blob: 286fb6269945218b233a8dea8da48f7a190a4ebf [file] [log] [blame]
From 23ad0a96cdb2c388f5fcd1f1e0290ed0a5965d7f Mon Sep 17 00:00:00 2001
From: Paul Zimmerman <Paul.Zimmerman@synopsys.com>
Date: Sat, 12 Feb 2011 14:07:20 -0800
Subject: [PATCH] xhci: Fix errors in the running total calculations in the
TRB math
commit 5807795bd4dececdf553719cc02869e633395787 upstream.
Calculations like
running_total = TRB_MAX_BUFF_SIZE -
(sg_dma_address(sg) & (TRB_MAX_BUFF_SIZE - 1));
if (running_total != 0)
num_trbs++;
are incorrect, because running_total can never be zero, so the if()
expression will never be true. I think the intention was that
running_total be in the range of 0 to TRB_MAX_BUFF_SIZE-1, not 1
to TRB_MAX_BUFF_SIZE. So adding a
running_total &= TRB_MAX_BUFF_SIZE - 1;
fixes the problem.
This patch should be queued for stable kernels back to 2.6.31.
[PG: 34 doesn't have count_isoc_trbs_needed() hence this has
only two of the three upstream running_total changes.]
Signed-off-by: Paul Zimmerman <paulz@synopsys.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
diff --git a/drivers/usb/host/xhci-ring.c b/drivers/usb/host/xhci-ring.c
index 8bf5563..e4afa99 100644
--- a/drivers/usb/host/xhci-ring.c
+++ b/drivers/usb/host/xhci-ring.c
@@ -1711,6 +1711,7 @@ static unsigned int count_sg_trbs_needed(struct xhci_hcd *xhci, struct urb *urb)
/* Scatter gather list entries may cross 64KB boundaries */
running_total = TRB_MAX_BUFF_SIZE -
(sg_dma_address(sg) & (TRB_MAX_BUFF_SIZE - 1));
+ running_total &= TRB_MAX_BUFF_SIZE - 1;
if (running_total != 0)
num_trbs++;
@@ -1978,6 +1979,7 @@ int xhci_queue_bulk_tx(struct xhci_hcd *xhci, gfp_t mem_flags,
/* How much data is (potentially) left before the 64KB boundary? */
running_total = TRB_MAX_BUFF_SIZE -
(urb->transfer_dma & (TRB_MAX_BUFF_SIZE - 1));
+ running_total &= TRB_MAX_BUFF_SIZE - 1;
/* If there's some data on this 64KB chunk, or we have to send a
* zero-length transfer, we need at least one TRB
--
1.7.4.4