fabi/xenia-rs
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix(cpu): review fixes — stswi/stswx two-line guard, dcbz/dcbz128 invalidate

Browse Source 
PPCBUG-160 partial: stswi's single invalidate_for_write(ea) only covered
the first cache line; with nb up to 32, the write span can cross a 128-byte
line boundary. Replace with two-call guard:
  first_line = ea & !RESERVATION_MASK
  last_line  = ea.wrapping_add(nb - 1) & !RESERVATION_MASK
  invalidate first; if last != first, invalidate last.

PPCBUG-160 partial: stswx had the same single-call gap; nb from XER[0:6]
can be up to 127 bytes. Same two-call guard applied; wrapped in `if nb > 0`
to guard against nb==0 underflow (XER TBC field is 0 when no bytes to store).

dcbz: zeroes 32 bytes at a 32-byte-aligned EA — touches exactly one 128-byte
cache line; add canonical single-call invalidate guard (was entirely missing).

dcbz128: zeroes 128 bytes at a 128-byte-aligned EA — one full reservation
line; add canonical single-call invalidate guard (was entirely missing).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This commit is contained in:
MechaCat02
2026-05-01 20:47:32 +02:00
parent d75c4edf67
commit c9f194dda1
1 changed files with 21 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

24
crates/xenia-cpu/src/interpreter.rs
View File

@@ -1464,7 +1464,12 @@ fn execute(ctx: &mut PpcContext, mem: &dyn MemoryAccess, instr: &DecodedInstr) -
let mut rs = instr.rs(); let mut rs = instr.rs();
let mut bytes_left = nb; let mut bytes_left = nb;
if let Some(t) = ctx.reservation_table.as_ref().filter(|t| t.is_enabled()) { if let Some(t) = ctx.reservation_table.as_ref().filter(|t| t.is_enabled()) {
if t.has_active_reservers() { t.invalidate_for_write(ea); } if t.has_active_reservers() {
let first_line = ea & !RESERVATION_MASK;
let last_line = ea.wrapping_add(nb - 1) & !RESERVATION_MASK;
t.invalidate_for_write(first_line);
if last_line != first_line { t.invalidate_for_write(last_line); }
}
} }
while bytes_left > 0 { while bytes_left > 0 {
let val = ctx.gpr[rs] as u32; let val = ctx.gpr[rs] as u32;
@@ -1600,6 +1605,9 @@ fn execute(ctx: &mut PpcContext, mem: &dyn MemoryAccess, instr: &DecodedInstr) -
// Zero 32 bytes at effective address // Zero 32 bytes at effective address
let ea = if instr.ra() == 0 { 0u64 } else { ctx.gpr[instr.ra()] }; let ea = if instr.ra() == 0 { 0u64 } else { ctx.gpr[instr.ra()] };
let ea = (ea.wrapping_add(ctx.gpr[instr.rb()]) as u32) & !31; let ea = (ea.wrapping_add(ctx.gpr[instr.rb()]) as u32) & !31;
if let Some(t) = ctx.reservation_table.as_ref().filter(|t| t.is_enabled()) {
if t.has_active_reservers() { t.invalidate_for_write(ea); }
}
for i in 0..8 { for i in 0..8 {
mem.write_u32(ea + i * 4, 0); mem.write_u32(ea + i * 4, 0);
} }
@@ -1609,6 +1617,9 @@ fn execute(ctx: &mut PpcContext, mem: &dyn MemoryAccess, instr: &DecodedInstr) -
// Zero 128 bytes // Zero 128 bytes
let ea = if instr.ra() == 0 { 0u64 } else { ctx.gpr[instr.ra()] }; let ea = if instr.ra() == 0 { 0u64 } else { ctx.gpr[instr.ra()] };
let ea = (ea.wrapping_add(ctx.gpr[instr.rb()]) as u32) & !127; let ea = (ea.wrapping_add(ctx.gpr[instr.rb()]) as u32) & !127;
if let Some(t) = ctx.reservation_table.as_ref().filter(|t| t.is_enabled()) {
if t.has_active_reservers() { t.invalidate_for_write(ea); }
}
for i in 0..32 { for i in 0..32 {
mem.write_u32(ea + i * 4, 0); mem.write_u32(ea + i * 4, 0);
} }
@@ -4418,8 +4429,15 @@ fn execute(ctx: &mut PpcContext, mem: &dyn MemoryAccess, instr: &DecodedInstr) -
let nb = ctx.xer() & 0x7F; let nb = ctx.xer() & 0x7F;
let mut rs = instr.rs(); let mut rs = instr.rs();
let mut bytes_left = nb; let mut bytes_left = nb;
if let Some(t) = ctx.reservation_table.as_ref().filter(|t| t.is_enabled()) { if nb > 0 {
if t.has_active_reservers() { t.invalidate_for_write(ea); } if let Some(t) = ctx.reservation_table.as_ref().filter(|t| t.is_enabled()) {
if t.has_active_reservers() {
let first_line = ea & !RESERVATION_MASK;
let last_line = ea.wrapping_add(nb - 1) & !RESERVATION_MASK;
t.invalidate_for_write(first_line);
if last_line != first_line { t.invalidate_for_write(last_line); }
}
}
} }
while bytes_left > 0 { while bytes_left > 0 {
let val = ctx.gpr[rs] as u32; let val = ctx.gpr[rs] as u32;