Patch "sparc64: Handle extremely large kernel TSB range flushes sanely." has been added to the 4.4-stable tree

Patch "sparc64: Handle extremely large kernel TSB range flushes sanely." has been added to the 4.4-stable tree

[gregkh]

This is a note to let you know that I've just added the patch titled

    sparc64: Handle extremely large kernel TSB range flushes sanely.

to the 4.4-stable tree which can be found at:
    http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary

The filename of the patch is:
     sparc64-handle-extremely-large-kernel-tsb-range-flushes-sanely.patch
and it can be found in the queue-4.4 subdirectory.

If you, or anyone else, feels it should not be added to the stable tree,
please let <stable@vger.kernel.org> know about it.


>From foo@baz Sat Nov 19 09:52:37 CET 2016
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 25 Oct 2016 19:43:17 -0700
Subject: sparc64: Handle extremely large kernel TSB range flushes sanely.

From: "David S. Miller" <davem@davemloft.net>


[ Upstream commit 849c498766060a16aad5b0e0d03206726e7d2fa4 ]

If the number of pages we are flushing is more than twice the number
of entries in the TSB, just scan the TSB table for matches rather
than probing each and every page in the range.

Based upon a patch and report by James Clarke.

Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
 arch/sparc/mm/tsb.c |   17 +++++++++++++++++
 1 file changed, 17 insertions(+)

--- a/arch/sparc/mm/tsb.c
+++ b/arch/sparc/mm/tsb.c
@@ -27,6 +27,20 @@ static inline int tag_compare(unsigned l
 	return (tag == (vaddr >> 22));
 }
 
+static void flush_tsb_kernel_range_scan(unsigned long start, unsigned long end)
+{
+	unsigned long idx;
+
+	for (idx = 0; idx < KERNEL_TSB_NENTRIES; idx++) {
+		struct tsb *ent = &swapper_tsb[idx];
+		unsigned long match = idx << 13;
+
+		match |= (ent->tag << 22);
+		if (match >= start && match < end)
+			ent->tag = (1UL << TSB_TAG_INVALID_BIT);
+	}
+}
+
 /* TSB flushes need only occur on the processor initiating the address
  * space modification, not on each cpu the address space has run on.
  * Only the TLB flush needs that treatment.
@@ -36,6 +50,9 @@ void flush_tsb_kernel_range(unsigned lon
 {
 	unsigned long v;
 
+	if ((end - start) >> PAGE_SHIFT >= 2 * KERNEL_TSB_NENTRIES)
+		return flush_tsb_kernel_range_scan(start, end);
+
 	for (v = start; v < end; v += PAGE_SIZE) {
 		unsigned long hash = tsb_hash(v, PAGE_SHIFT,
 					      KERNEL_TSB_NENTRIES);


Patches currently in stable-queue which might be from davem@davemloft.net are

queue-4.4/sparc64-delete-now-unused-user-copy-fixup-functions.patch
queue-4.4/net-__skb_flow_dissect-must-cap-its-return-value.patch
queue-4.4/tcp-take-care-of-truncations-done-by-sk_filter.patch
queue-4.4/net-clear-sk_err_soft-in-sk_clone_lock.patch
queue-4.4/sparc64-convert-copy_in_user-to-accurate-exception-reporting.patch
queue-4.4/sparc64-handle-extremely-large-kernel-tlb-range-flushes-more-gracefully.patch
queue-4.4/sparc-handle-negative-offsets-in-arch_jump_label_transform.patch
queue-4.4/sparc64-delete-__ret_efault.patch
queue-4.4/dctcp-avoid-bogus-doubling-of-cwnd-after-loss.patch
queue-4.4/sparc64-delete-now-unused-user-copy-assembler-helpers.patch
queue-4.4/net-mangle-zero-checksum-in-skb_checksum_help.patch
queue-4.4/sparc64-mm-fix-base-tsb-sizing-when-hugetlb-pages-are-used.patch
queue-4.4/ip6_tunnel-clear-ip6cb-in-ip6tunnel_xmit.patch
queue-4.4/sctp-assign-assoc_id-earlier-in-__sctp_connect.patch
queue-4.4/sparc64-convert-ng4copy_-from-to-_user-to-accurate-exception-reporting.patch
queue-4.4/ipv6-dccp-fix-out-of-bound-access-in-dccp_v6_err.patch
queue-4.4/sparc64-convert-u3copy_-from-to-_user-to-accurate-exception-reporting.patch
queue-4.4/sparc64-convert-ng2copy_-from-to-_user-to-accurate-exception-reporting.patch
queue-4.4/sparc64-fix-illegal-relative-branches-in-hypervisor-patched-tlb-code.patch
queue-4.4/sparc64-convert-gencopy_-from-to-_user-to-accurate-exception-reporting.patch
queue-4.4/sparc64-convert-u1copy_-from-to-_user-to-accurate-exception-reporting.patch
queue-4.4/sparc-don-t-leak-context-bits-into-thread-fault_address.patch
queue-4.4/sparc64-prepare-to-move-to-more-saner-user-copy-exception-handling.patch
queue-4.4/bgmac-stop-clearing-dma-receive-control-register-right-after-it-is-set.patch
queue-4.4/ipv6-dccp-add-missing-bind_conflict-to-dccp_ipv6_mapped.patch
queue-4.4/tcp-fix-potential-memory-corruption.patch
queue-4.4/sparc64-convert-ngcopy_-from-to-_user-to-accurate-exception-reporting.patch
queue-4.4/fib_trie-correct-proc-net-route-off-by-one-error.patch
queue-4.4/sparc64-fix-illegal-relative-branches-in-hypervisor-patched-tlb-cross-call-code.patch
queue-4.4/sparc64-handle-extremely-large-kernel-tsb-range-flushes-sanely.patch
queue-4.4/sparc64-fix-instruction-count-in-comment-for-__hypervisor_flush_tlb_pending.patch
queue-4.4/sparc-serial-sunhv-fix-a-double-lock-bug.patch
queue-4.4/dccp-do-not-send-reset-to-already-closed-sockets.patch
queue-4.4/ipv4-use-new_gw-for-redirect-neigh-lookup.patch
queue-4.4/dccp-fix-out-of-bound-access-in-dccp_v4_err.patch
queue-4.4/sock-fix-sendmmsg-for-partial-sendmsg.patch