openwrt/package/kernel/mac80211/patches/840-brcmfmac-avoid-writing-channel-out-of-allocated-arra.patch
Rafał Miłecki f714fe46a9 mac80211: pending brcmfmac patches cleaning channels management
They prepare brcmfmac for using wiphy_read_of_freq_limits.

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
2017-01-10 17:18:29 +01:00

101 lines
3.4 KiB
Diff

From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= <rafal@milecki.pl>
Date: Wed, 4 Jan 2017 12:09:41 +0100
Subject: [PATCH] brcmfmac: avoid writing channel out of allocated array
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Our code was assigning number of channels to the index variable by
default. If firmware reported channel we didn't predict this would
result in using that initial index value and writing out of array. This
never happened so far (we got a complete list of supported channels) but
it means possible memory corruption so we should handle it anyway.
This patch simply detects unexpected channel and ignores it.
As we don't try to create new entry now, it's also safe to drop hw_value
and center_freq assignment. For known channels we have these set anyway.
I decided to fix this issue by assigning NULL or a target channel to the
channel variable. This was one of possible ways, I prefefred this one as
it also avoids using channel[index] over and over.
Fixes: 58de92d2f95e ("brcmfmac: use static superset of channels for wiphy bands")
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
Acked-by: Arend van Spriel <arend.vanspriel@broadcom.com>
---
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
@@ -5913,7 +5913,6 @@ static int brcmf_construct_chaninfo(stru
u32 i, j;
u32 total;
u32 chaninfo;
- u32 index;
pbuf = kzalloc(BRCMF_DCMD_MEDLEN, GFP_KERNEL);
@@ -5961,33 +5960,36 @@ static int brcmf_construct_chaninfo(stru
ch.bw == BRCMU_CHAN_BW_80)
continue;
- channel = band->channels;
- index = band->n_channels;
+ channel = NULL;
for (j = 0; j < band->n_channels; j++) {
- if (channel[j].hw_value == ch.control_ch_num) {
- index = j;
+ if (band->channels[j].hw_value == ch.control_ch_num) {
+ channel = &band->channels[j];
break;
}
}
- channel[index].center_freq =
- ieee80211_channel_to_frequency(ch.control_ch_num,
- band->band);
- channel[index].hw_value = ch.control_ch_num;
+ if (!channel) {
+ /* It seems firmware supports some channel we never
+ * considered. Something new in IEEE standard?
+ */
+ brcmf_err("Ignoring unexpected firmware channel %d\n",
+ ch.control_ch_num);
+ continue;
+ }
/* assuming the chanspecs order is HT20,
* HT40 upper, HT40 lower, and VHT80.
*/
if (ch.bw == BRCMU_CHAN_BW_80) {
- channel[index].flags &= ~IEEE80211_CHAN_NO_80MHZ;
+ channel->flags &= ~IEEE80211_CHAN_NO_80MHZ;
} else if (ch.bw == BRCMU_CHAN_BW_40) {
- brcmf_update_bw40_channel_flag(&channel[index], &ch);
+ brcmf_update_bw40_channel_flag(channel, &ch);
} else {
/* enable the channel and disable other bandwidths
* for now as mentioned order assure they are enabled
* for subsequent chanspecs.
*/
- channel[index].flags = IEEE80211_CHAN_NO_HT40 |
- IEEE80211_CHAN_NO_80MHZ;
+ channel->flags = IEEE80211_CHAN_NO_HT40 |
+ IEEE80211_CHAN_NO_80MHZ;
ch.bw = BRCMU_CHAN_BW_20;
cfg->d11inf.encchspec(&ch);
chaninfo = ch.chspec;
@@ -5995,11 +5997,11 @@ static int brcmf_construct_chaninfo(stru
&chaninfo);
if (!err) {
if (chaninfo & WL_CHAN_RADAR)
- channel[index].flags |=
+ channel->flags |=
(IEEE80211_CHAN_RADAR |
IEEE80211_CHAN_NO_IR);
if (chaninfo & WL_CHAN_PASSIVE)
- channel[index].flags |=
+ channel->flags |=
IEEE80211_CHAN_NO_IR;
}
}