + } + return 0; +} + struct xc_sr_record { uint32_t type; @@ -490,14 +587,6 @@ int read_record_header(struct xc_sr_cont int read_record_data(struct xc_sr_context *ctx, int fd, struct xc_sr_rhdr *rhdr, struct xc_sr_record *rec); -/* - * This would ideally be private in restore.c, but is needed by - * x86_pv_localise_page() if we receive pagetables frames ahead of the - * contents of the frames they point at. - */ -int populate_pfns(struct xc_sr_context *ctx, unsigned int count, - const xen_pfn_t *original_pfns, const uint32_t *types); - /* Handle a STATIC_DATA_END record. */ int handle_static_data_end(struct xc_sr_context *ctx); --- a/tools/libxc/xc_sr_restore.c +++ b/tools/libxc/xc_sr_restore.c @@ -71,121 +71,6 @@ static int read_headers(struct xc_sr_con return 0; } -/* - * Is a pfn populated? - */ -static bool pfn_is_populated(const struct xc_sr_context *ctx, xen_pfn_t pfn) -{ - if ( pfn > ctx->restore.max_populated_pfn ) - return false; - return test_bit(pfn, ctx->restore.populated_pfns); -} - -/* - * Set a pfn as populated, expanding the tracking structures if needed. To - * avoid realloc()ing too excessively, the size increased to the nearest power - * of two large enough to contain the required pfn. - */ -static int pfn_set_populated(struct xc_sr_context *ctx, xen_pfn_t pfn) -{ - xc_interface *xch = ctx->xch; - - if ( pfn > ctx->restore.max_populated_pfn ) - { - xen_pfn_t new_max; - size_t old_sz, new_sz; - unsigned long *p; - - /* Round up to the nearest power of two larger than pfn, less 1. */ - new_max = pfn; - new_max |= new_max >> 1; - new_max |= new_max >> 2; - new_max |= new_max >> 4; - new_max |= new_max >> 8; - new_max |= new_max >> 16; -#ifdef __x86_64__ - new_max |= new_max >> 32; -#endif - - old_sz = bitmap_size(ctx->restore.max_populated_pfn + 1); - new_sz = bitmap_size(new_max + 1); - p = realloc(ctx->restore.populated_pfns, new_sz); - if ( !p ) - { - ERROR("Failed to realloc populated bitmap"); - errno = ENOMEM; - return -1; - } - - memset((uint8_t *)p + old_sz, 0x00, new_sz - old_sz); - - ctx->restore.populated_pfns = p; - ctx->restore.max_populated_pfn = new_max; - } - - assert(!test_bit(pfn, ctx->restore.populated_pfns)); - set_bit(pfn, ctx->restore.populated_pfns); - - return 0; -} - -/* - * Given a set of pfns, obtain memory from Xen to fill the physmap for the - * unpopulated subset. If types is NULL, no page type checking is performed - * and all unpopulated pfns are populated. - */ -int populate_pfns(struct xc_sr_context *ctx, unsigned int count, - const xen_pfn_t *original_pfns, const uint32_t *types) -{ - xc_interface *xch = ctx->xch; - xen_pfn_t *mfns = ctx->restore.m->pp_mfns, - *pfns = ctx->restore.m->pp_pfns; - unsigned int i, nr_pfns = 0; - int rc = -1; - - for ( i = 0; i < count; ++i ) - { - if ( (!types || - (types && page_type_has_stream_data(types[i]) == true)) && - !pfn_is_populated(ctx, original_pfns[i]) ) - { - rc = pfn_set_populated(ctx, original_pfns[i]); - if ( rc ) - goto err; - pfns[nr_pfns] = mfns[nr_pfns] = original_pfns[i]; - ++nr_pfns; - } - } - - if ( nr_pfns ) - { - rc = xc_domain_populate_physmap_exact( - xch, ctx->domid, nr_pfns, 0, 0, mfns); - if ( rc ) - { - PERROR("Failed to populate physmap"); - goto err; - } - - for ( i = 0; i < nr_pfns; ++i ) - { - if ( mfns[i] == INVALID_MFN ) - { - ERROR("Populate physmap failed for pfn %u", i); - rc = -1; - goto err; - } - - ctx->restore.ops.set_gfn(ctx, pfns[i], mfns[i]); - } - } - - rc = 0; - - err: - return rc; -} - static int handle_static_data_end_v2(struct xc_sr_context *ctx) { int rc = 0; @@ -328,7 +213,7 @@ static int map_guest_pages(struct xc_sr_ uint32_t i, p; int rc; - rc = populate_pfns(ctx, pages->count, m->pfns, m->types); + rc = ctx->restore.ops.populate_pfns(ctx, pages->count, m->pfns, m->types); if ( rc ) { ERROR("Failed to populate pfns for batch of %u pages", pages->count); @@ -929,10 +814,8 @@ static int setup(struct xc_sr_context *c if ( rc ) goto err; - ctx->restore.max_populated_pfn = (32 * 1024 / 4) - 1; - ctx->restore.populated_pfns = bitmap_alloc( - ctx->restore.max_populated_pfn + 1); - if ( !ctx->restore.populated_pfns ) + rc = !xc_sr_bitmap_resize(&ctx->restore.populated_pfns, 32 * 1024 / 4); + if ( rc ) { ERROR("Unable to allocate memory for populated_pfns bitmap"); rc = -1; @@ -977,7 +860,7 @@ static void cleanup(struct xc_sr_context free(ctx->restore.m); free(ctx->restore.buffered_records); - free(ctx->restore.populated_pfns); + xc_sr_bitmap_free(&ctx->restore.populated_pfns); if ( ctx->restore.ops.cleanup(ctx) ) PERROR("Failed to clean up"); @@ -1146,7 +1029,12 @@ int xc_domain_restore(xc_interface *xch, return -1; } + /* See xc_domain_getinfo */ + ctx.restore.max_pages = ctx.dominfo.max_memkb >> (PAGE_SHIFT-10); + ctx.restore.tot_pages = ctx.dominfo.nr_pages; ctx.restore.p2m_size = nr_pfns; + DPRINTF("dom %u p2m_size %lx max_pages %lx", + ctx.domid, ctx.restore.p2m_size, ctx.restore.max_pages); ctx.restore.ops = ctx.dominfo.hvm ? restore_ops_x86_hvm : restore_ops_x86_pv; --- a/tools/libxc/xc_sr_restore_x86_hvm.c +++ b/tools/libxc/xc_sr_restore_x86_hvm.c @@ -136,6 +136,8 @@ static int x86_hvm_localise_page(struct static int x86_hvm_setup(struct xc_sr_context *ctx) { xc_interface *xch = ctx->xch; + struct xc_sr_bitmap *bm; + unsigned long bits; if ( ctx->restore.guest_type != DHDR_TYPE_X86_HVM ) { @@ -161,7 +163,41 @@ static int x86_hvm_setup(struct xc_sr_co } #endif + bm = &ctx->x86.hvm.restore.attempted_1g; + bits = (ctx->restore.p2m_size >> SUPERPAGE_1GB_SHIFT) + 1; + if ( xc_sr_bitmap_resize(bm, bits) == false ) + goto out; + + bm = &ctx->x86.hvm.restore.attempted_2m; + bits = (ctx->restore.p2m_size >> SUPERPAGE_2MB_SHIFT) + 1; + if ( xc_sr_bitmap_resize(bm, bits) == false ) + goto out; + + bm = &ctx->x86.hvm.restore.allocated_pfns; + bits = ctx->restore.p2m_size + 1; + if ( xc_sr_bitmap_resize(bm, bits) == false ) + goto out; + + /* No superpage in 1st 2MB due to VGA hole */ +#define LAPIC_BASE_ADDRESS 0xfee00000u +#define ACPI_INFO_PHYSICAL_ADDRESS 0xfc000000u +#define LAPIC_BASE_PFN (LAPIC_BASE_ADDRESS >> XC_PAGE_SHIFT) +#define ACPI_INFO_PFN (ACPI_INFO_PHYSICAL_ADDRESS >> XC_PAGE_SHIFT) + bm = &ctx->x86.hvm.restore.attempted_1g; + xc_sr_set_bit(0, bm); + xc_sr_set_bit(LAPIC_BASE_PFN >> SUPERPAGE_1GB_SHIFT, bm); + xc_sr_set_bit(ACPI_INFO_PFN >> SUPERPAGE_1GB_SHIFT, bm); + + bm = &ctx->x86.hvm.restore.attempted_2m; + xc_sr_set_bit(0, bm); + xc_sr_set_bit(LAPIC_BASE_PFN >> SUPERPAGE_2MB_SHIFT, bm); + xc_sr_set_bit(ACPI_INFO_PFN >> SUPERPAGE_2MB_SHIFT, bm); + return 0; + +out: + ERROR("Unable to allocate memory for pfn bitmaps"); + return -1; } /* @@ -242,6 +278,9 @@ static int x86_hvm_stream_complete(struc static int x86_hvm_cleanup(struct xc_sr_context *ctx) { free(ctx->x86.hvm.restore.context.ptr); + xc_sr_bitmap_free(&ctx->x86.hvm.restore.attempted_1g); + xc_sr_bitmap_free(&ctx->x86.hvm.restore.attempted_2m); + xc_sr_bitmap_free(&ctx->x86.hvm.restore.allocated_pfns); free(ctx->x86.restore.cpuid.ptr); free(ctx->x86.restore.msr.ptr); @@ -249,6 +288,380 @@ static int x86_hvm_cleanup(struct xc_sr_ return 0; } +/* + * Set a pfn as allocated, expanding the tracking structures if needed. + */ +static int pfn_set_allocated(struct xc_sr_context *ctx, xen_pfn_t pfn) +{ + xc_interface *xch = ctx->xch; + + if ( !xc_sr_set_bit(pfn, &ctx->x86.hvm.restore.allocated_pfns) ) + { + ERROR("Failed to realloc allocated_pfns bitmap"); + errno = ENOMEM; + return -1; + } + return 0; +} + +/* + * Set a range of pfns as allocated, expanding the tracking structures if needed. + */ +static int pfn_set_long_allocated(struct xc_sr_context *ctx, xen_pfn_t base_pfn) +{ + xc_interface *xch = ctx->xch; + + if ( !xc_sr_set_long_bit(base_pfn, &ctx->x86.hvm.restore.allocated_pfns) ) + { + ERROR("Failed to realloc allocated_pfns bitmap"); + errno = ENOMEM; + return -1; + } + return 0; +} + +struct x86_hvm_sp { + xen_pfn_t pfn; + xen_pfn_t base_pfn; + unsigned long index; + unsigned long count; +}; + +/* + * Try to allocate a 1GB page for this pfn, but avoid Over-allocation. + * If this succeeds, mark the range of 2MB pages as busy. + */ +static bool x86_hvm_alloc_1g(struct xc_sr_context *ctx, struct x86_hvm_sp *sp) +{ + xc_interface *xch = ctx->xch; + struct xc_sr_bitmap *bm; + unsigned int order; + int i, done; + xen_pfn_t extent; + + bm = &ctx->x86.hvm.restore.attempted_1g; + + /* Only one attempt to avoid overlapping allocation */ + if ( xc_sr_test_and_set_bit(sp->index, bm) ) + return false; + + order = SUPERPAGE_1GB_SHIFT; + sp->count = SUPERPAGE_1GB_NR_PFNS; + + /* Allocate only if there is room for another superpage */ + if ( ctx->restore.tot_pages + sp->count > ctx->restore.max_pages ) + return false; + + extent = sp->base_pfn = (sp->pfn >> order) << order; + done = xc_domain_populate_physmap(xch, ctx->domid, 1, order, 0, &extent); + if ( done < 0 ) { + PERROR("populate_physmap failed."); + return false; + } + if ( done == 0 ) + return false; + + DPRINTF("1G %" PRI_xen_pfn "\n", sp->base_pfn); + + /* Mark all 2MB pages as done to avoid overlapping allocation */ + bm = &ctx->x86.hvm.restore.attempted_2m; + for ( i = 0; i < (SUPERPAGE_1GB_NR_PFNS/SUPERPAGE_2MB_NR_PFNS); i++ ) + xc_sr_set_bit((sp->base_pfn >> SUPERPAGE_2MB_SHIFT) + i, bm); + + return true; +} + +/* Allocate a 2MB page if x86_hvm_alloc_1g failed, avoid Over-allocation. */ +static bool x86_hvm_alloc_2m(struct xc_sr_context *ctx, struct x86_hvm_sp *sp) +{ + xc_interface *xch = ctx->xch; + struct xc_sr_bitmap *bm; + unsigned int order; + int done; + xen_pfn_t extent; + + bm = &ctx->x86.hvm.restore.attempted_2m; + + /* Only one attempt to avoid overlapping allocation */ + if ( xc_sr_test_and_set_bit(sp->index, bm) ) + return false; + + order = SUPERPAGE_2MB_SHIFT; + sp->count = SUPERPAGE_2MB_NR_PFNS; + + /* Allocate only if there is room for another superpage */ + if ( ctx->restore.tot_pages + sp->count > ctx->restore.max_pages ) + return false; + + extent = sp->base_pfn = (sp->pfn >> order) << order; + done = xc_domain_populate_physmap(xch, ctx->domid, 1, order, 0, &extent); + if ( done < 0 ) { + PERROR("populate_physmap failed."); + return false; + } + if ( done == 0 ) + return false; + + DPRINTF("2M %" PRI_xen_pfn "\n", sp->base_pfn); + return true; +} + +/* Allocate a single page if x86_hvm_alloc_2m failed. */ +static bool x86_hvm_alloc_4k(struct xc_sr_context *ctx, struct x86_hvm_sp *sp) +{ + xc_interface *xch = ctx->xch; + unsigned int order; + int done; + xen_pfn_t extent; + + order = 0; + sp->count = 1UL; + + /* Allocate only if there is room for another page */ + if ( ctx->restore.tot_pages + sp->count > ctx->restore.max_pages ) { + errno = E2BIG; + return false; + } + + extent = sp->base_pfn = (sp->pfn >> order) << order; + done = xc_domain_populate_physmap(xch, ctx->domid, 1, order, 0, &extent); + if ( done < 0 ) { + PERROR("populate_physmap failed."); + return false; + } + if ( done == 0 ) { + errno = ENOMEM; + return false; + } + + DPRINTF("4K %" PRI_xen_pfn "\n", sp->base_pfn); + return true; +} +/* + * Attempt to allocate a superpage where the pfn resides. + */ +static int x86_hvm_allocate_pfn(struct xc_sr_context *ctx, xen_pfn_t pfn) +{ + xc_interface *xch = ctx->xch; + bool success; + int rc = -1; + unsigned long idx_1g, idx_2m; + struct x86_hvm_sp sp = { + .pfn = pfn + }; + + if ( xc_sr_test_bit(pfn, &ctx->x86.hvm.restore.allocated_pfns) ) + return 0; + + idx_1g = pfn >> SUPERPAGE_1GB_SHIFT; + idx_2m = pfn >> SUPERPAGE_2MB_SHIFT; + if ( !xc_sr_bitmap_resize(&ctx->x86.hvm.restore.attempted_1g, idx_1g) ) + { + PERROR("Failed to realloc attempted_1g"); + return -1; + } + if ( !xc_sr_bitmap_resize(&ctx->x86.hvm.restore.attempted_2m, idx_2m) ) + { + PERROR("Failed to realloc attempted_2m"); + return -1; + } + + sp.index = idx_1g; + success = x86_hvm_alloc_1g(ctx, &sp); + + if ( success == false ) { + sp.index = idx_2m; + success = x86_hvm_alloc_2m(ctx, &sp); + } + + if ( success == false ) { + sp.index = 0; + success = x86_hvm_alloc_4k(ctx, &sp); + } + + if ( success == false ) + return -1; + + do { + if ( sp.count >= BITS_PER_LONG ) { + sp.count -= BITS_PER_LONG; + ctx->restore.tot_pages += BITS_PER_LONG; + rc = pfn_set_long_allocated(ctx, sp.base_pfn + sp.count); + } else { + sp.count--; + ctx->restore.tot_pages++; + rc = pfn_set_allocated(ctx, sp.base_pfn + sp.count); + } + if ( rc ) + break; + } while ( sp.count ); + + return rc; +} + +/* + * Deallocate memory. + * There was likely an optimistic superpage allocation. + * This means more pages may have been allocated past gap_end. + * This range is not freed now. Incoming higher pfns will release it. + */ +static int x86_hvm_punch_hole(struct xc_sr_context *ctx, + xen_pfn_t gap_start, xen_pfn_t gap_end) +{ + xc_interface *xch = ctx->xch; + struct xc_sr_bitmap *bm; + xen_pfn_t _pfn, pfn; + uint32_t domid, freed = 0; + int rc; + + bm = &ctx->x86.hvm.restore.allocated_pfns; + if ( !xc_sr_bitmap_resize(bm, gap_end) ) + { + PERROR("Failed to realloc allocated_pfns %" PRI_xen_pfn, gap_end); + return -1; + } + + pfn = gap_start >> SUPERPAGE_1GB_SHIFT; + do + { + xc_sr_set_bit(pfn, &ctx->x86.hvm.restore.attempted_1g); + } while (++pfn <= gap_end >> SUPERPAGE_1GB_SHIFT); + + pfn = gap_start >> SUPERPAGE_2MB_SHIFT; + do + { + xc_sr_set_bit(pfn, &ctx->x86.hvm.restore.attempted_2m); + } while (++pfn <= gap_end >> SUPERPAGE_2MB_SHIFT); + + pfn = gap_start; + + while ( pfn <= gap_end ) + { + if ( xc_sr_test_and_clear_bit(pfn, bm) ) + { + domid = ctx->domid; + _pfn = pfn; + rc = xc_domain_decrease_reservation_exact(xch, domid, 1, 0, &_pfn); + if ( rc ) + { + PERROR("Failed to release pfn %" PRI_xen_pfn, pfn); + return -1; + } + ctx->restore.tot_pages--; + freed++; + } + pfn++; + } + if ( freed ) + DPRINTF("freed %u between %" PRI_xen_pfn " %" PRI_xen_pfn "\n", + freed, gap_start, gap_end); + return 0; +} + +static int x86_hvm_unpopulate_page(struct xc_sr_context *ctx, xen_pfn_t pfn) +{ + xc_sr_test_and_clear_bit(pfn, &ctx->restore.populated_pfns); + return x86_hvm_punch_hole(ctx, pfn, pfn); +} + +static int x86_hvm_populate_page(struct xc_sr_context *ctx, xen_pfn_t pfn) +{ + xen_pfn_t gap_start, gap_end; + bool has_gap, first_iteration; + int rc; + + /* + * Check for a gap between the previous populated pfn and this pfn. + * In case a gap exists, it is required to punch a hole to release memory, + * starting after the previous pfn and before this pfn. + * + * But: this can be done only during the first iteration, which is the + * only place there superpage allocations are attempted. All following + * iterations lack the info to properly maintain prev_populated_pfn. + */ + has_gap = ctx->x86.hvm.restore.prev_populated_pfn + 1 < pfn; + first_iteration = ctx->x86.hvm.restore.iteration == 0; + if ( has_gap && first_iteration ) + { + gap_start = ctx->x86.hvm.restore.prev_populated_pfn + 1; + gap_end = pfn - 1; + + rc = x86_hvm_punch_hole(ctx, gap_start, gap_end); + if ( rc ) + goto err; + } + + rc = x86_hvm_allocate_pfn(ctx, pfn); + if ( rc ) + goto err; + rc = pfn_set_populated(ctx, pfn); + if ( rc ) + goto err; + ctx->x86.hvm.restore.prev_populated_pfn = pfn; + + rc = 0; +err: + return rc; +} + +/* + * Try to allocate superpages. + * This works without memory map because the pfns arrive in incremental order. + * All pfn numbers and their type are submitted. + * Only pfns with data will have also pfn content transmitted. + */ +static int x86_hvm_populate_pfns(struct xc_sr_context *ctx, unsigned count, + const xen_pfn_t *original_pfns, + const uint32_t *types) +{ + xc_interface *xch = ctx->xch; + xen_pfn_t pfn, min_pfn, max_pfn; + bool has_data, populated; + unsigned i = count; + int rc = 0; + + min_pfn = count ? original_pfns[0] : 0; + max_pfn = count ? original_pfns[count - 1] : 0; + DPRINTF("batch of %u pfns between %" PRI_xen_pfn " %" PRI_xen_pfn "\n", + count, min_pfn, max_pfn); + + /* + * There is no indicator for a new iteration. + * Simulate it by checking if a lower pfn is coming in. + * In the end it matters only to know if this iteration is the first one. + */ + if ( min_pfn < ctx->x86.hvm.restore.iteration_tracker_pfn ) + ctx->x86.hvm.restore.iteration++; + ctx->x86.hvm.restore.iteration_tracker_pfn = min_pfn; + + for ( i = 0; i < count; ++i ) + { + pfn = original_pfns[i]; + + has_data = page_type_has_stream_data(types[i]); + populated = pfn_is_populated(ctx, pfn); + + /* + * page has data, pfn populated: nothing to do + * page has data, pfn not populated: likely never seen before + * page has no data, pfn populated: likely ballooned out during migration + * page has no data, pfn not populated: nothing to do + */ + if ( has_data && !populated ) + { + rc = x86_hvm_populate_page(ctx, pfn); + } else if ( !has_data && populated ) + { + rc = x86_hvm_unpopulate_page(ctx, pfn); + } + if ( rc ) + break; + } + + return rc; +} + + struct xc_sr_restore_ops restore_ops_x86_hvm = { .pfn_is_valid = x86_hvm_pfn_is_valid, @@ -257,6 +670,7 @@ struct xc_sr_restore_ops restore_ops_x86 .set_page_type = x86_hvm_set_page_type, .localise_page = x86_hvm_localise_page, .setup = x86_hvm_setup, + .populate_pfns = x86_hvm_populate_pfns, .process_record = x86_hvm_process_record, .static_data_complete = x86_static_data_complete, .stream_complete = x86_hvm_stream_complete, --- a/tools/libxc/xc_sr_restore_x86_pv.c +++ b/tools/libxc/xc_sr_restore_x86_pv.c @@ -960,6 +960,64 @@ static void x86_pv_set_gfn(struct xc_sr_ } /* + * Given a set of pfns, obtain memory from Xen to fill the physmap for the + * unpopulated subset. If types is NULL, no page type checking is performed + * and all unpopulated pfns are populated. + */ +static int x86_pv_populate_pfns(struct xc_sr_context *ctx, unsigned count, + const xen_pfn_t *original_pfns, + const uint32_t *types) +{ + xc_interface *xch = ctx->xch; + xen_pfn_t *mfns = ctx->restore.m->pp_mfns, + *pfns = ctx->restore.m->pp_pfns; + unsigned i, nr_pfns = 0; + int rc = -1; + + for ( i = 0; i < count; ++i ) + { + if ( (!types || + (types && page_type_has_stream_data(types[i]) == true)) && + !pfn_is_populated(ctx, original_pfns[i]) ) + { + rc = pfn_set_populated(ctx, original_pfns[i]); + if ( rc ) + goto err; + pfns[nr_pfns] = mfns[nr_pfns] = original_pfns[i]; + ++nr_pfns; + } + } + + if ( nr_pfns ) + { + rc = xc_domain_populate_physmap_exact( + xch, ctx->domid, nr_pfns, 0, 0, mfns); + if ( rc ) + { + PERROR("Failed to populate physmap"); + goto err; + } + + for ( i = 0; i < nr_pfns; ++i ) + { + if ( mfns[i] == INVALID_MFN ) + { + ERROR("Populate physmap failed for pfn %u", i); + rc = -1; + goto err; + } + + ctx->restore.ops.set_gfn(ctx, pfns[i], mfns[i]); + } + } + + rc = 0; + + err: + return rc; +} + +/* * restore_ops function. Convert pfns back to mfns in pagetables. Possibly * needs to populate new frames if a PTE is found referring to a frame which * hasn't yet been seen from PAGE_DATA records. @@ -1003,7 +1061,7 @@ static int x86_pv_localise_page(struct x } } - if ( to_populate && populate_pfns(ctx, to_populate, pfns, NULL) ) + if ( to_populate && x86_pv_populate_pfns(ctx, to_populate, pfns, NULL) ) return -1; for ( i = 0; i < (PAGE_SIZE / sizeof(uint64_t)); ++i ) @@ -1193,6 +1251,7 @@ struct xc_sr_restore_ops restore_ops_x86 .set_gfn = x86_pv_set_gfn, .localise_page = x86_pv_localise_page, .setup = x86_pv_setup, + .populate_pfns = x86_pv_populate_pfns, .process_record = x86_pv_process_record, .static_data_complete = x86_static_data_complete, .stream_complete = x86_pv_stream_complete,
Locations
Projects
Search
Status Monitor
Help
OpenBuildService.org
Documentation
API Documentation
Code of Conduct
Contact
Support
@OBShq
Terms
openSUSE Build Service is sponsored by
The Open Build Service is an
openSUSE project
.
Sign Up
Log In
Places
Places
All Projects
Status Monitor
