mt76x0u: Probing issues on Raspberry Pi 3 B+

I found another bug in mt76usb SG code. We set sg->offset bigger than PAGE_SIZE that actually make sg point to memory in different page than sg->page. Correcting this with another patch that avoid using sg mapping with sg->length > PAGE_SIZE, fixed mt76x0u with AMD IOMMU:
https://bugzilla.kernel.org/show_bug.cgi?id=202241

I'm attaching 3 patches, they should also fix issue on rpi. It's also possible that only 2 patches are sufficient:

0001-mt76usb-do-not-set-urb-num_sgs-to-1-for-non-SG-usb-h.patch 0003-mt76usb-do-not-use-compound-head-page-for-SG-I-O.patch

to fix the issue, if doing dma_map_{page,sg} is fine with sg->offset == 0 and sg->length bigger than one page.

Please test, thanks.

Stanislaw
From 49654f00b5071602a9182f0cb488f346f01ec5ac Mon Sep 17 00:00:00 2001
From: Stanislaw Gruszka <[email protected]>
Date: Fri, 15 Feb 2019 07:39:00 +0100
Subject: [PATCH v5 1/3] mt76usb: do not set urb->num_sgs to 1 for non SG usb
host drivers

Track number of segments in mt76u_buf structure and do not submit urbs with urb->num_sgs = 1 if usb host driver sg_tablesize is zero.

This suppose fix problem of mt76 not working with some usb host controllers like dwc2.

---
 drivers/net/wireless/mediatek/mt76/mt76.h |  1 +
 drivers/net/wireless/mediatek/mt76/usb.c  | 57 +++++++++++++++++++------------
 2 files changed, 37 insertions(+), 21 deletions(-)

index 2bb9db4ed80a..97ad0270f8a6 100644
@@ -86,6 +86,7 @@ struct mt76_queue_buf {
 struct mt76u_buf {
 	struct mt76_dev *dev;
 	struct urb *urb;
+	int num_sgs;
 	size_t len;
 	bool done;
 };