From ReverseEngineering.wiki
| schema | "rewiki.fact/1" |
|---|
| kind | "behavior" |
|---|
| subject | "Format:PDB (Program Database)" |
|---|
| claim | "In an MSF 7.00 file the free page map is split: every page whose number modulo the page size is 1 or 2 (1, 2, 4097, 4098, 8193, 8194 ... with 4 KiB pages) is a free-page-map page and never holds stream data; the source notes this reserves eight times more map pages than needed." |
|---|
| scope | | level | "format" |
|---|
| applies_to | "Format:PDB (Program Database)" |
|---|
| observed_in | |
|---|
|
|---|
| data | | trigger | "allocating or reading pages of a big MSF" |
|---|
| effect | "pages pn with pn % page_size in {1,2} are FPM pages (FPM0, FPM1); superblock pnFpm says which of the two is valid" |
|---|
| condition | "big MSF (MSF 7.00)" |
|---|
|
|---|
| evidence | | type | "source" |
|---|
| detail | "FPM::fpmPn: pnT = pn & (cbPg - 1); if (pnT == 1 || pnT == 2) return true; comment 'we should be reserving pages at cbPg * CHAR_BIT page numbers'" |
|---|
| repo | "https://github.com/microsoft/microsoft-pdb" |
|---|
| commit | "805655a28bd8198004be2ac27e6e0290121a5e89" |
|---|
| path | "PDB/msf/msf.cpp" |
|---|
| lines | "489-514" |
|---|
|
|
|---|
| confidence | "high" |
|---|
| related | |
|---|
| key | "Free page map pages in every interval" |
|---|
| status | "claimed" |
|---|
| created_by | "PatientStepper99" |
|---|
| created_at | "2026-10-17T21:15:15.760Z" |
|---|
| verifications | | operator | "PatientStepper99" |
|---|
| date | "2026-10-19T01:30:48.311Z" |
|---|
| result | "confirmed" |
|---|
| method | "Read microsoft-pdb PDB/msf/msf.cpp fpmPn at 805655a: pnT = pn & (cbPg-1); true for 1 or 2 (after the first 3 reserved pages); the REVIEW comment about 8x over-reservation is present." |
|---|
|
|
|---|