diff options
Diffstat (limited to 'db2/log/log_put.c')
-rw-r--r-- | db2/log/log_put.c | 210 |
1 files changed, 121 insertions, 89 deletions
diff --git a/db2/log/log_put.c b/db2/log/log_put.c index 225595f33e..92d9563301 100644 --- a/db2/log/log_put.c +++ b/db2/log/log_put.c @@ -7,7 +7,7 @@ #include "config.h" #ifndef lint -static const char sccsid[] = "@(#)log_put.c 10.14 (Sleepycat) 9/23/97"; +static const char sccsid[] = "@(#)log_put.c 10.20 (Sleepycat) 11/2/97"; #endif /* not lint */ #ifndef NO_SYSTEM_INCLUDES @@ -29,9 +29,10 @@ static const char sccsid[] = "@(#)log_put.c 10.14 (Sleepycat) 9/23/97"; #include "common_ext.h" static int __log_fill __P((DB_LOG *, void *, u_int32_t)); +static int __log_flush __P((DB_LOG *, const DB_LSN *)); static int __log_newfd __P((DB_LOG *)); -static int __log_write __P((DB_LOG *, void *, u_int32_t)); static int __log_putr __P((DB_LOG *, const DBT *, u_int32_t)); +static int __log_write __P((DB_LOG *, void *, u_int32_t)); /* * log_put -- @@ -63,11 +64,8 @@ log_put(dblp, lsn, dbt, flags) } LOCK_LOGREGION(dblp); - ret = __log_put(dblp, lsn, dbt, flags); - UNLOCK_LOGREGION(dblp); - return (ret); } @@ -102,14 +100,10 @@ __log_put(dblp, lsn, dbt, flags) "log_put: record larger than maximum file size"); return (EINVAL); } - if (lp->b_off != 0) { - if ((ret = __log_write(dblp, lp->buf, lp->b_off)) != 0) - return (ret); - if ((ret = __db_fsync(dblp->lfd)) != 0) - return (ret); - lp->s_lsn.file = lp->lsn.file; - lp->s_lsn.offset = lp->lsn.offset - 1; - } + + /* Flush the log. */ + if ((ret = __log_flush(dblp, NULL)) != 0) + return (ret); /* * Save the last known offset from the previous file, we'll @@ -117,9 +111,15 @@ __log_put(dblp, lsn, dbt, flags) */ lastoff = lp->lsn.offset; + /* Point the current LSN to the new file. */ ++lp->lsn.file; lp->lsn.offset = 0; + + /* Reset the file write offset. */ lp->w_off = 0; + + /* Reset the first-unwritten LSN for the buffer. */ + lp->uw_lsn = lp->lsn; } else lastoff = 0; @@ -149,56 +149,54 @@ __log_put(dblp, lsn, dbt, flags) * Put out the checkpoint record (above). * Save the LSN of the checkpoint in the shared region. * Append the set of file name information into the log. - * Flush the current buffer contents to disk. - * Sync the log to disk. - * Save the time the checkpoint was written. - * Reset the bytes written since the last checkpoint. */ if (flags == DB_CHECKPOINT) { lp->c_lsn = *lsn; for (fnp = SH_TAILQ_FIRST(&dblp->lp->fq, __fname); fnp != NULL; fnp = SH_TAILQ_NEXT(fnp, q, __fname)) { - t.data = ADDR(dblp, fnp->name_off); + t.data = R_ADDR(dblp, fnp->name_off); t.size = strlen(t.data) + 1; memset(&fid_dbt, 0, sizeof(fid_dbt)); - fid_dbt.data = ADDR(dblp, fnp->fileid_off); + fid_dbt.data = R_ADDR(dblp, fnp->fileid_off); fid_dbt.size = DB_FILE_ID_LEN; if ((ret = __log_register_log(dblp, NULL, &r_unused, 0, &t, &fid_dbt, fnp->id, fnp->s_type)) != 0) return (ret); } - if (lp->b_off != 0 && - (ret = __log_write(dblp, lp->buf, lp->b_off)) != 0) - return (ret); - (void)time(&lp->chkpt); - lp->written = 0; - - if ((ret = __db_fsync(dblp->lfd)) != 0) - return (ret); - lp->s_lsn.file = lp->lsn.file; - lp->s_lsn.offset = lp->lsn.offset - 1; } - /* We always flush on a checkpoint. */ - if (flags == DB_FLUSH || flags == DB_CHECKPOINT) { - if (lp->b_off != 0 && - (ret = __log_write(dblp, lp->buf, lp->b_off)) != 0) + /* + * On a checkpoint or when flush is requested, we: + * Flush the current buffer contents to disk. + * Sync the log to disk. + */ + if (flags == DB_FLUSH || flags == DB_CHECKPOINT) + if ((ret = __log_flush(dblp, NULL)) != 0) return (ret); - if ((ret = __db_fsync(dblp->lfd)) != 0) - return (ret); - lp->s_lsn.file = lp->lsn.file; - lp->s_lsn.offset = lp->lsn.offset - 1; + /* + * On a checkpoint, we: + * Save the time the checkpoint was written. + * Reset the bytes written since the last checkpoint. + */ + if (flags == DB_CHECKPOINT) { + (void)time(&lp->chkpt); + lp->stat.st_wc_bytes = lp->stat.st_wc_mbytes = 0; } /* - * If we just did I/O, i.e., this LSN could have spanned the start of - * the in-core buffer, we remember it so that we can flush correctly - * during a sync. + * When an application calls the log_flush routine, we need to figure + * out if the current buffer needs to be flushed. The problem is that + * if a record spans buffers, it's possible for the record continued + * in the current buffer to have begun in a previous buffer. Each time + * we write a buffer, we update the first-unwritten LSN to point to the + * first LSN after that written buffer. If we have a spanning record, + * correct that value to be the LSN that started it all, here. */ if (lsn->offset < lp->w_off && lsn->offset + lp->len > lp->w_off) - lp->span_lsn = *lsn; + lp->uw_lsn = *lsn; + return (0); } @@ -248,6 +246,24 @@ log_flush(dblp, lsn) DB_LOG *dblp; const DB_LSN *lsn; { + int ret; + + LOCK_LOGREGION(dblp); + ret = __log_flush(dblp, lsn); + UNLOCK_LOGREGION(dblp); + return (ret); +} + +/* + * __log_flush -- + * Write all records less than or equal to the specified LSN; internal + * version. + */ +static int +__log_flush(dblp, lsn) + DB_LOG *dblp; + const DB_LSN *lsn; +{ DB_LSN t_lsn; LOG *lp; int ret; @@ -255,60 +271,64 @@ log_flush(dblp, lsn) ret = 0; lp = dblp->lp; - LOCK_LOGREGION(dblp); - - /* If no LSN specified, flush the entire log. */ + /* + * If no LSN specified, flush the entire log by setting the flush LSN + * to the last LSN written in the log. Otherwise, check that the LSN + * isn't a non-existent record for the log. + */ if (lsn == NULL) { t_lsn.file = lp->lsn.file; t_lsn.offset = lp->lsn.offset - lp->len; lsn = &t_lsn; - } - - /* If it's a non-existent record, it's an error. */ - if (lsn->file > lp->lsn.file || - (lsn->file == lp->lsn.file && lsn->offset > lp->lsn.offset)) { - __db_err(dblp->dbenv, "log_flush: LSN past current end-of-log"); - ret = EINVAL; - goto ret1; - } - - /* - * If it's from a previous file, we're done because we sync each - * file when we move to a new one. - */ - if (lsn->file < lp->lsn.file) - goto ret1; + } else + if (lsn->file > lp->lsn.file || + (lsn->file == lp->lsn.file && + lsn->offset > lp->lsn.offset - lp->len)) { + __db_err(dblp->dbenv, + "log_flush: LSN past current end-of-log"); + return (EINVAL); + } /* - * If it's less than the last-sync'd offset, we've already sync'd - * this LSN. + * If the LSN is less than the last-sync'd LSN, we're done. Note, + * the last-sync LSN saved in s_lsn is the LSN of the first byte + * that has not yet been written to disk, so the test is <, not <=. */ - if (lsn->offset <= lp->s_lsn.offset) - goto ret1; + if (lsn->file < lp->s_lsn.file || + (lsn->file == lp->s_lsn.file && lsn->offset < lp->s_lsn.offset)) + return (0); /* * We may need to write the current buffer. We have to write the - * current buffer if the sync LSN is greater than or equal to the - * saved spanning-LSN. + * current buffer if the flush LSN is greater than or equal to the + * first-unwritten LSN (uw_lsn). If we write the buffer, then we + * update the first-unwritten LSN. */ - if (lsn->file >= lp->span_lsn.file && - lsn->offset >= lp->span_lsn.offset) + if (lp->b_off != 0 && + lsn->file >= lp->uw_lsn.file && lsn->offset >= lp->uw_lsn.offset) if ((ret = __log_write(dblp, lp->buf, lp->b_off)) != 0) - goto ret1; + return (ret); - /* Acquire a file descriptor if we don't have one. */ - if (dblp->lfname != dblp->lp->lsn.file && - (ret = __log_newfd(dblp)) != 0) - goto ret1; + /* + * It's possible that this thread may never have written to this log + * file. Acquire a file descriptor if we don't already have one. + */ + if (dblp->lfname != dblp->lp->lsn.file) + if ((ret = __log_newfd(dblp)) != 0) + return (ret); + /* Sync all writes to disk. */ if ((ret = __db_fsync(dblp->lfd)) != 0) - goto ret1; + return (ret); + ++lp->stat.st_scount; - lp->s_lsn.file = lp->lsn.file; - lp->s_lsn.offset = lsn->offset; + /* + * Set the last-synced LSN, the first LSN after the last record + * that we know is on disk. + */ + lp->s_lsn = lp->uw_lsn; -ret1: UNLOCK_LOGREGION(dblp); - return (ret); + return (0); } /* @@ -385,17 +405,32 @@ __log_write(dblp, addr, len) * Seek to the offset in the file (someone may have written it * since we last did). */ - if ((ret = __db_lseek(dblp->lfd, 0, 0, lp->w_off, SEEK_SET)) != 0) + if ((ret = __db_seek(dblp->lfd, 0, 0, lp->w_off, SEEK_SET)) != 0) return (ret); if ((ret = __db_write(dblp->lfd, addr, len, &nw)) != 0) return (ret); if (nw != (int32_t)len) return (EIO); - /* Update the seek offset and reset the buffer offset. */ + /* + * Reset the buffer offset, update the seek offset, and update the + * first-unwritten LSN. + */ lp->b_off = 0; lp->w_off += len; - lp->written += len; + lp->uw_lsn.file = lp->lsn.file; + lp->uw_lsn.offset = lp->w_off; + + /* Update written statistics. */ + if ((lp->stat.st_w_bytes += len) >= MEGABYTE) { + lp->stat.st_w_bytes -= MEGABYTE; + ++lp->stat.st_w_mbytes; + } + if ((lp->stat.st_wc_bytes += len) >= MEGABYTE) { + lp->stat.st_wc_bytes -= MEGABYTE; + ++lp->stat.st_wc_mbytes; + } + ++lp->stat.st_wcount; return (0); } @@ -415,11 +450,8 @@ log_file(dblp, lsn, namep, len) char *p; LOCK_LOGREGION(dblp); - ret = __log_name(dblp, lsn->file, &p); - UNLOCK_LOGREGION(dblp); - if (ret != 0) return (ret); @@ -429,7 +461,7 @@ log_file(dblp, lsn, namep, len) return (ENOMEM); } (void)strcpy(namep, p); - free(p); + __db_free(p); return (0); } @@ -455,7 +487,7 @@ __log_newfd(dblp) dblp->lfname = dblp->lp->lsn.file; if ((ret = __log_name(dblp, dblp->lfname, &p)) != 0) return (ret); - if ((ret = __db_fdopen(p, + if ((ret = __db_open(p, DB_CREATE | DB_SEQUENTIAL, DB_CREATE | DB_SEQUENTIAL, dblp->lp->persist.mode, &dblp->lfd)) != 0) @@ -472,14 +504,14 @@ __log_newfd(dblp) * PUBLIC: int __log_name __P((DB_LOG *, int, char **)); */ int -__log_name(dblp, fileno, namep) +__log_name(dblp, filenumber, namep) DB_LOG *dblp; char **namep; - int fileno; + int filenumber; { char name[sizeof(LFNAME) + 10]; - (void)snprintf(name, sizeof(name), LFNAME, fileno); + (void)snprintf(name, sizeof(name), LFNAME, filenumber); return (__db_appname(dblp->dbenv, DB_APP_LOG, dblp->dir, name, NULL, namep)); } |