linux-wireless.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Arend van Spriel" <arend@broadcom.com>
To: gregkh@suse.de
Cc: "Roland Vossen" <rvossen@broadcom.com>,
	devel@linuxdriverproject.org, linux-wireless@vger.kernel.org,
	"Arend van Spriel" <arend@broadcom.com>
Subject: [PATCH 02/32] staging: brcm80211: new Kconfig item to enable IO reordering fix
Date: Fri, 13 May 2011 11:59:22 +0200	[thread overview]
Message-ID: <1305280792-9475-2-git-send-email-arend@broadcom.com> (raw)
In-Reply-To: <1305280792-9475-1-git-send-email-arend@broadcom.com>

From: Roland Vossen <rvossen@broadcom.com>

bcm4716 (which includes 4717 & 4718), plus 4706 on PCIe can reorder
transactions. As a fix, a read after write is performed on certain places
in the code. Older chips and the newer 5357 family don't require this fix.

Cc: devel@linuxdriverproject.org
Cc: linux-wireless@vger.kernel.org
Reviewed-by: Henry Ptasinski <henryp@broadcom.com>
Reviewed-by: Brett Rudley <brudley@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
---
 drivers/staging/brcm80211/Kconfig                  |    8 ++++++++
 .../staging/brcm80211/brcmsmac/phy/wlc_phy_cmn.c   |    4 ++--
 drivers/staging/brcm80211/include/bcmutils.h       |   11 +++++++++++
 3 files changed, 21 insertions(+), 2 deletions(-)

diff --git a/drivers/staging/brcm80211/Kconfig b/drivers/staging/brcm80211/Kconfig
index f4cf9b2..a11ffc7 100644
--- a/drivers/staging/brcm80211/Kconfig
+++ b/drivers/staging/brcm80211/Kconfig
@@ -36,3 +36,11 @@ config BRCMDBG
 	depends on BRCMSMAC || BRCMFMAC
 	---help---
 	  Selecting this enables additional code for debug purposes.
+
+config BRCM_FIX_IO_ORDER
+	bool "Broadcom I/O order fix"
+	default n
+	depends on BRCMSMAC
+	---help---
+	  Select this for for platforms running on bcm4706, 4716, 4717 or 4718.
+	  It resolves an I/O issue with the PCIe core used in these chips.
diff --git a/drivers/staging/brcm80211/brcmsmac/phy/wlc_phy_cmn.c b/drivers/staging/brcm80211/brcmsmac/phy/wlc_phy_cmn.c
index 6cba4df..b3c5f07 100644
--- a/drivers/staging/brcm80211/brcmsmac/phy/wlc_phy_cmn.c
+++ b/drivers/staging/brcm80211/brcmsmac/phy/wlc_phy_cmn.c
@@ -412,7 +412,7 @@ void write_phy_reg(phy_info_t *pi, u16 addr, u16 val)
 
 	regs = pi->regs;
 
-#ifdef __mips__
+#ifdef CONFIG_BRCM_FIX_IO_ORDER
 	W_REG(&regs->phyregaddr, addr);
 	(void)R_REG(&regs->phyregaddr);
 	W_REG(&regs->phyregdata, val);
@@ -427,7 +427,7 @@ void write_phy_reg(phy_info_t *pi, u16 addr, u16 val)
 			(void)R_REG(&regs->phyversion);
 		}
 	}
-#endif
+#endif /* CONFIG_BRCM_FIX_IO_ORDER */
 }
 
 void and_phy_reg(phy_info_t *pi, u16 addr, u16 val)
diff --git a/drivers/staging/brcm80211/include/bcmutils.h b/drivers/staging/brcm80211/include/bcmutils.h
index 224d4ce..a4a2a8d 100644
--- a/drivers/staging/brcm80211/include/bcmutils.h
+++ b/drivers/staging/brcm80211/include/bcmutils.h
@@ -367,6 +367,17 @@ extern void bcm_prpkt(const char *msg, struct sk_buff *p0);
 	} while (0)
 #endif				/* __BIG_ENDIAN */
 
+#ifdef CONFIG_BRCM_FIX_IO_ORDER
+/*
+ * bcm4716 (which includes 4717 & 4718), plus 4706 on PCIe can reorder
+ * transactions. As a fix, a read after write is performed on certain places
+ * in the code. Older chips and the newer 5357 family don't require this fix.
+ */
+#define W_REG_FLUSH(r, v)	{ W_REG((r), (v)); (void)R_REG(r); }
+#else
+#define W_REG_FLUSH(r, v)	W_REG((r), (v))
+#endif				/* CONFIG_BRCM_FIX_IO_ORDER */
+
 #define AND_REG(r, v)	W_REG((r), R_REG(r) & (v))
 #define OR_REG(r, v)	W_REG((r), R_REG(r) | (v))
 
-- 
1.7.4.1



  reply	other threads:[~2011-05-13 10:00 UTC|newest]

Thread overview: 40+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-05-13  9:59 [PATCH 01/32] staging: brcm80211: replaced IL_BIGENDIAN by __BIG_ENDIAN Arend van Spriel
2011-05-13  9:59 ` Arend van Spriel [this message]
2011-05-14 10:23   ` [PATCH 02/32] staging: brcm80211: new Kconfig item to enable IO reordering fix Christoph Hellwig
2011-05-14 15:46   ` Greg KH
2011-05-15 12:53     ` Arend van Spriel
2011-05-17 19:19       ` Greg KH
2011-05-13  9:59 ` [PATCH 03/32] staging: brcm80211: replaced #ifdef __mips__ sections by W_REG_FLUSH Arend van Spriel
2011-05-13  9:59 ` [PATCH 04/32] staging: brcm80211: got rid of #ifdef __mips__ sections in hnddma.c Arend van Spriel
2011-05-14 10:16   ` Jonas Gorski
2011-05-17 19:10     ` Greg KH
2011-05-13  9:59 ` [PATCH 05/32] staging: brcm80211: got rid of #ifdef __mips__ by new dma_spin_for_len() Arend van Spriel
2011-05-13  9:59 ` [PATCH 06/32] staging: brcm80211: remove unnecessary functions from wlc_main.c Arend van Spriel
2011-05-13  9:59 ` [PATCH 07/32] staging: brcm80211: remove unused functions and prototypes Arend van Spriel
2011-05-13  9:59 ` [PATCH 08/32] staging: brcm80211: make function definitions and prototype consistent Arend van Spriel
2011-05-13  9:59 ` [PATCH 09/32] staging: brcm80211: cleanup functions in nvram.c Arend van Spriel
2011-05-13  9:59 ` [PATCH 10/32] staging: brcm80211: fix makefiles for building drivers in kernel Arend van Spriel
2011-05-13  9:59 ` [PATCH 11/32] staging: brcm80211: Kconfig fix for 'Broadcom I/O order fix' Arend van Spriel
2011-05-13  9:59 ` [PATCH 12/32] staging: brcm80211: removed unused struct net_device_stats related code Arend van Spriel
2011-05-13  9:59 ` [PATCH 13/32] staging: brcm80211: removed counter related macros that expanded to NOP Arend van Spriel
2011-05-13  9:59 ` [PATCH 14/32] staging: brcm80211: removed stats WEXT support from fullmac Arend van Spriel
2011-05-13  9:59 ` [PATCH 15/32] staging: brcm80211: removed support for proprietary ioctl WLC_GET_PKTCNTS Arend van Spriel
2011-05-13  9:59 ` [PATCH 16/32] staging: brcm80211: removed counter related struct wl_cnt Arend van Spriel
2011-05-13  9:59 ` [PATCH 17/32] staging: brcm80211: removed more unused counter related structures Arend van Spriel
2011-05-13  9:59 ` [PATCH 18/32] staging: brcm80211: Fix for WPA GTK install issue in brcmfmac driver Arend van Spriel
2011-05-13 15:22   ` Grant Grundler
2011-05-13  9:59 ` [PATCH 19/32] staging: brcm80211: Fix for suspend " Arend van Spriel
2011-05-13 15:30   ` Grant Grundler
2011-05-13  9:59 ` [PATCH 20/32] staging: brcm80211: brcmfmac driver's interface name changed to "wlan" Arend van Spriel
2011-05-13  9:59 ` [PATCH 21/32] staging: brcm80211: Fixed wl_assoc_params structure miss match in brcmfmac driver Arend van Spriel
2011-05-13  9:59 ` [PATCH 22/32] staging: brcm80211: Optimized wl_inform_single_bss function " Arend van Spriel
2011-05-13  9:59 ` [PATCH 23/32] staging: brcm80211: Fixed bss (re)connect/disconnect issues " Arend van Spriel
2011-05-13  9:59 ` [PATCH 24/32] staging: brcm80211: Fixed ibss join/leave functionality " Arend van Spriel
2011-05-13  9:59 ` [PATCH 25/32] staging: brcm80211: Added support to change roaming parameters from " Arend van Spriel
2011-05-13  9:59 ` [PATCH 26/32] staging: brcm80211: Added support to change scan times " Arend van Spriel
2011-05-13  9:59 ` [PATCH 27/32] staging: brcm80211: Fixed scan notification functionality in " Arend van Spriel
2011-05-13  9:59 ` [PATCH 28/32] staging: brcm80211: Removed unused variable from " Arend van Spriel
2011-05-13  9:59 ` [PATCH 29/32] staging: brcm80211: cfg80211 up/down routines in brcmfmac driver are fixed Arend van Spriel
2011-05-13  9:59 ` [PATCH 30/32] staging: brcm80211: Better debug support added to brcmfmac driver Arend van Spriel
2011-05-13  9:59 ` [PATCH 31/32] staging: brcm80211: Corrected sdpcm_shared_t structure definition Arend van Spriel
2011-05-13  9:59 ` [PATCH 32/32] staging: brcm80211: fix checkpatch issue in wlc_channel.c Arend van Spriel

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1305280792-9475-2-git-send-email-arend@broadcom.com \
    --to=arend@broadcom.com \
    --cc=devel@linuxdriverproject.org \
    --cc=gregkh@suse.de \
    --cc=linux-wireless@vger.kernel.org \
    --cc=rvossen@broadcom.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).