[PATCH 2/6] vmtrunc: restore unmap_vmas zap_bytes

[Hugh Dickins]

The low-latency unmap_vmas patch silently moved the zap_bytes test
after the TLB finish and lockbreak and regather: why?  That not only
makes zap_bytes redundant (might as well use ZAP_BLOCK_SIZE), it makes
the unmap_vmas level redundant too - it's all about saving TLB flushes
when unmapping a series of small vmas.

Move zap_bytes test back before the lockbreak, and delete the curious
comment that a small zap block size doesn't matter: it's true need_flush
prevents TLB flush when no page has been unmapped, but unmapping pages
in small blocks involves many more TLB flushes than in large blocks.

Signed-off-by: Hugh Dickins <hugh@xxxxxxxxxxx>
---

 mm/memory.c |    9 +++------
 1 files changed, 3 insertions(+), 6 deletions(-)

--- trunc1/mm/memory.c	2004-10-03 01:07:01.396013328 +0100
+++ trunc2/mm/memory.c	2004-10-03 01:07:12.957255752 +0100
@@ -506,10 +506,6 @@ static void unmap_page_range(struct mmu_
 }
 
 #ifdef CONFIG_PREEMPT
-/*
- * It's not an issue to have a small zap block size - TLB flushes
- * only happen once normally, due to the tlb->need_flush optimization.
- */
 # define ZAP_BLOCK_SIZE	(8 * PAGE_SIZE)
 #else
 /* No preempt: go for improved straight-line efficiency */
@@ -588,6 +584,9 @@ int unmap_vmas(struct mmu_gather **tlbp,
 
 			start += block;
 			zap_bytes -= block;
+			if ((long)zap_bytes > 0)
+				continue;
+
 			if (!atomic) {
 				int fullmm = tlb_is_full_mm(*tlbp);
 				tlb_finish_mmu(*tlbp, tlb_start, start);
@@ -595,8 +594,6 @@ int unmap_vmas(struct mmu_gather **tlbp,
 				*tlbp = tlb_gather_mmu(mm, fullmm);
 				tlb_start_valid = 0;
 			}
-			if ((long)zap_bytes > 0)
-				continue;
 			zap_bytes = ZAP_BLOCK_SIZE;
 		}
 	}

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/