Date: Mon, 11 May 2009 05:16:57 +0000 (UTC) From: Alan Cox <alc@FreeBSD.org> To: src-committers@freebsd.org, svn-src-all@freebsd.org, svn-src-head@freebsd.org Subject: svn commit: r191986 - head/sys/kern Message-ID: <200905110516.n4B5Gv7O078608@svn.freebsd.org>
next in thread | raw e-mail | index | archive | help
Author: alc Date: Mon May 11 05:16:57 2009 New Revision: 191986 URL: http://svn.freebsd.org/changeset/base/191986 Log: Revert CVS revision 1.94 (svn r16840). Current pmap implementations don't suffer from the race condition that motivated revision 1.94. Consequently, the work-around that was implemented by revision 1.94 is no longer needed. Moreover, reverting this work-around eliminates the need for vfs_busy_pages() to acquire the page queues lock when preparing a buffer for read. Reviewed by: tegge Modified: head/sys/kern/vfs_bio.c Modified: head/sys/kern/vfs_bio.c ============================================================================== --- head/sys/kern/vfs_bio.c Mon May 11 04:57:40 2009 (r191985) +++ head/sys/kern/vfs_bio.c Mon May 11 05:16:57 2009 (r191986) @@ -3519,7 +3519,8 @@ retry: goto retry; } bogus = 0; - vm_page_lock_queues(); + if (clear_modify) + vm_page_lock_queues(); for (i = 0; i < bp->b_npages; i++) { m = bp->b_pages[i]; @@ -3542,17 +3543,18 @@ retry: * It may not work properly with small-block devices. * We need to find a better way. */ - pmap_remove_all(m); - if (clear_modify) + if (clear_modify) { + pmap_remove_write(m); vfs_page_set_valid(bp, foff, m); - else if (m->valid == VM_PAGE_BITS_ALL && + } else if (m->valid == VM_PAGE_BITS_ALL && (bp->b_flags & B_CACHE) == 0) { bp->b_pages[i] = bogus_page; bogus++; } foff = (foff + PAGE_SIZE) & ~(off_t)PAGE_MASK; } - vm_page_unlock_queues(); + if (clear_modify) + vm_page_unlock_queues(); VM_OBJECT_UNLOCK(obj); if (bogus) pmap_qenter(trunc_page((vm_offset_t)bp->b_data),
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200905110516.n4B5Gv7O078608>