From: Stanislaw Gruszka <sgruszka@redhat.com>
To: "John W. Linville" <linville@tuxdriver.com>
Cc: linux-wireless@vger.kernel.org, Stanislaw Gruszka <sgruszka@redhat.com>
Subject: [PATCH 03/24] iwlegacy: cleanup/fix memory barriers
Date: Mon, 13 Feb 2012 11:23:10 +0100 [thread overview]
Message-ID: <1329128611-6815-4-git-send-email-sgruszka@redhat.com> (raw)
In-Reply-To: <1329128611-6815-1-git-send-email-sgruszka@redhat.com>
wmb(), rmb() are not needed when writel(), readl() are used as
accessors for MMIO. We use them indirectly via iowrite32(),
ioread32().
What is needed mmiowb(), for synchronizing writes coming from
different CPUs on PCIe bridge (see in patch comments). This
fortunately is not needed on x86, where mmiowb() is just
defined as compiler barrier. As iwlegacy devices are most likely
not used on anything other than x86, this is not so important
fix.
Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com>
---
drivers/net/wireless/iwlegacy/common.c | 2 --
drivers/net/wireless/iwlegacy/common.h | 9 +++++++--
2 files changed, 7 insertions(+), 4 deletions(-)
diff --git a/drivers/net/wireless/iwlegacy/common.c b/drivers/net/wireless/iwlegacy/common.c
index 6dbd11f..50308da 100644
--- a/drivers/net/wireless/iwlegacy/common.c
+++ b/drivers/net/wireless/iwlegacy/common.c
@@ -179,7 +179,6 @@ il_read_targ_mem(struct il_priv *il, u32 addr)
_il_grab_nic_access(il);
_il_wr(il, HBUS_TARG_MEM_RADDR, addr);
- rmb();
value = _il_rd(il, HBUS_TARG_MEM_RDAT);
_il_release_nic_access(il);
@@ -196,7 +195,6 @@ il_write_targ_mem(struct il_priv *il, u32 addr, u32 val)
spin_lock_irqsave(&il->reg_lock, reg_flags);
if (likely(_il_grab_nic_access(il))) {
_il_wr(il, HBUS_TARG_MEM_WADDR, addr);
- wmb();
_il_wr(il, HBUS_TARG_MEM_WDAT, val);
_il_release_nic_access(il);
}
diff --git a/drivers/net/wireless/iwlegacy/common.h b/drivers/net/wireless/iwlegacy/common.h
index 1dfaa58..df9b5b4 100644
--- a/drivers/net/wireless/iwlegacy/common.h
+++ b/drivers/net/wireless/iwlegacy/common.h
@@ -2146,6 +2146,13 @@ static inline void
_il_release_nic_access(struct il_priv *il)
{
_il_clear_bit(il, CSR_GP_CNTRL, CSR_GP_CNTRL_REG_FLAG_MAC_ACCESS_REQ);
+ /*
+ * In above we are reading CSR_GP_CNTRL register, what will flush any
+ * previous writes, but still want write, which clear MAC_ACCESS_REQ
+ * bit, be performed on PCI bus before any other writes scheduled on
+ * different CPUs (after we drop reg_lock).
+ */
+ mmiowb();
}
static inline u32
@@ -2179,7 +2186,6 @@ static inline u32
_il_rd_prph(struct il_priv *il, u32 reg)
{
_il_wr(il, HBUS_TARG_PRPH_RADDR, reg | (3 << 24));
- rmb();
return _il_rd(il, HBUS_TARG_PRPH_RDAT);
}
@@ -2187,7 +2193,6 @@ static inline void
_il_wr_prph(struct il_priv *il, u32 addr, u32 val)
{
_il_wr(il, HBUS_TARG_PRPH_WADDR, ((addr & 0x0000FFFF) | (3 << 24)));
- wmb();
_il_wr(il, HBUS_TARG_PRPH_WDAT, val);
}
--
1.7.1
next prev parent reply other threads:[~2012-02-13 10:24 UTC|newest]
Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-02-13 10:23 [PATCH 00/24] iwlegacy update 2012-02-13 Stanislaw Gruszka
2012-02-13 10:23 ` [PATCH 01/24] iwlegacy: dump stack when fail to gain access to the device Stanislaw Gruszka
2012-02-13 10:23 ` [PATCH 02/24] iwlegacy: always check if got h/w access before write Stanislaw Gruszka
2012-02-13 10:23 ` Stanislaw Gruszka [this message]
2012-02-13 10:23 ` [PATCH 04/24] iwlegacy: use writeb,writel,readl directly Stanislaw Gruszka
2012-02-13 10:23 ` [PATCH 05/24] iwlegacy: regulatory_bands is not an ops Stanislaw Gruszka
2012-02-13 10:23 ` [PATCH 06/24] iwlegacy: gather all 4965 handlers in one place Stanislaw Gruszka
2012-02-13 10:23 ` [PATCH 07/24] iwlegacy: move debugfs_ops to il_priv Stanislaw Gruszka
2012-02-13 10:23 ` [PATCH 08/24] iwlegacy: remove temp_ops Stanislaw Gruszka
2012-02-13 10:23 ` [PATCH 09/24] iwlegacy: merge eeprom_ops into lib_ops Stanislaw Gruszka
2012-02-13 10:23 ` [PATCH 10/24] iwlegacy: remove il_apm_ops Stanislaw Gruszka
2012-02-13 10:23 ` [PATCH 11/24] iwlegacy: merge il_lib_ops into il_ops Stanislaw Gruszka
2012-02-13 10:23 ` [PATCH 12/24] iwlegacy: merge all ops structures into one Stanislaw Gruszka
2012-02-13 10:23 ` [PATCH 13/24] iwlegacy: get rid of tx/rx traffic log Stanislaw Gruszka
2012-02-13 10:23 ` [PATCH 14/24] iwlegacy: improve mac operation debuggability a bit Stanislaw Gruszka
2012-02-13 10:23 ` [PATCH 15/24] iwleagcy: remove old comments Stanislaw Gruszka
2012-02-13 10:23 ` [PATCH 16/24] iwleagcy: fix ident code damage Stanislaw Gruszka
2012-02-13 10:23 ` [PATCH 17/24] iwlegacy: do not grab nic access if rfkill Stanislaw Gruszka
2012-02-13 10:23 ` [PATCH 18/24] iwlegacy: check correct il_poll_bit error value Stanislaw Gruszka
2012-02-13 10:23 ` [PATCH 19/24] iwlegacy: small il4965_set_hw_ready cleanup Stanislaw Gruszka
2012-02-13 10:23 ` [PATCH 20/24] iwlegacy: only enable rfkill interrupt during UP Stanislaw Gruszka
2012-02-13 16:19 ` Dan Williams
2012-02-13 16:43 ` Stanislaw Gruszka
2012-02-14 7:50 ` [PATCH 20/24 v2] iwlegacy: enable only rfkill interrupt when rfkill switch is on during IFF_UP Stanislaw Gruszka
2012-02-13 10:23 ` [PATCH 21/24] iwlegacy: small queue initializations cleanup Stanislaw Gruszka
2012-02-13 10:23 ` [PATCH 22/24] iwlegacy: s/S_RF_KILL_HW/S_RFKILL/g Stanislaw Gruszka
2012-02-13 10:23 ` [PATCH 23/24] iwlegacy: s/il_txq_mem/il_free_txq_mem/g Stanislaw Gruszka
2012-02-13 10:23 ` [PATCH 24/24] iwlegacy: remove il_is_rfkill_hw Stanislaw Gruszka
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=1329128611-6815-4-git-send-email-sgruszka@redhat.com \
--to=sgruszka@redhat.com \
--cc=linux-wireless@vger.kernel.org \
--cc=linville@tuxdriver.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).