From: Stefan Richter Date: Sat, 25 Aug 2007 12:08:19 +0000 (+0200) Subject: firewire: fw-ohci: enforce read order for selfID generation X-Git-Tag: v2.6.24-rc1~825^2~9 X-Git-Url: https://err.no/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=ee71c2f9ee85117e41ef87357ed8f75d29369b98;p=linux-2.6 firewire: fw-ohci: enforce read order for selfID generation It seems unlikely, but access to self_id_cpu[0] could at least in theory be deferred until after the loop over self_id_cpu[1..n] or even after the subsequent reg_read. Enforce the desired order by a read barrier. Also prevent the reg_read from being reordered relative to the for loop. This isn't necessary if the loop's conditional printk counts as an implicit barrier, but better make it explicit. (self_id_cpu[] is a coherent DMA buffer.) Signed-off-by: Stefan Richter --- diff --git a/drivers/firewire/fw-ohci.c b/drivers/firewire/fw-ohci.c index e14c1ca781..a7947baeee 100644 --- a/drivers/firewire/fw-ohci.c +++ b/drivers/firewire/fw-ohci.c @@ -30,6 +30,7 @@ #include #include +#include #include "fw-transaction.h" #include "fw-ohci.h" @@ -926,12 +927,14 @@ static void bus_reset_tasklet(unsigned long data) self_id_count = (reg_read(ohci, OHCI1394_SelfIDCount) >> 3) & 0x3ff; generation = (le32_to_cpu(ohci->self_id_cpu[0]) >> 16) & 0xff; + rmb(); for (i = 1, j = 0; j < self_id_count; i += 2, j++) { if (ohci->self_id_cpu[i] != ~ohci->self_id_cpu[i + 1]) fw_error("inconsistent self IDs\n"); ohci->self_id_buffer[j] = le32_to_cpu(ohci->self_id_cpu[i]); } + rmb(); /* * Check the consistency of the self IDs we just read. The