lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <456145f7-9c6e-4231-aec4-9c02511b6550@suswa.mountain>
Date:   Mon, 20 Nov 2023 07:31:10 -0500
From:   Dan Carpenter <dan.carpenter@...aro.org>
To:     Ping-Ke Shih <pkshih@...ltek.com>
Cc:     Su Hui <suhui@...china.com>,
        "Jes.Sorensen@...il.com" <Jes.Sorensen@...il.com>,
        "kvalo@...nel.org" <kvalo@...nel.org>,
        "linux-wireless@...r.kernel.org" <linux-wireless@...r.kernel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "kernel-janitors@...r.kernel.org" <kernel-janitors@...r.kernel.org>
Subject: Re: [PATCH] wifi: rtl8xxxu: correct the error value of 'timeout'

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);
 	/*

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ