diff options
author | unknown <jonas@perch.ndb.mysql.com> | 2008-02-08 15:17:45 +0100 |
---|---|---|
committer | unknown <jonas@perch.ndb.mysql.com> | 2008-02-08 15:17:45 +0100 |
commit | 56165c20ec3cbfa1412dab7e9877d5e16ee0bbab (patch) | |
tree | 778c16dde653638a01639d3ab87f24e68dc2eae8 | |
parent | 25bddc4de1dbb0c40a9137ee9a5b2ffa15b4384d (diff) | |
download | mariadb-git-56165c20ec3cbfa1412dab7e9877d5e16ee0bbab.tar.gz |
ndb - bug#34378
Using more than 16g can cause record-pool ptr.i values to overflow
Fix by splitting memory into 2 zones, lo(16g)/hi(rest)
When record pools only use zone_lo, and datamemory, buffers etc...can use any
storage/ndb/src/kernel/blocks/lgman.cpp:
adopt to changed interface for Ndbd_mem_manager
storage/ndb/src/kernel/vm/Pool.cpp:
Always use ZONE_LO for record pools
as they use ptr.i == 19 bit page id + 13 bit page index
storage/ndb/src/kernel/vm/ndbd_malloc_impl.cpp:
Add zones to Ndbd_mem_manager
ZONE_LO = lower 16g
ZONE_HI = rest
storage/ndb/src/kernel/vm/ndbd_malloc_impl.hpp:
Add zones to Ndbd_mem_manager
ZONE_LO = lower 16g
ZONE_HI = rest
-rw-r--r-- | storage/ndb/src/kernel/blocks/lgman.cpp | 4 | ||||
-rw-r--r-- | storage/ndb/src/kernel/vm/Pool.cpp | 3 | ||||
-rw-r--r-- | storage/ndb/src/kernel/vm/ndbd_malloc_impl.cpp | 390 | ||||
-rw-r--r-- | storage/ndb/src/kernel/vm/ndbd_malloc_impl.hpp | 32 |
4 files changed, 335 insertions, 94 deletions
diff --git a/storage/ndb/src/kernel/blocks/lgman.cpp b/storage/ndb/src/kernel/blocks/lgman.cpp index fc28906f4db..53cb1e113e1 100644 --- a/storage/ndb/src/kernel/blocks/lgman.cpp +++ b/storage/ndb/src/kernel/blocks/lgman.cpp @@ -918,7 +918,7 @@ Lgman::alloc_logbuffer_memory(Ptr<Logfile_group> ptr, Uint32 bytes) { Uint32 ptrI; Uint32 cnt = pages > 64 ? 64 : pages; - m_ctx.m_mm.alloc(&ptrI, &cnt, 1); + m_ctx.m_mm.alloc_pages(RG_DISK_OPERATIONS, &ptrI, &cnt, 1); if (cnt) { Buffer_idx range; @@ -1037,7 +1037,7 @@ Lgman::free_logbuffer_memory(Ptr<Logfile_group> ptr) ndbrequire(map.next(it)); tmp[1] = *it.data; - m_ctx.m_mm.release(range.m_ptr_i, range.m_idx); + m_ctx.m_mm.release_pages(RG_DISK_OPERATIONS, range.m_ptr_i, range.m_idx); map.next(it); } map.release(); diff --git a/storage/ndb/src/kernel/vm/Pool.cpp b/storage/ndb/src/kernel/vm/Pool.cpp index 29c4aec4ee3..f252a601ac2 100644 --- a/storage/ndb/src/kernel/vm/Pool.cpp +++ b/storage/ndb/src/kernel/vm/Pool.cpp @@ -20,7 +20,8 @@ void* Pool_context::alloc_page(Uint32 type_id, Uint32 *i) { - return m_block->m_ctx.m_mm.alloc_page(type_id, i); + return m_block->m_ctx.m_mm.alloc_page(type_id, i, + Ndbd_mem_manager::NDB_ZONE_LO); } void diff --git a/storage/ndb/src/kernel/vm/ndbd_malloc_impl.cpp b/storage/ndb/src/kernel/vm/ndbd_malloc_impl.cpp index e771745a62d..e2100e66baa 100644 --- a/storage/ndb/src/kernel/vm/ndbd_malloc_impl.cpp +++ b/storage/ndb/src/kernel/vm/ndbd_malloc_impl.cpp @@ -25,19 +25,22 @@ extern EventLogger g_eventLogger; extern EventLogger g_eventLogger; #endif -#ifdef NDBD_MALLOC_METHOD -#if NDBD_MALLOC_METHOD == sbrk -static const char * f_method = "sbrk"; +static int f_method_idx = 0; +#ifdef NDBD_MALLOC_METHOD_SBRK +static const char * f_method = "SMsm"; #else -static const char * f_method = "malloc"; -#endif -#elif SIZEOF_CHARP == 8 -static const char * f_method = "sbrk"; -#else -static const char * f_method = "malloc"; +static const char * f_method = "MSms"; #endif #define MAX_CHUNKS 10 +#define ZONE_LO 0 +#define ZONE_HI 1 + +/** + * POOL_RECORD_BITS == 13 => 32 - 13 = 19 bits for page + */ +#define ZONE_LO_BOUND (1u << 19) + struct InitChunk { Uint32 m_cnt; @@ -54,28 +57,42 @@ do_malloc(Uint32 pages, InitChunk* chunk) pages += 1; void * ptr = 0; Uint32 sz = pages; - if (strcmp(f_method, "sbrk") == 0) + +retry: + char method = f_method[f_method_idx]; + switch(method){ + case 0: + return false; + case 'S': + case 's': { ptr = 0; while (ptr == 0) { ptr = sbrk(sizeof(Alloc_page) * sz); + if (ptr == (void*)-1) { + if (method == 'S') + { + f_method_idx++; + goto retry; + } + ptr = 0; sz = 1 + (9 * sz) / 10; if (pages >= 32 && sz < 32) { sz = pages; - f_method = "malloc"; - g_eventLogger.info("sbrk(%lld) failed, trying malloc", - (Uint64)(sizeof(Alloc_page) * sz)); - break; + f_method_idx++; + goto retry; } } } + break; } - if (strcmp(f_method, "malloc") == 0) + case 'M': + case 'm': { ptr = 0; while (ptr == 0) @@ -83,15 +100,26 @@ do_malloc(Uint32 pages, InitChunk* chunk) ptr = malloc(sizeof(Alloc_page) * sz); if (ptr == 0) { + if (method == 'M') + { + f_method_idx++; + goto retry; + } + sz = 1 + (9 * sz) / 10; if (pages >= 32 && sz < 32) { - return false; + f_method_idx++; + goto retry; } } } + break; } - + default: + return false; + } + chunk->m_cnt = sz; chunk->m_ptr = (Alloc_page*)ptr; const UintPtr align = sizeof(Alloc_page) - 1; @@ -151,6 +179,12 @@ Ndbd_mem_manager::Ndbd_mem_manager() } } +/** + * m_min = reserved + * m_curr = current + * m_max = max alloc, 0 = no limit + */ + void Ndbd_mem_manager::set_resource_limit(const Resource_limit& rl) { @@ -176,6 +210,40 @@ Ndbd_mem_manager::get_resource_limit(Uint32 id, Resource_limit& rl) const return false; } +static +inline +void +check_resource_limits(Resource_limit* rl) +{ +#ifdef VM_TRACE + Uint32 curr = 0; + Uint32 res_alloc = 0; + Uint32 shared_alloc = 0; + Uint32 sumres = 0; + for (Uint32 i = 1; i<XX_RL_COUNT; i++) + { + curr += rl[i].m_curr; + sumres += rl[i].m_min; + assert(rl[i].m_max == 0 || rl[i].m_curr <= rl[i].m_max); + if (rl[i].m_curr > rl[i].m_min) + { + shared_alloc += rl[i].m_curr - rl[i].m_min; + res_alloc += rl[i].m_min; + } + else + { + res_alloc += rl[i].m_curr; + } + } + assert(curr == rl[0].m_curr); + assert(res_alloc + shared_alloc == curr); + assert(res_alloc <= sumres); + assert(sumres == res_alloc + rl[0].m_min); + assert(rl[0].m_curr <= rl[0].m_max); +#endif +} + + bool Ndbd_mem_manager::init(bool alloc_less_memory) { @@ -292,6 +360,8 @@ Ndbd_mem_manager::init(bool alloc_less_memory) grow(chunks[i].m_start, chunks[i].m_cnt); } + check_resource_limits(m_resource_limit); + return true; } @@ -321,35 +391,68 @@ Ndbd_mem_manager::grow(Uint32 start, Uint32 cnt) cnt--; // last page is always marked as empty } - if (!m_used_bitmap_pages.get(start_bmp)) - { - if (start != (start_bmp << BPP_2LOG)) - { - ndbout_c("ndbd_malloc_impl.cpp:%d:grow(%d, %d) %d!=%d" - " - Unable to use due to bitmap pages missaligned!!", - __LINE__, start, cnt, start, (start_bmp << BPP_2LOG)); - g_eventLogger.error("ndbd_malloc_impl.cpp:%d:grow(%d, %d)" - " - Unable to use due to bitmap pages missaligned!!", - __LINE__, start, cnt); - return; - } + for (Uint32 i = 0; i<m_used_bitmap_pages.size(); i++) + if (m_used_bitmap_pages[i] == start_bmp) + goto found; + if (start != (start_bmp << BPP_2LOG)) + { + + ndbout_c("ndbd_malloc_impl.cpp:%d:grow(%d, %d) %d!=%d not using %uMb" + " - Unable to use due to bitmap pages missaligned!!", + __LINE__, start, cnt, start, (start_bmp << BPP_2LOG), + (cnt >> (20 - 15))); + g_eventLogger.error("ndbd_malloc_impl.cpp:%d:grow(%d, %d) not using %uMb" + " - Unable to use due to bitmap pages missaligned!!", + __LINE__, start, cnt, + (cnt >> (20 - 15))); + + dump(); + return; + } + #ifdef UNIT_TEST - ndbout_c("creating bitmap page %d", start_bmp); + ndbout_c("creating bitmap page %d", start_bmp); #endif - + + { Alloc_page* bmp = m_base_page + start; memset(bmp, 0, sizeof(Alloc_page)); - m_used_bitmap_pages.set(start_bmp); cnt--; start++; } - + m_used_bitmap_pages.push_back(start_bmp); + +found: if (cnt) { m_resource_limit[0].m_curr += cnt; m_resource_limit[0].m_max += cnt; - release(start, cnt); + if (start >= ZONE_LO_BOUND) + { + Uint64 mbytes = ((Uint64(cnt) * 32) + 1023) / 1024; + ndbout_c("Adding %uMb to ZONE_HI (%u,%u)", (Uint32)mbytes, start, cnt); + release(start, cnt); + } + else if (start + cnt <= ZONE_LO_BOUND) + { + Uint64 mbytes = ((Uint64(cnt)*32) + 1023) / 1024; + ndbout_c("Adding %uMb to ZONE_LO (%u,%u)", (Uint32)mbytes, start, cnt); + release(start, cnt); + } + else + { + Uint32 cnt0 = ZONE_LO_BOUND - start; + Uint32 cnt1 = start + cnt - ZONE_LO_BOUND; + Uint64 mbytes0 = ((Uint64(cnt0)*32) + 1023) / 1024; + Uint64 mbytes1 = ((Uint64(cnt1)*32) + 1023) / 1024; + ndbout_c("Adding %uMb to ZONE_LO (split %u,%u)", (Uint32)mbytes0, + start, cnt0); + ndbout_c("Adding %uMb to ZONE_HI (split %u,%u)", (Uint32)mbytes1, + ZONE_LO_BOUND, cnt1); + release(start, cnt0); + release(ZONE_LO_BOUND, cnt1); + } } } @@ -362,40 +465,58 @@ Ndbd_mem_manager::release(Uint32 start, Uint32 cnt) set(start, start+cnt-1); - release_impl(start, cnt); + Uint32 zone = start < ZONE_LO_BOUND ? 0 : 1; + release_impl(zone, start, cnt); } void -Ndbd_mem_manager::release_impl(Uint32 start, Uint32 cnt) +Ndbd_mem_manager::release_impl(Uint32 zone, Uint32 start, Uint32 cnt) { assert(start); Uint32 test = check(start-1, start+cnt); - if (test & 1) + if (start != ZONE_LO_BOUND && test & 1) { Free_page_data *fd = get_free_page_data(m_base_page + start - 1, start - 1); Uint32 sz = fd->m_size; Uint32 left = start - sz; - remove_free_list(left, fd->m_list); + remove_free_list(zone, left, fd->m_list); cnt += sz; start = left; } Uint32 right = start + cnt; - if (test & 2) + if (right != ZONE_LO_BOUND && test & 2) { Free_page_data *fd = get_free_page_data(m_base_page+right, right); Uint32 sz = fd->m_size; - remove_free_list(right, fd->m_list); + remove_free_list(zone, right, fd->m_list); cnt += sz; } - insert_free_list(start, cnt); + insert_free_list(zone, start, cnt); } void -Ndbd_mem_manager::alloc(Uint32* ret, Uint32 *pages, Uint32 min) +Ndbd_mem_manager::alloc(AllocZone zone, + Uint32* ret, Uint32 *pages, Uint32 min) +{ + if (zone == NDB_ZONE_ANY) + { + Uint32 save = * pages; + alloc_impl(ZONE_HI, ret, pages, min); + if (*pages) + return; + * pages = save; + } + + alloc_impl(ZONE_LO, ret, pages, min); +} + +void +Ndbd_mem_manager::alloc_impl(Uint32 zone, + Uint32* ret, Uint32 *pages, Uint32 min) { Int32 i; Uint32 start; @@ -407,19 +528,19 @@ Ndbd_mem_manager::alloc(Uint32* ret, Uint32 *pages, Uint32 min) for (i = list; i < 16; i++) { - if ((start = m_buddy_lists[i])) + if ((start = m_buddy_lists[zone][i])) { /* ---------------------------------------------------------------- */ /* PROPER AMOUNT OF PAGES WERE FOUND. NOW SPLIT THE FOUND */ /* AREA AND RETURN THE PART NOT NEEDED. */ /* ---------------------------------------------------------------- */ - Uint32 sz = remove_free_list(start, i); + Uint32 sz = remove_free_list(zone, start, i); Uint32 extra = sz - cnt; assert(sz >= cnt); if (extra) { - insert_free_list(start + cnt, extra); + insert_free_list(zone, start + cnt, extra); clear_and_set(start, start+cnt-1); } else @@ -427,8 +548,7 @@ Ndbd_mem_manager::alloc(Uint32* ret, Uint32 *pages, Uint32 min) clear(start, start+cnt-1); } * ret = start; - m_resource_limit[0].m_curr += cnt; - assert(m_resource_limit[0].m_curr <= m_resource_limit[0].m_max); + assert(m_resource_limit[0].m_curr + cnt <= m_resource_limit[0].m_max); return; } } @@ -442,13 +562,13 @@ Ndbd_mem_manager::alloc(Uint32* ret, Uint32 *pages, Uint32 min) assert((Int32)list >= min_list); for (i = list - 1; i >= min_list; i--) { - if ((start = m_buddy_lists[i])) + if ((start = m_buddy_lists[zone][i])) { - Uint32 sz = remove_free_list(start, i); + Uint32 sz = remove_free_list(zone, start, i); Uint32 extra = sz - cnt; if (sz > cnt) { - insert_free_list(start + cnt, extra); + insert_free_list(zone, start + cnt, extra); sz -= extra; clear_and_set(start, start+sz-1); } @@ -459,8 +579,7 @@ Ndbd_mem_manager::alloc(Uint32* ret, Uint32 *pages, Uint32 min) * ret = start; * pages = sz; - m_resource_limit[0].m_curr += sz; - assert(m_resource_limit[0].m_curr <= m_resource_limit[0].m_max); + assert(m_resource_limit[0].m_curr + sz <= m_resource_limit[0].m_max); return; } } @@ -468,12 +587,12 @@ Ndbd_mem_manager::alloc(Uint32* ret, Uint32 *pages, Uint32 min) } void -Ndbd_mem_manager::insert_free_list(Uint32 start, Uint32 size) +Ndbd_mem_manager::insert_free_list(Uint32 zone, Uint32 start, Uint32 size) { Uint32 list = ndb_log2(size) - 1; Uint32 last = start + size - 1; - Uint32 head = m_buddy_lists[list]; + Uint32 head = m_buddy_lists[zone][list]; Free_page_data* fd_first = get_free_page_data(m_base_page+start, start); fd_first->m_list = list; @@ -495,11 +614,11 @@ Ndbd_mem_manager::insert_free_list(Uint32 start, Uint32 size) fd->m_prev = start; } - m_buddy_lists[list] = start; + m_buddy_lists[zone][list] = start; } Uint32 -Ndbd_mem_manager::remove_free_list(Uint32 start, Uint32 list) +Ndbd_mem_manager::remove_free_list(Uint32 zone, Uint32 start, Uint32 list) { Free_page_data* fd = get_free_page_data(m_base_page+start, start); Uint32 size = fd->m_size; @@ -509,7 +628,7 @@ Ndbd_mem_manager::remove_free_list(Uint32 start, Uint32 list) if (prev) { - assert(m_buddy_lists[list] != start); + assert(m_buddy_lists[zone][list] != start); fd = get_free_page_data(m_base_page+prev, prev); assert(fd->m_next == start); assert(fd->m_list == list); @@ -517,8 +636,8 @@ Ndbd_mem_manager::remove_free_list(Uint32 start, Uint32 list) } else { - assert(m_buddy_lists[list] == start); - m_buddy_lists[list] = next; + assert(m_buddy_lists[zone][list] == start); + m_buddy_lists[zone][list] = next; } if (next) @@ -535,42 +654,62 @@ Ndbd_mem_manager::remove_free_list(Uint32 start, Uint32 list) void Ndbd_mem_manager::dump() const { - for(Uint32 i = 0; i<16; i++) + for (Uint32 zone = 0; zone < 2; zone ++) { - printf(" list: %d - ", i); - Uint32 head = m_buddy_lists[i]; - while(head) + for (Uint32 i = 0; i<16; i++) { - Free_page_data* fd = get_free_page_data(m_base_page+head, head); - printf("[ i: %d prev %d next %d list %d size %d ] ", - head, fd->m_prev, fd->m_next, fd->m_list, fd->m_size); - head = fd->m_next; + printf(" list: %d - ", i); + Uint32 head = m_buddy_lists[zone][i]; + while(head) + { + Free_page_data* fd = get_free_page_data(m_base_page+head, head); + printf("[ i: %d prev %d next %d list %d size %d ] ", + head, fd->m_prev, fd->m_next, fd->m_list, fd->m_size); + head = fd->m_next; + } + printf("EOL\n"); + } + + for (Uint32 i = 0; i<XX_RL_COUNT; i++) + { + printf("ri: %d min: %d curr: %d max: %d\n", + i, + m_resource_limit[i].m_min, + m_resource_limit[i].m_curr, + m_resource_limit[i].m_max); } - printf("EOL\n"); } } void* -Ndbd_mem_manager::alloc_page(Uint32 type, Uint32* i) +Ndbd_mem_manager::alloc_page(Uint32 type, Uint32* i, AllocZone zone) { Uint32 idx = type & RG_MASK; assert(idx && idx < XX_RL_COUNT); Resource_limit tot = m_resource_limit[0]; Resource_limit rl = m_resource_limit[idx]; - Uint32 add = (rl.m_curr < rl.m_min) ? 0 : 1; // Over min ? + Uint32 cnt = 1; + Uint32 res0 = (rl.m_curr < rl.m_min) ? 1 : 0; Uint32 limit = (rl.m_max == 0 || rl.m_curr < rl.m_max) ? 0 : 1; // Over limit Uint32 free = (tot.m_min + tot.m_curr < tot.m_max) ? 1 : 0; // Has free - if (likely(add == 0 || (limit == 0 && free == 1))) + assert(tot.m_min >= res0); + + if (likely(res0 == 1 || (limit == 0 && free == 1))) { - Uint32 cnt = 1; - alloc(i, &cnt, 1); - assert(cnt); - m_resource_limit[0].m_curr = tot.m_curr + add; - m_resource_limit[idx].m_curr = rl.m_curr + 1; - return m_base_page + *i; + alloc(zone, i, &cnt, 1); + if (likely(cnt)) + { + m_resource_limit[0].m_curr = tot.m_curr + cnt; + m_resource_limit[0].m_min = tot.m_min - res0; + m_resource_limit[idx].m_curr = rl.m_curr + cnt; + + check_resource_limits(m_resource_limit); + return m_base_page + *i; + } } + return 0; } @@ -582,10 +721,102 @@ Ndbd_mem_manager::release_page(Uint32 type, Uint32 i) Resource_limit tot = m_resource_limit[0]; Resource_limit rl = m_resource_limit[idx]; - Uint32 sub = (rl.m_curr < rl.m_min) ? 0 : 1; // Over min ? + Uint32 sub = (rl.m_curr <= rl.m_min) ? 1 : 0; // Over min ? release(i, 1); - m_resource_limit[0].m_curr = tot.m_curr - sub; + m_resource_limit[0].m_curr = tot.m_curr - 1; + m_resource_limit[0].m_min = tot.m_min + sub; m_resource_limit[idx].m_curr = rl.m_curr - 1; + + check_resource_limits(m_resource_limit); +} + +void +Ndbd_mem_manager::alloc_pages(Uint32 type, Uint32* i, Uint32 *cnt, Uint32 min) +{ + Uint32 idx = type & RG_MASK; + assert(idx && idx < XX_RL_COUNT); + Resource_limit tot = m_resource_limit[0]; + Resource_limit rl = m_resource_limit[idx]; + + Uint32 req = *cnt; + + Uint32 max = rl.m_max - rl.m_curr; + Uint32 res0 = rl.m_min - rl.m_curr; + Uint32 free_shared = tot.m_max - (tot.m_min + tot.m_curr); + + Uint32 res1; + if (rl.m_curr + req <= rl.m_min) + { + // all is reserved... + res0 = req; + res1 = 0; + } + else + { + req = rl.m_max ? max : req; + res0 = (rl.m_curr > rl.m_min) ? 0 : res0; + res1 = req - res0; + + if (unlikely(res1 > free_shared)) + { + res1 = free_shared; + req = res0 + res1; + } + } + + // req = pages to alloc + // res0 = portion that is reserved + // res1 = part that is over reserver + assert (res0 + res1 == req); + assert (tot.m_min >= res0); + + if (likely(req)) + { + // Hi order allocations can always use any zone + alloc(NDB_ZONE_ANY, i, &req, 1); + * cnt = req; + if (unlikely(req < res0)) // Got min than what was reserved :-( + { + res0 = req; + } + assert(tot.m_min >= res0); + assert(tot.m_curr + req <= tot.m_max); + + m_resource_limit[0].m_curr = tot.m_curr + req; + m_resource_limit[0].m_min = tot.m_min - res0; + m_resource_limit[idx].m_curr = rl.m_curr + req; + check_resource_limits(m_resource_limit); + return ; + } + * cnt = req; + return; +} + +void +Ndbd_mem_manager::release_pages(Uint32 type, Uint32 i, Uint32 cnt) +{ + Uint32 idx = type & RG_MASK; + assert(idx && idx < XX_RL_COUNT); + Resource_limit tot = m_resource_limit[0]; + Resource_limit rl = m_resource_limit[idx]; + + release(i, cnt); + + Uint32 currnew = rl.m_curr - cnt; + if (rl.m_curr > rl.m_min) + { + if (currnew < rl.m_min) + { + m_resource_limit[0].m_min = tot.m_min + (rl.m_min - currnew); + } + } + else + { + m_resource_limit[0].m_min = tot.m_min + cnt; + } + m_resource_limit[0].m_curr = tot.m_curr - cnt; + m_resource_limit[idx].m_curr = currnew; + check_resource_limits(m_resource_limit); } #ifdef UNIT_TEST @@ -781,3 +1012,4 @@ main(int argc, char** argv) template class Vector<Chunk>; #endif +template class Vector<Uint32>; diff --git a/storage/ndb/src/kernel/vm/ndbd_malloc_impl.hpp b/storage/ndb/src/kernel/vm/ndbd_malloc_impl.hpp index b89371b8d71..78e41f1cabd 100644 --- a/storage/ndb/src/kernel/vm/ndbd_malloc_impl.hpp +++ b/storage/ndb/src/kernel/vm/ndbd_malloc_impl.hpp @@ -20,6 +20,7 @@ #include <Bitmask.hpp> #include <assert.h> #include "Pool.hpp" +#include <Vector.hpp> /** * 13 -> 8192 words -> 32768 bytes @@ -59,16 +60,19 @@ public: bool init(bool allow_alloc_less_than_requested = true); void* get_memroot() const { return (void*)m_base_page;} - void alloc(Uint32* ret, Uint32 *pages, Uint32 min_requested); - void release(Uint32 start, Uint32 cnt); - void dump() const ; - void* alloc_page(Uint32 type, Uint32* i); + enum AllocZone + { + NDB_ZONE_LO = 0, // Only allocate with page_id < (1 << 13) + NDB_ZONE_ANY = 1 // Allocate with any page_id + }; + + void* alloc_page(Uint32 type, Uint32* i, enum AllocZone); void release_page(Uint32 type, Uint32 i); - void* alloc_pages(Uint32 type, Uint32* i, Uint32 *cnt, Uint32 min = 1); - void release_pages(Uint32 type, Uint32 i, void*p, Uint32 cnt); + void alloc_pages(Uint32 type, Uint32* i, Uint32 *cnt, Uint32 min = 1); + void release_pages(Uint32 type, Uint32 i, Uint32 cnt); /** * Compute 2log of size @@ -80,25 +84,29 @@ public: private: void grow(Uint32 start, Uint32 cnt); -#define XX_RL_COUNT 3 +#define XX_RL_COUNT 4 /** * Return pointer to free page data on page */ static Free_page_data* get_free_page_data(Alloc_page*, Uint32 idx); - Bitmask<1> m_used_bitmap_pages; + Vector<Uint32> m_used_bitmap_pages; - Uint32 m_buddy_lists[16]; + Uint32 m_buddy_lists[2][16]; Resource_limit m_resource_limit[XX_RL_COUNT]; // RG_COUNT in record_types.hpp Alloc_page * m_base_page; - void release_impl(Uint32 start, Uint32 cnt); - void insert_free_list(Uint32 start, Uint32 cnt); - Uint32 remove_free_list(Uint32 start, Uint32 list); + void release_impl(Uint32 zone, Uint32 start, Uint32 cnt); + void insert_free_list(Uint32 zone, Uint32 start, Uint32 cnt); + Uint32 remove_free_list(Uint32 zone, Uint32 start, Uint32 list); void set(Uint32 first, Uint32 last); void clear(Uint32 first, Uint32 last); void clear_and_set(Uint32 first, Uint32 last); Uint32 check(Uint32 first, Uint32 last); + + void alloc(AllocZone, Uint32* ret, Uint32 *pages, Uint32 min_requested); + void alloc_impl(Uint32 zone, Uint32* ret, Uint32 *pages, Uint32 min); + void release(Uint32 start, Uint32 cnt); }; inline |