From: Luis R. Rodriguez <lrodriguez@atheros.com>
To: ath9k-devel@lists.ath9k.org
Subject: [ath9k-devel] [RFC PATCH 2/2] ath9k: fix mixing of interrupt masks, remove ah->mask_reg
Date: Tue, 30 Mar 2010 09:52:19 -0700 [thread overview]
Message-ID: <20100330165219.GA25095@tux> (raw)
In-Reply-To: <20100330033924.7683.13491.stgit@mj.roinet.com>
On Mon, Mar 29, 2010 at 08:39:58PM -0700, Pavel Roskin wrote:
> ah->mask_reg is used inconsistently. ath9k_hw_init_chain_masks() uses
> it to cache the value it writes to AR_IMR. But once
> ath9k_hw_set_interrupts() is called, ah->mask_reg starts caching the
> interrupt mask in the format used in sc->imask, which differs in several
> bits.
>
> Use sc->imask instead of ah->mask_reg as omask ath9k_hw_set_interrupts()
> and ath9k_hw_updatetxtriglevel(). Remove ah->mask_reg, as it becomes
> write only. Use a local variable in ath9k_hw_init_chain_masks().
>
> Signed-off-by: Pavel Roskin <proski@gnu.org>
> Reported-by: Julia Lawall <julia@diku.dk>
> ---
>
> Adding ath9k.h to hw.c and mac.c could raise some eyebrows, but the
> alternative is moving imask from sc to ah, which would be more
> intrusive.
>
> I think sc and ah should be merged eventually, so all files that know
> the ah structure will know sc as well. There is no gain from hiding sc
> from the low-level code.
>
> drivers/net/wireless/ath/ath9k/hw.c | 19 +++++++++++--------
> drivers/net/wireless/ath/ath9k/hw.h | 1 -
> drivers/net/wireless/ath/ath9k/mac.c | 5 ++++-
> 3 files changed, 15 insertions(+), 10 deletions(-)
>
> diff --git a/drivers/net/wireless/ath/ath9k/hw.c b/drivers/net/wireless/ath/ath9k/hw.c
> index 77db932..791aeb9 100644
> --- a/drivers/net/wireless/ath/ath9k/hw.c
> +++ b/drivers/net/wireless/ath/ath9k/hw.c
> @@ -17,6 +17,7 @@
> #include <linux/io.h>
> #include <asm/unaligned.h>
>
> +#include "ath9k.h"
> #include "hw.h"
> #include "rc.h"
> #include "initvals.h"
> @@ -1120,23 +1121,25 @@ static void ath9k_hw_init_chain_masks(struct ath_hw *ah)
> static void ath9k_hw_init_interrupt_masks(struct ath_hw *ah,
> enum nl80211_iftype opmode)
> {
> - ah->mask_reg = AR_IMR_TXERR |
> + u32 imr_reg;
> +
> + imr_reg = AR_IMR_TXERR |
> AR_IMR_TXURN |
> AR_IMR_RXERR |
> AR_IMR_RXORN |
> AR_IMR_BCNMISC;
>
> if (ah->config.rx_intr_mitigation)
> - ah->mask_reg |= AR_IMR_RXINTM | AR_IMR_RXMINTR;
> + imr_reg |= AR_IMR_RXINTM | AR_IMR_RXMINTR;
> else
> - ah->mask_reg |= AR_IMR_RXOK;
> + imr_reg |= AR_IMR_RXOK;
>
> - ah->mask_reg |= AR_IMR_TXOK;
> + imr_reg |= AR_IMR_TXOK;
>
> if (opmode == NL80211_IFTYPE_AP)
> - ah->mask_reg |= AR_IMR_MIB;
> + imr_reg |= AR_IMR_MIB;
>
> - REG_WRITE(ah, AR_IMR, ah->mask_reg);
> + REG_WRITE(ah, AR_IMR, imr_reg);
> ah->imrs2_reg |= AR_IMR_S2_GTT;
> REG_WRITE(ah, AR_IMR_S2, ah->imrs2_reg);
>
> @@ -2839,10 +2842,11 @@ EXPORT_SYMBOL(ath9k_hw_getisr);
>
> enum ath9k_int ath9k_hw_set_interrupts(struct ath_hw *ah, enum ath9k_int ints)
> {
> - u32 omask = ah->mask_reg;
> u32 mask, mask2;
> struct ath9k_hw_capabilities *pCap = &ah->caps;
> struct ath_common *common = ath9k_hw_common(ah);
> + struct ath_softc *sc = (struct ath_softc *)common->priv;
This is exactly what I was trying to prevent, lets please try to figure out
another way, NACK for now.
Consider
struct ath9k_htc_priv *priv = (struct ath9k_htc_priv *) common->priv;
Luis
next prev parent reply other threads:[~2010-03-30 16:52 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-03-30 3:39 [ath9k-devel] [RFC PATCH 1/2] ath9k: rename symbols in enum ath9k_internal_frame_type to avoid confusion Pavel Roskin
2010-03-30 3:39 ` [ath9k-devel] [RFC PATCH 2/2] ath9k: fix mixing of interrupt masks, remove ah->mask_reg Pavel Roskin
2010-03-30 16:52 ` Luis R. Rodriguez [this message]
2010-03-31 21:54 ` Pavel Roskin
2010-03-30 16:57 ` [ath9k-devel] [RFC PATCH 1/2] ath9k: rename symbols in enum ath9k_internal_frame_type to avoid confusion Luis R. Rodriguez
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=20100330165219.GA25095@tux \
--to=lrodriguez@atheros.com \
--cc=ath9k-devel@lists.ath9k.org \
/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