Skip to content

M
MariaDB
Commit 9763d0d0 authored by jonas@perch.ndb.mysql.com's avatar jonas@perch.ndb.mysql.com
Browse files
Options

ndb - jamify (better) DbtupDiskAlloc

parent 4d552dbb
Hide whitespace changes
Inline Side-by-side
Showing with 148 additions and 67 deletions
storage/ndb/src/kernel/blocks/dbtup/Dbtup.hpp
View file @ 9763d0d0
......@@ -86,6 +86,7 @@ inline const Uint32* ALIGN_WORD(const void* ptr)
// DbtupDebug.cpp 30000
// DbtupVarAlloc.cpp 32000
// DbtupScan.cpp 33000
// DbtupDiskAlloc.cpp 35000
//------------------------------------------------------------------
/*
......
storage/ndb/src/kernel/blocks/dbtup/DbtupCommit.cpp
View file @ 9763d0d0
......@@ -108,8 +108,11 @@ void Dbtup::removeActiveOpList(Operationrec* const regOperPtr,
* Release copy tuple
*/
if(!regOperPtr->m_copy_tuple_location.isNull())
{
ljam();
c_undo_buffer.free_copy_tuple(&regOperPtr->m_copy_tuple_location);
}
if (regOperPtr->op_struct.in_active_list) {
regOperPtr->op_struct.in_active_list= false;
if (regOperPtr->nextActiveOp != RNIL) {
......@@ -172,6 +175,7 @@ Dbtup::dealloc_tuple(Signal* signal,
Uint32 extra_bits = Tuple_header::FREED;
if (bits & Tuple_header::DISK_PART)
{
ljam();
Local_key disk;
memcpy(&disk, ptr->get_disk_ref_ptr(regTabPtr), sizeof(disk));
PagePtr tmpptr;
......@@ -184,6 +188,7 @@ Dbtup::dealloc_tuple(Signal* signal,
if (! (bits & (Tuple_header::LCP_SKIP | Tuple_header::ALLOC)) &&
lcpScan_ptr_i != RNIL)
{
ljam();
ScanOpPtr scanOp;
c_scanOpPool.getPtr(scanOp, lcpScan_ptr_i);
Local_key rowid = regOperPtr->m_tuple_location;
......@@ -191,6 +196,7 @@ Dbtup::dealloc_tuple(Signal* signal,
rowid.m_page_no = page->frag_page_id;
if (rowid > scanpos)
{
ljam();
extra_bits = Tuple_header::LCP_KEEP; // Note REMOVE FREE
ptr->m_operation_ptr_i = lcp_keep_list;
regFragPtr->m_lcp_keep_list = rowid.ref();
......@@ -231,11 +237,13 @@ Dbtup::commit_operation(Signal* signal,
Uint32 mm_vars= regTabPtr->m_attributes[MM].m_no_of_varsize;
if(mm_vars == 0)
{
ljam();
memcpy(tuple_ptr, copy, 4*fixsize);
disk_ptr= (Tuple_header*)(((Uint32*)copy)+fixsize);
}
else
{
ljam();
/**
* Var_part_ref is only stored in *allocated* tuple
* so memcpy from copy, will over write it...
......@@ -260,6 +268,7 @@ Dbtup::commit_operation(Signal* signal,
if(copy_bits & Tuple_header::MM_SHRINK)
{
ljam();
vpagePtrP->shrink_entry(tmp.m_page_idx, (sz + 3) >> 2);
update_free_page_list(regFragPtr, vpagePtr);
}
......@@ -270,6 +279,7 @@ Dbtup::commit_operation(Signal* signal,
if (regTabPtr->m_no_of_disk_attributes &&
(copy_bits & Tuple_header::DISK_INLINE))
{
ljam();
Local_key key;
memcpy(&key, copy->get_disk_ref_ptr(regTabPtr), sizeof(Local_key));
Uint32 logfile_group_id= regFragPtr->m_logfile_group_id;
......@@ -280,22 +290,26 @@ Dbtup::commit_operation(Signal* signal,
Uint32 sz, *dst;
if(copy_bits & Tuple_header::DISK_ALLOC)
{
ljam();
disk_page_alloc(signal, regTabPtr, regFragPtr, &key, diskPagePtr, gci);
}
if(regTabPtr->m_attributes[DD].m_no_of_varsize == 0)
{
ljam();
sz= regTabPtr->m_offsets[DD].m_fix_header_size;
dst= ((Fix_page*)diskPagePtr.p)->get_ptr(key.m_page_idx, sz);
}
else
{
ljam();
dst= ((Var_page*)diskPagePtr.p)->get_ptr(key.m_page_idx);
sz= ((Var_page*)diskPagePtr.p)->get_entry_len(key.m_page_idx);
}
if(! (copy_bits & Tuple_header::DISK_ALLOC))
{
ljam();
disk_page_undo_update(diskPagePtr.p,
&key, dst, sz, gci, logfile_group_id);
}
......@@ -309,6 +323,7 @@ Dbtup::commit_operation(Signal* signal,
if(lcpScan_ptr_i != RNIL && (bits & Tuple_header::ALLOC))
{
ljam();
ScanOpPtr scanOp;
c_scanOpPool.getPtr(scanOp, lcpScan_ptr_i);
Local_key rowid = regOperPtr->m_tuple_location;
......@@ -316,6 +331,7 @@ Dbtup::commit_operation(Signal* signal,
rowid.m_page_no = pagePtr.p->frag_page_id;
if(rowid > scanpos)
{
ljam();
copy_bits |= Tuple_header::LCP_SKIP;
}
}
......@@ -374,7 +390,10 @@ Dbtup::disk_page_commit_callback(Signal* signal,
execTUP_COMMITREQ(signal);
if(signal->theData[0] == 0)
{
ljam();
c_lqh->tupcommit_conf_callback(signal, regOperPtr.p->userpointer);
}
}
void
......@@ -412,6 +431,7 @@ Dbtup::disk_page_log_buffer_callback(Signal* signal,
void
Dbtup::fix_commit_order(OperationrecPtr opPtr)
{
ljam();
ndbassert(!opPtr.p->is_first_operation());
OperationrecPtr firstPtr = opPtr;
while(firstPtr.p->prevActiveOp != RNIL)
......@@ -437,7 +457,10 @@ Dbtup::fix_commit_order(OperationrecPtr opPtr)
c_operation_pool.getPtr(seco)->prevActiveOp = opPtr.i;
c_operation_pool.getPtr(prev)->nextActiveOp = firstPtr.i;
if(next != RNIL)
{
ljam();
c_operation_pool.getPtr(next)->prevActiveOp = firstPtr.i;
}
}
/* ----------------------------------------------------------------- */
......@@ -502,6 +525,7 @@ void Dbtup::execTUP_COMMITREQ(Signal* signal)
bool get_page = false;
if(regOperPtr.p->op_struct.m_load_diskpage_on_commit)
{
ljam();
Page_cache_client::Request req;
ndbassert(regOperPtr.p->is_first_operation() &&
regOperPtr.p->is_last_operation());
......@@ -511,6 +535,7 @@ void Dbtup::execTUP_COMMITREQ(Signal* signal)
*/
if(!regOperPtr.p->m_copy_tuple_location.isNull())
{
ljam();
Tuple_header* tmp= (Tuple_header*)
c_undo_buffer.get_ptr(&regOperPtr.p->m_copy_tuple_location);
......@@ -520,23 +545,25 @@ void Dbtup::execTUP_COMMITREQ(Signal* signal)
if (unlikely(regOperPtr.p->op_struct.op_type == ZDELETE &&
tmp->m_header_bits & Tuple_header::DISK_ALLOC))
{
jam();
ljam();
/**
* Insert+Delete
*/
regOperPtr.p->op_struct.m_load_diskpage_on_commit = 0;
regOperPtr.p->op_struct.m_wait_log_buffer = 0;
disk_page_abort_prealloc(signal, regFragPtr.p,
regOperPtr.p->op_struct.m_load_diskpage_on_commit = 0;
regOperPtr.p->op_struct.m_wait_log_buffer = 0;
disk_page_abort_prealloc(signal, regFragPtr.p,
&req.m_page, req.m_page.m_page_idx);
c_lgman->free_log_space(regFragPtr.p->m_logfile_group_id,
c_lgman->free_log_space(regFragPtr.p->m_logfile_group_id,
regOperPtr.p->m_undo_buffer_space);
ndbout_c("insert+delete");
goto skip_disk;
ndbout_c("insert+delete");
ljamEntry();
goto skip_disk;
}
}
else
{
ljam();
// initial delete
ndbassert(regOperPtr.p->op_struct.op_type == ZDELETE);
memcpy(&req.m_page,
......@@ -560,11 +587,14 @@ void Dbtup::execTUP_COMMITREQ(Signal* signal)
/**
* Timeslice
*/
ljam();
signal->theData[0] = 1;
return;
case -1:
ndbrequire("NOT YET IMPLEMENTED" == 0);
break;
default:
ljam();
}
get_page = true;
......@@ -581,6 +611,7 @@ void Dbtup::execTUP_COMMITREQ(Signal* signal)
if(regOperPtr.p->op_struct.m_wait_log_buffer)
{
ljam();
ndbassert(regOperPtr.p->is_first_operation() &&
regOperPtr.p->is_last_operation());
......@@ -592,18 +623,23 @@ void Dbtup::execTUP_COMMITREQ(Signal* signal)
Logfile_client lgman(this, c_lgman, regFragPtr.p->m_logfile_group_id);
int res= lgman.get_log_buffer(signal, sz, &cb);
ljamEntry();
switch(res){
case 0:
ljam();
signal->theData[0] = 1;
return;
case -1:
ndbrequire("NOT YET IMPLEMENTED" == 0);
break;
default:
ljam();
}
}
if(!tuple_ptr)
{
ljam();
tuple_ptr = (Tuple_header*)
get_ptr(&page, &regOperPtr.p->m_tuple_location,regTabPtr.p);
}
......@@ -612,6 +648,7 @@ void Dbtup::execTUP_COMMITREQ(Signal* signal)
if(get_tuple_state(regOperPtr.p) == TUPLE_PREPARED)
{
ljam();
/**
* Execute all tux triggers at first commit
* since previous tuple is otherwise removed...
......@@ -637,6 +674,7 @@ void Dbtup::execTUP_COMMITREQ(Signal* signal)
if(regOperPtr.p->is_last_operation())
{
ljam();
/**
* Perform "real" commit
*/
......@@ -647,12 +685,14 @@ void Dbtup::execTUP_COMMITREQ(Signal* signal)
if(regOperPtr.p->op_struct.op_type != ZDELETE)
{
ljam();
commit_operation(signal, gci, tuple_ptr, page,
regOperPtr.p, regFragPtr.p, regTabPtr.p);
removeActiveOpList(regOperPtr.p, tuple_ptr);
}
else
{
ljam();
removeActiveOpList(regOperPtr.p, tuple_ptr);
if (get_page)
ndbassert(tuple_ptr->m_header_bits & Tuple_header::DISK_PART);
......@@ -662,6 +702,7 @@ void Dbtup::execTUP_COMMITREQ(Signal* signal)
}
else
{
ljam();
removeActiveOpList(regOperPtr.p, tuple_ptr);
}
......
storage/ndb/src/kernel/blocks/dbtup/DbtupDiskAlloc.cpp
View file @ 9763d0d0
......@@ -16,6 +16,10 @@
#define DBTUP_C
#include "Dbtup.hpp"
#define ljam() { jamLine(35000 + __LINE__); }
#define ljamEntry() { jamEntryLine(35000 + __LINE__); }
static bool f_undo_done = true;
static
......@@ -266,7 +270,7 @@ Dbtup::update_extent_pos(Disk_alloc_info& alloc,
Uint32 pos = alloc.calc_extent_pos(extentPtr.p);
if (old != pos)
{
jam();
ljam();
Local_extent_info_list old_list(c_extent_pool, alloc.m_free_extents[old]);
Local_extent_info_list new_list(c_extent_pool, alloc.m_free_extents[pos]);
old_list.remove(extentPtr);
......@@ -283,6 +287,7 @@ Dbtup::update_extent_pos(Disk_alloc_info& alloc,
void
Dbtup::restart_setup_page(Disk_alloc_info& alloc, PagePtr pagePtr)
{
ljam();
/**
* Link to extent, clear uncommitted_used_space
*/
......@@ -303,6 +308,7 @@ Dbtup::restart_setup_page(Disk_alloc_info& alloc, PagePtr pagePtr)
ddassert(real_free >= estimated);
if (real_free != estimated)
{
ljam();
extentPtr.p->m_free_space += (real_free - estimated);
update_extent_pos(alloc, extentPtr);
}
......@@ -318,7 +324,7 @@ Dbtup::restart_setup_page(Disk_alloc_info& alloc, PagePtr pagePtr)
unsigned uncommitted, committed;
uncommitted = committed = ~(unsigned)0;
int ret = tsman.get_page_free_bits(&page, &uncommitted, &committed);
jamEntry();
ljamEntry();
idx = alloc.calc_page_free_bits(real_free);
ddassert(idx == committed);
......@@ -375,6 +381,7 @@ Dbtup::disk_page_prealloc(Signal* signal,
key->m_file_no= tmp.p->m_file_no;
if (DBG_DISK)
ndbout << " found dirty page " << *key << endl;
ljam();
return 0; // Page in memory
}
}
......@@ -396,6 +403,7 @@ Dbtup::disk_page_prealloc(Signal* signal,
* key = req.p->m_key;
if (DBG_DISK)
ndbout << " found transit page " << *key << endl;
ljam();
return 0;
}
}
......@@ -405,6 +413,7 @@ Dbtup::disk_page_prealloc(Signal* signal,
*/
if (!c_page_request_pool.seize(req))
{
ljam();
err= 1;
//XXX set error code
ndbout_c("no free request");
......@@ -425,16 +434,16 @@ Dbtup::disk_page_prealloc(Signal* signal,
*/
if ((ext.i= alloc.m_curr_extent_info_ptr_i) != RNIL)
{
jam();
ljam();
c_extent_pool.getPtr(ext);
if ((pageBits= tsman.alloc_page_from_extent(&ext.p->m_key, bits)) >= 0)
{
jamEntry();
ljamEntry();
found= true;
}
else
{
jamEntry();
ljamEntry();
/**
* The current extent is not in a free list
* and since it couldn't accomadate the request
......@@ -453,14 +462,14 @@ Dbtup::disk_page_prealloc(Signal* signal,
Uint32 pos;
if ((pos= alloc.find_extent(sz)) != RNIL)
{
jam();
ljam();
Local_extent_info_list list(c_extent_pool, alloc.m_free_extents[pos]);
list.first(ext);
list.remove(ext);
}
else
{
jam();
ljam();
/**
* We need to alloc an extent
*/
......@@ -469,6 +478,7 @@ Dbtup::disk_page_prealloc(Signal* signal,
err = c_lgman->alloc_log_space(logfile_group_id,
sizeof(Disk_undo::AllocExtent)>>2);
ljamEntry();
if(unlikely(err))
{
return -err;
......@@ -477,7 +487,7 @@ Dbtup::disk_page_prealloc(Signal* signal,
if (!c_extent_pool.seize(ext))
{
jam();
ljam();
//XXX
err= 2;
#if NOT_YET_UNDO_ALLOC_EXTENT
......@@ -491,7 +501,7 @@ Dbtup::disk_page_prealloc(Signal* signal,
if ((err= tsman.alloc_extent(&ext.p->m_key)) < 0)
{
jamEntry();
ljamEntry();
#if NOT_YET_UNDO_ALLOC_EXTENT
c_lgman->free_log_space(logfile_group_id,
sizeof(Disk_undo::AllocExtent)>>2);
......@@ -543,7 +553,7 @@ Dbtup::disk_page_prealloc(Signal* signal,
alloc.m_curr_extent_info_ptr_i= ext.i;
ext.p->m_free_matrix_pos= RNIL;
pageBits= tsman.alloc_page_from_extent(&ext.p->m_key, bits);
jamEntry();
ljamEntry();
ddassert(pageBits >= 0);
}
......@@ -569,6 +579,7 @@ Dbtup::disk_page_prealloc(Signal* signal,
Uint32 newPageBits= alloc.calc_page_free_bits(new_size);
if (newPageBits != (Uint32)pageBits)
{
ljam();
ddassert(ext.p->m_free_page_count[pageBits] > 0);
ext.p->m_free_page_count[pageBits]--;
ext.p->m_free_page_count[newPageBits]++;
......@@ -596,6 +607,7 @@ Dbtup::disk_page_prealloc(Signal* signal,
int flags= Page_cache_client::ALLOC_REQ;
if (pageBits == 0)
{
ljam();
//XXX empty page -> fast to map
flags |= Page_cache_client::EMPTY_PAGE;
preq.m_callback.m_callbackFunction =
......@@ -603,15 +615,17 @@ Dbtup::disk_page_prealloc(Signal* signal,
}
int res= m_pgman.get_page(signal, preq, flags);
jamEntry();
ljamEntry();
switch(res)
{
case 0:
ljam();
break;
case -1:
ndbassert(false);
break;
default:
ljam();
execute(signal, preq.m_callback, res); // run callback
}
......@@ -623,6 +637,7 @@ Dbtup::disk_page_prealloc_dirty_page(Disk_alloc_info & alloc,
PagePtr pagePtr,
Uint32 old_idx, Uint32 sz)
{
ljam();
ddassert(pagePtr.p->list_index == old_idx);
Uint32 free= pagePtr.p->free_space;
......@@ -638,6 +653,7 @@ Dbtup::disk_page_prealloc_dirty_page(Disk_alloc_info & alloc,
if (old_idx != new_idx)
{
ljam();
LocalDLList<Page> old_list(*pool, alloc.m_dirty_pages[old_idx]);
LocalDLList<Page> new_list(*pool, alloc.m_dirty_pages[new_idx]);
old_list.remove(pagePtr);
......@@ -661,6 +677,7 @@ Dbtup::disk_page_prealloc_transit_page(Disk_alloc_info& alloc,
Ptr<Page_request> req,
Uint32 old_idx, Uint32 sz)
{
ljam();
ddassert(req.p->m_list_index == old_idx);
Uint32 free= req.p->m_estimated_free_space;
......@@ -675,6 +692,7 @@ Dbtup::disk_page_prealloc_transit_page(Disk_alloc_info& alloc,
if (old_idx != new_idx)
{
ljam();
Page_request_list::Head *lists = alloc.m_page_requests;
Local_page_request_list old_list(c_page_request_pool, lists[old_idx]);
Local_page_request_list new_list(c_page_request_pool, lists[new_idx]);
......@@ -699,6 +717,7 @@ void
Dbtup::disk_page_prealloc_callback(Signal* signal,
Uint32 page_request, Uint32 page_id)
{
ljamEntry();
//ndbout_c("disk_alloc_page_callback id: %d", page_id);
Ptr<Page_request> req;
......@@ -728,6 +747,7 @@ Dbtup::disk_page_prealloc_initial_callback(Signal*signal,
Uint32 page_request,
Uint32 page_id)
{
ljamEntry();
//ndbout_c("disk_alloc_page_callback_initial id: %d", page_id);
/**
* 1) lookup page request
......@@ -819,6 +839,7 @@ Dbtup::disk_page_prealloc_callback_common(Signal* signal,
if (old_idx != new_idx || free != real_free)
{
ljam();
Ptr<Extent_info> extentPtr;
c_extent_pool.getPtr(extentPtr, ext);
......@@ -826,6 +847,7 @@ Dbtup::disk_page_prealloc_callback_common(Signal* signal,
if (old_idx != new_idx)
{
ljam();
ddassert(extentPtr.p->m_free_page_count[old_idx]);
extentPtr.p->m_free_page_count[old_idx]--;
extentPtr.p->m_free_page_count[new_idx]++;
......@@ -844,9 +866,11 @@ Dbtup::disk_page_prealloc_callback_common(Signal* signal,
void
Dbtup::disk_page_set_dirty(PagePtr pagePtr)
{
ljam();
Uint32 idx = pagePtr.p->list_index;
if ((idx & 0x8000) == 0)
{
ljam();
/**
* Already in dirty list
*/
......@@ -878,7 +902,6 @@ Dbtup::disk_page_set_dirty(PagePtr pagePtr)
Uint32 used = pagePtr.p->uncommitted_used_space;
if (unlikely(pagePtr.p->m_restart_seq != globalData.m_restart_seq))
{
jam();
restart_setup_page(alloc, pagePtr);
idx = alloc.calc_page_free_bits(free);
used = 0;
......@@ -903,14 +926,14 @@ Dbtup::disk_page_set_dirty(PagePtr pagePtr)
// Make sure no one will allocate it...
tsman.unmap_page(&key, MAX_FREE_LIST - 1);
jamEntry();
ljamEntry();
}
void
Dbtup::disk_page_unmap_callback(Uint32 when,
Uint32 page_id, Uint32 dirty_count)
{
jamEntry();
ljamEntry();
Ptr<GlobalPage> gpage;
m_global_page_pool.getPtr(gpage, page_id);
PagePtr pagePtr;
......@@ -922,6 +945,7 @@ Dbtup::disk_page_unmap_callback(Uint32 when,
type != File_formats::PT_Tup_varsize_page) ||
f_undo_done == false))
{
ljam();
return ;
}
......@@ -941,7 +965,7 @@ Dbtup::disk_page_unmap_callback(Uint32 when,
/**
* Before pageout
*/
jam();
ljam();
if (DBG_DISK)
{
......@@ -962,7 +986,7 @@ Dbtup::disk_page_unmap_callback(Uint32 when,
if (dirty_count == 0)
{
jam();
ljam();
pagePtr.p->list_index = idx | 0x8000;
Local_key key;
......@@ -980,7 +1004,7 @@ Dbtup::disk_page_unmap_callback(Uint32 when,
fragPtr.p->m_tablespace_id);
tsman.unmap_page(&key, idx);
jamEntry();
ljamEntry();
}
}
else if (when == 1)
......@@ -988,7 +1012,7 @@ Dbtup::disk_page_unmap_callback(Uint32 when,
/**
* After page out
*/
jam();
ljam();
Local_key key;
key.m_page_no = pagePtr.p->m_page_no;
......@@ -1018,6 +1042,7 @@ Dbtup::disk_page_unmap_callback(Uint32 when,
<< endl;
}
tsman.update_page_free_bits(&key, alloc.calc_page_free_bits(real_free));
ljamEntry();
}
}
......@@ -1026,6 +1051,7 @@ Dbtup::disk_page_alloc(Signal* signal,
Tablerec* tabPtrP, Fragrecord* fragPtrP,
Local_key* key, PagePtr pagePtr, Uint32 gci)
{
ljam();
Uint32 logfile_group_id= fragPtrP->m_logfile_group_id;
Disk_alloc_info& alloc= fragPtrP->m_disk_alloc_info;
......@@ -1056,6 +1082,7 @@ Dbtup::disk_page_free(Signal *signal,
Tablerec *tabPtrP, Fragrecord * fragPtrP,
Local_key* key, PagePtr pagePtr, Uint32 gci)
{
ljam();
if (DBG_DISK)
ndbout << " disk_page_free " << *key << endl;
......@@ -1108,6 +1135,7 @@ Dbtup::disk_page_free(Signal *signal,
if (old_idx != new_idx)
{
ljam();
ddassert(extentPtr.p->m_free_page_count[old_idx]);
extentPtr.p->m_free_page_count[old_idx]--;
extentPtr.p->m_free_page_count[new_idx]++;
......@@ -1134,6 +1162,7 @@ void
Dbtup::disk_page_abort_prealloc(Signal *signal, Fragrecord* fragPtrP,
Local_key* key, Uint32 sz)
{
ljam();
Page_cache_client::Request req;
req.m_callback.m_callbackData= sz;
req.m_callback.m_callbackFunction =
......@@ -1143,13 +1172,17 @@ Dbtup::disk_page_abort_prealloc(Signal *signal, Fragrecord* fragPtrP,
memcpy(&req.m_page, key, sizeof(Local_key));
int res= m_pgman.get_page(signal, req, flags);
jamEntry();
ljamEntry();
switch(res)
{
case 0:
ljam();
break;
case -1:
ndbrequire(false);
break;
default:
ljam();
Ptr<GlobalPage> gpage;
m_global_page_pool.getPtr(gpage, (Uint32)res);
PagePtr pagePtr;
......@@ -1165,7 +1198,7 @@ Dbtup::disk_page_abort_prealloc_callback(Signal* signal,
Uint32 sz, Uint32 page_id)
{
//ndbout_c("disk_alloc_page_callback id: %d", page_id);
ljamEntry();
Ptr<GlobalPage> gpage;
m_global_page_pool.getPtr(gpage, page_id);
......@@ -1189,7 +1222,7 @@ Dbtup::disk_page_abort_prealloc_callback_1(Signal* signal,
PagePtr pagePtr,
Uint32 sz)
{
jam();
ljam();
disk_page_set_dirty(pagePtr);
Disk_alloc_info& alloc= fragPtrP->m_disk_alloc_info;
......@@ -1208,12 +1241,14 @@ Dbtup::disk_page_abort_prealloc_callback_1(Signal* signal,
c_extent_pool.getPtr(extentPtr, ext);
if (old_idx != new_idx)
{
ljam();
ddassert(extentPtr.p->m_free_page_count[old_idx]);
extentPtr.p->m_free_page_count[old_idx]--;
extentPtr.p->m_free_page_count[new_idx]++;
if (old_idx == page_idx)
{
ljam();
ArrayPool<Page> *pool= (ArrayPool<Page>*)&m_global_page_pool;
LocalDLList<Page> old_list(*pool, alloc.m_dirty_pages[old_idx]);
LocalDLList<Page> new_list(*pool, alloc.m_dirty_pages[new_idx]);
......@@ -1223,6 +1258,7 @@ Dbtup::disk_page_abort_prealloc_callback_1(Signal* signal,
}
else
{
ljam();
pagePtr.p->list_index = new_idx | 0x8000;
}
}
......@@ -1272,7 +1308,7 @@ Dbtup::disk_page_alloc_extent_log_buffer_callback(Signal* signal,
Uint64 lsn= lgman.add_entry(c, 1);
tsman.update_lsn(&key, lsn);
jamEntry();
ljamEntry();
}
#endif
......@@ -1280,6 +1316,7 @@ Uint64
Dbtup::disk_page_undo_alloc(Page* page, const Local_key* key,
Uint32 sz, Uint32 gci, Uint32 logfile_group_id)
{
ljam();
Logfile_client lgman(this, c_lgman, logfile_group_id);
Disk_undo::Alloc alloc;
......@@ -1291,7 +1328,7 @@ Dbtup::disk_page_undo_alloc(Page* page, const Local_key* key,
Uint64 lsn= lgman.add_entry(c, 1);
m_pgman.update_lsn(* key, lsn);
jamEntry();
ljamEntry();
return lsn;
}
......@@ -1301,6 +1338,7 @@ Dbtup::disk_page_undo_update(Page* page, const Local_key* key,
const Uint32* src, Uint32 sz,
Uint32 gci, Uint32 logfile_group_id)
{
ljam();
Logfile_client lgman(this, c_lgman, logfile_group_id);
Disk_undo::Update update;
......@@ -1321,7 +1359,7 @@ Dbtup::disk_page_undo_update(Page* page, const Local_key* key,
Uint64 lsn= lgman.add_entry(c, 3);
m_pgman.update_lsn(* key, lsn);
jamEntry();
ljamEntry();
return lsn;
}
......@@ -1331,6 +1369,7 @@ Dbtup::disk_page_undo_free(Page* page, const Local_key* key,
const Uint32* src, Uint32 sz,
Uint32 gci, Uint32 logfile_group_id)
{
ljam();
Logfile_client lgman(this, c_lgman, logfile_group_id);
Disk_undo::Free free;
......@@ -1351,7 +1390,7 @@ Dbtup::disk_page_undo_free(Page* page, const Local_key* key,
Uint64 lsn= lgman.add_entry(c, 3);
m_pgman.update_lsn(* key, lsn);
jamEntry();
ljamEntry();
return lsn;
}
......@@ -1377,7 +1416,7 @@ Dbtup::disk_restart_undo(Signal* signal, Uint64 lsn,
case File_formats::Undofile::UNDO_LCP_FIRST:
case File_formats::Undofile::UNDO_LCP:
{
jam();
ljam();
ndbrequire(len == 3);
Uint32 lcp = ptr[0];
Uint32 tableId = ptr[1] >> 16;
......@@ -1393,7 +1432,7 @@ Dbtup::disk_restart_undo(Signal* signal, Uint64 lsn,
}
case File_formats::Undofile::UNDO_TUP_ALLOC:
{
jam();
ljam();
Disk_undo::Alloc* rec= (Disk_undo::Alloc*)ptr;
preq.m_page.m_page_no = rec->m_page_no;
preq.m_page.m_file_no = rec->m_file_no_page_idx >> 16;
......@@ -1402,7 +1441,7 @@ Dbtup::disk_restart_undo(Signal* signal, Uint64 lsn,
}
case File_formats::Undofile::UNDO_TUP_UPDATE:
{
jam();
ljam();
Disk_undo::Update* rec= (Disk_undo::Update*)ptr;
preq.m_page.m_page_no = rec->m_page_no;
preq.m_page.m_file_no = rec->m_file_no_page_idx >> 16;
......@@ -1411,7 +1450,7 @@ Dbtup::disk_restart_undo(Signal* signal, Uint64 lsn,
}
case File_formats::Undofile::UNDO_TUP_FREE:
{
jam();
ljam();
Disk_undo::Free* rec= (Disk_undo::Free*)ptr;
preq.m_page.m_page_no = rec->m_page_no;
preq.m_page.m_file_no = rec->m_file_no_page_idx >> 16;
......@@ -1423,7 +1462,7 @@ Dbtup::disk_restart_undo(Signal* signal, Uint64 lsn,
*
*/
{
jam();
ljam();
Disk_undo::Create* rec= (Disk_undo::Create*)ptr;
Ptr<Tablerec> tabPtr;
tabPtr.i= rec->m_table;
......@@ -1442,7 +1481,7 @@ Dbtup::disk_restart_undo(Signal* signal, Uint64 lsn,
}
case File_formats::Undofile::UNDO_TUP_DROP:
{
jam();
ljam();
Disk_undo::Drop* rec = (Disk_undo::Drop*)ptr;
Ptr<Tablerec> tabPtr;
tabPtr.i= rec->m_table;
......@@ -1460,14 +1499,14 @@ Dbtup::disk_restart_undo(Signal* signal, Uint64 lsn,
return;
}
case File_formats::Undofile::UNDO_TUP_ALLOC_EXTENT:
jam();
ljam();
case File_formats::Undofile::UNDO_TUP_FREE_EXTENT:
jam();
ljam();
disk_restart_undo_next(signal);
return;
case File_formats::Undofile::UNDO_END:
jam();
ljam();
f_undo_done = true;
return;
default:
......@@ -1480,7 +1519,7 @@ Dbtup::disk_restart_undo(Signal* signal, Uint64 lsn,
int flags = 0;
int res= m_pgman.get_page(signal, preq, flags);
jamEntry();
ljamEntry();
switch(res)
{
case 0:
......@@ -1503,7 +1542,7 @@ Dbtup::disk_restart_undo_next(Signal* signal)
void
Dbtup::disk_restart_lcp_id(Uint32 tableId, Uint32 fragId, Uint32 lcpId)
{
jamEntry();
ljamEntry();
if (lcpId == RNIL)
{
......@@ -1534,23 +1573,23 @@ Dbtup::disk_restart_undo_lcp(Uint32 tableId, Uint32 fragId, Uint32 flag,
if (tabPtr.p->tableStatus == DEFINED)
{
jam();
ljam();
FragrecordPtr fragPtr;
getFragmentrec(fragPtr, fragId, tabPtr.p);
if (!fragPtr.isNull())
{
jam();
ljam();
switch(flag){
case Fragrecord::UC_CREATE:
jam();
ljam();
fragPtr.p->m_undo_complete |= flag;
return;
case Fragrecord::UC_LCP:
jam();
ljam();
if (fragPtr.p->m_undo_complete == 0 &&
fragPtr.p->m_restore_lcp_id == lcpId)
{
jam();
ljam();
fragPtr.p->m_undo_complete |= flag;
if (DBG_UNDO)
ndbout_c("table: %u fragment: %u lcp: %u -> done",
......@@ -1559,7 +1598,7 @@ Dbtup::disk_restart_undo_lcp(Uint32 tableId, Uint32 fragId, Uint32 flag,
return;
case Fragrecord::UC_SET_LCP:
{
jam();
ljam();
if (DBG_UNDO)
ndbout_c("table: %u fragment: %u restore to lcp: %u",
tableId, fragId, lcpId);
......@@ -1580,7 +1619,7 @@ Dbtup::disk_restart_undo_callback(Signal* signal,
Uint32 id,
Uint32 page_id)
{
jamEntry();
ljamEntry();
Ptr<GlobalPage> gpage;
m_global_page_pool.getPtr(gpage, page_id);
PagePtr pagePtr;
......@@ -1594,7 +1633,7 @@ Dbtup::disk_restart_undo_callback(Signal* signal,
pagePtr.p->nextList != RNIL ||
pagePtr.p->prevList != RNIL)
{
jam();
ljam();
update = true;
pagePtr.p->list_index |= 0x8000;
pagePtr.p->nextList = pagePtr.p->prevList = RNIL;
......@@ -1605,7 +1644,7 @@ Dbtup::disk_restart_undo_callback(Signal* signal,
if (tableId >= cnoOfTablerec)
{
jam();
ljam();
if (DBG_UNDO)
ndbout_c("UNDO table> %u", tableId);
disk_restart_undo_next(signal);
......@@ -1616,7 +1655,7 @@ Dbtup::disk_restart_undo_callback(Signal* signal,
if (undo->m_table_ptr.p->tableStatus != DEFINED)
{
jam();
ljam();
if (DBG_UNDO)
ndbout_c("UNDO !defined (%u) ", tableId);
disk_restart_undo_next(signal);
......@@ -1626,7 +1665,7 @@ Dbtup::disk_restart_undo_callback(Signal* signal,
getFragmentrec(undo->m_fragment_ptr, fragId, undo->m_table_ptr.p);
if(undo->m_fragment_ptr.isNull())
{
jam();
ljam();
if (DBG_UNDO)
ndbout_c("UNDO fragment null %u/%u", tableId, fragId);
disk_restart_undo_next(signal);
......@@ -1635,7 +1674,7 @@ Dbtup::disk_restart_undo_callback(Signal* signal,
if (undo->m_fragment_ptr.p->m_undo_complete)
{
jam();
ljam();
if (DBG_UNDO)
ndbout_c("UNDO undo complete %u/%u", tableId, fragId);
disk_restart_undo_next(signal);
......@@ -1654,7 +1693,7 @@ Dbtup::disk_restart_undo_callback(Signal* signal,
if (undo->m_lsn <= lsn)
{
jam();
ljam();
if (DBG_UNDO)
{
ndbout << "apply: " << undo->m_lsn << "(" << lsn << " )"
......@@ -1669,15 +1708,15 @@ Dbtup::disk_restart_undo_callback(Signal* signal,
*/
switch(undo->m_type){
case File_formats::Undofile::UNDO_TUP_ALLOC:
jam();
ljam();
disk_restart_undo_alloc(undo);
break;
case File_formats::Undofile::UNDO_TUP_UPDATE:
jam();
ljam();
disk_restart_undo_update(undo);
break;
case File_formats::Undofile::UNDO_TUP_FREE:
jam();
ljam();
disk_restart_undo_free(undo);
break;
default:
......@@ -1691,13 +1730,13 @@ Dbtup::disk_restart_undo_callback(Signal* signal,
lsn = undo->m_lsn - 1; // make sure undo isn't run again...
m_pgman.update_lsn(undo->m_key, lsn);
jamEntry();
ljamEntry();
disk_restart_undo_page_bits(signal, undo);
}
else if (DBG_UNDO)
{
jam();
ljam();
ndbout << "ignore: " << undo->m_lsn << "(" << lsn << " )"
<< key << " type: " << undo->m_type
<< " tab: " << tableId << endl;
......@@ -1783,7 +1822,7 @@ Dbtup::disk_restart_undo_page_bits(Signal* signal, Apply_undo* undo)
fragPtrP->m_tablespace_id);
tsman.restart_undo_page_free_bits(&undo->m_key, new_bits);
jamEntry();
ljamEntry();
}
int
......@@ -1799,7 +1838,7 @@ Dbtup::disk_restart_alloc_extent(Uint32 tableId, Uint32 fragId,
getFragmentrec(fragPtr, fragId, tabPtr.p);
if (fragPtr.p->m_undo_complete & Fragrecord::UC_CREATE)
{
jam();
ljam();
return -1;
}
......@@ -1819,7 +1858,7 @@ Dbtup::disk_restart_alloc_extent(Uint32 tableId, Uint32 fragId,
if (alloc.m_curr_extent_info_ptr_i != RNIL)
{
jam();
ljam();
Ptr<Extent_info> old;
c_extent_pool.getPtr(old, alloc.m_curr_extent_info_ptr_i);
ndbassert(old.p->m_free_matrix_pos == RNIL);
......@@ -1846,7 +1885,7 @@ void
Dbtup::disk_restart_page_bits(Uint32 tableId, Uint32 fragId,
const Local_key*, Uint32 bits)
{
jam();
ljam();
TablerecPtr tabPtr;
FragrecordPtr fragPtr;
tabPtr.i = tableId;
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment
GitLab Nexedi Edition | About GitLab | About Nexedi | 沪ICP备2021021310号-2 | 沪ICP备2021021310号-7