diff for duplicates of <20151102214424.GJ10656@dastard> diff --git a/a/1.txt b/N1/1.txt index 858a90c..23691ca 100644 --- a/a/1.txt +++ b/N1/1.txt @@ -40,14 +40,14 @@ pages and then writing to them again while cache hot. Oh, hell - dax_clear_blocks() is stupidly slow. A profile shows this loop spending most of the CPU time: - � � jbe ea - � de: clflus %ds:(%rax) - 84.67 � add %rsi,%rax - � cmp %rax,%rdx - � � ja de - � ea: add %r13,-0x38(%rbp) - � sub %r12,%r14 - � sub %r12,-0x40(%rbp) + ¿ ¿ jbe ea + ¿ de: clflus %ds:(%rax) + 84.67 ¿ add %rsi,%rax + ¿ cmp %rax,%rdx + ¿ ¿ ja de + ¿ ea: add %r13,-0x38(%rbp) + ¿ sub %r12,%r14 + ¿ sub %r12,-0x40(%rbp) That is the overhead of __arch_wb_cache_pmem() i.e. issuing CPU cache flushes after each memset. @@ -121,3 +121,8 @@ Dave. -- Dave Chinner david@fromorbit.com + +_______________________________________________ +xfs mailing list +xfs@oss.sgi.com +http://oss.sgi.com/mailman/listinfo/xfs diff --git a/a/content_digest b/N1/content_digest index 3b97888..6ec60b7 100644 --- a/a/content_digest +++ b/N1/content_digest @@ -9,12 +9,12 @@ "Subject\0Re: [PATCH 3/6] xfs: Don't use unwritten extents for DAX\0" "Date\0Tue, 3 Nov 2015 08:44:24 +1100\0" "To\0Brian Foster <bfoster@redhat.com>\0" - "Cc\0ross.zwisler@linux.intel.com" - jack@suse.cz + "Cc\0jack@suse.cz" + linux-nvdimm@lists.01.org xfs@oss.sgi.com linux-fsdevel@vger.kernel.org - dan.j.williams@intel.com - " linux-nvdimm@lists.01.org\0" + ross.zwisler@linux.intel.com + " dan.j.williams@intel.com\0" "\00:1\0" "b\0" "[add people to the cc list]\n" @@ -59,14 +59,14 @@ "Oh, hell - dax_clear_blocks() is stupidly slow. A profile shows this\n" "loop spending most of the CPU time:\n" "\n" - " \303\257\302\277\302\275 \303\257\302\277\302\275 jbe ea\n" - " \303\257\302\277\302\275 de: clflus %ds:(%rax)\n" - " 84.67 \303\257\302\277\302\275 add %rsi,%rax\n" - " \303\257\302\277\302\275 cmp %rax,%rdx\n" - " \303\257\302\277\302\275 \303\257\302\277\302\275 ja de\n" - " \303\257\302\277\302\275 ea: add %r13,-0x38(%rbp)\n" - " \303\257\302\277\302\275 sub %r12,%r14\n" - " \303\257\302\277\302\275 sub %r12,-0x40(%rbp)\n" + " \302\277 \302\277 jbe ea\n" + " \302\277 de: clflus %ds:(%rax)\n" + " 84.67 \302\277 add %rsi,%rax\n" + " \302\277 cmp %rax,%rdx\n" + " \302\277 \302\277 ja de\n" + " \302\277 ea: add %r13,-0x38(%rbp)\n" + " \302\277 sub %r12,%r14\n" + " \302\277 sub %r12,-0x40(%rbp)\n" "\n" "That is the overhead of __arch_wb_cache_pmem() i.e. issuing CPU\n" "cache flushes after each memset.\n" @@ -139,6 +139,11 @@ "Dave.\n" "-- \n" "Dave Chinner\n" - david@fromorbit.com + "david@fromorbit.com\n" + "\n" + "_______________________________________________\n" + "xfs mailing list\n" + "xfs@oss.sgi.com\n" + http://oss.sgi.com/mailman/listinfo/xfs -e3f5dfaf224d292464838b958739c697b4a02b403f1524113d76e4ebcad8f196 +1ef0d08c9d7f8801c27f93604d9b6b0adf33e0eef852b182e4570293701801b7
diff --git a/a/1.txt b/N2/1.txt index 858a90c..60325d2 100644 --- a/a/1.txt +++ b/N2/1.txt @@ -40,14 +40,14 @@ pages and then writing to them again while cache hot. Oh, hell - dax_clear_blocks() is stupidly slow. A profile shows this loop spending most of the CPU time: - � � jbe ea - � de: clflus %ds:(%rax) - 84.67 � add %rsi,%rax - � cmp %rax,%rdx - � � ja de - � ea: add %r13,-0x38(%rbp) - � sub %r12,%r14 - � sub %r12,-0x40(%rbp) + ¿ ¿ jbe ea + ¿ de: clflus %ds:(%rax) + 84.67 ¿ add %rsi,%rax + ¿ cmp %rax,%rdx + ¿ ¿ ja de + ¿ ea: add %r13,-0x38(%rbp) + ¿ sub %r12,%r14 + ¿ sub %r12,-0x40(%rbp) That is the overhead of __arch_wb_cache_pmem() i.e. issuing CPU cache flushes after each memset. @@ -121,3 +121,7 @@ Dave. -- Dave Chinner david@fromorbit.com +-- +To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in +the body of a message to majordomo@vger.kernel.org +More majordomo info at http://vger.kernel.org/majordomo-info.html diff --git a/a/content_digest b/N2/content_digest index 3b97888..a904192 100644 --- a/a/content_digest +++ b/N2/content_digest @@ -59,14 +59,14 @@ "Oh, hell - dax_clear_blocks() is stupidly slow. A profile shows this\n" "loop spending most of the CPU time:\n" "\n" - " \303\257\302\277\302\275 \303\257\302\277\302\275 jbe ea\n" - " \303\257\302\277\302\275 de: clflus %ds:(%rax)\n" - " 84.67 \303\257\302\277\302\275 add %rsi,%rax\n" - " \303\257\302\277\302\275 cmp %rax,%rdx\n" - " \303\257\302\277\302\275 \303\257\302\277\302\275 ja de\n" - " \303\257\302\277\302\275 ea: add %r13,-0x38(%rbp)\n" - " \303\257\302\277\302\275 sub %r12,%r14\n" - " \303\257\302\277\302\275 sub %r12,-0x40(%rbp)\n" + " \302\277 \302\277 jbe ea\n" + " \302\277 de: clflus %ds:(%rax)\n" + " 84.67 \302\277 add %rsi,%rax\n" + " \302\277 cmp %rax,%rdx\n" + " \302\277 \302\277 ja de\n" + " \302\277 ea: add %r13,-0x38(%rbp)\n" + " \302\277 sub %r12,%r14\n" + " \302\277 sub %r12,-0x40(%rbp)\n" "\n" "That is the overhead of __arch_wb_cache_pmem() i.e. issuing CPU\n" "cache flushes after each memset.\n" @@ -139,6 +139,10 @@ "Dave.\n" "-- \n" "Dave Chinner\n" - david@fromorbit.com + "david@fromorbit.com\n" + "--\n" + "To unsubscribe from this list: send the line \"unsubscribe linux-fsdevel\" in\n" + "the body of a message to majordomo@vger.kernel.org\n" + More majordomo info at http://vger.kernel.org/majordomo-info.html -e3f5dfaf224d292464838b958739c697b4a02b403f1524113d76e4ebcad8f196 +30bf1c3fddebcbec0ed76695f17bf747862a05c187674256a9b0d020dd76b6eb
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.