diff options
| author | Sergey Bugaev <bugaevc@gmail.com> | 2021-05-06 15:56:27 +0300 |
|---|---|---|
| committer | Samuel Thibault <samuel.thibault@ens-lyon.org> | 2021-05-06 23:58:42 +0200 |
| commit | 7b3815d91a49672864ad3cb2515eba01320d8360 (patch) | |
| tree | 2a178d699f6824bf0f4b762b0423eec098dbb9d8 | |
| parent | a9f1e487350cca14ac7df732ac24437822d4eac9 (diff) | |
libpager: Do not flush in-core pages on offer
pager_offer_page () is documented to may ignore the offered page if
the kernel already has a copy in core (indeed, that's what Mach does).
However the current behavior is the inverse of that: it asks the kernel
to flush (i.e. drop) its in-core copy, and replace it with the offered
one. Fix this by not doing that, and calling memory_object_data_supply ()
directly.
Message-Id: <20210506125631.79117-3-bugaevc@gmail.com>
| -rw-r--r-- | libpager/offer-page.c | 7 |
1 files changed, 0 insertions, 7 deletions
diff --git a/libpager/offer-page.c b/libpager/offer-page.c index ddea236a..0b0ca2eb 100644 --- a/libpager/offer-page.c +++ b/libpager/offer-page.c @@ -34,13 +34,6 @@ pager_offer_page (struct pager *p, goto release_out; short *pm_entry = &p->pagemap[offset / vm_page_size]; - - while (*pm_entry & PM_INCORE) - { - pthread_mutex_unlock (&p->interlock); - pager_flush_some (p, offset, vm_page_size, 1); - pthread_mutex_lock (&p->interlock); - } *pm_entry |= PM_INCORE; memory_object_data_supply (p->memobjcntl, offset, buf, vm_page_size, 0, |