| Message ID | 20231113054917.96894-1-suhui@nfschina.com (mailing list archive) |
|---|---|
| State | Changes Requested |
| Delegated to: | Kalle Valo |
| Headers | show |
| Series | wifi: rtl8xxxu: correct the error value of 'timeout' | expand |
> -----Original Message----- > From: Su Hui <suhui@nfschina.com> > Sent: Monday, November 13, 2023 1:49 PM > To: Jes.Sorensen@gmail.com; kvalo@kernel.org > Cc: Su Hui <suhui@nfschina.com>; linux-wireless@vger.kernel.org; linux-kernel@vger.kernel.org; > kernel-janitors@vger.kernel.org > Subject: [PATCH] wifi: rtl8xxxu: correct the error value of 'timeout' > > When 'rtl8xxxu_dma_agg_pages <= page_thresh', 'timeout' should equal to > 'page_thresh' rather than '4'. Change the code order to fix this problem. > > Fixes: 614e389f36a9 ("rtl8xxxu: gen1: Set aggregation timeout (REG_RXDMA_AGG_PG_TH + 1) as well") I think this should fix Fixes: fd83f1227826 ("rtl8xxxu: gen1: Add module parameters to adjust DMA aggregation parameters") > Signed-off-by: Su Hui <suhui@nfschina.com> > --- > .../net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c | 12 ++++++------ > 1 file changed, 6 insertions(+), 6 deletions(-) > > diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c > b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c > index 43ee7592bc6e..9cab8b1dc486 100644 > --- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c > +++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c > @@ -4757,6 +4757,12 @@ void rtl8xxxu_gen1_init_aggregation(struct rtl8xxxu_priv *priv) > * RxAggPageTimeout = 4 or 6 (absolute time 34ms/(2^6)) > */ > > + /* REG_RXDMA_AGG_PG_TH + 1 seems to be the timeout register on > + * gen2 chips and rtl8188eu. The rtl8723au seems unhappy if we > + * don't set it, so better set both. > + */ > + timeout = 4; > + > page_thresh = (priv->fops->rx_agg_buf_size / 512); > if (rtl8xxxu_dma_agg_pages >= 0) { > if (rtl8xxxu_dma_agg_pages <= page_thresh) The logic here is: page_thresh = (priv->fops->rx_agg_buf_size / 512); if (rtl8xxxu_dma_agg_pages >= 0) { if (rtl8xxxu_dma_agg_pages <= page_thresh) timeout = page_thresh; Do you know why 'timeout = page_thresh;'? Intuitively, units of 'timeout' and 'thresh' are different. Maybe, we should correct here instead?
On 2023/11/14 14:42, Ping-Ke Shih wrote: >> -----Original Message----- >> From: Su Hui <suhui@nfschina.com> >> Sent: Monday, November 13, 2023 1:49 PM >> To: Jes.Sorensen@gmail.com; kvalo@kernel.org >> Cc: Su Hui <suhui@nfschina.com>; linux-wireless@vger.kernel.org; linux-kernel@vger.kernel.org; >> kernel-janitors@vger.kernel.org >> Subject: [PATCH] wifi: rtl8xxxu: correct the error value of 'timeout' >> >> When 'rtl8xxxu_dma_agg_pages <= page_thresh', 'timeout' should equal to >> 'page_thresh' rather than '4'. Change the code order to fix this problem. >> >> Fixes: 614e389f36a9 ("rtl8xxxu: gen1: Set aggregation timeout (REG_RXDMA_AGG_PG_TH + 1) as well") > I think this should fix > Fixes: fd83f1227826 ("rtl8xxxu: gen1: Add module parameters to adjust DMA aggregation parameters") Agreed. Thanks for your reminder! > >> Signed-off-by: Su Hui <suhui@nfschina.com> >> --- >> .../net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c | 12 ++++++------ >> 1 file changed, 6 insertions(+), 6 deletions(-) >> >> diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c >> b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c >> index 43ee7592bc6e..9cab8b1dc486 100644 >> --- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c >> +++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c >> @@ -4757,6 +4757,12 @@ void rtl8xxxu_gen1_init_aggregation(struct rtl8xxxu_priv *priv) >> * RxAggPageTimeout = 4 or 6 (absolute time 34ms/(2^6)) >> */ >> >> + /* REG_RXDMA_AGG_PG_TH + 1 seems to be the timeout register on >> + * gen2 chips and rtl8188eu. The rtl8723au seems unhappy if we >> + * don't set it, so better set both. >> + */ >> + timeout = 4; >> + >> page_thresh = (priv->fops->rx_agg_buf_size / 512); >> if (rtl8xxxu_dma_agg_pages >= 0) { >> if (rtl8xxxu_dma_agg_pages <= page_thresh) > The logic here is: > > page_thresh = (priv->fops->rx_agg_buf_size / 512); > if (rtl8xxxu_dma_agg_pages >= 0) { > if (rtl8xxxu_dma_agg_pages <= page_thresh) > timeout = page_thresh; > > Do you know why 'timeout = page_thresh;'? Intuitively, units of 'timeout' and > 'thresh' are different. Maybe, we should correct here instead? I don't know the reason about this. :( I found this error because clang static checker complains about this code that value stored to 'timeout' is never read. Su Hui
On Tue, Nov 14, 2023 at 06:42:50AM +0000, Ping-Ke Shih wrote: > > diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c > > b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c > > index 43ee7592bc6e..9cab8b1dc486 100644 > > --- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c > > +++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c > > @@ -4757,6 +4757,12 @@ void rtl8xxxu_gen1_init_aggregation(struct rtl8xxxu_priv *priv) > > * RxAggPageTimeout = 4 or 6 (absolute time 34ms/(2^6)) > > */ > > > > + /* REG_RXDMA_AGG_PG_TH + 1 seems to be the timeout register on > > + * gen2 chips and rtl8188eu. The rtl8723au seems unhappy if we > > + * don't set it, so better set both. > > + */ > > + timeout = 4; > > + > > page_thresh = (priv->fops->rx_agg_buf_size / 512); > > if (rtl8xxxu_dma_agg_pages >= 0) { > > if (rtl8xxxu_dma_agg_pages <= page_thresh) > > The logic here is: > > page_thresh = (priv->fops->rx_agg_buf_size / 512); > if (rtl8xxxu_dma_agg_pages >= 0) { > if (rtl8xxxu_dma_agg_pages <= page_thresh) > timeout = page_thresh; > > Do you know why 'timeout = page_thresh;'? Intuitively, units of 'timeout' and > 'thresh' are different. Maybe, we should correct here instead? > Yeah. That's strange. I'm not convinced this fix is correct. I'm hesitant to suggest this but maybe the following is the correct fix? It just silences the warning but doesn't change runtime. I don't know. *shrug*. One thing that we could do is just leave the warning as-is until someone who knows better than we do can take a look at it. regards, dan carpenter diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c index 43ee7592bc6e..68d9b4a0ee63 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c +++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c @@ -4759,16 +4759,16 @@ void rtl8xxxu_gen1_init_aggregation(struct rtl8xxxu_priv *priv) page_thresh = (priv->fops->rx_agg_buf_size / 512); if (rtl8xxxu_dma_agg_pages >= 0) { - if (rtl8xxxu_dma_agg_pages <= page_thresh) - timeout = page_thresh; - else if (rtl8xxxu_dma_agg_pages <= 6) - dev_err(&priv->udev->dev, - "%s: dma_agg_pages=%i too small, minimum is 6\n", - __func__, rtl8xxxu_dma_agg_pages); - else - dev_err(&priv->udev->dev, - "%s: dma_agg_pages=%i larger than limit %i\n", - __func__, rtl8xxxu_dma_agg_pages, page_thresh); + if (rtl8xxxu_dma_agg_pages > page_thresh) { + if (rtl8xxxu_dma_agg_pages <= 6) + dev_err(&priv->udev->dev, + "%s: dma_agg_pages=%i too small, minimum is 6\n", + __func__, rtl8xxxu_dma_agg_pages); + else + dev_err(&priv->udev->dev, + "%s: dma_agg_pages=%i larger than limit %i\n", + __func__, rtl8xxxu_dma_agg_pages, page_thresh); + } } rtl8xxxu_write8(priv, REG_RXDMA_AGG_PG_TH, page_thresh); /*
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c index 43ee7592bc6e..9cab8b1dc486 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c +++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c @@ -4757,6 +4757,12 @@ void rtl8xxxu_gen1_init_aggregation(struct rtl8xxxu_priv *priv) * RxAggPageTimeout = 4 or 6 (absolute time 34ms/(2^6)) */ + /* REG_RXDMA_AGG_PG_TH + 1 seems to be the timeout register on + * gen2 chips and rtl8188eu. The rtl8723au seems unhappy if we + * don't set it, so better set both. + */ + timeout = 4; + page_thresh = (priv->fops->rx_agg_buf_size / 512); if (rtl8xxxu_dma_agg_pages >= 0) { if (rtl8xxxu_dma_agg_pages <= page_thresh) @@ -4771,12 +4777,6 @@ void rtl8xxxu_gen1_init_aggregation(struct rtl8xxxu_priv *priv) __func__, rtl8xxxu_dma_agg_pages, page_thresh); } rtl8xxxu_write8(priv, REG_RXDMA_AGG_PG_TH, page_thresh); - /* - * REG_RXDMA_AGG_PG_TH + 1 seems to be the timeout register on - * gen2 chips and rtl8188eu. The rtl8723au seems unhappy if we - * don't set it, so better set both. - */ - timeout = 4; if (rtl8xxxu_dma_agg_timeout >= 0) { if (rtl8xxxu_dma_agg_timeout <= 127)
When 'rtl8xxxu_dma_agg_pages <= page_thresh', 'timeout' should equal to 'page_thresh' rather than '4'. Change the code order to fix this problem. Fixes: 614e389f36a9 ("rtl8xxxu: gen1: Set aggregation timeout (REG_RXDMA_AGG_PG_TH + 1) as well") Signed-off-by: Su Hui <suhui@nfschina.com> --- .../net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-)