Commit 2ce4d4c9 authored by Jack Steiner's avatar Jack Steiner Committed by Linus Torvalds

gru: fix GRU interrupt race at deallocate

Fix a race where an interrupt could be received for a GRU context that has
been deallocated.
Signed-off-by: default avatarJack Steiner <steiner@sgi.com>
Signed-off-by: default avatarAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: default avatarLinus Torvalds <torvalds@linux-foundation.org>
parent 74ccd095
...@@ -360,7 +360,8 @@ static void gru_preload_tlb(struct gru_state *gru, ...@@ -360,7 +360,8 @@ static void gru_preload_tlb(struct gru_state *gru,
* < 0 = error code * < 0 = error code
* *
*/ */
static int gru_try_dropin(struct gru_thread_state *gts, static int gru_try_dropin(struct gru_state *gru,
struct gru_thread_state *gts,
struct gru_tlb_fault_handle *tfh, struct gru_tlb_fault_handle *tfh,
struct gru_instruction_bits *cbk) struct gru_instruction_bits *cbk)
{ {
...@@ -432,7 +433,7 @@ static int gru_try_dropin(struct gru_thread_state *gts, ...@@ -432,7 +433,7 @@ static int gru_try_dropin(struct gru_thread_state *gts,
} }
if (unlikely(cbe) && pageshift == PAGE_SHIFT) { if (unlikely(cbe) && pageshift == PAGE_SHIFT) {
gru_preload_tlb(gts->ts_gru, gts, atomic, vaddr, asid, write, tlb_preload_count, tfh, cbe); gru_preload_tlb(gru, gts, atomic, vaddr, asid, write, tlb_preload_count, tfh, cbe);
gru_flush_cache_cbe(cbe); gru_flush_cache_cbe(cbe);
} }
...@@ -442,7 +443,7 @@ static int gru_try_dropin(struct gru_thread_state *gts, ...@@ -442,7 +443,7 @@ static int gru_try_dropin(struct gru_thread_state *gts,
gru_dbg(grudev, gru_dbg(grudev,
"%s: gid %d, gts 0x%p, tfh 0x%p, vaddr 0x%lx, asid 0x%x, indexway 0x%x," "%s: gid %d, gts 0x%p, tfh 0x%p, vaddr 0x%lx, asid 0x%x, indexway 0x%x,"
" rw %d, ps %d, gpa 0x%lx\n", " rw %d, ps %d, gpa 0x%lx\n",
atomic ? "atomic" : "non-atomic", gts->ts_gru->gs_gid, gts, tfh, vaddr, asid, atomic ? "atomic" : "non-atomic", gru->gs_gid, gts, tfh, vaddr, asid,
indexway, write, pageshift, gpa); indexway, write, pageshift, gpa);
STAT(tlb_dropin); STAT(tlb_dropin);
return 0; return 0;
...@@ -528,6 +529,7 @@ static irqreturn_t gru_intr(int chiplet, int blade) ...@@ -528,6 +529,7 @@ static irqreturn_t gru_intr(int chiplet, int blade)
struct gru_tlb_fault_map imap, dmap; struct gru_tlb_fault_map imap, dmap;
struct gru_thread_state *gts; struct gru_thread_state *gts;
struct gru_tlb_fault_handle *tfh = NULL; struct gru_tlb_fault_handle *tfh = NULL;
struct completion *cmp;
int cbrnum, ctxnum; int cbrnum, ctxnum;
STAT(intr); STAT(intr);
...@@ -547,9 +549,11 @@ static irqreturn_t gru_intr(int chiplet, int blade) ...@@ -547,9 +549,11 @@ static irqreturn_t gru_intr(int chiplet, int blade)
for_each_cbr_in_tfm(cbrnum, dmap.fault_bits) { for_each_cbr_in_tfm(cbrnum, dmap.fault_bits) {
STAT(intr_cbr); STAT(intr_cbr);
complete(gru->gs_blade->bs_async_wq); cmp = gru->gs_blade->bs_async_wq;
if (cmp)
complete(cmp);
gru_dbg(grudev, "gid %d, cbr_done %d, done %d\n", gru_dbg(grudev, "gid %d, cbr_done %d, done %d\n",
gru->gs_gid, cbrnum, gru->gs_blade->bs_async_wq->done); gru->gs_gid, cbrnum, cmp ? cmp->done : -1);
} }
for_each_cbr_in_tfm(cbrnum, imap.fault_bits) { for_each_cbr_in_tfm(cbrnum, imap.fault_bits) {
...@@ -566,6 +570,12 @@ static irqreturn_t gru_intr(int chiplet, int blade) ...@@ -566,6 +570,12 @@ static irqreturn_t gru_intr(int chiplet, int blade)
ctxnum = tfh->ctxnum; ctxnum = tfh->ctxnum;
gts = gru->gs_gts[ctxnum]; gts = gru->gs_gts[ctxnum];
/* Spurious interrupts can cause this. Ignore. */
if (!gts) {
STAT(intr_spurious);
continue;
}
/* /*
* This is running in interrupt context. Trylock the mmap_sem. * This is running in interrupt context. Trylock the mmap_sem.
* If it fails, retry the fault in user context. * If it fails, retry the fault in user context.
...@@ -573,7 +583,7 @@ static irqreturn_t gru_intr(int chiplet, int blade) ...@@ -573,7 +583,7 @@ static irqreturn_t gru_intr(int chiplet, int blade)
if (!gts->ts_force_cch_reload && if (!gts->ts_force_cch_reload &&
down_read_trylock(&gts->ts_mm->mmap_sem)) { down_read_trylock(&gts->ts_mm->mmap_sem)) {
gts->ustats.fmm_tlbdropin++; gts->ustats.fmm_tlbdropin++;
gru_try_dropin(gts, tfh, NULL); gru_try_dropin(gru, gts, tfh, NULL);
up_read(&gts->ts_mm->mmap_sem); up_read(&gts->ts_mm->mmap_sem);
} else { } else {
tfh_user_polling_mode(tfh); tfh_user_polling_mode(tfh);
...@@ -619,7 +629,7 @@ static int gru_user_dropin(struct gru_thread_state *gts, ...@@ -619,7 +629,7 @@ static int gru_user_dropin(struct gru_thread_state *gts,
wait_event(gms->ms_wait_queue, wait_event(gms->ms_wait_queue,
atomic_read(&gms->ms_range_active) == 0); atomic_read(&gms->ms_range_active) == 0);
prefetchw(tfh); /* Helps on hdw, required for emulator */ prefetchw(tfh); /* Helps on hdw, required for emulator */
ret = gru_try_dropin(gts, tfh, cb); ret = gru_try_dropin(gts->ts_gru, gts, tfh, cb);
if (ret <= 0) if (ret <= 0)
return ret; return ret;
STAT(call_os_wait_queue); STAT(call_os_wait_queue);
......
...@@ -66,6 +66,7 @@ static int statistics_show(struct seq_file *s, void *p) ...@@ -66,6 +66,7 @@ static int statistics_show(struct seq_file *s, void *p)
printstat(s, intr); printstat(s, intr);
printstat(s, intr_cbr); printstat(s, intr_cbr);
printstat(s, intr_tfh); printstat(s, intr_tfh);
printstat(s, intr_spurious);
printstat(s, intr_mm_lock_failed); printstat(s, intr_mm_lock_failed);
printstat(s, call_os); printstat(s, call_os);
printstat(s, call_os_wait_queue); printstat(s, call_os_wait_queue);
......
...@@ -192,6 +192,7 @@ struct gru_stats_s { ...@@ -192,6 +192,7 @@ struct gru_stats_s {
atomic_long_t intr; atomic_long_t intr;
atomic_long_t intr_cbr; atomic_long_t intr_cbr;
atomic_long_t intr_tfh; atomic_long_t intr_tfh;
atomic_long_t intr_spurious;
atomic_long_t intr_mm_lock_failed; atomic_long_t intr_mm_lock_failed;
atomic_long_t call_os; atomic_long_t call_os;
atomic_long_t call_os_wait_queue; atomic_long_t call_os_wait_queue;
......
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